system
stringclasses
7 values
input
stringlengths
10
94.5k
output
stringlengths
14
55.5k
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
/* Jaitirth Jacob - 13CO125 Vidit Bhargava - 13CO151 */ #include <stdio.h> #include <stdlib.h> #include <time.h> #define ITERATIONS 4 //Repeat the experiment for greater accuracy __global__ void add(int *a, int *b, int *c, int tpb) { //Find the correct thread index in the grid int i = blockIdx.x * tpb + threadIdx.x; c[i] = a[i] + b[i]; } #define N 1000000 //Array Size #define min_threads 16 #define max_threads 1024 int main(void) { int *a,*b,*c; int *d_a, *d_b, *d_c; int size = N * sizeof(int); a = (int *)malloc(size); b = (int *)malloc(size); c = (int *)malloc(size); //Allocate on device cudaMalloc((void **)&d_a, size); cudaMalloc((void **)&d_b, size); cudaMalloc((void **)&d_c, size); srand(time(NULL)); //Populate a and b for (int i = 0; i < N; ++i) { a[i] = rand()%20; b[i] = rand()%37; } int numBlocks; cudaEvent_t start, copy, exec, result; //Events for measuring time //To calculate average over a number of iterations float t1[7], t2[7], t3[7], total[7]; for (int i = 0; i < 7; ++i) { t1[i]=0; t2[i]=0; t3[i]=0; total[i]=0; } printf("t1: time for copying arrays\n"); printf("t2: time for kernel execution\n"); printf("t3: time for copying result back\n\n"); printf("All times in milliseconds\n"); printf("TPB\t\tNB\t\tt1\t\tt2\t\tt3\t\ttotal\t\n"); int count; for (int i = 0; i < ITERATIONS; ++i) { count=0; for (int threadsPerBlock = min_threads; threadsPerBlock <= max_threads; threadsPerBlock*=2) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; cudaEventCreate(&start); cudaEventCreate(&copy); cudaEventCreate(&exec); cudaEventCreate(&result); cudaEventRecord(start); //Copy inputs to device cudaMemcpy(d_a, a, size, cudaMemcpyHostToDevice); cudaMemcpy(d_b, b, size, cudaMemcpyHostToDevice); cudaEventRecord(copy); cudaEventSynchronize(copy); //Launch add() kernel on GPU add<<<numBlocks,threadsPerBlock>>>(d_a, d_b, d_c, threadsPerBlock); cudaEventRecord(exec); cudaEventSynchronize(exec); //Copy result back to host cudaMemcpy(c, d_c, size, cudaMemcpyDeviceToHost); cudaEventRecord(result); cudaEventSynchronize(result); float temp1=0, temp2=0, temp3=0, temptotal; cudaEventElapsedTime(&temp1, start, copy); cudaEventElapsedTime(&temp2, copy, exec); cudaEventElapsedTime(&temp3, exec, result); cudaEventElapsedTime(&temptotal, start, result); t1[count] += temp1; t2[count] += temp2; t3[count] += temp3; total[count] += temptotal; cudaEventDestroy(start); cudaEventDestroy(copy); cudaEventDestroy(exec); cudaEventDestroy(result); count++; } } int threadsPerBlock = min_threads; for (int i = 0; i < 7; ++i) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; t1[i]/=(float)ITERATIONS; t2[i]/=(float)ITERATIONS; t3[i]/=(float)ITERATIONS; total[i]/=(float)ITERATIONS; printf("%d\t\t%d\t\t%.5f\t\t%.5f\t\t%.5f\t\t%.5f\t\t\n", threadsPerBlock, numBlocks, t1[i], t2[i], t3[i], total[i]); threadsPerBlock*=2; } //Cleanup free(a); free(b); free(c); cudaFree(d_a); cudaFree(d_b); cudaFree(d_c); return 0; }
/* Jaitirth Jacob - 13CO125 Vidit Bhargava - 13CO151 */ #include <hip/hip_runtime.h> #include <stdio.h> #include <stdlib.h> #include <time.h> #define ITERATIONS 4 //Repeat the experiment for greater accuracy __global__ void add(int *a, int *b, int *c, int tpb) { //Find the correct thread index in the grid int i = blockIdx.x * tpb + threadIdx.x; c[i] = a[i] + b[i]; } #define N 1000000 //Array Size #define min_threads 16 #define max_threads 1024 int main(void) { int *a,*b,*c; int *d_a, *d_b, *d_c; int size = N * sizeof(int); a = (int *)malloc(size); b = (int *)malloc(size); c = (int *)malloc(size); //Allocate on device hipMalloc((void **)&d_a, size); hipMalloc((void **)&d_b, size); hipMalloc((void **)&d_c, size); srand(time(NULL)); //Populate a and b for (int i = 0; i < N; ++i) { a[i] = rand()%20; b[i] = rand()%37; } int numBlocks; hipEvent_t start, copy, exec, result; //Events for measuring time //To calculate average over a number of iterations float t1[7], t2[7], t3[7], total[7]; for (int i = 0; i < 7; ++i) { t1[i]=0; t2[i]=0; t3[i]=0; total[i]=0; } printf("t1: time for copying arrays\n"); printf("t2: time for kernel execution\n"); printf("t3: time for copying result back\n\n"); printf("All times in milliseconds\n"); printf("TPB\t\tNB\t\tt1\t\tt2\t\tt3\t\ttotal\t\n"); int count; for (int i = 0; i < ITERATIONS; ++i) { count=0; for (int threadsPerBlock = min_threads; threadsPerBlock <= max_threads; threadsPerBlock*=2) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; hipEventCreate(&start); hipEventCreate(&copy); hipEventCreate(&exec); hipEventCreate(&result); hipEventRecord(start); //Copy inputs to device hipMemcpy(d_a, a, size, hipMemcpyHostToDevice); hipMemcpy(d_b, b, size, hipMemcpyHostToDevice); hipEventRecord(copy); hipEventSynchronize(copy); //Launch add() kernel on GPU add<<<numBlocks,threadsPerBlock>>>(d_a, d_b, d_c, threadsPerBlock); hipEventRecord(exec); hipEventSynchronize(exec); //Copy result back to host hipMemcpy(c, d_c, size, hipMemcpyDeviceToHost); hipEventRecord(result); hipEventSynchronize(result); float temp1=0, temp2=0, temp3=0, temptotal; hipEventElapsedTime(&temp1, start, copy); hipEventElapsedTime(&temp2, copy, exec); hipEventElapsedTime(&temp3, exec, result); hipEventElapsedTime(&temptotal, start, result); t1[count] += temp1; t2[count] += temp2; t3[count] += temp3; total[count] += temptotal; hipEventDestroy(start); hipEventDestroy(copy); hipEventDestroy(exec); hipEventDestroy(result); count++; } } int threadsPerBlock = min_threads; for (int i = 0; i < 7; ++i) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; t1[i]/=(float)ITERATIONS; t2[i]/=(float)ITERATIONS; t3[i]/=(float)ITERATIONS; total[i]/=(float)ITERATIONS; printf("%d\t\t%d\t\t%.5f\t\t%.5f\t\t%.5f\t\t%.5f\t\t\n", threadsPerBlock, numBlocks, t1[i], t2[i], t3[i], total[i]); threadsPerBlock*=2; } //Cleanup free(a); free(b); free(c); hipFree(d_a); hipFree(d_b); hipFree(d_c); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
/* Jaitirth Jacob - 13CO125 Vidit Bhargava - 13CO151 */ #include <hip/hip_runtime.h> #include <stdio.h> #include <stdlib.h> #include <time.h> #define ITERATIONS 4 //Repeat the experiment for greater accuracy __global__ void add(int *a, int *b, int *c, int tpb) { //Find the correct thread index in the grid int i = blockIdx.x * tpb + threadIdx.x; c[i] = a[i] + b[i]; } #define N 1000000 //Array Size #define min_threads 16 #define max_threads 1024 int main(void) { int *a,*b,*c; int *d_a, *d_b, *d_c; int size = N * sizeof(int); a = (int *)malloc(size); b = (int *)malloc(size); c = (int *)malloc(size); //Allocate on device hipMalloc((void **)&d_a, size); hipMalloc((void **)&d_b, size); hipMalloc((void **)&d_c, size); srand(time(NULL)); //Populate a and b for (int i = 0; i < N; ++i) { a[i] = rand()%20; b[i] = rand()%37; } int numBlocks; hipEvent_t start, copy, exec, result; //Events for measuring time //To calculate average over a number of iterations float t1[7], t2[7], t3[7], total[7]; for (int i = 0; i < 7; ++i) { t1[i]=0; t2[i]=0; t3[i]=0; total[i]=0; } printf("t1: time for copying arrays\n"); printf("t2: time for kernel execution\n"); printf("t3: time for copying result back\n\n"); printf("All times in milliseconds\n"); printf("TPB\t\tNB\t\tt1\t\tt2\t\tt3\t\ttotal\t\n"); int count; for (int i = 0; i < ITERATIONS; ++i) { count=0; for (int threadsPerBlock = min_threads; threadsPerBlock <= max_threads; threadsPerBlock*=2) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; hipEventCreate(&start); hipEventCreate(&copy); hipEventCreate(&exec); hipEventCreate(&result); hipEventRecord(start); //Copy inputs to device hipMemcpy(d_a, a, size, hipMemcpyHostToDevice); hipMemcpy(d_b, b, size, hipMemcpyHostToDevice); hipEventRecord(copy); hipEventSynchronize(copy); //Launch add() kernel on GPU add<<<numBlocks,threadsPerBlock>>>(d_a, d_b, d_c, threadsPerBlock); hipEventRecord(exec); hipEventSynchronize(exec); //Copy result back to host hipMemcpy(c, d_c, size, hipMemcpyDeviceToHost); hipEventRecord(result); hipEventSynchronize(result); float temp1=0, temp2=0, temp3=0, temptotal; hipEventElapsedTime(&temp1, start, copy); hipEventElapsedTime(&temp2, copy, exec); hipEventElapsedTime(&temp3, exec, result); hipEventElapsedTime(&temptotal, start, result); t1[count] += temp1; t2[count] += temp2; t3[count] += temp3; total[count] += temptotal; hipEventDestroy(start); hipEventDestroy(copy); hipEventDestroy(exec); hipEventDestroy(result); count++; } } int threadsPerBlock = min_threads; for (int i = 0; i < 7; ++i) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; t1[i]/=(float)ITERATIONS; t2[i]/=(float)ITERATIONS; t3[i]/=(float)ITERATIONS; total[i]/=(float)ITERATIONS; printf("%d\t\t%d\t\t%.5f\t\t%.5f\t\t%.5f\t\t%.5f\t\t\n", threadsPerBlock, numBlocks, t1[i], t2[i], t3[i], total[i]); threadsPerBlock*=2; } //Cleanup free(a); free(b); free(c); hipFree(d_a); hipFree(d_b); hipFree(d_c); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z3addPiS_S_i .globl _Z3addPiS_S_i .p2align 8 .type _Z3addPiS_S_i,@function _Z3addPiS_S_i: s_clause 0x2 s_load_b32 s2, s[0:1], 0x18 s_load_b128 s[4:7], s[0:1], 0x0 s_load_b64 s[0:1], s[0:1], 0x10 s_waitcnt lgkmcnt(0) v_mad_u64_u32 v[1:2], null, s15, s2, v[0:1] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v2, 31, v1 v_lshlrev_b64 v[0:1], 2, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v2, vcc_lo, s4, v0 v_add_co_ci_u32_e32 v3, vcc_lo, s5, v1, vcc_lo v_add_co_u32 v4, vcc_lo, s6, v0 v_add_co_ci_u32_e32 v5, vcc_lo, s7, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s0, v0 global_load_b32 v2, v[2:3], off global_load_b32 v3, v[4:5], off v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo s_waitcnt vmcnt(0) v_add_nc_u32_e32 v2, v3, v2 global_store_b32 v[0:1], v2, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z3addPiS_S_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 28 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 6 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z3addPiS_S_i, .Lfunc_end0-_Z3addPiS_S_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 28 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z3addPiS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z3addPiS_S_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 6 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
/* Jaitirth Jacob - 13CO125 Vidit Bhargava - 13CO151 */ #include <hip/hip_runtime.h> #include <stdio.h> #include <stdlib.h> #include <time.h> #define ITERATIONS 4 //Repeat the experiment for greater accuracy __global__ void add(int *a, int *b, int *c, int tpb) { //Find the correct thread index in the grid int i = blockIdx.x * tpb + threadIdx.x; c[i] = a[i] + b[i]; } #define N 1000000 //Array Size #define min_threads 16 #define max_threads 1024 int main(void) { int *a,*b,*c; int *d_a, *d_b, *d_c; int size = N * sizeof(int); a = (int *)malloc(size); b = (int *)malloc(size); c = (int *)malloc(size); //Allocate on device hipMalloc((void **)&d_a, size); hipMalloc((void **)&d_b, size); hipMalloc((void **)&d_c, size); srand(time(NULL)); //Populate a and b for (int i = 0; i < N; ++i) { a[i] = rand()%20; b[i] = rand()%37; } int numBlocks; hipEvent_t start, copy, exec, result; //Events for measuring time //To calculate average over a number of iterations float t1[7], t2[7], t3[7], total[7]; for (int i = 0; i < 7; ++i) { t1[i]=0; t2[i]=0; t3[i]=0; total[i]=0; } printf("t1: time for copying arrays\n"); printf("t2: time for kernel execution\n"); printf("t3: time for copying result back\n\n"); printf("All times in milliseconds\n"); printf("TPB\t\tNB\t\tt1\t\tt2\t\tt3\t\ttotal\t\n"); int count; for (int i = 0; i < ITERATIONS; ++i) { count=0; for (int threadsPerBlock = min_threads; threadsPerBlock <= max_threads; threadsPerBlock*=2) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; hipEventCreate(&start); hipEventCreate(&copy); hipEventCreate(&exec); hipEventCreate(&result); hipEventRecord(start); //Copy inputs to device hipMemcpy(d_a, a, size, hipMemcpyHostToDevice); hipMemcpy(d_b, b, size, hipMemcpyHostToDevice); hipEventRecord(copy); hipEventSynchronize(copy); //Launch add() kernel on GPU add<<<numBlocks,threadsPerBlock>>>(d_a, d_b, d_c, threadsPerBlock); hipEventRecord(exec); hipEventSynchronize(exec); //Copy result back to host hipMemcpy(c, d_c, size, hipMemcpyDeviceToHost); hipEventRecord(result); hipEventSynchronize(result); float temp1=0, temp2=0, temp3=0, temptotal; hipEventElapsedTime(&temp1, start, copy); hipEventElapsedTime(&temp2, copy, exec); hipEventElapsedTime(&temp3, exec, result); hipEventElapsedTime(&temptotal, start, result); t1[count] += temp1; t2[count] += temp2; t3[count] += temp3; total[count] += temptotal; hipEventDestroy(start); hipEventDestroy(copy); hipEventDestroy(exec); hipEventDestroy(result); count++; } } int threadsPerBlock = min_threads; for (int i = 0; i < 7; ++i) { numBlocks = (N + threadsPerBlock - 1)/threadsPerBlock; t1[i]/=(float)ITERATIONS; t2[i]/=(float)ITERATIONS; t3[i]/=(float)ITERATIONS; total[i]/=(float)ITERATIONS; printf("%d\t\t%d\t\t%.5f\t\t%.5f\t\t%.5f\t\t%.5f\t\t\n", threadsPerBlock, numBlocks, t1[i], t2[i], t3[i], total[i]); threadsPerBlock*=2; } //Cleanup free(a); free(b); free(c); hipFree(d_a); hipFree(d_b); hipFree(d_c); return 0; }
.text .file "q3.hip" .globl _Z18__device_stub__addPiS_S_i # -- Begin function _Z18__device_stub__addPiS_S_i .p2align 4, 0x90 .type _Z18__device_stub__addPiS_S_i,@function _Z18__device_stub__addPiS_S_i: # @_Z18__device_stub__addPiS_S_i .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) movl %ecx, 4(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 4(%rsp), %rax movq %rax, 104(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z3addPiS_S_i, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z18__device_stub__addPiS_S_i, .Lfunc_end0-_Z18__device_stub__addPiS_S_i .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function main .LCPI1_0: .long 0x3e800000 # float 0.25 .text .globl main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 subq $328, %rsp # imm = 0x148 .cfi_def_cfa_offset 384 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movl $4000000, %edi # imm = 0x3D0900 callq malloc movq %rax, %r12 movl $4000000, %edi # imm = 0x3D0900 callq malloc movq %rax, %r15 movl $4000000, %edi # imm = 0x3D0900 callq malloc movq %rax, 104(%rsp) # 8-byte Spill leaq 48(%rsp), %rdi movl $4000000, %esi # imm = 0x3D0900 callq hipMalloc leaq 40(%rsp), %rdi movl $4000000, %esi # imm = 0x3D0900 callq hipMalloc leaq 32(%rsp), %rdi movl $4000000, %esi # imm = 0x3D0900 callq hipMalloc xorl %ebx, %ebx xorl %edi, %edi callq time movl %eax, %edi callq srand .p2align 4, 0x90 .LBB1_1: # =>This Inner Loop Header: Depth=1 callq rand cltq imulq $1717986919, %rax, %rcx # imm = 0x66666667 movq %rcx, %rdx shrq $63, %rdx shrq $32, %rcx sarl $3, %ecx addl %edx, %ecx shll $2, %ecx leal (%rcx,%rcx,4), %ecx subl %ecx, %eax movl %eax, (%r12,%rbx,4) callq rand cltq imulq $-580400985, %rax, %rcx # imm = 0xDD67C8A7 shrq $32, %rcx addl %eax, %ecx movl %ecx, %edx shrl $31, %edx sarl $5, %ecx addl %edx, %ecx leal (%rcx,%rcx,8), %edx leal (%rcx,%rdx,4), %ecx subl %ecx, %eax movl %eax, (%r15,%rbx,4) incq %rbx cmpq $1000000, %rbx # imm = 0xF4240 jne .LBB1_1 # %bb.2: movabsq $4294967296, %r14 # imm = 0x100000000 xorps %xmm0, %xmm0 movaps %xmm0, 288(%rsp) movups %xmm0, 300(%rsp) movaps %xmm0, 256(%rsp) movups %xmm0, 268(%rsp) movaps %xmm0, 224(%rsp) movups %xmm0, 236(%rsp) movaps %xmm0, 192(%rsp) movups %xmm0, 204(%rsp) movl $.Lstr, %edi callq puts@PLT movl $.Lstr.1, %edi callq puts@PLT movl $.Lstr.2, %edi callq puts@PLT movl $.Lstr.3, %edi callq puts@PLT movl $.Lstr.4, %edi callq puts@PLT xorl %eax, %eax jmp .LBB1_3 .p2align 4, 0x90 .LBB1_7: # in Loop: Header=BB1_3 Depth=1 movq 152(%rsp), %rax # 8-byte Reload incl %eax cmpl $4, %eax je .LBB1_8 .LBB1_3: # %.preheader71 # =>This Loop Header: Depth=1 # Child Loop BB1_4 Depth 2 movq %rax, 152(%rsp) # 8-byte Spill movl $16, %r13d xorl %ebp, %ebp jmp .LBB1_4 .p2align 4, 0x90 .LBB1_6: # in Loop: Header=BB1_4 Depth=2 movq 8(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 8(%rsp), %rdi callq hipEventSynchronize movq 32(%rsp), %rsi movl $4000000, %edx # imm = 0x3D0900 movq 104(%rsp), %rdi # 8-byte Reload movl $2, %ecx callq hipMemcpy movq (%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq (%rsp), %rdi callq hipEventSynchronize movl $0, 112(%rsp) movl $0, 72(%rsp) movl $0, 56(%rsp) movq 24(%rsp), %rsi movq 16(%rsp), %rdx leaq 112(%rsp), %rdi callq hipEventElapsedTime movq 16(%rsp), %rsi movq 8(%rsp), %rdx leaq 72(%rsp), %rdi callq hipEventElapsedTime movq 8(%rsp), %rsi movq (%rsp), %rdx leaq 56(%rsp), %rdi callq hipEventElapsedTime movq 24(%rsp), %rsi movq (%rsp), %rdx leaq 88(%rsp), %rdi callq hipEventElapsedTime movss 112(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 288(%rsp,%rbp,4), %xmm0 movss %xmm0, 288(%rsp,%rbp,4) movss 72(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 256(%rsp,%rbp,4), %xmm0 movss %xmm0, 256(%rsp,%rbp,4) movss 56(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 224(%rsp,%rbp,4), %xmm0 movss %xmm0, 224(%rsp,%rbp,4) movss 88(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 192(%rsp,%rbp,4), %xmm0 movss %xmm0, 192(%rsp,%rbp,4) movq 24(%rsp), %rdi callq hipEventDestroy movq 16(%rsp), %rdi callq hipEventDestroy movq 8(%rsp), %rdi callq hipEventDestroy movq (%rsp), %rdi callq hipEventDestroy incq %rbp addl %r13d, %r13d cmpq $7, %rbp je .LBB1_7 .LBB1_4: # Parent Loop BB1_3 Depth=1 # => This Inner Loop Header: Depth=2 leal 999999(%r13), %eax xorl %edx, %edx divl %r13d movl %eax, %ebx leaq 24(%rsp), %rdi callq hipEventCreate leaq 16(%rsp), %rdi callq hipEventCreate leaq 8(%rsp), %rdi callq hipEventCreate movq %rsp, %rdi callq hipEventCreate movq 24(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 48(%rsp), %rdi movl $4000000, %edx # imm = 0x3D0900 movq %r12, %rsi movl $1, %ecx callq hipMemcpy movq 40(%rsp), %rdi movl $4000000, %edx # imm = 0x3D0900 movq %r15, %rsi movl $1, %ecx callq hipMemcpy movq 16(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 16(%rsp), %rdi callq hipEventSynchronize orq %r14, %rbx movl %r13d, %edx orq %r14, %rdx movq %rbx, %rdi movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_6 # %bb.5: # in Loop: Header=BB1_4 Depth=2 movq 48(%rsp), %rax movq 40(%rsp), %rcx movq 32(%rsp), %rdx movq %rax, 88(%rsp) movq %rcx, 184(%rsp) movq %rdx, 176(%rsp) movl %r13d, 100(%rsp) leaq 88(%rsp), %rax movq %rax, 112(%rsp) leaq 184(%rsp), %rax movq %rax, 120(%rsp) leaq 176(%rsp), %rax movq %rax, 128(%rsp) leaq 100(%rsp), %rax movq %rax, 136(%rsp) leaq 72(%rsp), %rdi leaq 56(%rsp), %rsi leaq 168(%rsp), %rdx leaq 160(%rsp), %rcx callq __hipPopCallConfiguration movq 72(%rsp), %rsi movl 80(%rsp), %edx movq 56(%rsp), %rcx movl 64(%rsp), %r8d movl $_Z3addPiS_S_i, %edi leaq 112(%rsp), %r9 pushq 160(%rsp) .cfi_adjust_cfa_offset 8 pushq 176(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 jmp .LBB1_6 .LBB1_8: # %.preheader.preheader movl $16, %ebx xorl %r14d, %r14d .p2align 4, 0x90 .LBB1_9: # %.preheader # =>This Inner Loop Header: Depth=1 leal 999999(%rbx), %eax xorl %edx, %edx divl %ebx movss 288(%rsp,%r14,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss .LCPI1_0(%rip), %xmm4 # xmm4 = mem[0],zero,zero,zero mulss %xmm4, %xmm0 movss %xmm0, 288(%rsp,%r14,4) movss 256(%rsp,%r14,4), %xmm1 # xmm1 = mem[0],zero,zero,zero mulss %xmm4, %xmm1 movss %xmm1, 256(%rsp,%r14,4) movss 224(%rsp,%r14,4), %xmm2 # xmm2 = mem[0],zero,zero,zero mulss %xmm4, %xmm2 movss %xmm2, 224(%rsp,%r14,4) movss 192(%rsp,%r14,4), %xmm3 # xmm3 = mem[0],zero,zero,zero mulss %xmm4, %xmm3 movss %xmm3, 192(%rsp,%r14,4) cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 cvtss2sd %xmm2, %xmm2 cvtss2sd %xmm3, %xmm3 movl $.L.str.5, %edi movl %ebx, %esi movl %eax, %edx movb $4, %al callq printf addl %ebx, %ebx incq %r14 cmpq $7, %r14 jne .LBB1_9 # %bb.10: movq %r12, %rdi callq free movq %r15, %rdi callq free movq 104(%rsp), %rdi # 8-byte Reload callq free movq 48(%rsp), %rdi callq hipFree movq 40(%rsp), %rdi callq hipFree movq 32(%rsp), %rdi callq hipFree xorl %eax, %eax addq $328, %rsp # imm = 0x148 .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end1: .size main, .Lfunc_end1-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB2_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB2_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z3addPiS_S_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end2: .size __hip_module_ctor, .Lfunc_end2-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB3_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB3_2: retq .Lfunc_end3: .size __hip_module_dtor, .Lfunc_end3-__hip_module_dtor .cfi_endproc # -- End function .type _Z3addPiS_S_i,@object # @_Z3addPiS_S_i .section .rodata,"a",@progbits .globl _Z3addPiS_S_i .p2align 3, 0x0 _Z3addPiS_S_i: .quad _Z18__device_stub__addPiS_S_i .size _Z3addPiS_S_i, 8 .type .L.str.5,@object # @.str.5 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.5: .asciz "%d\t\t%d\t\t%.5f\t\t%.5f\t\t%.5f\t\t%.5f\t\t\n" .size .L.str.5, 34 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z3addPiS_S_i" .size .L__unnamed_1, 14 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "t1: time for copying arrays" .size .Lstr, 28 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "t2: time for kernel execution" .size .Lstr.1, 30 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "t3: time for copying result back\n" .size .Lstr.2, 34 .type .Lstr.3,@object # @str.3 .Lstr.3: .asciz "All times in milliseconds" .size .Lstr.3, 26 .type .Lstr.4,@object # @str.4 .Lstr.4: .asciz "TPB\t\tNB\t\tt1\t\tt2\t\tt3\t\ttotal\t" .size .Lstr.4, 28 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z18__device_stub__addPiS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z3addPiS_S_i .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z3addPiS_S_i .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R6, SR_CTAID.X ; /* 0x0000000000067919 */ /* 0x000e220000002500 */ /*0020*/ HFMA2.MMA R7, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff077435 */ /* 0x000fe200000001ff */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe40000000a00 */ /*0040*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e240000002100 */ /*0050*/ IMAD R6, R6, c[0x0][0x178], R3 ; /* 0x00005e0006067a24 */ /* 0x001fca00078e0203 */ /*0060*/ IMAD.WIDE R2, R6, R7, c[0x0][0x160] ; /* 0x0000580006027625 */ /* 0x000fc800078e0207 */ /*0070*/ IMAD.WIDE R4, R6.reuse, R7.reuse, c[0x0][0x168] ; /* 0x00005a0006047625 */ /* 0x0c0fe400078e0207 */ /*0080*/ LDG.E R2, [R2.64] ; /* 0x0000000402027981 */ /* 0x000ea8000c1e1900 */ /*0090*/ LDG.E R5, [R4.64] ; /* 0x0000000404057981 */ /* 0x000ea2000c1e1900 */ /*00a0*/ IMAD.WIDE R6, R6, R7, c[0x0][0x170] ; /* 0x00005c0006067625 */ /* 0x000fe200078e0207 */ /*00b0*/ IADD3 R9, R2, R5, RZ ; /* 0x0000000502097210 */ /* 0x004fca0007ffe0ff */ /*00c0*/ STG.E [R6.64], R9 ; /* 0x0000000906007986 */ /* 0x000fe2000c101904 */ /*00d0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*00e0*/ BRA 0xe0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*00f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0100*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0110*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0120*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0130*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0140*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0150*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0160*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0170*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z3addPiS_S_i .globl _Z3addPiS_S_i .p2align 8 .type _Z3addPiS_S_i,@function _Z3addPiS_S_i: s_clause 0x2 s_load_b32 s2, s[0:1], 0x18 s_load_b128 s[4:7], s[0:1], 0x0 s_load_b64 s[0:1], s[0:1], 0x10 s_waitcnt lgkmcnt(0) v_mad_u64_u32 v[1:2], null, s15, s2, v[0:1] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v2, 31, v1 v_lshlrev_b64 v[0:1], 2, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v2, vcc_lo, s4, v0 v_add_co_ci_u32_e32 v3, vcc_lo, s5, v1, vcc_lo v_add_co_u32 v4, vcc_lo, s6, v0 v_add_co_ci_u32_e32 v5, vcc_lo, s7, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s0, v0 global_load_b32 v2, v[2:3], off global_load_b32 v3, v[4:5], off v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo s_waitcnt vmcnt(0) v_add_nc_u32_e32 v2, v3, v2 global_store_b32 v[0:1], v2, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z3addPiS_S_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 28 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 6 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z3addPiS_S_i, .Lfunc_end0-_Z3addPiS_S_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 28 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z3addPiS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z3addPiS_S_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 6 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_0016390e_00000000-6_q3.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2060: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2060: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z27__device_stub__Z3addPiS_S_iPiS_S_i .type _Z27__device_stub__Z3addPiS_S_iPiS_S_i, @function _Z27__device_stub__Z3addPiS_S_iPiS_S_i: .LFB2082: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movl %ecx, 4(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) leaq 4(%rsp), %rax movq %rax, 120(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 136(%rsp), %rax subq %fs:40, %rax jne .L8 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 168 pushq 40(%rsp) .cfi_def_cfa_offset 176 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z3addPiS_S_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2082: .size _Z27__device_stub__Z3addPiS_S_iPiS_S_i, .-_Z27__device_stub__Z3addPiS_S_iPiS_S_i .globl _Z3addPiS_S_i .type _Z3addPiS_S_i, @function _Z3addPiS_S_i: .LFB2083: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z27__device_stub__Z3addPiS_S_iPiS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2083: .size _Z3addPiS_S_i, .-_Z3addPiS_S_i .section .rodata.str1.1,"aMS",@progbits,1 .LC1: .string "t1: time for copying arrays\n" .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC2: .string "t2: time for kernel execution\n" .align 8 .LC3: .string "t3: time for copying result back\n\n" .section .rodata.str1.1 .LC4: .string "All times in milliseconds\n" .LC5: .string "TPB\t\tNB\t\tt1\t\tt2\t\tt3\t\ttotal\t\n" .section .rodata.str1.8 .align 8 .LC7: .string "%d\t\t%d\t\t%.5f\t\t%.5f\t\t%.5f\t\t%.5f\t\t\n" .text .globl main .type main, @function main: .LFB2057: .cfi_startproc endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $264, %rsp .cfi_def_cfa_offset 320 movq %fs:40, %rax movq %rax, 248(%rsp) xorl %eax, %eax movl $4000000, %edi call malloc@PLT movq %rax, %r14 movl $4000000, %edi call malloc@PLT movq %rax, %r13 movl $4000000, %edi call malloc@PLT movq %rax, %r15 leaq 32(%rsp), %rdi movl $4000000, %esi call cudaMalloc@PLT leaq 40(%rsp), %rdi movl $4000000, %esi call cudaMalloc@PLT leaq 48(%rsp), %rdi movl $4000000, %esi call cudaMalloc@PLT movl $0, %edi call time@PLT movl %eax, %edi call srand@PLT movl $0, %ebx .L12: call rand@PLT movslq %eax, %rdx imulq $1717986919, %rdx, %rdx sarq $35, %rdx movl %eax, %ecx sarl $31, %ecx subl %ecx, %edx leal (%rdx,%rdx,4), %edx sall $2, %edx subl %edx, %eax movl %eax, (%r14,%rbx) call rand@PLT movslq %eax, %rdx imulq $-580400985, %rdx, %rdx shrq $32, %rdx addl %eax, %edx sarl $5, %edx movl %eax, %ecx sarl $31, %ecx subl %ecx, %edx leal (%rdx,%rdx,8), %ecx leal (%rdx,%rcx,4), %edx subl %edx, %eax movl %eax, 0(%r13,%rbx) addq $4, %rbx cmpq $4000000, %rbx jne .L12 movl $0, %eax .L13: movl $0x00000000, 112(%rsp,%rax) movl $0x00000000, 144(%rsp,%rax) movl $0x00000000, 176(%rsp,%rax) movl $0x00000000, 208(%rsp,%rax) addq $4, %rax cmpq $28, %rax jne .L13 leaq .LC1(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq .LC2(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq .LC3(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq .LC4(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq .LC5(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movl $4, 12(%rsp) leaq 56(%rsp), %rax movq %rax, (%rsp) jmp .L14 .L15: movl $0, %esi movq 72(%rsp), %rdi call cudaEventRecord@PLT movq 72(%rsp), %rdi call cudaEventSynchronize@PLT movl $2, %ecx movl $4000000, %edx movq 48(%rsp), %rsi movq %r15, %rdi call cudaMemcpy@PLT movl $0, %esi movq 80(%rsp), %rdi call cudaEventRecord@PLT movq 80(%rsp), %rdi call cudaEventSynchronize@PLT movl $0x00000000, 24(%rsp) movl $0x00000000, 28(%rsp) movl $0x00000000, 88(%rsp) leaq 24(%rsp), %rdi movq 64(%rsp), %rdx movq 56(%rsp), %rsi call cudaEventElapsedTime@PLT leaq 28(%rsp), %rdi movq 72(%rsp), %rdx movq 64(%rsp), %rsi call cudaEventElapsedTime@PLT leaq 88(%rsp), %rdi movq 80(%rsp), %rdx movq 72(%rsp), %rsi call cudaEventElapsedTime@PLT leaq 100(%rsp), %rdi movq 80(%rsp), %rdx movq 56(%rsp), %rsi call cudaEventElapsedTime@PLT movss 112(%rsp,%rbx), %xmm0 addss 24(%rsp), %xmm0 movss %xmm0, 112(%rsp,%rbx) movss 144(%rsp,%rbx), %xmm0 addss 28(%rsp), %xmm0 movss %xmm0, 144(%rsp,%rbx) movss 176(%rsp,%rbx), %xmm0 addss 88(%rsp), %xmm0 movss %xmm0, 176(%rsp,%rbx) movss 208(%rsp,%rbx), %xmm0 addss 100(%rsp), %xmm0 movss %xmm0, 208(%rsp,%rbx) movq 56(%rsp), %rdi call cudaEventDestroy@PLT movq 64(%rsp), %rdi call cudaEventDestroy@PLT movq 72(%rsp), %rdi call cudaEventDestroy@PLT movq 80(%rsp), %rdi call cudaEventDestroy@PLT addl %ebp, %ebp addq $4, %rbx cmpq $28, %rbx je .L25 .L16: leal 999999(%rbp), %eax cltd idivl %ebp movl %eax, %r12d movq (%rsp), %rdi call cudaEventCreate@PLT leaq 64(%rsp), %rdi call cudaEventCreate@PLT leaq 72(%rsp), %rdi call cudaEventCreate@PLT leaq 80(%rsp), %rdi call cudaEventCreate@PLT movl $0, %esi movq 56(%rsp), %rdi call cudaEventRecord@PLT movl $1, %ecx movl $4000000, %edx movq %r14, %rsi movq 32(%rsp), %rdi call cudaMemcpy@PLT movl $1, %ecx movl $4000000, %edx movq %r13, %rsi movq 40(%rsp), %rdi call cudaMemcpy@PLT movl $0, %esi movq 64(%rsp), %rdi call cudaEventRecord@PLT movq 64(%rsp), %rdi call cudaEventSynchronize@PLT movl %ebp, 100(%rsp) movl $1, 104(%rsp) movl $1, 108(%rsp) movl %r12d, 88(%rsp) movl $1, 92(%rsp) movl $1, 96(%rsp) movl $0, %r9d movl $0, %r8d movq 100(%rsp), %rdx movl $1, %ecx movq 88(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax jne .L15 movl %ebp, %ecx movq 48(%rsp), %rdx movq 40(%rsp), %rsi movq 32(%rsp), %rdi call _Z27__device_stub__Z3addPiS_S_iPiS_S_i jmp .L15 .L25: subl $1, 12(%rsp) je .L19 .L14: movl $0, %ebx movl $16, %ebp jmp .L16 .L19: movl $0, %ebx movl $16, %ebp leaq .LC7(%rip), %r12 .L17: movss .LC6(%rip), %xmm0 mulss 112(%rsp,%rbx), %xmm0 movss %xmm0, 112(%rsp,%rbx) movss .LC6(%rip), %xmm1 mulss 144(%rsp,%rbx), %xmm1 movss %xmm1, 144(%rsp,%rbx) movss .LC6(%rip), %xmm2 mulss 176(%rsp,%rbx), %xmm2 movss %xmm2, 176(%rsp,%rbx) movss .LC6(%rip), %xmm3 mulss 208(%rsp,%rbx), %xmm3 movss %xmm3, 208(%rsp,%rbx) cvtss2sd %xmm0, %xmm0 leal 999999(%rbp), %eax cltd idivl %ebp cvtss2sd %xmm3, %xmm3 cvtss2sd %xmm2, %xmm2 cvtss2sd %xmm1, %xmm1 movl %eax, %ecx movl %ebp, %edx movq %r12, %rsi movl $2, %edi movl $4, %eax call __printf_chk@PLT addl %ebp, %ebp addq $4, %rbx cmpq $28, %rbx jne .L17 movq %r14, %rdi call free@PLT movq %r13, %rdi call free@PLT movq %r15, %rdi call free@PLT movq 32(%rsp), %rdi call cudaFree@PLT movq 40(%rsp), %rdi call cudaFree@PLT movq 48(%rsp), %rdi call cudaFree@PLT movq 248(%rsp), %rax subq %fs:40, %rax jne .L26 movl $0, %eax addq $264, %rsp .cfi_remember_state .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .L26: .cfi_restore_state call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size main, .-main .section .rodata.str1.1 .LC8: .string "_Z3addPiS_S_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2085: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC8(%rip), %rdx movq %rdx, %rcx leaq _Z3addPiS_S_i(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .section .rodata.cst4,"aM",@progbits,4 .align 4 .LC6: .long 1048576000 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "q3.hip" .globl _Z18__device_stub__addPiS_S_i # -- Begin function _Z18__device_stub__addPiS_S_i .p2align 4, 0x90 .type _Z18__device_stub__addPiS_S_i,@function _Z18__device_stub__addPiS_S_i: # @_Z18__device_stub__addPiS_S_i .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) movl %ecx, 4(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 4(%rsp), %rax movq %rax, 104(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z3addPiS_S_i, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z18__device_stub__addPiS_S_i, .Lfunc_end0-_Z18__device_stub__addPiS_S_i .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function main .LCPI1_0: .long 0x3e800000 # float 0.25 .text .globl main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 subq $328, %rsp # imm = 0x148 .cfi_def_cfa_offset 384 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movl $4000000, %edi # imm = 0x3D0900 callq malloc movq %rax, %r12 movl $4000000, %edi # imm = 0x3D0900 callq malloc movq %rax, %r15 movl $4000000, %edi # imm = 0x3D0900 callq malloc movq %rax, 104(%rsp) # 8-byte Spill leaq 48(%rsp), %rdi movl $4000000, %esi # imm = 0x3D0900 callq hipMalloc leaq 40(%rsp), %rdi movl $4000000, %esi # imm = 0x3D0900 callq hipMalloc leaq 32(%rsp), %rdi movl $4000000, %esi # imm = 0x3D0900 callq hipMalloc xorl %ebx, %ebx xorl %edi, %edi callq time movl %eax, %edi callq srand .p2align 4, 0x90 .LBB1_1: # =>This Inner Loop Header: Depth=1 callq rand cltq imulq $1717986919, %rax, %rcx # imm = 0x66666667 movq %rcx, %rdx shrq $63, %rdx shrq $32, %rcx sarl $3, %ecx addl %edx, %ecx shll $2, %ecx leal (%rcx,%rcx,4), %ecx subl %ecx, %eax movl %eax, (%r12,%rbx,4) callq rand cltq imulq $-580400985, %rax, %rcx # imm = 0xDD67C8A7 shrq $32, %rcx addl %eax, %ecx movl %ecx, %edx shrl $31, %edx sarl $5, %ecx addl %edx, %ecx leal (%rcx,%rcx,8), %edx leal (%rcx,%rdx,4), %ecx subl %ecx, %eax movl %eax, (%r15,%rbx,4) incq %rbx cmpq $1000000, %rbx # imm = 0xF4240 jne .LBB1_1 # %bb.2: movabsq $4294967296, %r14 # imm = 0x100000000 xorps %xmm0, %xmm0 movaps %xmm0, 288(%rsp) movups %xmm0, 300(%rsp) movaps %xmm0, 256(%rsp) movups %xmm0, 268(%rsp) movaps %xmm0, 224(%rsp) movups %xmm0, 236(%rsp) movaps %xmm0, 192(%rsp) movups %xmm0, 204(%rsp) movl $.Lstr, %edi callq puts@PLT movl $.Lstr.1, %edi callq puts@PLT movl $.Lstr.2, %edi callq puts@PLT movl $.Lstr.3, %edi callq puts@PLT movl $.Lstr.4, %edi callq puts@PLT xorl %eax, %eax jmp .LBB1_3 .p2align 4, 0x90 .LBB1_7: # in Loop: Header=BB1_3 Depth=1 movq 152(%rsp), %rax # 8-byte Reload incl %eax cmpl $4, %eax je .LBB1_8 .LBB1_3: # %.preheader71 # =>This Loop Header: Depth=1 # Child Loop BB1_4 Depth 2 movq %rax, 152(%rsp) # 8-byte Spill movl $16, %r13d xorl %ebp, %ebp jmp .LBB1_4 .p2align 4, 0x90 .LBB1_6: # in Loop: Header=BB1_4 Depth=2 movq 8(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 8(%rsp), %rdi callq hipEventSynchronize movq 32(%rsp), %rsi movl $4000000, %edx # imm = 0x3D0900 movq 104(%rsp), %rdi # 8-byte Reload movl $2, %ecx callq hipMemcpy movq (%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq (%rsp), %rdi callq hipEventSynchronize movl $0, 112(%rsp) movl $0, 72(%rsp) movl $0, 56(%rsp) movq 24(%rsp), %rsi movq 16(%rsp), %rdx leaq 112(%rsp), %rdi callq hipEventElapsedTime movq 16(%rsp), %rsi movq 8(%rsp), %rdx leaq 72(%rsp), %rdi callq hipEventElapsedTime movq 8(%rsp), %rsi movq (%rsp), %rdx leaq 56(%rsp), %rdi callq hipEventElapsedTime movq 24(%rsp), %rsi movq (%rsp), %rdx leaq 88(%rsp), %rdi callq hipEventElapsedTime movss 112(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 288(%rsp,%rbp,4), %xmm0 movss %xmm0, 288(%rsp,%rbp,4) movss 72(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 256(%rsp,%rbp,4), %xmm0 movss %xmm0, 256(%rsp,%rbp,4) movss 56(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 224(%rsp,%rbp,4), %xmm0 movss %xmm0, 224(%rsp,%rbp,4) movss 88(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero addss 192(%rsp,%rbp,4), %xmm0 movss %xmm0, 192(%rsp,%rbp,4) movq 24(%rsp), %rdi callq hipEventDestroy movq 16(%rsp), %rdi callq hipEventDestroy movq 8(%rsp), %rdi callq hipEventDestroy movq (%rsp), %rdi callq hipEventDestroy incq %rbp addl %r13d, %r13d cmpq $7, %rbp je .LBB1_7 .LBB1_4: # Parent Loop BB1_3 Depth=1 # => This Inner Loop Header: Depth=2 leal 999999(%r13), %eax xorl %edx, %edx divl %r13d movl %eax, %ebx leaq 24(%rsp), %rdi callq hipEventCreate leaq 16(%rsp), %rdi callq hipEventCreate leaq 8(%rsp), %rdi callq hipEventCreate movq %rsp, %rdi callq hipEventCreate movq 24(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 48(%rsp), %rdi movl $4000000, %edx # imm = 0x3D0900 movq %r12, %rsi movl $1, %ecx callq hipMemcpy movq 40(%rsp), %rdi movl $4000000, %edx # imm = 0x3D0900 movq %r15, %rsi movl $1, %ecx callq hipMemcpy movq 16(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 16(%rsp), %rdi callq hipEventSynchronize orq %r14, %rbx movl %r13d, %edx orq %r14, %rdx movq %rbx, %rdi movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_6 # %bb.5: # in Loop: Header=BB1_4 Depth=2 movq 48(%rsp), %rax movq 40(%rsp), %rcx movq 32(%rsp), %rdx movq %rax, 88(%rsp) movq %rcx, 184(%rsp) movq %rdx, 176(%rsp) movl %r13d, 100(%rsp) leaq 88(%rsp), %rax movq %rax, 112(%rsp) leaq 184(%rsp), %rax movq %rax, 120(%rsp) leaq 176(%rsp), %rax movq %rax, 128(%rsp) leaq 100(%rsp), %rax movq %rax, 136(%rsp) leaq 72(%rsp), %rdi leaq 56(%rsp), %rsi leaq 168(%rsp), %rdx leaq 160(%rsp), %rcx callq __hipPopCallConfiguration movq 72(%rsp), %rsi movl 80(%rsp), %edx movq 56(%rsp), %rcx movl 64(%rsp), %r8d movl $_Z3addPiS_S_i, %edi leaq 112(%rsp), %r9 pushq 160(%rsp) .cfi_adjust_cfa_offset 8 pushq 176(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 jmp .LBB1_6 .LBB1_8: # %.preheader.preheader movl $16, %ebx xorl %r14d, %r14d .p2align 4, 0x90 .LBB1_9: # %.preheader # =>This Inner Loop Header: Depth=1 leal 999999(%rbx), %eax xorl %edx, %edx divl %ebx movss 288(%rsp,%r14,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss .LCPI1_0(%rip), %xmm4 # xmm4 = mem[0],zero,zero,zero mulss %xmm4, %xmm0 movss %xmm0, 288(%rsp,%r14,4) movss 256(%rsp,%r14,4), %xmm1 # xmm1 = mem[0],zero,zero,zero mulss %xmm4, %xmm1 movss %xmm1, 256(%rsp,%r14,4) movss 224(%rsp,%r14,4), %xmm2 # xmm2 = mem[0],zero,zero,zero mulss %xmm4, %xmm2 movss %xmm2, 224(%rsp,%r14,4) movss 192(%rsp,%r14,4), %xmm3 # xmm3 = mem[0],zero,zero,zero mulss %xmm4, %xmm3 movss %xmm3, 192(%rsp,%r14,4) cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 cvtss2sd %xmm2, %xmm2 cvtss2sd %xmm3, %xmm3 movl $.L.str.5, %edi movl %ebx, %esi movl %eax, %edx movb $4, %al callq printf addl %ebx, %ebx incq %r14 cmpq $7, %r14 jne .LBB1_9 # %bb.10: movq %r12, %rdi callq free movq %r15, %rdi callq free movq 104(%rsp), %rdi # 8-byte Reload callq free movq 48(%rsp), %rdi callq hipFree movq 40(%rsp), %rdi callq hipFree movq 32(%rsp), %rdi callq hipFree xorl %eax, %eax addq $328, %rsp # imm = 0x148 .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end1: .size main, .Lfunc_end1-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB2_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB2_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z3addPiS_S_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end2: .size __hip_module_ctor, .Lfunc_end2-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB3_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB3_2: retq .Lfunc_end3: .size __hip_module_dtor, .Lfunc_end3-__hip_module_dtor .cfi_endproc # -- End function .type _Z3addPiS_S_i,@object # @_Z3addPiS_S_i .section .rodata,"a",@progbits .globl _Z3addPiS_S_i .p2align 3, 0x0 _Z3addPiS_S_i: .quad _Z18__device_stub__addPiS_S_i .size _Z3addPiS_S_i, 8 .type .L.str.5,@object # @.str.5 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.5: .asciz "%d\t\t%d\t\t%.5f\t\t%.5f\t\t%.5f\t\t%.5f\t\t\n" .size .L.str.5, 34 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z3addPiS_S_i" .size .L__unnamed_1, 14 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "t1: time for copying arrays" .size .Lstr, 28 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "t2: time for kernel execution" .size .Lstr.1, 30 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "t3: time for copying result back\n" .size .Lstr.2, 34 .type .Lstr.3,@object # @str.3 .Lstr.3: .asciz "All times in milliseconds" .size .Lstr.3, 26 .type .Lstr.4,@object # @str.4 .Lstr.4: .asciz "TPB\t\tNB\t\tt1\t\tt2\t\tt3\t\ttotal\t" .size .Lstr.4, 28 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z18__device_stub__addPiS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z3addPiS_S_i .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include <cuda_runtime_api.h> #include <iostream> #include <list> #include "tbb/concurrent_queue.h" #include "math.h" #define nThreadsPerBlocks 512 __global__ void assign(int *d_a, int *d_b, int size) { int index = threadIdx.x; if (index < size) { d_b[index] = d_a[0] ; } } void cudaPrintError(std::string m) { cudaError_t err = cudaGetLastError(); if (err != cudaSuccess) { std::cerr<<m<<" : "<<cudaGetErrorString(err)<<std::endl; } } int main(int argc, char* argv[]){ int GPU1 = 0; int GPU2 = 1; int size = 200; int *a; int *d_a0; int *b, *c; int *d_b1; cudaSetDevice(GPU1); a = (int *) malloc(1 * sizeof(int)); a[0] = 15 ; cudaMalloc((void**)&d_a0, 1*sizeof(int)) ; cudaMemcpy(d_a0, a, 1*sizeof(int), cudaMemcpyHostToDevice); cudaPrintError("after GPU1"); cudaDeviceEnablePeerAccess(GPU1,0); cudaSetDevice(GPU2); cudaMalloc((void**)&d_b1, size*sizeof(int)) ; cudaDeviceEnablePeerAccess(GPU1,0); cudaPrintError("after GPU2"); assign<<<1, nThreadsPerBlocks>>>(d_a0,d_b1,size); cudaPrintError("after KERNEL"); c = (int *) malloc(size * sizeof(int)); cudaMemcpy(c, d_b1, size*sizeof(int), cudaMemcpyDeviceToHost); cudaPrintError("after getData"); for(int i=0; i<size; i++ ) std::cout << i << " " << c[i] << std::endl; cudaFree(d_a0); cudaFree(d_b1); return 0; }
code for sm_80 Function : _Z6assignPiS_i .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R4, SR_TID.X ; /* 0x0000000000047919 */ /* 0x000e240000002100 */ /*0020*/ ISETP.GE.AND P0, PT, R4, c[0x0][0x170], PT ; /* 0x00005c0004007a0c */ /* 0x001fda0003f06270 */ /*0030*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0040*/ MOV R3, c[0x0][0x164] ; /* 0x0000590000037a02 */ /* 0x000fe20000000f00 */ /*0050*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x160] ; /* 0x00005800ff027624 */ /* 0x000fe200078e00ff */ /*0060*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc80000000a00 */ /*0070*/ LDG.E R3, [R2.64] ; /* 0x0000000402037981 */ /* 0x000ea2000c1e1900 */ /*0080*/ HFMA2.MMA R5, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff057435 */ /* 0x000fd400000001ff */ /*0090*/ IMAD.WIDE R4, R4, R5, c[0x0][0x168] ; /* 0x00005a0004047625 */ /* 0x000fca00078e0205 */ /*00a0*/ STG.E [R4.64], R3 ; /* 0x0000000304007986 */ /* 0x004fe2000c101904 */ /*00b0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*00c0*/ BRA 0xc0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*00d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0100*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0110*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0120*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0130*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0140*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0150*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0160*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0170*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <cuda_runtime_api.h> #include <iostream> #include <list> #include "tbb/concurrent_queue.h" #include "math.h" #define nThreadsPerBlocks 512 __global__ void assign(int *d_a, int *d_b, int size) { int index = threadIdx.x; if (index < size) { d_b[index] = d_a[0] ; } } void cudaPrintError(std::string m) { cudaError_t err = cudaGetLastError(); if (err != cudaSuccess) { std::cerr<<m<<" : "<<cudaGetErrorString(err)<<std::endl; } } int main(int argc, char* argv[]){ int GPU1 = 0; int GPU2 = 1; int size = 200; int *a; int *d_a0; int *b, *c; int *d_b1; cudaSetDevice(GPU1); a = (int *) malloc(1 * sizeof(int)); a[0] = 15 ; cudaMalloc((void**)&d_a0, 1*sizeof(int)) ; cudaMemcpy(d_a0, a, 1*sizeof(int), cudaMemcpyHostToDevice); cudaPrintError("after GPU1"); cudaDeviceEnablePeerAccess(GPU1,0); cudaSetDevice(GPU2); cudaMalloc((void**)&d_b1, size*sizeof(int)) ; cudaDeviceEnablePeerAccess(GPU1,0); cudaPrintError("after GPU2"); assign<<<1, nThreadsPerBlocks>>>(d_a0,d_b1,size); cudaPrintError("after KERNEL"); c = (int *) malloc(size * sizeof(int)); cudaMemcpy(c, d_b1, size*sizeof(int), cudaMemcpyDeviceToHost); cudaPrintError("after getData"); for(int i=0; i<size; i++ ) std::cout << i << " " << c[i] << std::endl; cudaFree(d_a0); cudaFree(d_b1); return 0; }
.file "tmpxft_000eecf1_00000000-6_test_cuda_peerTopeer.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB12824: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE12824: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string " : " .text .globl _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .type _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE, @function _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE: .LFB12820: .cfi_startproc endbr64 pushq %rbp .cfi_def_cfa_offset 16 .cfi_offset 6, -16 pushq %rbx .cfi_def_cfa_offset 24 .cfi_offset 3, -24 subq $8, %rsp .cfi_def_cfa_offset 32 movq %rdi, %rbp call cudaGetLastError@PLT testl %eax, %eax jne .L11 .L3: addq $8, %rsp .cfi_remember_state .cfi_def_cfa_offset 24 popq %rbx .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 ret .L11: .cfi_restore_state movl %eax, %ebx movq 8(%rbp), %rdx movq 0(%rbp), %rsi leaq _ZSt4cerr(%rip), %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movq %rax, %rbp movl $3, %edx leaq .LC0(%rip), %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %ebx, %edi call cudaGetErrorString@PLT movq %rax, %rbx testq %rax, %rax je .L12 movq %rax, %rdi call strlen@PLT movq %rax, %rdx movq %rbx, %rsi movq %rbp, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT .L6: movq 0(%rbp), %rax movq -24(%rax), %rax movq 240(%rbp,%rax), %rbx testq %rbx, %rbx je .L13 cmpb $0, 56(%rbx) je .L8 movzbl 67(%rbx), %esi .L9: movsbl %sil, %esi movq %rbp, %rdi call _ZNSo3putEc@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT jmp .L3 .L12: movq 0(%rbp), %rax movq %rbp, %rdi addq -24(%rax), %rdi movl 32(%rdi), %esi orl $1, %esi call _ZNSt9basic_iosIcSt11char_traitsIcEE5clearESt12_Ios_Iostate@PLT jmp .L6 .L13: call _ZSt16__throw_bad_castv@PLT .L8: movq %rbx, %rdi call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq (%rbx), %rax movl $10, %esi movq %rbx, %rdi call *48(%rax) movl %eax, %esi jmp .L9 .cfi_endproc .LFE12820: .size _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE, .-_Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .globl _Z28__device_stub__Z6assignPiS_iPiS_i .type _Z28__device_stub__Z6assignPiS_iPiS_i, @function _Z28__device_stub__Z6assignPiS_iPiS_i: .LFB12846: .cfi_startproc endbr64 subq $136, %rsp .cfi_def_cfa_offset 144 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movl %edx, 12(%rsp) movq %fs:40, %rax movq %rax, 120(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 12(%rsp), %rax movq %rax, 112(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L18 .L14: movq 120(%rsp), %rax subq %fs:40, %rax jne .L19 addq $136, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L18: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z6assignPiS_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L14 .L19: call __stack_chk_fail@PLT .cfi_endproc .LFE12846: .size _Z28__device_stub__Z6assignPiS_iPiS_i, .-_Z28__device_stub__Z6assignPiS_iPiS_i .globl _Z6assignPiS_i .type _Z6assignPiS_i, @function _Z6assignPiS_i: .LFB12847: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z28__device_stub__Z6assignPiS_iPiS_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE12847: .size _Z6assignPiS_i, .-_Z6assignPiS_i .section .rodata.str1.1 .LC1: .string "_Z6assignPiS_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB12849: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC1(%rip), %rdx movq %rdx, %rcx leaq _Z6assignPiS_i(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE12849: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .rodata._ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_.str1.8,"aMS",@progbits,1 .align 8 .LC2: .string "basic_string: construction from null is not valid" .section .text._ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_,"axG",@progbits,_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC5IS3_EEPKcRKS3_,comdat .align 2 .weak _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_ .type _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_, @function _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_: .LFB13261: .cfi_startproc endbr64 pushq %r13 .cfi_def_cfa_offset 16 .cfi_offset 13, -16 pushq %r12 .cfi_def_cfa_offset 24 .cfi_offset 12, -24 pushq %rbp .cfi_def_cfa_offset 32 .cfi_offset 6, -32 pushq %rbx .cfi_def_cfa_offset 40 .cfi_offset 3, -40 subq $24, %rsp .cfi_def_cfa_offset 64 movq %fs:40, %rax movq %rax, 8(%rsp) xorl %eax, %eax leaq 16(%rdi), %r12 movq %r12, (%rdi) testq %rsi, %rsi je .L33 movq %rdi, %rbx movq %rsi, %r13 movq %rsi, %rdi call strlen@PLT movq %rax, %rbp movq %rax, (%rsp) cmpq $15, %rax ja .L34 cmpq $1, %rax jne .L29 movzbl 0(%r13), %eax movb %al, 16(%rbx) .L30: movq (%rsp), %rax movq %rax, 8(%rbx) movq (%rbx), %rdx movb $0, (%rdx,%rax) movq 8(%rsp), %rax subq %fs:40, %rax jne .L35 addq $24, %rsp .cfi_remember_state .cfi_def_cfa_offset 40 popq %rbx .cfi_def_cfa_offset 32 popq %rbp .cfi_def_cfa_offset 24 popq %r12 .cfi_def_cfa_offset 16 popq %r13 .cfi_def_cfa_offset 8 ret .L33: .cfi_restore_state movq 8(%rsp), %rax subq %fs:40, %rax jne .L36 leaq .LC2(%rip), %rdi call _ZSt19__throw_logic_errorPKc@PLT .L36: call __stack_chk_fail@PLT .L34: movq %rsp, %rsi movl $0, %edx movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE9_M_createERmm@PLT movq %rax, %r12 movq %rax, (%rbx) movq (%rsp), %rax movq %rax, 16(%rbx) .L28: movq %rbp, %rdx movq %r13, %rsi movq %r12, %rdi call memcpy@PLT jmp .L30 .L29: testq %rax, %rax je .L30 jmp .L28 .L35: call __stack_chk_fail@PLT .cfi_endproc .LFE13261: .size _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_, .-_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_ .weak _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .set _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_,_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_ .section .rodata.str1.1 .LC3: .string "after GPU1" .LC4: .string "after GPU2" .LC5: .string "after KERNEL" .LC6: .string "after getData" .LC7: .string " " .text .globl main .type main, @function main: .LFB12821: .cfi_startproc .cfi_personality 0x9b,DW.ref.__gxx_personality_v0 .cfi_lsda 0x1b,.LLSDA12821 endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $104, %rsp .cfi_def_cfa_offset 160 movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax movl $0, %edi .LEHB0: call cudaSetDevice@PLT movl $4, %edi call malloc@PLT movq %rax, %rbx movl $15, (%rax) leaq 8(%rsp), %rdi movl $4, %esi call cudaMalloc@PLT movl $1, %ecx movl $4, %edx movq %rbx, %rsi movq 8(%rsp), %rdi call cudaMemcpy@PLT leaq 36(%rsp), %rdx leaq 48(%rsp), %rbx leaq .LC3(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE0: movq %rbx, %rdi .LEHB1: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE1: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $0, %esi movl $0, %edi .LEHB2: call cudaDeviceEnablePeerAccess@PLT movl $1, %edi call cudaSetDevice@PLT leaq 16(%rsp), %rdi movl $800, %esi call cudaMalloc@PLT movl $0, %esi movl $0, %edi call cudaDeviceEnablePeerAccess@PLT leaq 36(%rsp), %rdx leaq .LC4(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE2: movq %rbx, %rdi .LEHB3: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE3: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $512, 36(%rsp) movl $1, 40(%rsp) movl $1, 44(%rsp) movl $1, 24(%rsp) movl $1, 28(%rsp) movl $1, 32(%rsp) movl $0, %r9d movl $0, %r8d movq 36(%rsp), %rdx movl $1, %ecx movq 24(%rsp), %rdi movl $1, %esi .LEHB4: call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L60 .L38: leaq 36(%rsp), %rdx leaq 48(%rsp), %rbx leaq .LC5(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE4: movq %rbx, %rdi .LEHB5: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE5: jmp .L61 .L60: movl $200, %edx movq 16(%rsp), %rsi movq 8(%rsp), %rdi .LEHB6: call _Z28__device_stub__Z6assignPiS_iPiS_i jmp .L38 .L61: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $800, %edi call malloc@PLT movq %rax, %r12 movl $2, %ecx movl $800, %edx movq 16(%rsp), %rsi movq %rax, %rdi call cudaMemcpy@PLT leaq 36(%rsp), %rdx leaq .LC6(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE6: movq %rbx, %rdi .LEHB7: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE7: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $0, %ebp leaq _ZSt4cout(%rip), %r14 leaq .LC7(%rip), %r13 jmp .L43 .L64: movq 88(%rsp), %rax subq %fs:40, %rax jne .L62 .LEHB8: call _ZSt16__throw_bad_castv@PLT .L62: call __stack_chk_fail@PLT .L65: movzbl 67(%r15), %esi .L42: movsbl %sil, %esi movq %rbx, %rdi call _ZNSo3putEc@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT addq $1, %rbp cmpq $200, %rbp je .L63 .L43: movl %ebp, %esi movq %r14, %rdi call _ZNSolsEi@PLT movq %rax, %rbx movl $1, %edx movq %r13, %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl (%r12,%rbp,4), %esi movq %rbx, %rdi call _ZNSolsEi@PLT movq %rax, %rbx movq (%rax), %rax movq -24(%rax), %rax movq 240(%rbx,%rax), %r15 testq %r15, %r15 je .L64 cmpb $0, 56(%r15) jne .L65 movq %r15, %rdi call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq (%r15), %rax movl $10, %esi movq %r15, %rdi call *48(%rax) movl %eax, %esi jmp .L42 .L63: movq 8(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call cudaFree@PLT movq 88(%rsp), %rax subq %fs:40, %rax jne .L66 movl $0, %eax addq $104, %rsp .cfi_remember_state .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .L53: .cfi_restore_state endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L45 call __stack_chk_fail@PLT .L45: movq %rbx, %rdi call _Unwind_Resume@PLT .L54: endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L47 call __stack_chk_fail@PLT .L47: movq %rbx, %rdi call _Unwind_Resume@PLT .L55: endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L49 call __stack_chk_fail@PLT .L49: movq %rbx, %rdi call _Unwind_Resume@PLT .L56: endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L51 call __stack_chk_fail@PLT .L51: movq %rbx, %rdi call _Unwind_Resume@PLT .LEHE8: .L66: call __stack_chk_fail@PLT .cfi_endproc .LFE12821: .globl __gxx_personality_v0 .section .gcc_except_table,"a",@progbits .LLSDA12821: .byte 0xff .byte 0xff .byte 0x1 .uleb128 .LLSDACSE12821-.LLSDACSB12821 .LLSDACSB12821: .uleb128 .LEHB0-.LFB12821 .uleb128 .LEHE0-.LEHB0 .uleb128 0 .uleb128 0 .uleb128 .LEHB1-.LFB12821 .uleb128 .LEHE1-.LEHB1 .uleb128 .L53-.LFB12821 .uleb128 0 .uleb128 .LEHB2-.LFB12821 .uleb128 .LEHE2-.LEHB2 .uleb128 0 .uleb128 0 .uleb128 .LEHB3-.LFB12821 .uleb128 .LEHE3-.LEHB3 .uleb128 .L54-.LFB12821 .uleb128 0 .uleb128 .LEHB4-.LFB12821 .uleb128 .LEHE4-.LEHB4 .uleb128 0 .uleb128 0 .uleb128 .LEHB5-.LFB12821 .uleb128 .LEHE5-.LEHB5 .uleb128 .L55-.LFB12821 .uleb128 0 .uleb128 .LEHB6-.LFB12821 .uleb128 .LEHE6-.LEHB6 .uleb128 0 .uleb128 0 .uleb128 .LEHB7-.LFB12821 .uleb128 .LEHE7-.LEHB7 .uleb128 .L56-.LFB12821 .uleb128 0 .uleb128 .LEHB8-.LFB12821 .uleb128 .LEHE8-.LEHB8 .uleb128 0 .uleb128 0 .LLSDACSE12821: .text .size main, .-main .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .hidden DW.ref.__gxx_personality_v0 .weak DW.ref.__gxx_personality_v0 .section .data.rel.local.DW.ref.__gxx_personality_v0,"awG",@progbits,DW.ref.__gxx_personality_v0,comdat .align 8 .type DW.ref.__gxx_personality_v0, @object .size DW.ref.__gxx_personality_v0, 8 DW.ref.__gxx_personality_v0: .quad __gxx_personality_v0 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include <cuda_runtime_api.h> #include <iostream> #include <list> #include "tbb/concurrent_queue.h" #include "math.h" #define nThreadsPerBlocks 512 __global__ void assign(int *d_a, int *d_b, int size) { int index = threadIdx.x; if (index < size) { d_b[index] = d_a[0] ; } } void cudaPrintError(std::string m) { cudaError_t err = cudaGetLastError(); if (err != cudaSuccess) { std::cerr<<m<<" : "<<cudaGetErrorString(err)<<std::endl; } } int main(int argc, char* argv[]){ int GPU1 = 0; int GPU2 = 1; int size = 200; int *a; int *d_a0; int *b, *c; int *d_b1; cudaSetDevice(GPU1); a = (int *) malloc(1 * sizeof(int)); a[0] = 15 ; cudaMalloc((void**)&d_a0, 1*sizeof(int)) ; cudaMemcpy(d_a0, a, 1*sizeof(int), cudaMemcpyHostToDevice); cudaPrintError("after GPU1"); cudaDeviceEnablePeerAccess(GPU1,0); cudaSetDevice(GPU2); cudaMalloc((void**)&d_b1, size*sizeof(int)) ; cudaDeviceEnablePeerAccess(GPU1,0); cudaPrintError("after GPU2"); assign<<<1, nThreadsPerBlocks>>>(d_a0,d_b1,size); cudaPrintError("after KERNEL"); c = (int *) malloc(size * sizeof(int)); cudaMemcpy(c, d_b1, size*sizeof(int), cudaMemcpyDeviceToHost); cudaPrintError("after getData"); for(int i=0; i<size; i++ ) std::cout << i << " " << c[i] << std::endl; cudaFree(d_a0); cudaFree(d_b1); return 0; }
#include <hip/hip_runtime.h> #include <hip/hip_runtime_api.h> #include <iostream> #include <list> #include "tbb/concurrent_queue.h" #include "math.h" #define nThreadsPerBlocks 512 __global__ void assign(int *d_a, int *d_b, int size) { int index = threadIdx.x; if (index < size) { d_b[index] = d_a[0] ; } } void cudaPrintError(std::string m) { hipError_t err = hipGetLastError(); if (err != hipSuccess) { std::cerr<<m<<" : "<<hipGetErrorString(err)<<std::endl; } } int main(int argc, char* argv[]){ int GPU1 = 0; int GPU2 = 1; int size = 200; int *a; int *d_a0; int *b, *c; int *d_b1; hipSetDevice(GPU1); a = (int *) malloc(1 * sizeof(int)); a[0] = 15 ; hipMalloc((void**)&d_a0, 1*sizeof(int)) ; hipMemcpy(d_a0, a, 1*sizeof(int), hipMemcpyHostToDevice); cudaPrintError("after GPU1"); hipDeviceEnablePeerAccess(GPU1,0); hipSetDevice(GPU2); hipMalloc((void**)&d_b1, size*sizeof(int)) ; hipDeviceEnablePeerAccess(GPU1,0); cudaPrintError("after GPU2"); assign<<<1, nThreadsPerBlocks>>>(d_a0,d_b1,size); cudaPrintError("after KERNEL"); c = (int *) malloc(size * sizeof(int)); hipMemcpy(c, d_b1, size*sizeof(int), hipMemcpyDeviceToHost); cudaPrintError("after getData"); for(int i=0; i<size; i++ ) std::cout << i << " " << c[i] << std::endl; hipFree(d_a0); hipFree(d_b1); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include <hip/hip_runtime_api.h> #include <iostream> #include <list> #include "tbb/concurrent_queue.h" #include "math.h" #define nThreadsPerBlocks 512 __global__ void assign(int *d_a, int *d_b, int size) { int index = threadIdx.x; if (index < size) { d_b[index] = d_a[0] ; } } void cudaPrintError(std::string m) { hipError_t err = hipGetLastError(); if (err != hipSuccess) { std::cerr<<m<<" : "<<hipGetErrorString(err)<<std::endl; } } int main(int argc, char* argv[]){ int GPU1 = 0; int GPU2 = 1; int size = 200; int *a; int *d_a0; int *b, *c; int *d_b1; hipSetDevice(GPU1); a = (int *) malloc(1 * sizeof(int)); a[0] = 15 ; hipMalloc((void**)&d_a0, 1*sizeof(int)) ; hipMemcpy(d_a0, a, 1*sizeof(int), hipMemcpyHostToDevice); cudaPrintError("after GPU1"); hipDeviceEnablePeerAccess(GPU1,0); hipSetDevice(GPU2); hipMalloc((void**)&d_b1, size*sizeof(int)) ; hipDeviceEnablePeerAccess(GPU1,0); cudaPrintError("after GPU2"); assign<<<1, nThreadsPerBlocks>>>(d_a0,d_b1,size); cudaPrintError("after KERNEL"); c = (int *) malloc(size * sizeof(int)); hipMemcpy(c, d_b1, size*sizeof(int), hipMemcpyDeviceToHost); cudaPrintError("after getData"); for(int i=0; i<size; i++ ) std::cout << i << " " << c[i] << std::endl; hipFree(d_a0); hipFree(d_b1); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z6assignPiS_i .globl _Z6assignPiS_i .p2align 8 .type _Z6assignPiS_i,@function _Z6assignPiS_i: s_load_b32 s2, s[0:1], 0x10 s_waitcnt lgkmcnt(0) v_cmp_gt_i32_e32 vcc_lo, s2, v0 s_and_saveexec_b32 s2, vcc_lo s_cbranch_execz .LBB0_2 s_load_b128 s[0:3], s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_load_b32 s0, s[0:1], 0x0 s_waitcnt lgkmcnt(0) v_dual_mov_b32 v1, s0 :: v_dual_lshlrev_b32 v0, 2, v0 global_store_b32 v0, v1, s[2:3] .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z6assignPiS_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 20 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 2 .amdhsa_next_free_sgpr 4 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z6assignPiS_i, .Lfunc_end0-_Z6assignPiS_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .offset: 16 .size: 4 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 20 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z6assignPiS_i .private_segment_fixed_size: 0 .sgpr_count: 6 .sgpr_spill_count: 0 .symbol: _Z6assignPiS_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 2 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include <hip/hip_runtime_api.h> #include <iostream> #include <list> #include "tbb/concurrent_queue.h" #include "math.h" #define nThreadsPerBlocks 512 __global__ void assign(int *d_a, int *d_b, int size) { int index = threadIdx.x; if (index < size) { d_b[index] = d_a[0] ; } } void cudaPrintError(std::string m) { hipError_t err = hipGetLastError(); if (err != hipSuccess) { std::cerr<<m<<" : "<<hipGetErrorString(err)<<std::endl; } } int main(int argc, char* argv[]){ int GPU1 = 0; int GPU2 = 1; int size = 200; int *a; int *d_a0; int *b, *c; int *d_b1; hipSetDevice(GPU1); a = (int *) malloc(1 * sizeof(int)); a[0] = 15 ; hipMalloc((void**)&d_a0, 1*sizeof(int)) ; hipMemcpy(d_a0, a, 1*sizeof(int), hipMemcpyHostToDevice); cudaPrintError("after GPU1"); hipDeviceEnablePeerAccess(GPU1,0); hipSetDevice(GPU2); hipMalloc((void**)&d_b1, size*sizeof(int)) ; hipDeviceEnablePeerAccess(GPU1,0); cudaPrintError("after GPU2"); assign<<<1, nThreadsPerBlocks>>>(d_a0,d_b1,size); cudaPrintError("after KERNEL"); c = (int *) malloc(size * sizeof(int)); hipMemcpy(c, d_b1, size*sizeof(int), hipMemcpyDeviceToHost); cudaPrintError("after getData"); for(int i=0; i<size; i++ ) std::cout << i << " " << c[i] << std::endl; hipFree(d_a0); hipFree(d_b1); return 0; }
.text .file "test_cuda_peerTopeer.hip" # Start of file scope inline assembly .globl _ZSt21ios_base_library_initv # End of file scope inline assembly .globl _Z21__device_stub__assignPiS_i # -- Begin function _Z21__device_stub__assignPiS_i .p2align 4, 0x90 .type _Z21__device_stub__assignPiS_i,@function _Z21__device_stub__assignPiS_i: # @_Z21__device_stub__assignPiS_i .cfi_startproc # %bb.0: subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movl %edx, 12(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 12(%rsp), %rax movq %rax, 96(%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z6assignPiS_i, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $120, %rsp .cfi_adjust_cfa_offset -120 retq .Lfunc_end0: .size _Z21__device_stub__assignPiS_i, .Lfunc_end0-_Z21__device_stub__assignPiS_i .cfi_endproc # -- End function .globl _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE # -- Begin function _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .p2align 4, 0x90 .type _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE,@function _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE: # @_Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r14 .cfi_def_cfa_offset 24 pushq %rbx .cfi_def_cfa_offset 32 .cfi_offset %rbx, -32 .cfi_offset %r14, -24 .cfi_offset %rbp, -16 movq %rdi, %rbx callq hipGetLastError testl %eax, %eax je .LBB1_9 # %bb.1: movl %eax, %ebp movq (%rbx), %rsi movq 8(%rbx), %rdx movl $_ZSt4cerr, %edi callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l movq %rax, %rbx movl $.L.str, %esi movl $3, %edx movq %rax, %rdi callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l movl %ebp, %edi callq hipGetErrorString testq %rax, %rax je .LBB1_2 # %bb.3: movq %rax, %rdi movq %rax, %r14 callq strlen movq %rbx, %rdi movq %r14, %rsi movq %rax, %rdx callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l jmp .LBB1_4 .LBB1_9: popq %rbx .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .LBB1_2: .cfi_def_cfa_offset 32 movq (%rbx), %rax movq -24(%rax), %rax movq %rbx, %rdi addq %rax, %rdi movl 32(%rbx,%rax), %esi orl $1, %esi callq _ZNSt9basic_iosIcSt11char_traitsIcEE5clearESt12_Ios_Iostate .LBB1_4: # %_ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_PKc.exit movq (%rbx), %rax movq -24(%rax), %rax movq 240(%rbx,%rax), %r14 testq %r14, %r14 je .LBB1_10 # %bb.5: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i cmpb $0, 56(%r14) je .LBB1_7 # %bb.6: movzbl 67(%r14), %eax jmp .LBB1_8 .LBB1_7: movq %r14, %rdi callq _ZNKSt5ctypeIcE13_M_widen_initEv movq (%r14), %rax movq %r14, %rdi movl $10, %esi callq *48(%rax) .LBB1_8: # %_ZSt4endlIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_.exit movsbl %al, %esi movq %rbx, %rdi callq _ZNSo3putEc movq %rax, %rdi popq %rbx .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 jmp _ZNSo5flushEv # TAILCALL .LBB1_10: .cfi_def_cfa_offset 32 callq _ZSt16__throw_bad_castv .Lfunc_end1: .size _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE, .Lfunc_end1-_Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .Lfunc_begin0: .cfi_startproc .cfi_personality 3, __gxx_personality_v0 .cfi_lsda 3, .Lexception0 # %bb.0: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_.exit pushq %r15 .cfi_def_cfa_offset 16 pushq %r14 .cfi_def_cfa_offset 24 pushq %r12 .cfi_def_cfa_offset 32 pushq %rbx .cfi_def_cfa_offset 40 subq $248, %rsp .cfi_def_cfa_offset 288 .cfi_offset %rbx, -40 .cfi_offset %r12, -32 .cfi_offset %r14, -24 .cfi_offset %r15, -16 .cfi_escape 0x2e, 0x00 xorl %edi, %edi callq hipSetDevice .cfi_escape 0x2e, 0x00 movl $4, %edi callq malloc movq %rax, %rbx movl $15, (%rax) .cfi_escape 0x2e, 0x00 leaq 16(%rsp), %rdi movl $4, %esi callq hipMalloc movq 16(%rsp), %rdi .cfi_escape 0x2e, 0x00 movl $4, %edx movq %rbx, %rsi movl $1, %ecx callq hipMemcpy leaq 144(%rsp), %r14 movq %r14, 128(%rsp) movabsq $5784627922081179233, %rbx # imm = 0x5047207265746661 movq %rbx, 144(%rsp) movw $12629, 152(%rsp) # imm = 0x3155 movq $10, 136(%rsp) movb $0, 154(%rsp) .Ltmp0: .cfi_escape 0x2e, 0x00 leaq 128(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp1: # %bb.1: movq 128(%rsp), %rdi cmpq %r14, %rdi je .LBB2_3 # %bb.2: # %.critedge.i.i .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_3: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit .cfi_escape 0x2e, 0x00 xorl %edi, %edi xorl %esi, %esi callq hipDeviceEnablePeerAccess .cfi_escape 0x2e, 0x00 movl $1, %edi callq hipSetDevice .cfi_escape 0x2e, 0x00 leaq 8(%rsp), %rdi movl $800, %esi # imm = 0x320 callq hipMalloc .cfi_escape 0x2e, 0x00 xorl %edi, %edi xorl %esi, %esi callq hipDeviceEnablePeerAccess leaq 112(%rsp), %r14 movq %r14, 96(%rsp) movq %rbx, 112(%rsp) movw $12885, 120(%rsp) # imm = 0x3255 movq $10, 104(%rsp) movb $0, 122(%rsp) .Ltmp3: .cfi_escape 0x2e, 0x00 leaq 96(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp4: # %bb.4: movq 96(%rsp), %rdi cmpq %r14, %rdi je .LBB2_6 # %bb.5: # %.critedge.i.i43 .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_6: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit45 movabsq $4294967297, %rdi # imm = 0x100000001 leaq 511(%rdi), %rdx .cfi_escape 0x2e, 0x00 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB2_8 # %bb.7: movq 16(%rsp), %rax movq 8(%rsp), %rcx movq %rax, 216(%rsp) movq %rcx, 208(%rsp) movl $200, 28(%rsp) leaq 216(%rsp), %rax movq %rax, 224(%rsp) leaq 208(%rsp), %rax movq %rax, 232(%rsp) leaq 28(%rsp), %rax movq %rax, 240(%rsp) .cfi_escape 0x2e, 0x00 leaq 192(%rsp), %rdi leaq 176(%rsp), %rsi leaq 168(%rsp), %rdx leaq 160(%rsp), %rcx callq __hipPopCallConfiguration movq 192(%rsp), %rsi movl 200(%rsp), %edx movq 176(%rsp), %rcx movl 184(%rsp), %r8d .cfi_escape 0x2e, 0x10 leaq 224(%rsp), %r9 movl $_Z6assignPiS_i, %edi pushq 160(%rsp) .cfi_adjust_cfa_offset 8 pushq 176(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB2_8: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_.exit58 leaq 80(%rsp), %r14 movq %r14, 64(%rsp) movabsq $4993120287570814561, %rax # imm = 0x454B207265746661 movq %rax, 80(%rsp) movl $1279610450, 88(%rsp) # imm = 0x4C454E52 movq $12, 72(%rsp) movb $0, 92(%rsp) .Ltmp6: .cfi_escape 0x2e, 0x00 leaq 64(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp7: # %bb.9: movq 64(%rsp), %rdi cmpq %r14, %rdi je .LBB2_11 # %bb.10: # %.critedge.i.i59 .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_11: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit61 .cfi_escape 0x2e, 0x00 movl $800, %edi # imm = 0x320 callq malloc movq %rax, %rbx movq 8(%rsp), %rsi .cfi_escape 0x2e, 0x00 movl $800, %edx # imm = 0x320 movq %rax, %rdi movl $2, %ecx callq hipMemcpy leaq 48(%rsp), %r14 movq %r14, 32(%rsp) movabsq $7306844596132406881, %rax # imm = 0x6567207265746661 movq %rax, 48(%rsp) movabsq $7022344665615918880, %rax # imm = 0x6174614474656720 movq %rax, 53(%rsp) movq $13, 40(%rsp) movb $0, 61(%rsp) .Ltmp9: .cfi_escape 0x2e, 0x00 leaq 32(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp10: # %bb.12: movq 32(%rsp), %rdi cmpq %r14, %rdi je .LBB2_14 # %bb.13: # %.critedge.i.i69 .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_14: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit71.preheader xorl %r14d, %r14d jmp .LBB2_15 .p2align 4, 0x90 .LBB2_29: # in Loop: Header=BB2_15 Depth=1 .cfi_escape 0x2e, 0x00 movq %r15, %rdi movq %rax, %r12 callq _ZNKSt5ctypeIcE13_M_widen_initEv movq (%r15), %rax .cfi_escape 0x2e, 0x00 movq %r15, %rdi movl $10, %esi callq *48(%rax) movl %eax, %ecx movq %r12, %rax .LBB2_30: # %_ZSt4endlIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_.exit # in Loop: Header=BB2_15 Depth=1 .cfi_escape 0x2e, 0x00 movsbl %cl, %esi movq %rax, %rdi callq _ZNSo3putEc .cfi_escape 0x2e, 0x00 movq %rax, %rdi callq _ZNSo5flushEv incq %r14 cmpq $200, %r14 je .LBB2_31 .LBB2_15: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit71 # =>This Inner Loop Header: Depth=1 .cfi_escape 0x2e, 0x00 movl $_ZSt4cout, %edi movl %r14d, %esi callq _ZNSolsEi movq %rax, %r15 .cfi_escape 0x2e, 0x00 movl $.L.str.5, %esi movl $1, %edx movq %rax, %rdi callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l movl (%rbx,%r14,4), %esi .cfi_escape 0x2e, 0x00 movq %r15, %rdi callq _ZNSolsEi movq (%rax), %rcx movq -24(%rcx), %rcx movq 240(%rax,%rcx), %r15 testq %r15, %r15 je .LBB2_16 # %bb.27: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i # in Loop: Header=BB2_15 Depth=1 cmpb $0, 56(%r15) je .LBB2_29 # %bb.28: # in Loop: Header=BB2_15 Depth=1 movzbl 67(%r15), %ecx jmp .LBB2_30 .LBB2_31: movq 16(%rsp), %rdi .cfi_escape 0x2e, 0x00 callq hipFree movq 8(%rsp), %rdi .cfi_escape 0x2e, 0x00 callq hipFree xorl %eax, %eax addq $248, %rsp .cfi_def_cfa_offset 40 popq %rbx .cfi_def_cfa_offset 32 popq %r12 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 retq .LBB2_16: .cfi_def_cfa_offset 288 .cfi_escape 0x2e, 0x00 callq _ZSt16__throw_bad_castv .LBB2_25: .Ltmp11: movq %rax, %rbx movq 32(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.26: # %.critedge.i.i75 .cfi_escape 0x2e, 0x00 jmp .LBB2_19 .LBB2_23: .Ltmp8: movq %rax, %rbx movq 64(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.24: # %.critedge.i.i72 .cfi_escape 0x2e, 0x00 jmp .LBB2_19 .LBB2_21: .Ltmp5: movq %rax, %rbx movq 96(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.22: # %.critedge.i.i49 .cfi_escape 0x2e, 0x00 jmp .LBB2_19 .LBB2_17: .Ltmp2: movq %rax, %rbx movq 128(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.18: # %.critedge.i.i46 .cfi_escape 0x2e, 0x00 .LBB2_19: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit48 callq _ZdlPv .LBB2_20: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit48 .cfi_escape 0x2e, 0x00 movq %rbx, %rdi callq _Unwind_Resume@PLT .Lfunc_end2: .size main, .Lfunc_end2-main .cfi_endproc .section .gcc_except_table,"a",@progbits .p2align 2, 0x0 GCC_except_table2: .Lexception0: .byte 255 # @LPStart Encoding = omit .byte 255 # @TType Encoding = omit .byte 1 # Call site Encoding = uleb128 .uleb128 .Lcst_end0-.Lcst_begin0 .Lcst_begin0: .uleb128 .Lfunc_begin0-.Lfunc_begin0 # >> Call Site 1 << .uleb128 .Ltmp0-.Lfunc_begin0 # Call between .Lfunc_begin0 and .Ltmp0 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp0-.Lfunc_begin0 # >> Call Site 2 << .uleb128 .Ltmp1-.Ltmp0 # Call between .Ltmp0 and .Ltmp1 .uleb128 .Ltmp2-.Lfunc_begin0 # jumps to .Ltmp2 .byte 0 # On action: cleanup .uleb128 .Ltmp1-.Lfunc_begin0 # >> Call Site 3 << .uleb128 .Ltmp3-.Ltmp1 # Call between .Ltmp1 and .Ltmp3 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp3-.Lfunc_begin0 # >> Call Site 4 << .uleb128 .Ltmp4-.Ltmp3 # Call between .Ltmp3 and .Ltmp4 .uleb128 .Ltmp5-.Lfunc_begin0 # jumps to .Ltmp5 .byte 0 # On action: cleanup .uleb128 .Ltmp4-.Lfunc_begin0 # >> Call Site 5 << .uleb128 .Ltmp6-.Ltmp4 # Call between .Ltmp4 and .Ltmp6 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp6-.Lfunc_begin0 # >> Call Site 6 << .uleb128 .Ltmp7-.Ltmp6 # Call between .Ltmp6 and .Ltmp7 .uleb128 .Ltmp8-.Lfunc_begin0 # jumps to .Ltmp8 .byte 0 # On action: cleanup .uleb128 .Ltmp7-.Lfunc_begin0 # >> Call Site 7 << .uleb128 .Ltmp9-.Ltmp7 # Call between .Ltmp7 and .Ltmp9 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp9-.Lfunc_begin0 # >> Call Site 8 << .uleb128 .Ltmp10-.Ltmp9 # Call between .Ltmp9 and .Ltmp10 .uleb128 .Ltmp11-.Lfunc_begin0 # jumps to .Ltmp11 .byte 0 # On action: cleanup .uleb128 .Ltmp10-.Lfunc_begin0 # >> Call Site 9 << .uleb128 .Lfunc_end2-.Ltmp10 # Call between .Ltmp10 and .Lfunc_end2 .byte 0 # has no landing pad .byte 0 # On action: cleanup .Lcst_end0: .p2align 2, 0x0 # -- End function .text .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB3_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB3_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z6assignPiS_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end3: .size __hip_module_ctor, .Lfunc_end3-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB4_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB4_2: retq .Lfunc_end4: .size __hip_module_dtor, .Lfunc_end4-__hip_module_dtor .cfi_endproc # -- End function .type _Z6assignPiS_i,@object # @_Z6assignPiS_i .section .rodata,"a",@progbits .globl _Z6assignPiS_i .p2align 3, 0x0 _Z6assignPiS_i: .quad _Z21__device_stub__assignPiS_i .size _Z6assignPiS_i, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz " : " .size .L.str, 4 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "after GPU1" .size .L.str.1, 11 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "after GPU2" .size .L.str.2, 11 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "after KERNEL" .size .L.str.3, 13 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "after getData" .size .L.str.4, 14 .type .L.str.5,@object # @.str.5 .L.str.5: .asciz " " .size .L.str.5, 2 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z6assignPiS_i" .size .L__unnamed_1, 15 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z21__device_stub__assignPiS_i .addrsig_sym __gxx_personality_v0 .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Unwind_Resume .addrsig_sym _Z6assignPiS_i .addrsig_sym _ZSt4cerr .addrsig_sym _ZSt4cout .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z6assignPiS_i .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R4, SR_TID.X ; /* 0x0000000000047919 */ /* 0x000e240000002100 */ /*0020*/ ISETP.GE.AND P0, PT, R4, c[0x0][0x170], PT ; /* 0x00005c0004007a0c */ /* 0x001fda0003f06270 */ /*0030*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0040*/ MOV R3, c[0x0][0x164] ; /* 0x0000590000037a02 */ /* 0x000fe20000000f00 */ /*0050*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x160] ; /* 0x00005800ff027624 */ /* 0x000fe200078e00ff */ /*0060*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc80000000a00 */ /*0070*/ LDG.E R3, [R2.64] ; /* 0x0000000402037981 */ /* 0x000ea2000c1e1900 */ /*0080*/ HFMA2.MMA R5, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff057435 */ /* 0x000fd400000001ff */ /*0090*/ IMAD.WIDE R4, R4, R5, c[0x0][0x168] ; /* 0x00005a0004047625 */ /* 0x000fca00078e0205 */ /*00a0*/ STG.E [R4.64], R3 ; /* 0x0000000304007986 */ /* 0x004fe2000c101904 */ /*00b0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*00c0*/ BRA 0xc0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*00d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0100*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0110*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0120*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0130*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0140*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0150*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0160*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0170*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z6assignPiS_i .globl _Z6assignPiS_i .p2align 8 .type _Z6assignPiS_i,@function _Z6assignPiS_i: s_load_b32 s2, s[0:1], 0x10 s_waitcnt lgkmcnt(0) v_cmp_gt_i32_e32 vcc_lo, s2, v0 s_and_saveexec_b32 s2, vcc_lo s_cbranch_execz .LBB0_2 s_load_b128 s[0:3], s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_load_b32 s0, s[0:1], 0x0 s_waitcnt lgkmcnt(0) v_dual_mov_b32 v1, s0 :: v_dual_lshlrev_b32 v0, 2, v0 global_store_b32 v0, v1, s[2:3] .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z6assignPiS_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 20 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 2 .amdhsa_next_free_sgpr 4 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z6assignPiS_i, .Lfunc_end0-_Z6assignPiS_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .offset: 16 .size: 4 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 20 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z6assignPiS_i .private_segment_fixed_size: 0 .sgpr_count: 6 .sgpr_spill_count: 0 .symbol: _Z6assignPiS_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 2 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_000eecf1_00000000-6_test_cuda_peerTopeer.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB12824: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE12824: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string " : " .text .globl _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .type _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE, @function _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE: .LFB12820: .cfi_startproc endbr64 pushq %rbp .cfi_def_cfa_offset 16 .cfi_offset 6, -16 pushq %rbx .cfi_def_cfa_offset 24 .cfi_offset 3, -24 subq $8, %rsp .cfi_def_cfa_offset 32 movq %rdi, %rbp call cudaGetLastError@PLT testl %eax, %eax jne .L11 .L3: addq $8, %rsp .cfi_remember_state .cfi_def_cfa_offset 24 popq %rbx .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 ret .L11: .cfi_restore_state movl %eax, %ebx movq 8(%rbp), %rdx movq 0(%rbp), %rsi leaq _ZSt4cerr(%rip), %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movq %rax, %rbp movl $3, %edx leaq .LC0(%rip), %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %ebx, %edi call cudaGetErrorString@PLT movq %rax, %rbx testq %rax, %rax je .L12 movq %rax, %rdi call strlen@PLT movq %rax, %rdx movq %rbx, %rsi movq %rbp, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT .L6: movq 0(%rbp), %rax movq -24(%rax), %rax movq 240(%rbp,%rax), %rbx testq %rbx, %rbx je .L13 cmpb $0, 56(%rbx) je .L8 movzbl 67(%rbx), %esi .L9: movsbl %sil, %esi movq %rbp, %rdi call _ZNSo3putEc@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT jmp .L3 .L12: movq 0(%rbp), %rax movq %rbp, %rdi addq -24(%rax), %rdi movl 32(%rdi), %esi orl $1, %esi call _ZNSt9basic_iosIcSt11char_traitsIcEE5clearESt12_Ios_Iostate@PLT jmp .L6 .L13: call _ZSt16__throw_bad_castv@PLT .L8: movq %rbx, %rdi call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq (%rbx), %rax movl $10, %esi movq %rbx, %rdi call *48(%rax) movl %eax, %esi jmp .L9 .cfi_endproc .LFE12820: .size _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE, .-_Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .globl _Z28__device_stub__Z6assignPiS_iPiS_i .type _Z28__device_stub__Z6assignPiS_iPiS_i, @function _Z28__device_stub__Z6assignPiS_iPiS_i: .LFB12846: .cfi_startproc endbr64 subq $136, %rsp .cfi_def_cfa_offset 144 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movl %edx, 12(%rsp) movq %fs:40, %rax movq %rax, 120(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 12(%rsp), %rax movq %rax, 112(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L18 .L14: movq 120(%rsp), %rax subq %fs:40, %rax jne .L19 addq $136, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L18: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z6assignPiS_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L14 .L19: call __stack_chk_fail@PLT .cfi_endproc .LFE12846: .size _Z28__device_stub__Z6assignPiS_iPiS_i, .-_Z28__device_stub__Z6assignPiS_iPiS_i .globl _Z6assignPiS_i .type _Z6assignPiS_i, @function _Z6assignPiS_i: .LFB12847: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z28__device_stub__Z6assignPiS_iPiS_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE12847: .size _Z6assignPiS_i, .-_Z6assignPiS_i .section .rodata.str1.1 .LC1: .string "_Z6assignPiS_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB12849: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC1(%rip), %rdx movq %rdx, %rcx leaq _Z6assignPiS_i(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE12849: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .rodata._ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_.str1.8,"aMS",@progbits,1 .align 8 .LC2: .string "basic_string: construction from null is not valid" .section .text._ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_,"axG",@progbits,_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC5IS3_EEPKcRKS3_,comdat .align 2 .weak _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_ .type _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_, @function _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_: .LFB13261: .cfi_startproc endbr64 pushq %r13 .cfi_def_cfa_offset 16 .cfi_offset 13, -16 pushq %r12 .cfi_def_cfa_offset 24 .cfi_offset 12, -24 pushq %rbp .cfi_def_cfa_offset 32 .cfi_offset 6, -32 pushq %rbx .cfi_def_cfa_offset 40 .cfi_offset 3, -40 subq $24, %rsp .cfi_def_cfa_offset 64 movq %fs:40, %rax movq %rax, 8(%rsp) xorl %eax, %eax leaq 16(%rdi), %r12 movq %r12, (%rdi) testq %rsi, %rsi je .L33 movq %rdi, %rbx movq %rsi, %r13 movq %rsi, %rdi call strlen@PLT movq %rax, %rbp movq %rax, (%rsp) cmpq $15, %rax ja .L34 cmpq $1, %rax jne .L29 movzbl 0(%r13), %eax movb %al, 16(%rbx) .L30: movq (%rsp), %rax movq %rax, 8(%rbx) movq (%rbx), %rdx movb $0, (%rdx,%rax) movq 8(%rsp), %rax subq %fs:40, %rax jne .L35 addq $24, %rsp .cfi_remember_state .cfi_def_cfa_offset 40 popq %rbx .cfi_def_cfa_offset 32 popq %rbp .cfi_def_cfa_offset 24 popq %r12 .cfi_def_cfa_offset 16 popq %r13 .cfi_def_cfa_offset 8 ret .L33: .cfi_restore_state movq 8(%rsp), %rax subq %fs:40, %rax jne .L36 leaq .LC2(%rip), %rdi call _ZSt19__throw_logic_errorPKc@PLT .L36: call __stack_chk_fail@PLT .L34: movq %rsp, %rsi movl $0, %edx movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE9_M_createERmm@PLT movq %rax, %r12 movq %rax, (%rbx) movq (%rsp), %rax movq %rax, 16(%rbx) .L28: movq %rbp, %rdx movq %r13, %rsi movq %r12, %rdi call memcpy@PLT jmp .L30 .L29: testq %rax, %rax je .L30 jmp .L28 .L35: call __stack_chk_fail@PLT .cfi_endproc .LFE13261: .size _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_, .-_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_ .weak _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .set _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_,_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_ .section .rodata.str1.1 .LC3: .string "after GPU1" .LC4: .string "after GPU2" .LC5: .string "after KERNEL" .LC6: .string "after getData" .LC7: .string " " .text .globl main .type main, @function main: .LFB12821: .cfi_startproc .cfi_personality 0x9b,DW.ref.__gxx_personality_v0 .cfi_lsda 0x1b,.LLSDA12821 endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $104, %rsp .cfi_def_cfa_offset 160 movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax movl $0, %edi .LEHB0: call cudaSetDevice@PLT movl $4, %edi call malloc@PLT movq %rax, %rbx movl $15, (%rax) leaq 8(%rsp), %rdi movl $4, %esi call cudaMalloc@PLT movl $1, %ecx movl $4, %edx movq %rbx, %rsi movq 8(%rsp), %rdi call cudaMemcpy@PLT leaq 36(%rsp), %rdx leaq 48(%rsp), %rbx leaq .LC3(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE0: movq %rbx, %rdi .LEHB1: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE1: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $0, %esi movl $0, %edi .LEHB2: call cudaDeviceEnablePeerAccess@PLT movl $1, %edi call cudaSetDevice@PLT leaq 16(%rsp), %rdi movl $800, %esi call cudaMalloc@PLT movl $0, %esi movl $0, %edi call cudaDeviceEnablePeerAccess@PLT leaq 36(%rsp), %rdx leaq .LC4(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE2: movq %rbx, %rdi .LEHB3: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE3: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $512, 36(%rsp) movl $1, 40(%rsp) movl $1, 44(%rsp) movl $1, 24(%rsp) movl $1, 28(%rsp) movl $1, 32(%rsp) movl $0, %r9d movl $0, %r8d movq 36(%rsp), %rdx movl $1, %ecx movq 24(%rsp), %rdi movl $1, %esi .LEHB4: call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L60 .L38: leaq 36(%rsp), %rdx leaq 48(%rsp), %rbx leaq .LC5(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE4: movq %rbx, %rdi .LEHB5: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE5: jmp .L61 .L60: movl $200, %edx movq 16(%rsp), %rsi movq 8(%rsp), %rdi .LEHB6: call _Z28__device_stub__Z6assignPiS_iPiS_i jmp .L38 .L61: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $800, %edi call malloc@PLT movq %rax, %r12 movl $2, %ecx movl $800, %edx movq 16(%rsp), %rsi movq %rax, %rdi call cudaMemcpy@PLT leaq 36(%rsp), %rdx leaq .LC6(%rip), %rsi movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC1IS3_EEPKcRKS3_ .LEHE6: movq %rbx, %rdi .LEHB7: call _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .LEHE7: movq %rbx, %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movl $0, %ebp leaq _ZSt4cout(%rip), %r14 leaq .LC7(%rip), %r13 jmp .L43 .L64: movq 88(%rsp), %rax subq %fs:40, %rax jne .L62 .LEHB8: call _ZSt16__throw_bad_castv@PLT .L62: call __stack_chk_fail@PLT .L65: movzbl 67(%r15), %esi .L42: movsbl %sil, %esi movq %rbx, %rdi call _ZNSo3putEc@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT addq $1, %rbp cmpq $200, %rbp je .L63 .L43: movl %ebp, %esi movq %r14, %rdi call _ZNSolsEi@PLT movq %rax, %rbx movl $1, %edx movq %r13, %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl (%r12,%rbp,4), %esi movq %rbx, %rdi call _ZNSolsEi@PLT movq %rax, %rbx movq (%rax), %rax movq -24(%rax), %rax movq 240(%rbx,%rax), %r15 testq %r15, %r15 je .L64 cmpb $0, 56(%r15) jne .L65 movq %r15, %rdi call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq (%r15), %rax movl $10, %esi movq %r15, %rdi call *48(%rax) movl %eax, %esi jmp .L42 .L63: movq 8(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call cudaFree@PLT movq 88(%rsp), %rax subq %fs:40, %rax jne .L66 movl $0, %eax addq $104, %rsp .cfi_remember_state .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .L53: .cfi_restore_state endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L45 call __stack_chk_fail@PLT .L45: movq %rbx, %rdi call _Unwind_Resume@PLT .L54: endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L47 call __stack_chk_fail@PLT .L47: movq %rbx, %rdi call _Unwind_Resume@PLT .L55: endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L49 call __stack_chk_fail@PLT .L49: movq %rbx, %rdi call _Unwind_Resume@PLT .L56: endbr64 movq %rax, %rbx leaq 48(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 88(%rsp), %rax subq %fs:40, %rax je .L51 call __stack_chk_fail@PLT .L51: movq %rbx, %rdi call _Unwind_Resume@PLT .LEHE8: .L66: call __stack_chk_fail@PLT .cfi_endproc .LFE12821: .globl __gxx_personality_v0 .section .gcc_except_table,"a",@progbits .LLSDA12821: .byte 0xff .byte 0xff .byte 0x1 .uleb128 .LLSDACSE12821-.LLSDACSB12821 .LLSDACSB12821: .uleb128 .LEHB0-.LFB12821 .uleb128 .LEHE0-.LEHB0 .uleb128 0 .uleb128 0 .uleb128 .LEHB1-.LFB12821 .uleb128 .LEHE1-.LEHB1 .uleb128 .L53-.LFB12821 .uleb128 0 .uleb128 .LEHB2-.LFB12821 .uleb128 .LEHE2-.LEHB2 .uleb128 0 .uleb128 0 .uleb128 .LEHB3-.LFB12821 .uleb128 .LEHE3-.LEHB3 .uleb128 .L54-.LFB12821 .uleb128 0 .uleb128 .LEHB4-.LFB12821 .uleb128 .LEHE4-.LEHB4 .uleb128 0 .uleb128 0 .uleb128 .LEHB5-.LFB12821 .uleb128 .LEHE5-.LEHB5 .uleb128 .L55-.LFB12821 .uleb128 0 .uleb128 .LEHB6-.LFB12821 .uleb128 .LEHE6-.LEHB6 .uleb128 0 .uleb128 0 .uleb128 .LEHB7-.LFB12821 .uleb128 .LEHE7-.LEHB7 .uleb128 .L56-.LFB12821 .uleb128 0 .uleb128 .LEHB8-.LFB12821 .uleb128 .LEHE8-.LEHB8 .uleb128 0 .uleb128 0 .LLSDACSE12821: .text .size main, .-main .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .hidden DW.ref.__gxx_personality_v0 .weak DW.ref.__gxx_personality_v0 .section .data.rel.local.DW.ref.__gxx_personality_v0,"awG",@progbits,DW.ref.__gxx_personality_v0,comdat .align 8 .type DW.ref.__gxx_personality_v0, @object .size DW.ref.__gxx_personality_v0, 8 DW.ref.__gxx_personality_v0: .quad __gxx_personality_v0 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "test_cuda_peerTopeer.hip" # Start of file scope inline assembly .globl _ZSt21ios_base_library_initv # End of file scope inline assembly .globl _Z21__device_stub__assignPiS_i # -- Begin function _Z21__device_stub__assignPiS_i .p2align 4, 0x90 .type _Z21__device_stub__assignPiS_i,@function _Z21__device_stub__assignPiS_i: # @_Z21__device_stub__assignPiS_i .cfi_startproc # %bb.0: subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movl %edx, 12(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 12(%rsp), %rax movq %rax, 96(%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z6assignPiS_i, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $120, %rsp .cfi_adjust_cfa_offset -120 retq .Lfunc_end0: .size _Z21__device_stub__assignPiS_i, .Lfunc_end0-_Z21__device_stub__assignPiS_i .cfi_endproc # -- End function .globl _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE # -- Begin function _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .p2align 4, 0x90 .type _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE,@function _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE: # @_Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r14 .cfi_def_cfa_offset 24 pushq %rbx .cfi_def_cfa_offset 32 .cfi_offset %rbx, -32 .cfi_offset %r14, -24 .cfi_offset %rbp, -16 movq %rdi, %rbx callq hipGetLastError testl %eax, %eax je .LBB1_9 # %bb.1: movl %eax, %ebp movq (%rbx), %rsi movq 8(%rbx), %rdx movl $_ZSt4cerr, %edi callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l movq %rax, %rbx movl $.L.str, %esi movl $3, %edx movq %rax, %rdi callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l movl %ebp, %edi callq hipGetErrorString testq %rax, %rax je .LBB1_2 # %bb.3: movq %rax, %rdi movq %rax, %r14 callq strlen movq %rbx, %rdi movq %r14, %rsi movq %rax, %rdx callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l jmp .LBB1_4 .LBB1_9: popq %rbx .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .LBB1_2: .cfi_def_cfa_offset 32 movq (%rbx), %rax movq -24(%rax), %rax movq %rbx, %rdi addq %rax, %rdi movl 32(%rbx,%rax), %esi orl $1, %esi callq _ZNSt9basic_iosIcSt11char_traitsIcEE5clearESt12_Ios_Iostate .LBB1_4: # %_ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_PKc.exit movq (%rbx), %rax movq -24(%rax), %rax movq 240(%rbx,%rax), %r14 testq %r14, %r14 je .LBB1_10 # %bb.5: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i cmpb $0, 56(%r14) je .LBB1_7 # %bb.6: movzbl 67(%r14), %eax jmp .LBB1_8 .LBB1_7: movq %r14, %rdi callq _ZNKSt5ctypeIcE13_M_widen_initEv movq (%r14), %rax movq %r14, %rdi movl $10, %esi callq *48(%rax) .LBB1_8: # %_ZSt4endlIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_.exit movsbl %al, %esi movq %rbx, %rdi callq _ZNSo3putEc movq %rax, %rdi popq %rbx .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 jmp _ZNSo5flushEv # TAILCALL .LBB1_10: .cfi_def_cfa_offset 32 callq _ZSt16__throw_bad_castv .Lfunc_end1: .size _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE, .Lfunc_end1-_Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .Lfunc_begin0: .cfi_startproc .cfi_personality 3, __gxx_personality_v0 .cfi_lsda 3, .Lexception0 # %bb.0: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_.exit pushq %r15 .cfi_def_cfa_offset 16 pushq %r14 .cfi_def_cfa_offset 24 pushq %r12 .cfi_def_cfa_offset 32 pushq %rbx .cfi_def_cfa_offset 40 subq $248, %rsp .cfi_def_cfa_offset 288 .cfi_offset %rbx, -40 .cfi_offset %r12, -32 .cfi_offset %r14, -24 .cfi_offset %r15, -16 .cfi_escape 0x2e, 0x00 xorl %edi, %edi callq hipSetDevice .cfi_escape 0x2e, 0x00 movl $4, %edi callq malloc movq %rax, %rbx movl $15, (%rax) .cfi_escape 0x2e, 0x00 leaq 16(%rsp), %rdi movl $4, %esi callq hipMalloc movq 16(%rsp), %rdi .cfi_escape 0x2e, 0x00 movl $4, %edx movq %rbx, %rsi movl $1, %ecx callq hipMemcpy leaq 144(%rsp), %r14 movq %r14, 128(%rsp) movabsq $5784627922081179233, %rbx # imm = 0x5047207265746661 movq %rbx, 144(%rsp) movw $12629, 152(%rsp) # imm = 0x3155 movq $10, 136(%rsp) movb $0, 154(%rsp) .Ltmp0: .cfi_escape 0x2e, 0x00 leaq 128(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp1: # %bb.1: movq 128(%rsp), %rdi cmpq %r14, %rdi je .LBB2_3 # %bb.2: # %.critedge.i.i .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_3: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit .cfi_escape 0x2e, 0x00 xorl %edi, %edi xorl %esi, %esi callq hipDeviceEnablePeerAccess .cfi_escape 0x2e, 0x00 movl $1, %edi callq hipSetDevice .cfi_escape 0x2e, 0x00 leaq 8(%rsp), %rdi movl $800, %esi # imm = 0x320 callq hipMalloc .cfi_escape 0x2e, 0x00 xorl %edi, %edi xorl %esi, %esi callq hipDeviceEnablePeerAccess leaq 112(%rsp), %r14 movq %r14, 96(%rsp) movq %rbx, 112(%rsp) movw $12885, 120(%rsp) # imm = 0x3255 movq $10, 104(%rsp) movb $0, 122(%rsp) .Ltmp3: .cfi_escape 0x2e, 0x00 leaq 96(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp4: # %bb.4: movq 96(%rsp), %rdi cmpq %r14, %rdi je .LBB2_6 # %bb.5: # %.critedge.i.i43 .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_6: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit45 movabsq $4294967297, %rdi # imm = 0x100000001 leaq 511(%rdi), %rdx .cfi_escape 0x2e, 0x00 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB2_8 # %bb.7: movq 16(%rsp), %rax movq 8(%rsp), %rcx movq %rax, 216(%rsp) movq %rcx, 208(%rsp) movl $200, 28(%rsp) leaq 216(%rsp), %rax movq %rax, 224(%rsp) leaq 208(%rsp), %rax movq %rax, 232(%rsp) leaq 28(%rsp), %rax movq %rax, 240(%rsp) .cfi_escape 0x2e, 0x00 leaq 192(%rsp), %rdi leaq 176(%rsp), %rsi leaq 168(%rsp), %rdx leaq 160(%rsp), %rcx callq __hipPopCallConfiguration movq 192(%rsp), %rsi movl 200(%rsp), %edx movq 176(%rsp), %rcx movl 184(%rsp), %r8d .cfi_escape 0x2e, 0x10 leaq 224(%rsp), %r9 movl $_Z6assignPiS_i, %edi pushq 160(%rsp) .cfi_adjust_cfa_offset 8 pushq 176(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB2_8: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEC2IS3_EEPKcRKS3_.exit58 leaq 80(%rsp), %r14 movq %r14, 64(%rsp) movabsq $4993120287570814561, %rax # imm = 0x454B207265746661 movq %rax, 80(%rsp) movl $1279610450, 88(%rsp) # imm = 0x4C454E52 movq $12, 72(%rsp) movb $0, 92(%rsp) .Ltmp6: .cfi_escape 0x2e, 0x00 leaq 64(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp7: # %bb.9: movq 64(%rsp), %rdi cmpq %r14, %rdi je .LBB2_11 # %bb.10: # %.critedge.i.i59 .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_11: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit61 .cfi_escape 0x2e, 0x00 movl $800, %edi # imm = 0x320 callq malloc movq %rax, %rbx movq 8(%rsp), %rsi .cfi_escape 0x2e, 0x00 movl $800, %edx # imm = 0x320 movq %rax, %rdi movl $2, %ecx callq hipMemcpy leaq 48(%rsp), %r14 movq %r14, 32(%rsp) movabsq $7306844596132406881, %rax # imm = 0x6567207265746661 movq %rax, 48(%rsp) movabsq $7022344665615918880, %rax # imm = 0x6174614474656720 movq %rax, 53(%rsp) movq $13, 40(%rsp) movb $0, 61(%rsp) .Ltmp9: .cfi_escape 0x2e, 0x00 leaq 32(%rsp), %rdi callq _Z14cudaPrintErrorNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE .Ltmp10: # %bb.12: movq 32(%rsp), %rdi cmpq %r14, %rdi je .LBB2_14 # %bb.13: # %.critedge.i.i69 .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB2_14: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit71.preheader xorl %r14d, %r14d jmp .LBB2_15 .p2align 4, 0x90 .LBB2_29: # in Loop: Header=BB2_15 Depth=1 .cfi_escape 0x2e, 0x00 movq %r15, %rdi movq %rax, %r12 callq _ZNKSt5ctypeIcE13_M_widen_initEv movq (%r15), %rax .cfi_escape 0x2e, 0x00 movq %r15, %rdi movl $10, %esi callq *48(%rax) movl %eax, %ecx movq %r12, %rax .LBB2_30: # %_ZSt4endlIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_.exit # in Loop: Header=BB2_15 Depth=1 .cfi_escape 0x2e, 0x00 movsbl %cl, %esi movq %rax, %rdi callq _ZNSo3putEc .cfi_escape 0x2e, 0x00 movq %rax, %rdi callq _ZNSo5flushEv incq %r14 cmpq $200, %r14 je .LBB2_31 .LBB2_15: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit71 # =>This Inner Loop Header: Depth=1 .cfi_escape 0x2e, 0x00 movl $_ZSt4cout, %edi movl %r14d, %esi callq _ZNSolsEi movq %rax, %r15 .cfi_escape 0x2e, 0x00 movl $.L.str.5, %esi movl $1, %edx movq %rax, %rdi callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l movl (%rbx,%r14,4), %esi .cfi_escape 0x2e, 0x00 movq %r15, %rdi callq _ZNSolsEi movq (%rax), %rcx movq -24(%rcx), %rcx movq 240(%rax,%rcx), %r15 testq %r15, %r15 je .LBB2_16 # %bb.27: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i # in Loop: Header=BB2_15 Depth=1 cmpb $0, 56(%r15) je .LBB2_29 # %bb.28: # in Loop: Header=BB2_15 Depth=1 movzbl 67(%r15), %ecx jmp .LBB2_30 .LBB2_31: movq 16(%rsp), %rdi .cfi_escape 0x2e, 0x00 callq hipFree movq 8(%rsp), %rdi .cfi_escape 0x2e, 0x00 callq hipFree xorl %eax, %eax addq $248, %rsp .cfi_def_cfa_offset 40 popq %rbx .cfi_def_cfa_offset 32 popq %r12 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 retq .LBB2_16: .cfi_def_cfa_offset 288 .cfi_escape 0x2e, 0x00 callq _ZSt16__throw_bad_castv .LBB2_25: .Ltmp11: movq %rax, %rbx movq 32(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.26: # %.critedge.i.i75 .cfi_escape 0x2e, 0x00 jmp .LBB2_19 .LBB2_23: .Ltmp8: movq %rax, %rbx movq 64(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.24: # %.critedge.i.i72 .cfi_escape 0x2e, 0x00 jmp .LBB2_19 .LBB2_21: .Ltmp5: movq %rax, %rbx movq 96(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.22: # %.critedge.i.i49 .cfi_escape 0x2e, 0x00 jmp .LBB2_19 .LBB2_17: .Ltmp2: movq %rax, %rbx movq 128(%rsp), %rdi cmpq %r14, %rdi je .LBB2_20 # %bb.18: # %.critedge.i.i46 .cfi_escape 0x2e, 0x00 .LBB2_19: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit48 callq _ZdlPv .LBB2_20: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit48 .cfi_escape 0x2e, 0x00 movq %rbx, %rdi callq _Unwind_Resume@PLT .Lfunc_end2: .size main, .Lfunc_end2-main .cfi_endproc .section .gcc_except_table,"a",@progbits .p2align 2, 0x0 GCC_except_table2: .Lexception0: .byte 255 # @LPStart Encoding = omit .byte 255 # @TType Encoding = omit .byte 1 # Call site Encoding = uleb128 .uleb128 .Lcst_end0-.Lcst_begin0 .Lcst_begin0: .uleb128 .Lfunc_begin0-.Lfunc_begin0 # >> Call Site 1 << .uleb128 .Ltmp0-.Lfunc_begin0 # Call between .Lfunc_begin0 and .Ltmp0 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp0-.Lfunc_begin0 # >> Call Site 2 << .uleb128 .Ltmp1-.Ltmp0 # Call between .Ltmp0 and .Ltmp1 .uleb128 .Ltmp2-.Lfunc_begin0 # jumps to .Ltmp2 .byte 0 # On action: cleanup .uleb128 .Ltmp1-.Lfunc_begin0 # >> Call Site 3 << .uleb128 .Ltmp3-.Ltmp1 # Call between .Ltmp1 and .Ltmp3 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp3-.Lfunc_begin0 # >> Call Site 4 << .uleb128 .Ltmp4-.Ltmp3 # Call between .Ltmp3 and .Ltmp4 .uleb128 .Ltmp5-.Lfunc_begin0 # jumps to .Ltmp5 .byte 0 # On action: cleanup .uleb128 .Ltmp4-.Lfunc_begin0 # >> Call Site 5 << .uleb128 .Ltmp6-.Ltmp4 # Call between .Ltmp4 and .Ltmp6 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp6-.Lfunc_begin0 # >> Call Site 6 << .uleb128 .Ltmp7-.Ltmp6 # Call between .Ltmp6 and .Ltmp7 .uleb128 .Ltmp8-.Lfunc_begin0 # jumps to .Ltmp8 .byte 0 # On action: cleanup .uleb128 .Ltmp7-.Lfunc_begin0 # >> Call Site 7 << .uleb128 .Ltmp9-.Ltmp7 # Call between .Ltmp7 and .Ltmp9 .byte 0 # has no landing pad .byte 0 # On action: cleanup .uleb128 .Ltmp9-.Lfunc_begin0 # >> Call Site 8 << .uleb128 .Ltmp10-.Ltmp9 # Call between .Ltmp9 and .Ltmp10 .uleb128 .Ltmp11-.Lfunc_begin0 # jumps to .Ltmp11 .byte 0 # On action: cleanup .uleb128 .Ltmp10-.Lfunc_begin0 # >> Call Site 9 << .uleb128 .Lfunc_end2-.Ltmp10 # Call between .Ltmp10 and .Lfunc_end2 .byte 0 # has no landing pad .byte 0 # On action: cleanup .Lcst_end0: .p2align 2, 0x0 # -- End function .text .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB3_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB3_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z6assignPiS_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end3: .size __hip_module_ctor, .Lfunc_end3-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB4_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB4_2: retq .Lfunc_end4: .size __hip_module_dtor, .Lfunc_end4-__hip_module_dtor .cfi_endproc # -- End function .type _Z6assignPiS_i,@object # @_Z6assignPiS_i .section .rodata,"a",@progbits .globl _Z6assignPiS_i .p2align 3, 0x0 _Z6assignPiS_i: .quad _Z21__device_stub__assignPiS_i .size _Z6assignPiS_i, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz " : " .size .L.str, 4 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "after GPU1" .size .L.str.1, 11 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "after GPU2" .size .L.str.2, 11 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "after KERNEL" .size .L.str.3, 13 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "after getData" .size .L.str.4, 14 .type .L.str.5,@object # @.str.5 .L.str.5: .asciz " " .size .L.str.5, 2 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z6assignPiS_i" .size .L__unnamed_1, 15 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z21__device_stub__assignPiS_i .addrsig_sym __gxx_personality_v0 .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Unwind_Resume .addrsig_sym _Z6assignPiS_i .addrsig_sym _ZSt4cerr .addrsig_sym _ZSt4cout .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include "includes.h" __global__ void sgemm_kernel_B(const float *A, const float *B, float *C, int N, int M, int K, float alpha, float beta) { int col = blockIdx.x * blockDim.x + threadIdx.x; int row = blockIdx.y * blockDim.y + threadIdx.y; float sum = 0.f; for (int i = 0; i < K; ++i) sum += A[row * K + i] * B[i * K + col]; C[row * M + col] = alpha * sum + beta * C[row * M + col]; }
code for sm_80 Function : _Z14sgemm_kernel_BPKfS0_Pfiiiff .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ MOV R0, c[0x0][0x180] ; /* 0x0000600000007a02 */ /* 0x000fe20000000f00 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ HFMA2.MMA R24, -RZ, RZ, 0, 0 ; /* 0x00000000ff187435 */ /* 0x000fe200000001ff */ /*0050*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e220000002100 */ /*0060*/ ISETP.GE.AND P0, PT, R0, 0x1, PT ; /* 0x000000010000780c */ /* 0x000fc60003f06270 */ /*0070*/ S2R R4, SR_CTAID.Y ; /* 0x0000000000047919 */ /* 0x000e680000002600 */ /*0080*/ S2R R5, SR_TID.Y ; /* 0x0000000000057919 */ /* 0x000e620000002200 */ /*0090*/ IMAD R2, R2, c[0x0][0x0], R3 ; /* 0x0000000002027a24 */ /* 0x001fe400078e0203 */ /*00a0*/ IMAD R3, R4, c[0x0][0x4], R5 ; /* 0x0000010004037a24 */ /* 0x002fc600078e0205 */ /*00b0*/ @!P0 BRA 0xbd0 ; /* 0x00000b1000008947 */ /* 0x000fea0003800000 */ /*00c0*/ IADD3 R4, R0.reuse, -0x1, RZ ; /* 0xffffffff00047810 */ /* 0x040fe40007ffe0ff */ /*00d0*/ LOP3.LUT R5, R0, 0x3, RZ, 0xc0, !PT ; /* 0x0000000300057812 */ /* 0x000fe400078ec0ff */ /*00e0*/ ISETP.GE.U32.AND P0, PT, R4, 0x3, PT ; /* 0x000000030400780c */ /* 0x000fe40003f06070 */ /*00f0*/ MOV R24, RZ ; /* 0x000000ff00187202 */ /* 0x000fe40000000f00 */ /*0100*/ MOV R4, RZ ; /* 0x000000ff00047202 */ /* 0x000fd20000000f00 */ /*0110*/ @!P0 BRA 0xad0 ; /* 0x000009b000008947 */ /* 0x000fea0003800000 */ /*0120*/ IADD3 R6, -R5, c[0x0][0x180], RZ ; /* 0x0000600005067a10 */ /* 0x000fe20007ffe1ff */ /*0130*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0140*/ ULDC.64 UR6, c[0x0][0x160] ; /* 0x0000580000067ab9 */ /* 0x000fe20000000a00 */ /*0150*/ HFMA2.MMA R4, -RZ, RZ, 0, 0 ; /* 0x00000000ff047435 */ /* 0x000fe200000001ff */ /*0160*/ ISETP.GT.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f04270 */ /*0170*/ IMAD R7, R3, c[0x0][0x180], RZ ; /* 0x0000600003077a24 */ /* 0x000fe200078e02ff */ /*0180*/ MOV R24, RZ ; /* 0x000000ff00187202 */ /* 0x000fca0000000f00 */ /*0190*/ IMAD.WIDE R8, R2, R9, c[0x0][0x168] ; /* 0x00005a0002087625 */ /* 0x000fcc00078e0209 */ /*01a0*/ @!P0 BRA 0x940 ; /* 0x0000079000008947 */ /* 0x000fea0003800000 */ /*01b0*/ ISETP.GT.AND P1, PT, R6, 0xc, PT ; /* 0x0000000c0600780c */ /* 0x000fe40003f24270 */ /*01c0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fd60003f0f070 */ /*01d0*/ @!P1 BRA 0x680 ; /* 0x000004a000009947 */ /* 0x000fea0003800000 */ /*01e0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*01f0*/ MOV R14, UR6 ; /* 0x00000006000e7c02 */ /* 0x000fe20008000f00 */ /*0200*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x0000a2000c1e1900 */ /*0210*/ MOV R15, UR7 ; /* 0x00000007000f7c02 */ /* 0x000fca0008000f00 */ /*0220*/ IMAD.WIDE R14, R7, 0x4, R14 ; /* 0x00000004070e7825 */ /* 0x000fca00078e020e */ /*0230*/ LDG.E R10, [R14.64] ; /* 0x000000040e0a7981 */ /* 0x000ea2000c1e1900 */ /*0240*/ IMAD.WIDE R8, R0, 0x4, R8 ; /* 0x0000000400087825 */ /* 0x001fc600078e0208 */ /*0250*/ LDG.E R18, [R14.64+0x4] ; /* 0x000004040e127981 */ /* 0x000ee6000c1e1900 */ /*0260*/ IMAD.WIDE R22, R0.reuse, 0x4, R8 ; /* 0x0000000400167825 */ /* 0x040fe200078e0208 */ /*0270*/ LDG.E R19, [R8.64] ; /* 0x0000000408137981 */ /* 0x0000e8000c1e1900 */ /*0280*/ LDG.E R28, [R22.64] ; /* 0x00000004161c7981 */ /* 0x000322000c1e1900 */ /*0290*/ IMAD.WIDE R26, R0, 0x4, R22 ; /* 0x00000004001a7825 */ /* 0x000fc600078e0216 */ /*02a0*/ LDG.E R29, [R14.64+0x8] ; /* 0x000008040e1d7981 */ /* 0x000f26000c1e1900 */ /*02b0*/ IMAD.WIDE R12, R0.reuse, 0x4, R26 ; /* 0x00000004000c7825 */ /* 0x040fe200078e021a */ /*02c0*/ LDG.E R16, [R26.64] ; /* 0x000000041a107981 */ /* 0x000b28000c1e1900 */ /*02d0*/ LDG.E R17, [R14.64+0xc] ; /* 0x00000c040e117981 */ /* 0x000f28000c1e1900 */ /*02e0*/ LDG.E R20, [R14.64+0x10] ; /* 0x000010040e147981 */ /* 0x000f28000c1e1900 */ /*02f0*/ LDG.E R21, [R12.64] ; /* 0x000000040c157981 */ /* 0x000328000c1e1900 */ /*0300*/ LDG.E R8, [R14.64+0x14] ; /* 0x000014040e087981 */ /* 0x001f28000c1e1900 */ /*0310*/ LDG.E R26, [R14.64+0x1c] ; /* 0x00001c040e1a7981 */ /* 0x020f62000c1e1900 */ /*0320*/ IMAD.WIDE R12, R0, 0x4, R12 ; /* 0x00000004000c7825 */ /* 0x002fca00078e020c */ /*0330*/ LDG.E R9, [R12.64] ; /* 0x000000040c097981 */ /* 0x000562000c1e1900 */ /*0340*/ IMAD.WIDE R22, R0, 0x4, R12 ; /* 0x0000000400167825 */ /* 0x000fc800078e020c */ /*0350*/ FFMA R12, R11, R10, R24 ; /* 0x0000000a0b0c7223 */ /* 0x004fe40000000018 */ /*0360*/ LDG.E R10, [R14.64+0x18] ; /* 0x000018040e0a7981 */ /* 0x000ea2000c1e1900 */ /*0370*/ IMAD.WIDE R24, R0, 0x4, R22 ; /* 0x0000000400187825 */ /* 0x000fc600078e0216 */ /*0380*/ LDG.E R11, [R22.64] ; /* 0x00000004160b7981 */ /* 0x0000a8000c1e1900 */ /*0390*/ LDG.E R27, [R24.64] ; /* 0x00000004181b7981 */ /* 0x0002a2000c1e1900 */ /*03a0*/ FFMA R12, R19, R18, R12 ; /* 0x00000012130c7223 */ /* 0x008fe4000000000c */ /*03b0*/ IMAD.WIDE R18, R0, 0x4, R24 ; /* 0x0000000400127825 */ /* 0x000fe200078e0218 */ /*03c0*/ LDG.E R23, [R14.64+0x20] ; /* 0x000020040e177981 */ /* 0x001ee6000c1e1900 */ /*03d0*/ FFMA R28, R28, R29, R12 ; /* 0x0000001d1c1c7223 */ /* 0x010fc4000000000c */ /*03e0*/ IMAD.WIDE R12, R0, 0x4, R18 ; /* 0x00000004000c7825 */ /* 0x000fe200078e0212 */ /*03f0*/ LDG.E R25, [R14.64+0x24] ; /* 0x000024040e197981 */ /* 0x002f26000c1e1900 */ /*0400*/ FFMA R28, R16, R17, R28 ; /* 0x00000011101c7223 */ /* 0x000fe2000000001c */ /*0410*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x0000e2000c1e1900 */ /*0420*/ IMAD.WIDE R16, R0, 0x4, R12 ; /* 0x0000000400107825 */ /* 0x000fc600078e020c */ /*0430*/ LDG.E R12, [R12.64] ; /* 0x000000040c0c7981 */ /* 0x000322000c1e1900 */ /*0440*/ FFMA R28, R21, R20, R28 ; /* 0x00000014151c7223 */ /* 0x000fc6000000001c */ /*0450*/ LDG.E R22, [R16.64] ; /* 0x0000000410167981 */ /* 0x0002e2000c1e1900 */ /*0460*/ IMAD.WIDE R20, R0, 0x4, R16 ; /* 0x0000000400147825 */ /* 0x000fc600078e0210 */ /*0470*/ LDG.E R29, [R14.64+0x28] ; /* 0x000028040e1d7981 */ /* 0x000f28000c1e1900 */ /*0480*/ LDG.E R19, [R14.64+0x2c] ; /* 0x00002c040e137981 */ /* 0x001f28000c1e1900 */ /*0490*/ LDG.E R24, [R14.64+0x30] ; /* 0x000030040e187981 */ /* 0x000f22000c1e1900 */ /*04a0*/ FFMA R28, R9, R8, R28 ; /* 0x00000008091c7223 */ /* 0x020fe4000000001c */ /*04b0*/ IMAD.WIDE R8, R0, 0x4, R20 ; /* 0x0000000400087825 */ /* 0x000fc400078e0214 */ /*04c0*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x000168000c1e1900 */ /*04d0*/ LDG.E R21, [R14.64+0x38] ; /* 0x000038040e157981 */ /* 0x001f62000c1e1900 */ /*04e0*/ FFMA R28, R11, R10, R28 ; /* 0x0000000a0b1c7223 */ /* 0x004fe4000000001c */ /*04f0*/ IMAD.WIDE R10, R0, 0x4, R8 ; /* 0x00000004000a7825 */ /* 0x000fe400078e0208 */ /*0500*/ LDG.E R8, [R8.64] ; /* 0x0000000408087981 */ /* 0x0000a4000c1e1900 */ /*0510*/ FFMA R13, R27, R26, R28 ; /* 0x0000001a1b0d7223 */ /* 0x002fc4000000001c */ /*0520*/ IMAD.WIDE R26, R0.reuse, 0x4, R10 ; /* 0x00000004001a7825 */ /* 0x040fe400078e020a */ /*0530*/ LDG.E R10, [R10.64] ; /* 0x000000040a0a7981 */ /* 0x0002a8000c1e1900 */ /*0540*/ LDG.E R9, [R14.64+0x34] ; /* 0x000034040e097981 */ /* 0x001ea2000c1e1900 */ /*0550*/ IMAD.WIDE R16, R0, 0x4, R26 ; /* 0x0000000400107825 */ /* 0x000fc600078e021a */ /*0560*/ LDG.E R28, [R26.64] ; /* 0x000000041a1c7981 */ /* 0x0000a8000c1e1900 */ /*0570*/ LDG.E R11, [R16.64] ; /* 0x00000004100b7981 */ /* 0x002ea8000c1e1900 */ /*0580*/ LDG.E R26, [R14.64+0x3c] ; /* 0x00003c040e1a7981 */ /* 0x001ea2000c1e1900 */ /*0590*/ FFMA R13, R18, R23, R13 ; /* 0x00000017120d7223 */ /* 0x008fc8000000000d */ /*05a0*/ FFMA R12, R12, R25, R13 ; /* 0x000000190c0c7223 */ /* 0x010fe2000000000d */ /*05b0*/ IADD3 R6, R6, -0x10, RZ ; /* 0xfffffff006067810 */ /* 0x000fc60007ffe0ff */ /*05c0*/ FFMA R12, R22, R29, R12 ; /* 0x0000001d160c7223 */ /* 0x000fe2000000000c */ /*05d0*/ ISETP.GT.AND P1, PT, R6, 0xc, PT ; /* 0x0000000c0600780c */ /* 0x000fc60003f24270 */ /*05e0*/ FFMA R19, R20, R19, R12 ; /* 0x0000001314137223 */ /* 0x020fe2000000000c */ /*05f0*/ UIADD3 UR6, UP0, UR6, 0x40, URZ ; /* 0x0000004006067890 */ /* 0x000fe2000ff1e03f */ /*0600*/ IADD3 R4, R4, 0x10, RZ ; /* 0x0000001004047810 */ /* 0x000fc60007ffe0ff */ /*0610*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0620*/ FFMA R8, R8, R24, R19 ; /* 0x0000001808087223 */ /* 0x004fc80000000013 */ /*0630*/ FFMA R8, R10, R9, R8 ; /* 0x000000090a087223 */ /* 0x000fc80000000008 */ /*0640*/ FFMA R8, R28, R21, R8 ; /* 0x000000151c087223 */ /* 0x000fc80000000008 */ /*0650*/ FFMA R24, R11, R26, R8 ; /* 0x0000001a0b187223 */ /* 0x000fe40000000008 */ /*0660*/ IMAD.WIDE R8, R0, 0x4, R16 ; /* 0x0000000400087825 */ /* 0x000fe200078e0210 */ /*0670*/ @P1 BRA 0x1f0 ; /* 0xfffffb7000001947 */ /* 0x000fea000383ffff */ /*0680*/ ISETP.GT.AND P1, PT, R6, 0x4, PT ; /* 0x000000040600780c */ /* 0x000fda0003f24270 */ /*0690*/ @!P1 BRA 0x920 ; /* 0x0000028000009947 */ /* 0x000fea0003800000 */ /*06a0*/ IMAD.WIDE R16, R0, 0x4, R8 ; /* 0x0000000400107825 */ /* 0x000fe200078e0208 */ /*06b0*/ MOV R10, UR6 ; /* 0x00000006000a7c02 */ /* 0x000fe20008000f00 */ /*06c0*/ LDG.E R23, [R8.64] ; /* 0x0000000408177981 */ /* 0x0000a2000c1e1900 */ /*06d0*/ MOV R11, UR7 ; /* 0x00000007000b7c02 */ /* 0x000fc60008000f00 */ /*06e0*/ IMAD.WIDE R12, R0.reuse, 0x4, R16 ; /* 0x00000004000c7825 */ /* 0x040fe400078e0210 */ /*06f0*/ LDG.E R16, [R16.64] ; /* 0x0000000410107981 */ /* 0x0002e4000c1e1900 */ /*0700*/ IMAD.WIDE R10, R7, 0x4, R10 ; /* 0x00000004070a7825 */ /* 0x000fe400078e020a */ /*0710*/ LDG.E R26, [R12.64] ; /* 0x000000040c1a7981 */ /* 0x000964000c1e1900 */ /*0720*/ IMAD.WIDE R14, R0.reuse, 0x4, R12 ; /* 0x00000004000e7825 */ /* 0x040fe400078e020c */ /*0730*/ LDG.E R22, [R10.64] ; /* 0x000000040a167981 */ /* 0x000ea8000c1e1900 */ /*0740*/ LDG.E R25, [R10.64+0x4] ; /* 0x000004040a197981 */ /* 0x000ee2000c1e1900 */ /*0750*/ IMAD.WIDE R18, R0, 0x4, R14 ; /* 0x0000000400127825 */ /* 0x000fc600078e020e */ /*0760*/ LDG.E R27, [R10.64+0x8] ; /* 0x000008040a1b7981 */ /* 0x000f66000c1e1900 */ /*0770*/ IMAD.WIDE R20, R0.reuse, 0x4, R18 ; /* 0x0000000400147825 */ /* 0x040fe200078e0212 */ /*0780*/ LDG.E R14, [R14.64] ; /* 0x000000040e0e7981 */ /* 0x000368000c1e1900 */ /*0790*/ LDG.E R29, [R10.64+0xc] ; /* 0x00000c040a1d7981 */ /* 0x000f62000c1e1900 */ /*07a0*/ IMAD.WIDE R8, R0, 0x4, R20 ; /* 0x0000000400087825 */ /* 0x001fc600078e0214 */ /*07b0*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x000168000c1e1900 */ /*07c0*/ LDG.E R28, [R10.64+0x10] ; /* 0x000010040a1c7981 */ /* 0x000f62000c1e1900 */ /*07d0*/ IMAD.WIDE R12, R0, 0x4, R8 ; /* 0x00000004000c7825 */ /* 0x010fc600078e0208 */ /*07e0*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x000f28000c1e1900 */ /*07f0*/ LDG.E R15, [R10.64+0x14] ; /* 0x000014040a0f7981 */ /* 0x002f28000c1e1900 */ /*0800*/ LDG.E R17, [R8.64] ; /* 0x0000000408117981 */ /* 0x000328000c1e1900 */ /*0810*/ LDG.E R19, [R10.64+0x1c] ; /* 0x00001c040a137981 */ /* 0x001f28000c1e1900 */ /*0820*/ LDG.E R8, [R10.64+0x18] ; /* 0x000018040a087981 */ /* 0x002f28000c1e1900 */ /*0830*/ LDG.E R9, [R12.64] ; /* 0x000000040c097981 */ /* 0x000f22000c1e1900 */ /*0840*/ UIADD3 UR6, UP0, UR6, 0x20, URZ ; /* 0x0000002006067890 */ /* 0x000fe2000ff1e03f */ /*0850*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fc40003f0e170 */ /*0860*/ IADD3 R4, R4, 0x8, RZ ; /* 0x0000000804047810 */ /* 0x000fe40007ffe0ff */ /*0870*/ IADD3 R6, R6, -0x8, RZ ; /* 0xfffffff806067810 */ /* 0x000fe20007ffe0ff */ /*0880*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0890*/ FFMA R22, R23, R22, R24 ; /* 0x0000001617167223 */ /* 0x004fc80000000018 */ /*08a0*/ FFMA R16, R16, R25, R22 ; /* 0x0000001910107223 */ /* 0x008fc80000000016 */ /*08b0*/ FFMA R16, R26, R27, R16 ; /* 0x0000001b1a107223 */ /* 0x020fc80000000010 */ /*08c0*/ FFMA R29, R14, R29, R16 ; /* 0x0000001d0e1d7223 */ /* 0x000fc80000000010 */ /*08d0*/ FFMA R18, R18, R28, R29 ; /* 0x0000001c12127223 */ /* 0x000fc8000000001d */ /*08e0*/ FFMA R15, R20, R15, R18 ; /* 0x0000000f140f7223 */ /* 0x010fc80000000012 */ /*08f0*/ FFMA R8, R17, R8, R15 ; /* 0x0000000811087223 */ /* 0x000fc8000000000f */ /*0900*/ FFMA R24, R9, R19, R8 ; /* 0x0000001309187223 */ /* 0x000fe40000000008 */ /*0910*/ IMAD.WIDE R8, R0, 0x4, R12 ; /* 0x0000000400087825 */ /* 0x000fc800078e020c */ /*0920*/ ISETP.NE.OR P0, PT, R6, RZ, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0000705670 */ /*0930*/ @!P0 BRA 0xad0 ; /* 0x0000019000008947 */ /* 0x000fea0003800000 */ /*0940*/ MOV R12, UR6 ; /* 0x00000006000c7c02 */ /* 0x000fe20008000f00 */ /*0950*/ IMAD.WIDE R10, R0, 0x4, R8 ; /* 0x00000004000a7825 */ /* 0x000fe200078e0208 */ /*0960*/ MOV R13, UR7 ; /* 0x00000007000d7c02 */ /* 0x000fe20008000f00 */ /*0970*/ LDG.E R9, [R8.64] ; /* 0x0000000408097981 */ /* 0x000ea8000c1e1900 */ /*0980*/ IMAD.WIDE R12, R7, 0x4, R12 ; /* 0x00000004070c7825 */ /* 0x000fc800078e020c */ /*0990*/ IMAD.WIDE R14, R0.reuse, 0x4, R10 ; /* 0x00000004000e7825 */ /* 0x040fe200078e020a */ /*09a0*/ LDG.E R18, [R12.64] ; /* 0x000000040c127981 */ /* 0x000ea8000c1e1900 */ /*09b0*/ LDG.E R11, [R10.64] ; /* 0x000000040a0b7981 */ /* 0x000ee2000c1e1900 */ /*09c0*/ IMAD.WIDE R16, R0, 0x4, R14 ; /* 0x0000000400107825 */ /* 0x000fc600078e020e */ /*09d0*/ LDG.E R19, [R12.64+0x4] ; /* 0x000004040c137981 */ /* 0x000ee8000c1e1900 */ /*09e0*/ LDG.E R21, [R14.64] ; /* 0x000000040e157981 */ /* 0x000f28000c1e1900 */ /*09f0*/ LDG.E R20, [R12.64+0x8] ; /* 0x000008040c147981 */ /* 0x000f28000c1e1900 */ /*0a00*/ LDG.E R22, [R12.64+0xc] ; /* 0x00000c040c167981 */ /* 0x000f68000c1e1900 */ /*0a10*/ LDG.E R23, [R16.64] ; /* 0x0000000410177981 */ /* 0x000f62000c1e1900 */ /*0a20*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc80007ffe0ff */ /*0a30*/ ISETP.NE.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f05270 */ /*0a40*/ UIADD3 UR6, UP0, UR6, 0x10, URZ ; /* 0x0000001006067890 */ /* 0x000fe2000ff1e03f */ /*0a50*/ IADD3 R4, R4, 0x4, RZ ; /* 0x0000000404047810 */ /* 0x000fc60007ffe0ff */ /*0a60*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0a70*/ FFMA R18, R9, R18, R24 ; /* 0x0000001209127223 */ /* 0x004fc80000000018 */ /*0a80*/ FFMA R18, R11, R19, R18 ; /* 0x000000130b127223 */ /* 0x008fe40000000012 */ /*0a90*/ IMAD.WIDE R8, R0, 0x4, R16 ; /* 0x0000000400087825 */ /* 0x000fc800078e0210 */ /*0aa0*/ FFMA R18, R21, R20, R18 ; /* 0x0000001415127223 */ /* 0x010fc80000000012 */ /*0ab0*/ FFMA R24, R23, R22, R18 ; /* 0x0000001617187223 */ /* 0x020fe20000000012 */ /*0ac0*/ @P0 BRA 0x940 ; /* 0xfffffe7000000947 */ /* 0x000fea000383ffff */ /*0ad0*/ ISETP.NE.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fda0003f05270 */ /*0ae0*/ @!P0 BRA 0xbd0 ; /* 0x000000e000008947 */ /* 0x000fea0003800000 */ /*0af0*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0b00*/ IMAD R6, R3, c[0x0][0x180], R4 ; /* 0x0000600003067a24 */ /* 0x000fe400078e0204 */ /*0b10*/ IMAD R4, R4, c[0x0][0x180], R2 ; /* 0x0000600004047a24 */ /* 0x000fce00078e0202 */ /*0b20*/ IMAD.WIDE R6, R6, R9, c[0x0][0x160] ; /* 0x0000580006067625 */ /* 0x000fc800078e0209 */ /*0b30*/ IMAD.WIDE R8, R4, R9, c[0x0][0x168] ; /* 0x00005a0004087625 */ /* 0x000fca00078e0209 */ /*0b40*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x0000a8000c1e1900 */ /*0b50*/ LDG.E R4, [R6.64] ; /* 0x0000000406047981 */ /* 0x0002a2000c1e1900 */ /*0b60*/ IADD3 R5, R5, -0x1, RZ ; /* 0xffffffff05057810 */ /* 0x000fc80007ffe0ff */ /*0b70*/ ISETP.NE.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fe20003f05270 */ /*0b80*/ IMAD.WIDE R8, R0, 0x4, R8 ; /* 0x0000000400087825 */ /* 0x001fe200078e0208 */ /*0b90*/ IADD3 R6, P1, R6, 0x4, RZ ; /* 0x0000000406067810 */ /* 0x002fc80007f3e0ff */ /*0ba0*/ IADD3.X R7, RZ, R7, RZ, P1, !PT ; /* 0x00000007ff077210 */ /* 0x000fe20000ffe4ff */ /*0bb0*/ FFMA R24, R11, R4, R24 ; /* 0x000000040b187223 */ /* 0x004fcc0000000018 */ /*0bc0*/ @P0 BRA 0xb40 ; /* 0xffffff7000000947 */ /* 0x000fea000383ffff */ /*0bd0*/ MOV R5, 0x4 ; /* 0x0000000400057802 */ /* 0x000fe20000000f00 */ /*0be0*/ IMAD R2, R3, c[0x0][0x17c], R2 ; /* 0x00005f0003027a24 */ /* 0x000fc800078e0202 */ /*0bf0*/ IMAD.WIDE R2, R2, R5, c[0x0][0x170] ; /* 0x00005c0002027625 */ /* 0x000fca00078e0205 */ /*0c00*/ LDG.E R0, [R2.64] ; /* 0x0000000402007981 */ /* 0x000ea4000c1e1900 */ /*0c10*/ FMUL R5, R0, c[0x0][0x188] ; /* 0x0000620000057a20 */ /* 0x004fc80000400000 */ /*0c20*/ FFMA R5, R24, c[0x0][0x184], R5 ; /* 0x0000610018057a23 */ /* 0x000fca0000000005 */ /*0c30*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe2000c101904 */ /*0c40*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0c50*/ BRA 0xc50; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0c60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ca0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ce0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include "includes.h" __global__ void sgemm_kernel_B(const float *A, const float *B, float *C, int N, int M, int K, float alpha, float beta) { int col = blockIdx.x * blockDim.x + threadIdx.x; int row = blockIdx.y * blockDim.y + threadIdx.y; float sum = 0.f; for (int i = 0; i < K; ++i) sum += A[row * K + i] * B[i * K + col]; C[row * M + col] = alpha * sum + beta * C[row * M + col]; }
.file "tmpxft_0012fe85_00000000-6_sgemm_kernel_B.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff .type _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff, @function _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff: .LFB2051: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) movss %xmm0, 8(%rsp) movss %xmm1, 4(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 20(%rsp), %rax movq %rax, 136(%rsp) leaq 16(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%rsp) leaq 8(%rsp), %rax movq %rax, 160(%rsp) leaq 4(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 184(%rsp), %rax subq %fs:40, %rax jne .L8 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z14sgemm_kernel_BPKfS0_Pfiiiff(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff, .-_Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff .globl _Z14sgemm_kernel_BPKfS0_Pfiiiff .type _Z14sgemm_kernel_BPKfS0_Pfiiiff, @function _Z14sgemm_kernel_BPKfS0_Pfiiiff: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z14sgemm_kernel_BPKfS0_Pfiiiff, .-_Z14sgemm_kernel_BPKfS0_Pfiiiff .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z14sgemm_kernel_BPKfS0_Pfiiiff" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z14sgemm_kernel_BPKfS0_Pfiiiff(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include "includes.h" __global__ void sgemm_kernel_B(const float *A, const float *B, float *C, int N, int M, int K, float alpha, float beta) { int col = blockIdx.x * blockDim.x + threadIdx.x; int row = blockIdx.y * blockDim.y + threadIdx.y; float sum = 0.f; for (int i = 0; i < K; ++i) sum += A[row * K + i] * B[i * K + col]; C[row * M + col] = alpha * sum + beta * C[row * M + col]; }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void sgemm_kernel_B(const float *A, const float *B, float *C, int N, int M, int K, float alpha, float beta) { int col = blockIdx.x * blockDim.x + threadIdx.x; int row = blockIdx.y * blockDim.y + threadIdx.y; float sum = 0.f; for (int i = 0; i < K; ++i) sum += A[row * K + i] * B[i * K + col]; C[row * M + col] = alpha * sum + beta * C[row * M + col]; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void sgemm_kernel_B(const float *A, const float *B, float *C, int N, int M, int K, float alpha, float beta) { int col = blockIdx.x * blockDim.x + threadIdx.x; int row = blockIdx.y * blockDim.y + threadIdx.y; float sum = 0.f; for (int i = 0; i < K; ++i) sum += A[row * K + i] * B[i * K + col]; C[row * M + col] = alpha * sum + beta * C[row * M + col]; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z14sgemm_kernel_BPKfS0_Pfiiiff .globl _Z14sgemm_kernel_BPKfS0_Pfiiiff .p2align 8 .type _Z14sgemm_kernel_BPKfS0_Pfiiiff,@function _Z14sgemm_kernel_BPKfS0_Pfiiiff: s_clause 0x1 s_load_b32 s3, s[0:1], 0x3c s_load_b32 s2, s[0:1], 0x20 v_and_b32_e32 v2, 0x3ff, v0 v_bfe_u32 v3, v0, 10, 10 s_waitcnt lgkmcnt(0) s_and_b32 s4, s3, 0xffff s_lshr_b32 s3, s3, 16 s_delay_alu instid0(VALU_DEP_1) v_mad_u64_u32 v[0:1], null, s14, s4, v[2:3] v_mad_u64_u32 v[1:2], null, s15, s3, v[3:4] s_cmp_lt_i32 s2, 1 s_cbranch_scc1 .LBB0_3 s_load_b128 s[4:7], s[0:1], 0x0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_4) v_mul_lo_u32 v2, v1, s2 v_mov_b32_e32 v6, 0 v_mov_b32_e32 v4, v0 s_mov_b32 s3, s2 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[2:3], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v2, vcc_lo, s4, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s5, v3, vcc_lo .p2align 6 .LBB0_2: v_ashrrev_i32_e32 v5, 31, v4 s_add_i32 s3, s3, -1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) s_cmp_eq_u32 s3, 0 v_lshlrev_b64 v[7:8], 2, v[4:5] v_add_nc_u32_e32 v4, s2, v4 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_u32 v7, vcc_lo, s6, v7 v_add_co_ci_u32_e32 v8, vcc_lo, s7, v8, vcc_lo global_load_b32 v5, v[2:3], off global_load_b32 v7, v[7:8], off v_add_co_u32 v2, vcc_lo, v2, 4 v_add_co_ci_u32_e32 v3, vcc_lo, 0, v3, vcc_lo s_waitcnt vmcnt(0) v_fmac_f32_e32 v6, v5, v7 s_cbranch_scc0 .LBB0_2 s_branch .LBB0_4 .LBB0_3: v_mov_b32_e32 v6, 0 .LBB0_4: s_clause 0x2 s_load_b32 s4, s[0:1], 0x1c s_load_b64 s[2:3], s[0:1], 0x10 s_load_b64 s[0:1], s[0:1], 0x24 s_waitcnt lgkmcnt(0) v_mad_u64_u32 v[2:3], null, v1, s4, v[0:1] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[0:1], 2, v[2:3] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v0, vcc_lo, s2, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s3, v1, vcc_lo global_load_b32 v2, v[0:1], off s_waitcnt vmcnt(0) v_mul_f32_e32 v2, s1, v2 s_delay_alu instid0(VALU_DEP_1) v_fmac_f32_e32 v2, s0, v6 global_store_b32 v[0:1], v2, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z14sgemm_kernel_BPKfS0_Pfiiiff .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 304 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 9 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z14sgemm_kernel_BPKfS0_Pfiiiff, .Lfunc_end0-_Z14sgemm_kernel_BPKfS0_Pfiiiff .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value - .offset: 28 .size: 4 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: by_value - .offset: 36 .size: 4 .value_kind: by_value - .offset: 40 .size: 4 .value_kind: by_value - .offset: 48 .size: 4 .value_kind: hidden_block_count_x - .offset: 52 .size: 4 .value_kind: hidden_block_count_y - .offset: 56 .size: 4 .value_kind: hidden_block_count_z - .offset: 60 .size: 2 .value_kind: hidden_group_size_x - .offset: 62 .size: 2 .value_kind: hidden_group_size_y - .offset: 64 .size: 2 .value_kind: hidden_group_size_z - .offset: 66 .size: 2 .value_kind: hidden_remainder_x - .offset: 68 .size: 2 .value_kind: hidden_remainder_y - .offset: 70 .size: 2 .value_kind: hidden_remainder_z - .offset: 88 .size: 8 .value_kind: hidden_global_offset_x - .offset: 96 .size: 8 .value_kind: hidden_global_offset_y - .offset: 104 .size: 8 .value_kind: hidden_global_offset_z - .offset: 112 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 304 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z14sgemm_kernel_BPKfS0_Pfiiiff .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z14sgemm_kernel_BPKfS0_Pfiiiff.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 9 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void sgemm_kernel_B(const float *A, const float *B, float *C, int N, int M, int K, float alpha, float beta) { int col = blockIdx.x * blockDim.x + threadIdx.x; int row = blockIdx.y * blockDim.y + threadIdx.y; float sum = 0.f; for (int i = 0; i < K; ++i) sum += A[row * K + i] * B[i * K + col]; C[row * M + col] = alpha * sum + beta * C[row * M + col]; }
.text .file "sgemm_kernel_B.hip" .globl _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff # -- Begin function _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .p2align 4, 0x90 .type _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff,@function _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff: # @_Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) movss %xmm0, 8(%rsp) movss %xmm1, 4(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 20(%rsp), %rax movq %rax, 120(%rsp) leaq 16(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 4(%rsp), %rax movq %rax, 152(%rsp) leaq 56(%rsp), %rdi leaq 40(%rsp), %rsi leaq 32(%rsp), %rdx leaq 24(%rsp), %rcx callq __hipPopCallConfiguration movq 56(%rsp), %rsi movl 64(%rsp), %edx movq 40(%rsp), %rcx movl 48(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z14sgemm_kernel_BPKfS0_Pfiiiff, %edi pushq 24(%rsp) .cfi_adjust_cfa_offset 8 pushq 40(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end0: .size _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff, .Lfunc_end0-_Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z14sgemm_kernel_BPKfS0_Pfiiiff, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z14sgemm_kernel_BPKfS0_Pfiiiff,@object # @_Z14sgemm_kernel_BPKfS0_Pfiiiff .section .rodata,"a",@progbits .globl _Z14sgemm_kernel_BPKfS0_Pfiiiff .p2align 3, 0x0 _Z14sgemm_kernel_BPKfS0_Pfiiiff: .quad _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .size _Z14sgemm_kernel_BPKfS0_Pfiiiff, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z14sgemm_kernel_BPKfS0_Pfiiiff" .size .L__unnamed_1, 32 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z14sgemm_kernel_BPKfS0_Pfiiiff .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z14sgemm_kernel_BPKfS0_Pfiiiff .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ MOV R0, c[0x0][0x180] ; /* 0x0000600000007a02 */ /* 0x000fe20000000f00 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ HFMA2.MMA R24, -RZ, RZ, 0, 0 ; /* 0x00000000ff187435 */ /* 0x000fe200000001ff */ /*0050*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e220000002100 */ /*0060*/ ISETP.GE.AND P0, PT, R0, 0x1, PT ; /* 0x000000010000780c */ /* 0x000fc60003f06270 */ /*0070*/ S2R R4, SR_CTAID.Y ; /* 0x0000000000047919 */ /* 0x000e680000002600 */ /*0080*/ S2R R5, SR_TID.Y ; /* 0x0000000000057919 */ /* 0x000e620000002200 */ /*0090*/ IMAD R2, R2, c[0x0][0x0], R3 ; /* 0x0000000002027a24 */ /* 0x001fe400078e0203 */ /*00a0*/ IMAD R3, R4, c[0x0][0x4], R5 ; /* 0x0000010004037a24 */ /* 0x002fc600078e0205 */ /*00b0*/ @!P0 BRA 0xbd0 ; /* 0x00000b1000008947 */ /* 0x000fea0003800000 */ /*00c0*/ IADD3 R4, R0.reuse, -0x1, RZ ; /* 0xffffffff00047810 */ /* 0x040fe40007ffe0ff */ /*00d0*/ LOP3.LUT R5, R0, 0x3, RZ, 0xc0, !PT ; /* 0x0000000300057812 */ /* 0x000fe400078ec0ff */ /*00e0*/ ISETP.GE.U32.AND P0, PT, R4, 0x3, PT ; /* 0x000000030400780c */ /* 0x000fe40003f06070 */ /*00f0*/ MOV R24, RZ ; /* 0x000000ff00187202 */ /* 0x000fe40000000f00 */ /*0100*/ MOV R4, RZ ; /* 0x000000ff00047202 */ /* 0x000fd20000000f00 */ /*0110*/ @!P0 BRA 0xad0 ; /* 0x000009b000008947 */ /* 0x000fea0003800000 */ /*0120*/ IADD3 R6, -R5, c[0x0][0x180], RZ ; /* 0x0000600005067a10 */ /* 0x000fe20007ffe1ff */ /*0130*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0140*/ ULDC.64 UR6, c[0x0][0x160] ; /* 0x0000580000067ab9 */ /* 0x000fe20000000a00 */ /*0150*/ HFMA2.MMA R4, -RZ, RZ, 0, 0 ; /* 0x00000000ff047435 */ /* 0x000fe200000001ff */ /*0160*/ ISETP.GT.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f04270 */ /*0170*/ IMAD R7, R3, c[0x0][0x180], RZ ; /* 0x0000600003077a24 */ /* 0x000fe200078e02ff */ /*0180*/ MOV R24, RZ ; /* 0x000000ff00187202 */ /* 0x000fca0000000f00 */ /*0190*/ IMAD.WIDE R8, R2, R9, c[0x0][0x168] ; /* 0x00005a0002087625 */ /* 0x000fcc00078e0209 */ /*01a0*/ @!P0 BRA 0x940 ; /* 0x0000079000008947 */ /* 0x000fea0003800000 */ /*01b0*/ ISETP.GT.AND P1, PT, R6, 0xc, PT ; /* 0x0000000c0600780c */ /* 0x000fe40003f24270 */ /*01c0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fd60003f0f070 */ /*01d0*/ @!P1 BRA 0x680 ; /* 0x000004a000009947 */ /* 0x000fea0003800000 */ /*01e0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*01f0*/ MOV R14, UR6 ; /* 0x00000006000e7c02 */ /* 0x000fe20008000f00 */ /*0200*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x0000a2000c1e1900 */ /*0210*/ MOV R15, UR7 ; /* 0x00000007000f7c02 */ /* 0x000fca0008000f00 */ /*0220*/ IMAD.WIDE R14, R7, 0x4, R14 ; /* 0x00000004070e7825 */ /* 0x000fca00078e020e */ /*0230*/ LDG.E R10, [R14.64] ; /* 0x000000040e0a7981 */ /* 0x000ea2000c1e1900 */ /*0240*/ IMAD.WIDE R8, R0, 0x4, R8 ; /* 0x0000000400087825 */ /* 0x001fc600078e0208 */ /*0250*/ LDG.E R18, [R14.64+0x4] ; /* 0x000004040e127981 */ /* 0x000ee6000c1e1900 */ /*0260*/ IMAD.WIDE R22, R0.reuse, 0x4, R8 ; /* 0x0000000400167825 */ /* 0x040fe200078e0208 */ /*0270*/ LDG.E R19, [R8.64] ; /* 0x0000000408137981 */ /* 0x0000e8000c1e1900 */ /*0280*/ LDG.E R28, [R22.64] ; /* 0x00000004161c7981 */ /* 0x000322000c1e1900 */ /*0290*/ IMAD.WIDE R26, R0, 0x4, R22 ; /* 0x00000004001a7825 */ /* 0x000fc600078e0216 */ /*02a0*/ LDG.E R29, [R14.64+0x8] ; /* 0x000008040e1d7981 */ /* 0x000f26000c1e1900 */ /*02b0*/ IMAD.WIDE R12, R0.reuse, 0x4, R26 ; /* 0x00000004000c7825 */ /* 0x040fe200078e021a */ /*02c0*/ LDG.E R16, [R26.64] ; /* 0x000000041a107981 */ /* 0x000b28000c1e1900 */ /*02d0*/ LDG.E R17, [R14.64+0xc] ; /* 0x00000c040e117981 */ /* 0x000f28000c1e1900 */ /*02e0*/ LDG.E R20, [R14.64+0x10] ; /* 0x000010040e147981 */ /* 0x000f28000c1e1900 */ /*02f0*/ LDG.E R21, [R12.64] ; /* 0x000000040c157981 */ /* 0x000328000c1e1900 */ /*0300*/ LDG.E R8, [R14.64+0x14] ; /* 0x000014040e087981 */ /* 0x001f28000c1e1900 */ /*0310*/ LDG.E R26, [R14.64+0x1c] ; /* 0x00001c040e1a7981 */ /* 0x020f62000c1e1900 */ /*0320*/ IMAD.WIDE R12, R0, 0x4, R12 ; /* 0x00000004000c7825 */ /* 0x002fca00078e020c */ /*0330*/ LDG.E R9, [R12.64] ; /* 0x000000040c097981 */ /* 0x000562000c1e1900 */ /*0340*/ IMAD.WIDE R22, R0, 0x4, R12 ; /* 0x0000000400167825 */ /* 0x000fc800078e020c */ /*0350*/ FFMA R12, R11, R10, R24 ; /* 0x0000000a0b0c7223 */ /* 0x004fe40000000018 */ /*0360*/ LDG.E R10, [R14.64+0x18] ; /* 0x000018040e0a7981 */ /* 0x000ea2000c1e1900 */ /*0370*/ IMAD.WIDE R24, R0, 0x4, R22 ; /* 0x0000000400187825 */ /* 0x000fc600078e0216 */ /*0380*/ LDG.E R11, [R22.64] ; /* 0x00000004160b7981 */ /* 0x0000a8000c1e1900 */ /*0390*/ LDG.E R27, [R24.64] ; /* 0x00000004181b7981 */ /* 0x0002a2000c1e1900 */ /*03a0*/ FFMA R12, R19, R18, R12 ; /* 0x00000012130c7223 */ /* 0x008fe4000000000c */ /*03b0*/ IMAD.WIDE R18, R0, 0x4, R24 ; /* 0x0000000400127825 */ /* 0x000fe200078e0218 */ /*03c0*/ LDG.E R23, [R14.64+0x20] ; /* 0x000020040e177981 */ /* 0x001ee6000c1e1900 */ /*03d0*/ FFMA R28, R28, R29, R12 ; /* 0x0000001d1c1c7223 */ /* 0x010fc4000000000c */ /*03e0*/ IMAD.WIDE R12, R0, 0x4, R18 ; /* 0x00000004000c7825 */ /* 0x000fe200078e0212 */ /*03f0*/ LDG.E R25, [R14.64+0x24] ; /* 0x000024040e197981 */ /* 0x002f26000c1e1900 */ /*0400*/ FFMA R28, R16, R17, R28 ; /* 0x00000011101c7223 */ /* 0x000fe2000000001c */ /*0410*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x0000e2000c1e1900 */ /*0420*/ IMAD.WIDE R16, R0, 0x4, R12 ; /* 0x0000000400107825 */ /* 0x000fc600078e020c */ /*0430*/ LDG.E R12, [R12.64] ; /* 0x000000040c0c7981 */ /* 0x000322000c1e1900 */ /*0440*/ FFMA R28, R21, R20, R28 ; /* 0x00000014151c7223 */ /* 0x000fc6000000001c */ /*0450*/ LDG.E R22, [R16.64] ; /* 0x0000000410167981 */ /* 0x0002e2000c1e1900 */ /*0460*/ IMAD.WIDE R20, R0, 0x4, R16 ; /* 0x0000000400147825 */ /* 0x000fc600078e0210 */ /*0470*/ LDG.E R29, [R14.64+0x28] ; /* 0x000028040e1d7981 */ /* 0x000f28000c1e1900 */ /*0480*/ LDG.E R19, [R14.64+0x2c] ; /* 0x00002c040e137981 */ /* 0x001f28000c1e1900 */ /*0490*/ LDG.E R24, [R14.64+0x30] ; /* 0x000030040e187981 */ /* 0x000f22000c1e1900 */ /*04a0*/ FFMA R28, R9, R8, R28 ; /* 0x00000008091c7223 */ /* 0x020fe4000000001c */ /*04b0*/ IMAD.WIDE R8, R0, 0x4, R20 ; /* 0x0000000400087825 */ /* 0x000fc400078e0214 */ /*04c0*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x000168000c1e1900 */ /*04d0*/ LDG.E R21, [R14.64+0x38] ; /* 0x000038040e157981 */ /* 0x001f62000c1e1900 */ /*04e0*/ FFMA R28, R11, R10, R28 ; /* 0x0000000a0b1c7223 */ /* 0x004fe4000000001c */ /*04f0*/ IMAD.WIDE R10, R0, 0x4, R8 ; /* 0x00000004000a7825 */ /* 0x000fe400078e0208 */ /*0500*/ LDG.E R8, [R8.64] ; /* 0x0000000408087981 */ /* 0x0000a4000c1e1900 */ /*0510*/ FFMA R13, R27, R26, R28 ; /* 0x0000001a1b0d7223 */ /* 0x002fc4000000001c */ /*0520*/ IMAD.WIDE R26, R0.reuse, 0x4, R10 ; /* 0x00000004001a7825 */ /* 0x040fe400078e020a */ /*0530*/ LDG.E R10, [R10.64] ; /* 0x000000040a0a7981 */ /* 0x0002a8000c1e1900 */ /*0540*/ LDG.E R9, [R14.64+0x34] ; /* 0x000034040e097981 */ /* 0x001ea2000c1e1900 */ /*0550*/ IMAD.WIDE R16, R0, 0x4, R26 ; /* 0x0000000400107825 */ /* 0x000fc600078e021a */ /*0560*/ LDG.E R28, [R26.64] ; /* 0x000000041a1c7981 */ /* 0x0000a8000c1e1900 */ /*0570*/ LDG.E R11, [R16.64] ; /* 0x00000004100b7981 */ /* 0x002ea8000c1e1900 */ /*0580*/ LDG.E R26, [R14.64+0x3c] ; /* 0x00003c040e1a7981 */ /* 0x001ea2000c1e1900 */ /*0590*/ FFMA R13, R18, R23, R13 ; /* 0x00000017120d7223 */ /* 0x008fc8000000000d */ /*05a0*/ FFMA R12, R12, R25, R13 ; /* 0x000000190c0c7223 */ /* 0x010fe2000000000d */ /*05b0*/ IADD3 R6, R6, -0x10, RZ ; /* 0xfffffff006067810 */ /* 0x000fc60007ffe0ff */ /*05c0*/ FFMA R12, R22, R29, R12 ; /* 0x0000001d160c7223 */ /* 0x000fe2000000000c */ /*05d0*/ ISETP.GT.AND P1, PT, R6, 0xc, PT ; /* 0x0000000c0600780c */ /* 0x000fc60003f24270 */ /*05e0*/ FFMA R19, R20, R19, R12 ; /* 0x0000001314137223 */ /* 0x020fe2000000000c */ /*05f0*/ UIADD3 UR6, UP0, UR6, 0x40, URZ ; /* 0x0000004006067890 */ /* 0x000fe2000ff1e03f */ /*0600*/ IADD3 R4, R4, 0x10, RZ ; /* 0x0000001004047810 */ /* 0x000fc60007ffe0ff */ /*0610*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0620*/ FFMA R8, R8, R24, R19 ; /* 0x0000001808087223 */ /* 0x004fc80000000013 */ /*0630*/ FFMA R8, R10, R9, R8 ; /* 0x000000090a087223 */ /* 0x000fc80000000008 */ /*0640*/ FFMA R8, R28, R21, R8 ; /* 0x000000151c087223 */ /* 0x000fc80000000008 */ /*0650*/ FFMA R24, R11, R26, R8 ; /* 0x0000001a0b187223 */ /* 0x000fe40000000008 */ /*0660*/ IMAD.WIDE R8, R0, 0x4, R16 ; /* 0x0000000400087825 */ /* 0x000fe200078e0210 */ /*0670*/ @P1 BRA 0x1f0 ; /* 0xfffffb7000001947 */ /* 0x000fea000383ffff */ /*0680*/ ISETP.GT.AND P1, PT, R6, 0x4, PT ; /* 0x000000040600780c */ /* 0x000fda0003f24270 */ /*0690*/ @!P1 BRA 0x920 ; /* 0x0000028000009947 */ /* 0x000fea0003800000 */ /*06a0*/ IMAD.WIDE R16, R0, 0x4, R8 ; /* 0x0000000400107825 */ /* 0x000fe200078e0208 */ /*06b0*/ MOV R10, UR6 ; /* 0x00000006000a7c02 */ /* 0x000fe20008000f00 */ /*06c0*/ LDG.E R23, [R8.64] ; /* 0x0000000408177981 */ /* 0x0000a2000c1e1900 */ /*06d0*/ MOV R11, UR7 ; /* 0x00000007000b7c02 */ /* 0x000fc60008000f00 */ /*06e0*/ IMAD.WIDE R12, R0.reuse, 0x4, R16 ; /* 0x00000004000c7825 */ /* 0x040fe400078e0210 */ /*06f0*/ LDG.E R16, [R16.64] ; /* 0x0000000410107981 */ /* 0x0002e4000c1e1900 */ /*0700*/ IMAD.WIDE R10, R7, 0x4, R10 ; /* 0x00000004070a7825 */ /* 0x000fe400078e020a */ /*0710*/ LDG.E R26, [R12.64] ; /* 0x000000040c1a7981 */ /* 0x000964000c1e1900 */ /*0720*/ IMAD.WIDE R14, R0.reuse, 0x4, R12 ; /* 0x00000004000e7825 */ /* 0x040fe400078e020c */ /*0730*/ LDG.E R22, [R10.64] ; /* 0x000000040a167981 */ /* 0x000ea8000c1e1900 */ /*0740*/ LDG.E R25, [R10.64+0x4] ; /* 0x000004040a197981 */ /* 0x000ee2000c1e1900 */ /*0750*/ IMAD.WIDE R18, R0, 0x4, R14 ; /* 0x0000000400127825 */ /* 0x000fc600078e020e */ /*0760*/ LDG.E R27, [R10.64+0x8] ; /* 0x000008040a1b7981 */ /* 0x000f66000c1e1900 */ /*0770*/ IMAD.WIDE R20, R0.reuse, 0x4, R18 ; /* 0x0000000400147825 */ /* 0x040fe200078e0212 */ /*0780*/ LDG.E R14, [R14.64] ; /* 0x000000040e0e7981 */ /* 0x000368000c1e1900 */ /*0790*/ LDG.E R29, [R10.64+0xc] ; /* 0x00000c040a1d7981 */ /* 0x000f62000c1e1900 */ /*07a0*/ IMAD.WIDE R8, R0, 0x4, R20 ; /* 0x0000000400087825 */ /* 0x001fc600078e0214 */ /*07b0*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x000168000c1e1900 */ /*07c0*/ LDG.E R28, [R10.64+0x10] ; /* 0x000010040a1c7981 */ /* 0x000f62000c1e1900 */ /*07d0*/ IMAD.WIDE R12, R0, 0x4, R8 ; /* 0x00000004000c7825 */ /* 0x010fc600078e0208 */ /*07e0*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x000f28000c1e1900 */ /*07f0*/ LDG.E R15, [R10.64+0x14] ; /* 0x000014040a0f7981 */ /* 0x002f28000c1e1900 */ /*0800*/ LDG.E R17, [R8.64] ; /* 0x0000000408117981 */ /* 0x000328000c1e1900 */ /*0810*/ LDG.E R19, [R10.64+0x1c] ; /* 0x00001c040a137981 */ /* 0x001f28000c1e1900 */ /*0820*/ LDG.E R8, [R10.64+0x18] ; /* 0x000018040a087981 */ /* 0x002f28000c1e1900 */ /*0830*/ LDG.E R9, [R12.64] ; /* 0x000000040c097981 */ /* 0x000f22000c1e1900 */ /*0840*/ UIADD3 UR6, UP0, UR6, 0x20, URZ ; /* 0x0000002006067890 */ /* 0x000fe2000ff1e03f */ /*0850*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fc40003f0e170 */ /*0860*/ IADD3 R4, R4, 0x8, RZ ; /* 0x0000000804047810 */ /* 0x000fe40007ffe0ff */ /*0870*/ IADD3 R6, R6, -0x8, RZ ; /* 0xfffffff806067810 */ /* 0x000fe20007ffe0ff */ /*0880*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0890*/ FFMA R22, R23, R22, R24 ; /* 0x0000001617167223 */ /* 0x004fc80000000018 */ /*08a0*/ FFMA R16, R16, R25, R22 ; /* 0x0000001910107223 */ /* 0x008fc80000000016 */ /*08b0*/ FFMA R16, R26, R27, R16 ; /* 0x0000001b1a107223 */ /* 0x020fc80000000010 */ /*08c0*/ FFMA R29, R14, R29, R16 ; /* 0x0000001d0e1d7223 */ /* 0x000fc80000000010 */ /*08d0*/ FFMA R18, R18, R28, R29 ; /* 0x0000001c12127223 */ /* 0x000fc8000000001d */ /*08e0*/ FFMA R15, R20, R15, R18 ; /* 0x0000000f140f7223 */ /* 0x010fc80000000012 */ /*08f0*/ FFMA R8, R17, R8, R15 ; /* 0x0000000811087223 */ /* 0x000fc8000000000f */ /*0900*/ FFMA R24, R9, R19, R8 ; /* 0x0000001309187223 */ /* 0x000fe40000000008 */ /*0910*/ IMAD.WIDE R8, R0, 0x4, R12 ; /* 0x0000000400087825 */ /* 0x000fc800078e020c */ /*0920*/ ISETP.NE.OR P0, PT, R6, RZ, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0000705670 */ /*0930*/ @!P0 BRA 0xad0 ; /* 0x0000019000008947 */ /* 0x000fea0003800000 */ /*0940*/ MOV R12, UR6 ; /* 0x00000006000c7c02 */ /* 0x000fe20008000f00 */ /*0950*/ IMAD.WIDE R10, R0, 0x4, R8 ; /* 0x00000004000a7825 */ /* 0x000fe200078e0208 */ /*0960*/ MOV R13, UR7 ; /* 0x00000007000d7c02 */ /* 0x000fe20008000f00 */ /*0970*/ LDG.E R9, [R8.64] ; /* 0x0000000408097981 */ /* 0x000ea8000c1e1900 */ /*0980*/ IMAD.WIDE R12, R7, 0x4, R12 ; /* 0x00000004070c7825 */ /* 0x000fc800078e020c */ /*0990*/ IMAD.WIDE R14, R0.reuse, 0x4, R10 ; /* 0x00000004000e7825 */ /* 0x040fe200078e020a */ /*09a0*/ LDG.E R18, [R12.64] ; /* 0x000000040c127981 */ /* 0x000ea8000c1e1900 */ /*09b0*/ LDG.E R11, [R10.64] ; /* 0x000000040a0b7981 */ /* 0x000ee2000c1e1900 */ /*09c0*/ IMAD.WIDE R16, R0, 0x4, R14 ; /* 0x0000000400107825 */ /* 0x000fc600078e020e */ /*09d0*/ LDG.E R19, [R12.64+0x4] ; /* 0x000004040c137981 */ /* 0x000ee8000c1e1900 */ /*09e0*/ LDG.E R21, [R14.64] ; /* 0x000000040e157981 */ /* 0x000f28000c1e1900 */ /*09f0*/ LDG.E R20, [R12.64+0x8] ; /* 0x000008040c147981 */ /* 0x000f28000c1e1900 */ /*0a00*/ LDG.E R22, [R12.64+0xc] ; /* 0x00000c040c167981 */ /* 0x000f68000c1e1900 */ /*0a10*/ LDG.E R23, [R16.64] ; /* 0x0000000410177981 */ /* 0x000f62000c1e1900 */ /*0a20*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc80007ffe0ff */ /*0a30*/ ISETP.NE.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f05270 */ /*0a40*/ UIADD3 UR6, UP0, UR6, 0x10, URZ ; /* 0x0000001006067890 */ /* 0x000fe2000ff1e03f */ /*0a50*/ IADD3 R4, R4, 0x4, RZ ; /* 0x0000000404047810 */ /* 0x000fc60007ffe0ff */ /*0a60*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0a70*/ FFMA R18, R9, R18, R24 ; /* 0x0000001209127223 */ /* 0x004fc80000000018 */ /*0a80*/ FFMA R18, R11, R19, R18 ; /* 0x000000130b127223 */ /* 0x008fe40000000012 */ /*0a90*/ IMAD.WIDE R8, R0, 0x4, R16 ; /* 0x0000000400087825 */ /* 0x000fc800078e0210 */ /*0aa0*/ FFMA R18, R21, R20, R18 ; /* 0x0000001415127223 */ /* 0x010fc80000000012 */ /*0ab0*/ FFMA R24, R23, R22, R18 ; /* 0x0000001617187223 */ /* 0x020fe20000000012 */ /*0ac0*/ @P0 BRA 0x940 ; /* 0xfffffe7000000947 */ /* 0x000fea000383ffff */ /*0ad0*/ ISETP.NE.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fda0003f05270 */ /*0ae0*/ @!P0 BRA 0xbd0 ; /* 0x000000e000008947 */ /* 0x000fea0003800000 */ /*0af0*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0b00*/ IMAD R6, R3, c[0x0][0x180], R4 ; /* 0x0000600003067a24 */ /* 0x000fe400078e0204 */ /*0b10*/ IMAD R4, R4, c[0x0][0x180], R2 ; /* 0x0000600004047a24 */ /* 0x000fce00078e0202 */ /*0b20*/ IMAD.WIDE R6, R6, R9, c[0x0][0x160] ; /* 0x0000580006067625 */ /* 0x000fc800078e0209 */ /*0b30*/ IMAD.WIDE R8, R4, R9, c[0x0][0x168] ; /* 0x00005a0004087625 */ /* 0x000fca00078e0209 */ /*0b40*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x0000a8000c1e1900 */ /*0b50*/ LDG.E R4, [R6.64] ; /* 0x0000000406047981 */ /* 0x0002a2000c1e1900 */ /*0b60*/ IADD3 R5, R5, -0x1, RZ ; /* 0xffffffff05057810 */ /* 0x000fc80007ffe0ff */ /*0b70*/ ISETP.NE.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fe20003f05270 */ /*0b80*/ IMAD.WIDE R8, R0, 0x4, R8 ; /* 0x0000000400087825 */ /* 0x001fe200078e0208 */ /*0b90*/ IADD3 R6, P1, R6, 0x4, RZ ; /* 0x0000000406067810 */ /* 0x002fc80007f3e0ff */ /*0ba0*/ IADD3.X R7, RZ, R7, RZ, P1, !PT ; /* 0x00000007ff077210 */ /* 0x000fe20000ffe4ff */ /*0bb0*/ FFMA R24, R11, R4, R24 ; /* 0x000000040b187223 */ /* 0x004fcc0000000018 */ /*0bc0*/ @P0 BRA 0xb40 ; /* 0xffffff7000000947 */ /* 0x000fea000383ffff */ /*0bd0*/ MOV R5, 0x4 ; /* 0x0000000400057802 */ /* 0x000fe20000000f00 */ /*0be0*/ IMAD R2, R3, c[0x0][0x17c], R2 ; /* 0x00005f0003027a24 */ /* 0x000fc800078e0202 */ /*0bf0*/ IMAD.WIDE R2, R2, R5, c[0x0][0x170] ; /* 0x00005c0002027625 */ /* 0x000fca00078e0205 */ /*0c00*/ LDG.E R0, [R2.64] ; /* 0x0000000402007981 */ /* 0x000ea4000c1e1900 */ /*0c10*/ FMUL R5, R0, c[0x0][0x188] ; /* 0x0000620000057a20 */ /* 0x004fc80000400000 */ /*0c20*/ FFMA R5, R24, c[0x0][0x184], R5 ; /* 0x0000610018057a23 */ /* 0x000fca0000000005 */ /*0c30*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe2000c101904 */ /*0c40*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0c50*/ BRA 0xc50; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0c60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ca0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ce0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z14sgemm_kernel_BPKfS0_Pfiiiff .globl _Z14sgemm_kernel_BPKfS0_Pfiiiff .p2align 8 .type _Z14sgemm_kernel_BPKfS0_Pfiiiff,@function _Z14sgemm_kernel_BPKfS0_Pfiiiff: s_clause 0x1 s_load_b32 s3, s[0:1], 0x3c s_load_b32 s2, s[0:1], 0x20 v_and_b32_e32 v2, 0x3ff, v0 v_bfe_u32 v3, v0, 10, 10 s_waitcnt lgkmcnt(0) s_and_b32 s4, s3, 0xffff s_lshr_b32 s3, s3, 16 s_delay_alu instid0(VALU_DEP_1) v_mad_u64_u32 v[0:1], null, s14, s4, v[2:3] v_mad_u64_u32 v[1:2], null, s15, s3, v[3:4] s_cmp_lt_i32 s2, 1 s_cbranch_scc1 .LBB0_3 s_load_b128 s[4:7], s[0:1], 0x0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_4) v_mul_lo_u32 v2, v1, s2 v_mov_b32_e32 v6, 0 v_mov_b32_e32 v4, v0 s_mov_b32 s3, s2 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[2:3], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v2, vcc_lo, s4, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s5, v3, vcc_lo .p2align 6 .LBB0_2: v_ashrrev_i32_e32 v5, 31, v4 s_add_i32 s3, s3, -1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) s_cmp_eq_u32 s3, 0 v_lshlrev_b64 v[7:8], 2, v[4:5] v_add_nc_u32_e32 v4, s2, v4 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_u32 v7, vcc_lo, s6, v7 v_add_co_ci_u32_e32 v8, vcc_lo, s7, v8, vcc_lo global_load_b32 v5, v[2:3], off global_load_b32 v7, v[7:8], off v_add_co_u32 v2, vcc_lo, v2, 4 v_add_co_ci_u32_e32 v3, vcc_lo, 0, v3, vcc_lo s_waitcnt vmcnt(0) v_fmac_f32_e32 v6, v5, v7 s_cbranch_scc0 .LBB0_2 s_branch .LBB0_4 .LBB0_3: v_mov_b32_e32 v6, 0 .LBB0_4: s_clause 0x2 s_load_b32 s4, s[0:1], 0x1c s_load_b64 s[2:3], s[0:1], 0x10 s_load_b64 s[0:1], s[0:1], 0x24 s_waitcnt lgkmcnt(0) v_mad_u64_u32 v[2:3], null, v1, s4, v[0:1] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[0:1], 2, v[2:3] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v0, vcc_lo, s2, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s3, v1, vcc_lo global_load_b32 v2, v[0:1], off s_waitcnt vmcnt(0) v_mul_f32_e32 v2, s1, v2 s_delay_alu instid0(VALU_DEP_1) v_fmac_f32_e32 v2, s0, v6 global_store_b32 v[0:1], v2, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z14sgemm_kernel_BPKfS0_Pfiiiff .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 304 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 9 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z14sgemm_kernel_BPKfS0_Pfiiiff, .Lfunc_end0-_Z14sgemm_kernel_BPKfS0_Pfiiiff .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value - .offset: 28 .size: 4 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: by_value - .offset: 36 .size: 4 .value_kind: by_value - .offset: 40 .size: 4 .value_kind: by_value - .offset: 48 .size: 4 .value_kind: hidden_block_count_x - .offset: 52 .size: 4 .value_kind: hidden_block_count_y - .offset: 56 .size: 4 .value_kind: hidden_block_count_z - .offset: 60 .size: 2 .value_kind: hidden_group_size_x - .offset: 62 .size: 2 .value_kind: hidden_group_size_y - .offset: 64 .size: 2 .value_kind: hidden_group_size_z - .offset: 66 .size: 2 .value_kind: hidden_remainder_x - .offset: 68 .size: 2 .value_kind: hidden_remainder_y - .offset: 70 .size: 2 .value_kind: hidden_remainder_z - .offset: 88 .size: 8 .value_kind: hidden_global_offset_x - .offset: 96 .size: 8 .value_kind: hidden_global_offset_y - .offset: 104 .size: 8 .value_kind: hidden_global_offset_z - .offset: 112 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 304 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z14sgemm_kernel_BPKfS0_Pfiiiff .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z14sgemm_kernel_BPKfS0_Pfiiiff.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 9 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_0012fe85_00000000-6_sgemm_kernel_B.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff .type _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff, @function _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff: .LFB2051: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) movss %xmm0, 8(%rsp) movss %xmm1, 4(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 20(%rsp), %rax movq %rax, 136(%rsp) leaq 16(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%rsp) leaq 8(%rsp), %rax movq %rax, 160(%rsp) leaq 4(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 184(%rsp), %rax subq %fs:40, %rax jne .L8 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z14sgemm_kernel_BPKfS0_Pfiiiff(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff, .-_Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff .globl _Z14sgemm_kernel_BPKfS0_Pfiiiff .type _Z14sgemm_kernel_BPKfS0_Pfiiiff, @function _Z14sgemm_kernel_BPKfS0_Pfiiiff: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z45__device_stub__Z14sgemm_kernel_BPKfS0_PfiiiffPKfS0_Pfiiiff addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z14sgemm_kernel_BPKfS0_Pfiiiff, .-_Z14sgemm_kernel_BPKfS0_Pfiiiff .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z14sgemm_kernel_BPKfS0_Pfiiiff" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z14sgemm_kernel_BPKfS0_Pfiiiff(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "sgemm_kernel_B.hip" .globl _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff # -- Begin function _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .p2align 4, 0x90 .type _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff,@function _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff: # @_Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) movss %xmm0, 8(%rsp) movss %xmm1, 4(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 20(%rsp), %rax movq %rax, 120(%rsp) leaq 16(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 4(%rsp), %rax movq %rax, 152(%rsp) leaq 56(%rsp), %rdi leaq 40(%rsp), %rsi leaq 32(%rsp), %rdx leaq 24(%rsp), %rcx callq __hipPopCallConfiguration movq 56(%rsp), %rsi movl 64(%rsp), %edx movq 40(%rsp), %rcx movl 48(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z14sgemm_kernel_BPKfS0_Pfiiiff, %edi pushq 24(%rsp) .cfi_adjust_cfa_offset 8 pushq 40(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end0: .size _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff, .Lfunc_end0-_Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z14sgemm_kernel_BPKfS0_Pfiiiff, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z14sgemm_kernel_BPKfS0_Pfiiiff,@object # @_Z14sgemm_kernel_BPKfS0_Pfiiiff .section .rodata,"a",@progbits .globl _Z14sgemm_kernel_BPKfS0_Pfiiiff .p2align 3, 0x0 _Z14sgemm_kernel_BPKfS0_Pfiiiff: .quad _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .size _Z14sgemm_kernel_BPKfS0_Pfiiiff, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z14sgemm_kernel_BPKfS0_Pfiiiff" .size .L__unnamed_1, 32 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z29__device_stub__sgemm_kernel_BPKfS0_Pfiiiff .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z14sgemm_kernel_BPKfS0_Pfiiiff .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include "includes.h" // helper for CUDA error handling __global__ void getMeanImage( const double* images, double* meanImage, std::size_t imageNum, std::size_t pixelNum ) { std::size_t col = blockIdx.x * blockDim.x + threadIdx.x; if(col >= pixelNum) { return; } meanImage[col] = 0.0; for(std::size_t row = 0; row < imageNum; ++row) { meanImage[col] += images[row*pixelNum + col]; } meanImage[col] /= imageNum; }
code for sm_80 Function : _Z12getMeanImagePKdPdmm .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R4, SR_CTAID.X ; /* 0x0000000000047919 */ /* 0x000e280000002500 */ /*0020*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R4, R4, c[0x0][0x0], R5 ; /* 0x0000000004047a24 */ /* 0x001fca00078e0205 */ /*0040*/ ISETP.GE.U32.AND P0, PT, R4, c[0x0][0x178], PT ; /* 0x00005e0004007a0c */ /* 0x000fc80003f06070 */ /*0050*/ ISETP.GE.U32.AND.EX P0, PT, RZ, c[0x0][0x17c], PT, P0 ; /* 0x00005f00ff007a0c */ /* 0x000fda0003f06100 */ /*0060*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0070*/ IMAD.SHL.U32 R8, R4, 0x8, RZ ; /* 0x0000000804087824 */ /* 0x000fe200078e00ff */ /*0080*/ SHF.R.U32.HI R11, RZ, 0x1d, R4 ; /* 0x0000001dff0b7819 */ /* 0x000fe20000011604 */ /*0090*/ ULDC.64 UR10, c[0x0][0x118] ; /* 0x00004600000a7ab9 */ /* 0x000fe20000000a00 */ /*00a0*/ CS2R R12, SRZ ; /* 0x00000000000c7805 */ /* 0x000fe4000001ff00 */ /*00b0*/ IADD3 R2, P0, R8, c[0x0][0x168], RZ ; /* 0x00005a0008027a10 */ /* 0x000fc80007f1e0ff */ /*00c0*/ IADD3.X R3, R11, c[0x0][0x16c], RZ, P0, !PT ; /* 0x00005b000b037a10 */ /* 0x000fe400007fe4ff */ /*00d0*/ ISETP.NE.U32.AND P0, PT, RZ, c[0x0][0x170], PT ; /* 0x00005c00ff007a0c */ /* 0x000fc60003f05070 */ /*00e0*/ STG.E.64 [R2.64], RZ ; /* 0x000000ff02007986 */ /* 0x0001e2000c101b0a */ /*00f0*/ ISETP.NE.AND.EX P0, PT, RZ, c[0x0][0x174], PT, P0 ; /* 0x00005d00ff007a0c */ /* 0x000fda0003f05300 */ /*0100*/ @!P0 BRA 0x5e0 ; /* 0x000004d000008947 */ /* 0x000fea0003800000 */ /*0110*/ IMAD.MOV.U32 R7, RZ, RZ, 0x1 ; /* 0x00000001ff077424 */ /* 0x001fe200078e00ff */ /*0120*/ CS2R R12, SRZ ; /* 0x00000000000c7805 */ /* 0x000fe2000001ff00 */ /*0130*/ IMAD.MOV.U32 R9, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff097624 */ /* 0x000fe400078e00ff */ /*0140*/ IMAD.MOV.U32 R0, RZ, RZ, 0x3 ; /* 0x00000003ff007424 */ /* 0x000fe200078e00ff */ /*0150*/ IADD3 R7, P0, -R7, c[0x0][0x170], RZ ; /* 0x00005c0007077a10 */ /* 0x000fc80007f1e1ff */ /*0160*/ ISETP.GE.U32.AND P1, PT, R7, 0x3, PT ; /* 0x000000030700780c */ /* 0x000fe40003f26070 */ /*0170*/ IADD3.X R7, R9, -0x1, RZ, P0, !PT ; /* 0xffffffff09077810 */ /* 0x000fe200007fe4ff */ /*0180*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0190*/ LOP3.LUT R6, R0, c[0x0][0x170], RZ, 0xc0, !PT ; /* 0x00005c0000067a12 */ /* 0x000fe400078ec0ff */ /*01a0*/ ISETP.GE.U32.AND.EX P1, PT, R7, RZ, PT, P1 ; /* 0x000000ff0700720c */ /* 0x000fe20003f26110 */ /*01b0*/ IMAD.MOV.U32 R7, RZ, RZ, RZ ; /* 0x000000ffff077224 */ /* 0x000fe200078e00ff */ /*01c0*/ ISETP.NE.U32.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fc80003f05070 */ /*01d0*/ ISETP.NE.AND.EX P0, PT, RZ, RZ, PT, P0 ; /* 0x000000ffff00720c */ /* 0x000fce0003f05300 */ /*01e0*/ @!P1 BRA 0x460 ; /* 0x0000027000009947 */ /* 0x000fec0003800000 */ /*01f0*/ IADD3 R10, P2, R6, -c[0x0][0x170], RZ ; /* 0x80005c00060a7a10 */ /* 0x000fe20007f5e0ff */ /*0200*/ UMOV UR6, 0x8 ; /* 0x0000000800067882 */ /* 0x000fe20000000000 */ /*0210*/ IADD3 R22, P1, R8, c[0x0][0x160], RZ ; /* 0x0000580008167a10 */ /* 0x000fe20007f3e0ff */ /*0220*/ ULDC.64 UR8, c[0x0][0x178] ; /* 0x00005e0000087ab9 */ /* 0x000fe20000000a00 */ /*0230*/ IMAD.MOV.U32 R7, RZ, RZ, RZ ; /* 0x000000ffff077224 */ /* 0x000fe200078e00ff */ /*0240*/ UIMAD.WIDE.U32 UR4, UR6, UR8, URZ ; /* 0x00000008060472a5 */ /* 0x000fe2000f8e003f */ /*0250*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0260*/ UIMAD UR6, UR6, UR9, URZ ; /* 0x00000009060672a4 */ /* 0x000fe2000f8e023f */ /*0270*/ CS2R R12, SRZ ; /* 0x00000000000c7805 */ /* 0x000fe2000001ff00 */ /*0280*/ IADD3.X R23, R11, c[0x0][0x164], RZ, P1, !PT ; /* 0x000059000b177a10 */ /* 0x000fe20000ffe4ff */ /*0290*/ IMAD.X R8, RZ, RZ, ~c[0x0][0x174], P2 ; /* 0x80005d00ff087624 */ /* 0x000fe200010e06ff */ /*02a0*/ UIADD3 UR6, UR5, UR6, URZ ; /* 0x0000000605067290 */ /* 0x000fc6000fffe03f */ /*02b0*/ LDG.E.64 R14, [R22.64] ; /* 0x0000000a160e7981 */ /* 0x000ea2000c1e1b00 */ /*02c0*/ IADD3 R16, P1, R22, UR4, RZ ; /* 0x0000000416107c10 */ /* 0x000fc8000ff3e0ff */ /*02d0*/ IADD3.X R17, R23, UR6, RZ, P1, !PT ; /* 0x0000000617117c10 */ /* 0x000fe20008ffe4ff */ /*02e0*/ DADD R12, R14, R12 ; /* 0x000000000e0c7229 */ /* 0x006e0e000000000c */ /*02f0*/ STG.E.64 [R2.64], R12 ; /* 0x0000000c02007986 */ /* 0x0011e8000c101b0a */ /*0300*/ LDG.E.64 R14, [R16.64] ; /* 0x0000000a100e7981 */ /* 0x000ea2000c1e1b00 */ /*0310*/ IADD3 R24, P1, R16, UR4, RZ ; /* 0x0000000410187c10 */ /* 0x000fc8000ff3e0ff */ /*0320*/ IADD3.X R25, R17, UR6, RZ, P1, !PT ; /* 0x0000000611197c10 */ /* 0x000fe20008ffe4ff */ /*0330*/ DADD R18, R12, R14 ; /* 0x000000000c127229 */ /* 0x004e4e000000000e */ /*0340*/ STG.E.64 [R2.64], R18 ; /* 0x0000001202007986 */ /* 0x0023e8000c101b0a */ /*0350*/ LDG.E.64 R20, [R24.64] ; /* 0x0000000a18147981 */ /* 0x000ea2000c1e1b00 */ /*0360*/ IADD3 R14, P1, R24, UR4, RZ ; /* 0x00000004180e7c10 */ /* 0x000fc8000ff3e0ff */ /*0370*/ IADD3.X R15, R25, UR6, RZ, P1, !PT ; /* 0x00000006190f7c10 */ /* 0x000fe20008ffe4ff */ /*0380*/ DADD R20, R18, R20 ; /* 0x0000000012147229 */ /* 0x004e8e0000000014 */ /*0390*/ STG.E.64 [R2.64], R20 ; /* 0x0000001402007986 */ /* 0x0043e8000c101b0a */ /*03a0*/ LDG.E.64 R12, [R14.64] ; /* 0x0000000a0e0c7981 */ /* 0x001ea2000c1e1b00 */ /*03b0*/ IADD3 R7, P1, R7, 0x4, RZ ; /* 0x0000000407077810 */ /* 0x000fc80007f3e0ff */ /*03c0*/ IADD3 R11, P2, R7, R10, RZ ; /* 0x0000000a070b7210 */ /* 0x000fe20007f5e0ff */ /*03d0*/ IMAD.X R9, RZ, RZ, R9, P1 ; /* 0x000000ffff097224 */ /* 0x000fc600008e0609 */ /*03e0*/ ISETP.NE.U32.AND P1, PT, R11, RZ, PT ; /* 0x000000ff0b00720c */ /* 0x000fe20003f25070 */ /*03f0*/ IMAD.X R11, R9, 0x1, R8, P2 ; /* 0x00000001090b7824 */ /* 0x000fe200010e0608 */ /*0400*/ IADD3 R22, P2, R14, UR4, RZ ; /* 0x000000040e167c10 */ /* 0x000fc8000ff5e0ff */ /*0410*/ ISETP.NE.AND.EX P1, PT, R11, RZ, PT, P1 ; /* 0x000000ff0b00720c */ /* 0x000fe40003f25310 */ /*0420*/ IADD3.X R23, R15, UR6, RZ, P2, !PT ; /* 0x000000060f177c10 */ /* 0x000fe200097fe4ff */ /*0430*/ DADD R12, R20, R12 ; /* 0x00000000140c7229 */ /* 0x004e0e000000000c */ /*0440*/ STG.E.64 [R2.64], R12 ; /* 0x0000000c02007986 */ /* 0x0013e6000c101b0a */ /*0450*/ @P1 BRA 0x2b0 ; /* 0xfffffe5000001947 */ /* 0x000fea000383ffff */ /*0460*/ @!P0 BRA 0x5e0 ; /* 0x0000017000008947 */ /* 0x000fea0003800000 */ /*0470*/ IMAD R8, R9, c[0x0][0x178], RZ ; /* 0x00005e0009087a24 */ /* 0x000fe200078e02ff */ /*0480*/ IADD3 R6, P1, RZ, -R6, RZ ; /* 0x80000006ff067210 */ /* 0x000fe20007f3e0ff */ /*0490*/ IMAD.MOV.U32 R5, RZ, RZ, RZ ; /* 0x000000ffff057224 */ /* 0x000fe400078e00ff */ /*04a0*/ IMAD R9, R7.reuse, c[0x0][0x17c], R8 ; /* 0x00005f0007097a24 */ /* 0x040fe400078e0208 */ /*04b0*/ IMAD.WIDE.U32 R4, R7, c[0x0][0x178], R4 ; /* 0x00005e0007047a25 */ /* 0x000fc800078e0004 */ /*04c0*/ IMAD.MOV.U32 R10, RZ, RZ, c[0x0][0x178] ; /* 0x00005e00ff0a7624 */ /* 0x000fe200078e00ff */ /*04d0*/ LEA R7, P0, R4, c[0x0][0x160], 0x3 ; /* 0x0000580004077a11 */ /* 0x000fe200078018ff */ /*04e0*/ IMAD.IADD R5, R5, 0x1, R9 ; /* 0x0000000105057824 */ /* 0x000fc600078e0209 */ /*04f0*/ SHF.L.U64.HI R9, R10, R0, c[0x0][0x17c] ; /* 0x00005f000a097619 */ /* 0x000fe20000010200 */ /*0500*/ IMAD.X R0, RZ, RZ, -0x1, P1 ; /* 0xffffffffff007424 */ /* 0x000fe200008e06ff */ /*0510*/ LEA.HI.X R8, R4, c[0x0][0x164], R5, 0x3, P0 ; /* 0x0000590004087a11 */ /* 0x000fc600000f1c05 */ /*0520*/ IMAD.MOV.U32 R4, RZ, RZ, R7 ; /* 0x000000ffff047224 */ /* 0x000fe400078e0007 */ /*0530*/ IMAD.MOV.U32 R5, RZ, RZ, R8 ; /* 0x000000ffff057224 */ /* 0x000fcc00078e0008 */ /*0540*/ LDG.E.64 R4, [R4.64] ; /* 0x0000000a04047981 */ /* 0x000ea2000c1e1b00 */ /*0550*/ IADD3 R6, P0, R6, 0x1, RZ ; /* 0x0000000106067810 */ /* 0x000fe40007f1e0ff */ /*0560*/ LEA R7, P1, R10, R7, 0x3 ; /* 0x000000070a077211 */ /* 0x000fc600078218ff */ /*0570*/ IMAD.X R0, RZ, RZ, R0, P0 ; /* 0x000000ffff007224 */ /* 0x000fe200000e0600 */ /*0580*/ ISETP.NE.U32.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f05070 */ /*0590*/ IMAD.X R8, R8, 0x1, R9, P1 ; /* 0x0000000108087824 */ /* 0x000fc600008e0609 */ /*05a0*/ ISETP.NE.AND.EX P0, PT, R0, RZ, PT, P0 ; /* 0x000000ff0000720c */ /* 0x000fe20003f05300 */ /*05b0*/ DADD R12, R4, R12 ; /* 0x00000000040c7229 */ /* 0x007e0e000000000c */ /*05c0*/ STG.E.64 [R2.64], R12 ; /* 0x0000000c02007986 */ /* 0x0011ea000c101b0a */ /*05d0*/ @P0 BRA 0x520 ; /* 0xffffff4000000947 */ /* 0x000fea000383ffff */ /*05e0*/ I2F.F64.U64 R8, c[0x0][0x170] ; /* 0x00005c0000087b12 */ /* 0x001e220000301800 */ /*05f0*/ IMAD.MOV.U32 R4, RZ, RZ, 0x1 ; /* 0x00000001ff047424 */ /* 0x000fe200078e00ff */ /*0600*/ FSETP.GEU.AND P1, PT, |R13|, 6.5827683646048100446e-37, PT ; /* 0x036000000d00780b */ /* 0x000fe20003f2e200 */ /*0610*/ BSSY B0, 0x750 ; /* 0x0000013000007945 */ /* 0x000fea0003800000 */ /*0620*/ MUFU.RCP64H R5, R9 ; /* 0x0000000900057308 */ /* 0x001e240000001800 */ /*0630*/ DFMA R6, -R8, R4, 1 ; /* 0x3ff000000806742b */ /* 0x001e0c0000000104 */ /*0640*/ DFMA R6, R6, R6, R6 ; /* 0x000000060606722b */ /* 0x001e0c0000000006 */ /*0650*/ DFMA R6, R4, R6, R4 ; /* 0x000000060406722b */ /* 0x001e0c0000000004 */ /*0660*/ DFMA R4, -R8, R6, 1 ; /* 0x3ff000000804742b */ /* 0x001e0c0000000106 */ /*0670*/ DFMA R4, R6, R4, R6 ; /* 0x000000040604722b */ /* 0x001e0c0000000006 */ /*0680*/ DMUL R6, R4, R12 ; /* 0x0000000c04067228 */ /* 0x001e0c0000000000 */ /*0690*/ DFMA R10, -R8, R6, R12 ; /* 0x00000006080a722b */ /* 0x001e0c000000010c */ /*06a0*/ DFMA R4, R4, R10, R6 ; /* 0x0000000a0404722b */ /* 0x001e140000000006 */ /*06b0*/ FFMA R0, RZ, R9, R5 ; /* 0x00000009ff007223 */ /* 0x001fca0000000005 */ /*06c0*/ FSETP.GT.AND P0, PT, |R0|, 1.469367938527859385e-39, PT ; /* 0x001000000000780b */ /* 0x000fda0003f04200 */ /*06d0*/ @P0 BRA P1, 0x740 ; /* 0x0000006000000947 */ /* 0x000fea0000800000 */ /*06e0*/ IMAD.MOV.U32 R6, RZ, RZ, R12 ; /* 0x000000ffff067224 */ /* 0x000fe200078e000c */ /*06f0*/ MOV R0, 0x720 ; /* 0x0000072000007802 */ /* 0x000fe20000000f00 */ /*0700*/ IMAD.MOV.U32 R7, RZ, RZ, R13 ; /* 0x000000ffff077224 */ /* 0x000fe400078e000d */ /*0710*/ CALL.REL.NOINC 0x770 ; /* 0x0000005000007944 */ /* 0x002fea0003c00000 */ /*0720*/ IMAD.MOV.U32 R4, RZ, RZ, R12 ; /* 0x000000ffff047224 */ /* 0x000fe400078e000c */ /*0730*/ IMAD.MOV.U32 R5, RZ, RZ, R13 ; /* 0x000000ffff057224 */ /* 0x000fe400078e000d */ /*0740*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0750*/ STG.E.64 [R2.64], R4 ; /* 0x0000000402007986 */ /* 0x000fe2000c101b0a */ /*0760*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0770*/ FSETP.GEU.AND P0, PT, |R9|.reuse, 1.469367938527859385e-39, PT ; /* 0x001000000900780b */ /* 0x040fe20003f0e200 */ /*0780*/ IMAD.MOV.U32 R13, RZ, RZ, 0x1ca00000 ; /* 0x1ca00000ff0d7424 */ /* 0x000fe200078e00ff */ /*0790*/ LOP3.LUT R4, R9, 0x800fffff, RZ, 0xc0, !PT ; /* 0x800fffff09047812 */ /* 0x000fe200078ec0ff */ /*07a0*/ IMAD.MOV.U32 R16, RZ, RZ, 0x1 ; /* 0x00000001ff107424 */ /* 0x000fe200078e00ff */ /*07b0*/ FSETP.GEU.AND P2, PT, |R7|.reuse, 1.469367938527859385e-39, PT ; /* 0x001000000700780b */ /* 0x040fe20003f4e200 */ /*07c0*/ IMAD.MOV.U32 R10, RZ, RZ, R6 ; /* 0x000000ffff0a7224 */ /* 0x000fe200078e0006 */ /*07d0*/ LOP3.LUT R5, R4, 0x3ff00000, RZ, 0xfc, !PT ; /* 0x3ff0000004057812 */ /* 0x000fe200078efcff */ /*07e0*/ IMAD.MOV.U32 R4, RZ, RZ, R8 ; /* 0x000000ffff047224 */ /* 0x000fe200078e0008 */ /*07f0*/ LOP3.LUT R12, R7, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff00000070c7812 */ /* 0x000fe200078ec0ff */ /*0800*/ BSSY B1, 0xd10 ; /* 0x0000050000017945 */ /* 0x000fe20003800000 */ /*0810*/ LOP3.LUT R15, R9, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff00000090f7812 */ /* 0x000fc600078ec0ff */ /*0820*/ @!P0 DMUL R4, R8, 8.98846567431157953865e+307 ; /* 0x7fe0000008048828 */ /* 0x000e220000000000 */ /*0830*/ ISETP.GE.U32.AND P1, PT, R12, R15, PT ; /* 0x0000000f0c00720c */ /* 0x000fc60003f26070 */ /*0840*/ @!P2 LOP3.LUT R11, R9, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff00000090ba812 */ /* 0x000fe200078ec0ff */ /*0850*/ @!P2 IMAD.MOV.U32 R20, RZ, RZ, RZ ; /* 0x000000ffff14a224 */ /* 0x000fe200078e00ff */ /*0860*/ MUFU.RCP64H R17, R5 ; /* 0x0000000500117308 */ /* 0x001e240000001800 */ /*0870*/ @!P2 ISETP.GE.U32.AND P3, PT, R12, R11, PT ; /* 0x0000000b0c00a20c */ /* 0x000fe40003f66070 */ /*0880*/ SEL R11, R13.reuse, 0x63400000, !P1 ; /* 0x634000000d0b7807 */ /* 0x040fe40004800000 */ /*0890*/ @!P2 SEL R21, R13, 0x63400000, !P3 ; /* 0x634000000d15a807 */ /* 0x000fe40005800000 */ /*08a0*/ LOP3.LUT R11, R11, 0x800fffff, R7, 0xf8, !PT ; /* 0x800fffff0b0b7812 */ /* 0x000fc400078ef807 */ /*08b0*/ @!P2 LOP3.LUT R21, R21, 0x80000000, R7, 0xf8, !PT ; /* 0x800000001515a812 */ /* 0x000fc800078ef807 */ /*08c0*/ @!P2 LOP3.LUT R21, R21, 0x100000, RZ, 0xfc, !PT ; /* 0x001000001515a812 */ /* 0x000fe200078efcff */ /*08d0*/ DFMA R18, R16, -R4, 1 ; /* 0x3ff000001012742b */ /* 0x001e0a0000000804 */ /*08e0*/ @!P2 DFMA R10, R10, 2, -R20 ; /* 0x400000000a0aa82b */ /* 0x000fc80000000814 */ /*08f0*/ DFMA R18, R18, R18, R18 ; /* 0x000000121212722b */ /* 0x001e0c0000000012 */ /*0900*/ DFMA R18, R16, R18, R16 ; /* 0x000000121012722b */ /* 0x0010640000000010 */ /*0910*/ IMAD.MOV.U32 R16, RZ, RZ, R12 ; /* 0x000000ffff107224 */ /* 0x001fe200078e000c */ /*0920*/ @!P2 LOP3.LUT R16, R11, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff000000b10a812 */ /* 0x000fe200078ec0ff */ /*0930*/ IMAD.MOV.U32 R17, RZ, RZ, R15 ; /* 0x000000ffff117224 */ /* 0x000fe200078e000f */ /*0940*/ @!P0 LOP3.LUT R17, R5, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff0000005118812 */ /* 0x000fe200078ec0ff */ /*0950*/ DFMA R20, R18, -R4, 1 ; /* 0x3ff000001214742b */ /* 0x002e060000000804 */ /*0960*/ IADD3 R22, R17, -0x1, RZ ; /* 0xffffffff11167810 */ /* 0x000fc60007ffe0ff */ /*0970*/ DFMA R18, R18, R20, R18 ; /* 0x000000141212722b */ /* 0x0010640000000012 */ /*0980*/ IADD3 R20, R16, -0x1, RZ ; /* 0xffffffff10147810 */ /* 0x001fc80007ffe0ff */ /*0990*/ ISETP.GT.U32.AND P0, PT, R20, 0x7feffffe, PT ; /* 0x7feffffe1400780c */ /* 0x000fe20003f04070 */ /*09a0*/ DMUL R14, R18, R10 ; /* 0x0000000a120e7228 */ /* 0x002e060000000000 */ /*09b0*/ ISETP.GT.U32.OR P0, PT, R22, 0x7feffffe, P0 ; /* 0x7feffffe1600780c */ /* 0x000fc60000704470 */ /*09c0*/ DFMA R20, R14, -R4, R10 ; /* 0x800000040e14722b */ /* 0x001e0c000000000a */ /*09d0*/ DFMA R14, R18, R20, R14 ; /* 0x00000014120e722b */ /* 0x001048000000000e */ /*09e0*/ @P0 BRA 0xbb0 ; /* 0x000001c000000947 */ /* 0x000fea0003800000 */ /*09f0*/ LOP3.LUT R7, R9, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff0000009077812 */ /* 0x003fc800078ec0ff */ /*0a00*/ ISETP.GE.U32.AND P0, PT, R12.reuse, R7, PT ; /* 0x000000070c00720c */ /* 0x040fe20003f06070 */ /*0a10*/ IMAD.IADD R6, R12, 0x1, -R7 ; /* 0x000000010c067824 */ /* 0x000fc600078e0a07 */ /*0a20*/ SEL R13, R13, 0x63400000, !P0 ; /* 0x634000000d0d7807 */ /* 0x000fe40004000000 */ /*0a30*/ IMNMX R6, R6, -0x46a00000, !PT ; /* 0xb960000006067817 */ /* 0x000fc80007800200 */ /*0a40*/ IMNMX R6, R6, 0x46a00000, PT ; /* 0x46a0000006067817 */ /* 0x000fca0003800200 */ /*0a50*/ IMAD.IADD R16, R6, 0x1, -R13 ; /* 0x0000000106107824 */ /* 0x000fe400078e0a0d */ /*0a60*/ IMAD.MOV.U32 R6, RZ, RZ, RZ ; /* 0x000000ffff067224 */ /* 0x000fc600078e00ff */ /*0a70*/ IADD3 R7, R16, 0x7fe00000, RZ ; /* 0x7fe0000010077810 */ /* 0x000fcc0007ffe0ff */ /*0a80*/ DMUL R12, R14, R6 ; /* 0x000000060e0c7228 */ /* 0x000e140000000000 */ /*0a90*/ FSETP.GTU.AND P0, PT, |R13|, 1.469367938527859385e-39, PT ; /* 0x001000000d00780b */ /* 0x001fda0003f0c200 */ /*0aa0*/ @P0 BRA 0xd00 ; /* 0x0000025000000947 */ /* 0x000fea0003800000 */ /*0ab0*/ DFMA R4, R14, -R4, R10 ; /* 0x800000040e04722b */ /* 0x000e22000000000a */ /*0ac0*/ IMAD.MOV.U32 R6, RZ, RZ, RZ ; /* 0x000000ffff067224 */ /* 0x000fd200078e00ff */ /*0ad0*/ FSETP.NEU.AND P0, PT, R5.reuse, RZ, PT ; /* 0x000000ff0500720b */ /* 0x041fe40003f0d000 */ /*0ae0*/ LOP3.LUT R9, R5, 0x80000000, R9, 0x48, !PT ; /* 0x8000000005097812 */ /* 0x000fc800078e4809 */ /*0af0*/ LOP3.LUT R7, R9, R7, RZ, 0xfc, !PT ; /* 0x0000000709077212 */ /* 0x000fce00078efcff */ /*0b00*/ @!P0 BRA 0xd00 ; /* 0x000001f000008947 */ /* 0x000fea0003800000 */ /*0b10*/ IMAD.MOV R5, RZ, RZ, -R16 ; /* 0x000000ffff057224 */ /* 0x000fe200078e0a10 */ /*0b20*/ DMUL.RP R6, R14, R6 ; /* 0x000000060e067228 */ /* 0x000e220000008000 */ /*0b30*/ IMAD.MOV.U32 R4, RZ, RZ, RZ ; /* 0x000000ffff047224 */ /* 0x000fcc00078e00ff */ /*0b40*/ DFMA R4, R12, -R4, R14 ; /* 0x800000040c04722b */ /* 0x000e46000000000e */ /*0b50*/ LOP3.LUT R9, R7, R9, RZ, 0x3c, !PT ; /* 0x0000000907097212 */ /* 0x001fc600078e3cff */ /*0b60*/ IADD3 R4, -R16, -0x43300000, RZ ; /* 0xbcd0000010047810 */ /* 0x002fc80007ffe1ff */ /*0b70*/ FSETP.NEU.AND P0, PT, |R5|, R4, PT ; /* 0x000000040500720b */ /* 0x000fc80003f0d200 */ /*0b80*/ FSEL R12, R6, R12, !P0 ; /* 0x0000000c060c7208 */ /* 0x000fe40004000000 */ /*0b90*/ FSEL R13, R9, R13, !P0 ; /* 0x0000000d090d7208 */ /* 0x000fe20004000000 */ /*0ba0*/ BRA 0xd00 ; /* 0x0000015000007947 */ /* 0x000fea0003800000 */ /*0bb0*/ DSETP.NAN.AND P0, PT, R6, R6, PT ; /* 0x000000060600722a */ /* 0x003e1c0003f08000 */ /*0bc0*/ @P0 BRA 0xce0 ; /* 0x0000011000000947 */ /* 0x001fea0003800000 */ /*0bd0*/ DSETP.NAN.AND P0, PT, R8, R8, PT ; /* 0x000000080800722a */ /* 0x000e1c0003f08000 */ /*0be0*/ @P0 BRA 0xcb0 ; /* 0x000000c000000947 */ /* 0x001fea0003800000 */ /*0bf0*/ ISETP.NE.AND P0, PT, R16, R17, PT ; /* 0x000000111000720c */ /* 0x000fe20003f05270 */ /*0c00*/ IMAD.MOV.U32 R12, RZ, RZ, 0x0 ; /* 0x00000000ff0c7424 */ /* 0x000fe400078e00ff */ /*0c10*/ IMAD.MOV.U32 R13, RZ, RZ, -0x80000 ; /* 0xfff80000ff0d7424 */ /* 0x000fd400078e00ff */ /*0c20*/ @!P0 BRA 0xd00 ; /* 0x000000d000008947 */ /* 0x000fea0003800000 */ /*0c30*/ ISETP.NE.AND P0, PT, R16, 0x7ff00000, PT ; /* 0x7ff000001000780c */ /* 0x000fe40003f05270 */ /*0c40*/ LOP3.LUT R13, R7, 0x80000000, R9, 0x48, !PT ; /* 0x80000000070d7812 */ /* 0x000fe400078e4809 */ /*0c50*/ ISETP.EQ.OR P0, PT, R17, RZ, !P0 ; /* 0x000000ff1100720c */ /* 0x000fda0004702670 */ /*0c60*/ @P0 LOP3.LUT R4, R13, 0x7ff00000, RZ, 0xfc, !PT ; /* 0x7ff000000d040812 */ /* 0x000fe200078efcff */ /*0c70*/ @!P0 IMAD.MOV.U32 R12, RZ, RZ, RZ ; /* 0x000000ffff0c8224 */ /* 0x000fe400078e00ff */ /*0c80*/ @P0 IMAD.MOV.U32 R12, RZ, RZ, RZ ; /* 0x000000ffff0c0224 */ /* 0x000fe400078e00ff */ /*0c90*/ @P0 IMAD.MOV.U32 R13, RZ, RZ, R4 ; /* 0x000000ffff0d0224 */ /* 0x000fe200078e0004 */ /*0ca0*/ BRA 0xd00 ; /* 0x0000005000007947 */ /* 0x000fea0003800000 */ /*0cb0*/ LOP3.LUT R13, R9, 0x80000, RZ, 0xfc, !PT ; /* 0x00080000090d7812 */ /* 0x000fe200078efcff */ /*0cc0*/ IMAD.MOV.U32 R12, RZ, RZ, R8 ; /* 0x000000ffff0c7224 */ /* 0x000fe200078e0008 */ /*0cd0*/ BRA 0xd00 ; /* 0x0000002000007947 */ /* 0x000fea0003800000 */ /*0ce0*/ LOP3.LUT R13, R7, 0x80000, RZ, 0xfc, !PT ; /* 0x00080000070d7812 */ /* 0x000fe200078efcff */ /*0cf0*/ IMAD.MOV.U32 R12, RZ, RZ, R6 ; /* 0x000000ffff0c7224 */ /* 0x000fe400078e0006 */ /*0d00*/ BSYNC B1 ; /* 0x0000000000017941 */ /* 0x000fea0003800000 */ /*0d10*/ IMAD.MOV.U32 R4, RZ, RZ, R0 ; /* 0x000000ffff047224 */ /* 0x000fe400078e0000 */ /*0d20*/ IMAD.MOV.U32 R5, RZ, RZ, 0x0 ; /* 0x00000000ff057424 */ /* 0x000fc800078e00ff */ /*0d30*/ RET.REL.NODEC R4 0x0 ; /* 0xfffff2c004007950 */ /* 0x000fea0003c3ffff */ /*0d40*/ BRA 0xd40; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0d50*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0da0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0db0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0dc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0dd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0de0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0df0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include "includes.h" // helper for CUDA error handling __global__ void getMeanImage( const double* images, double* meanImage, std::size_t imageNum, std::size_t pixelNum ) { std::size_t col = blockIdx.x * blockDim.x + threadIdx.x; if(col >= pixelNum) { return; } meanImage[col] = 0.0; for(std::size_t row = 0; row < imageNum; ++row) { meanImage[col] += images[row*pixelNum + col]; } meanImage[col] /= imageNum; }
.file "tmpxft_0010e20e_00000000-6_getMeanImage.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm .type _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm, @function _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movq %rcx, (%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) movq %rsp, %rax movq %rax, 120(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 136(%rsp), %rax subq %fs:40, %rax jne .L8 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 168 pushq 40(%rsp) .cfi_def_cfa_offset 176 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z12getMeanImagePKdPdmm(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm, .-_Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm .globl _Z12getMeanImagePKdPdmm .type _Z12getMeanImagePKdPdmm, @function _Z12getMeanImagePKdPdmm: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z12getMeanImagePKdPdmm, .-_Z12getMeanImagePKdPdmm .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z12getMeanImagePKdPdmm" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z12getMeanImagePKdPdmm(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include "includes.h" // helper for CUDA error handling __global__ void getMeanImage( const double* images, double* meanImage, std::size_t imageNum, std::size_t pixelNum ) { std::size_t col = blockIdx.x * blockDim.x + threadIdx.x; if(col >= pixelNum) { return; } meanImage[col] = 0.0; for(std::size_t row = 0; row < imageNum; ++row) { meanImage[col] += images[row*pixelNum + col]; } meanImage[col] /= imageNum; }
#include <hip/hip_runtime.h> #include "includes.h" // helper for CUDA error handling __global__ void getMeanImage( const double* images, double* meanImage, std::size_t imageNum, std::size_t pixelNum ) { std::size_t col = blockIdx.x * blockDim.x + threadIdx.x; if(col >= pixelNum) { return; } meanImage[col] = 0.0; for(std::size_t row = 0; row < imageNum; ++row) { meanImage[col] += images[row*pixelNum + col]; } meanImage[col] /= imageNum; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include "includes.h" // helper for CUDA error handling __global__ void getMeanImage( const double* images, double* meanImage, std::size_t imageNum, std::size_t pixelNum ) { std::size_t col = blockIdx.x * blockDim.x + threadIdx.x; if(col >= pixelNum) { return; } meanImage[col] = 0.0; for(std::size_t row = 0; row < imageNum; ++row) { meanImage[col] += images[row*pixelNum + col]; } meanImage[col] /= imageNum; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z12getMeanImagePKdPdmm .globl _Z12getMeanImagePKdPdmm .p2align 8 .type _Z12getMeanImagePKdPdmm,@function _Z12getMeanImagePKdPdmm: s_clause 0x1 s_load_b32 s4, s[0:1], 0x2c s_load_b64 s[2:3], s[0:1], 0x18 s_waitcnt lgkmcnt(0) s_and_b32 s4, s4, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s4, v[0:1] v_mov_b32_e32 v2, 0 s_mov_b32 s4, exec_lo v_cmpx_gt_u64_e64 s[2:3], v[1:2] s_cbranch_execz .LBB0_5 s_load_b128 s[4:7], s[0:1], 0x8 v_lshlrev_b64 v[4:5], 3, v[1:2] v_mov_b32_e32 v2, 0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_mov_b32_e32 v3, v2 s_waitcnt lgkmcnt(0) v_add_co_u32 v0, vcc_lo, s4, v4 s_delay_alu instid0(VALU_DEP_4) v_add_co_ci_u32_e32 v1, vcc_lo, s5, v5, vcc_lo s_cmp_eq_u64 s[6:7], 0 global_store_b64 v[0:1], v[2:3], off s_cbranch_scc1 .LBB0_4 global_load_b64 v[2:3], v[0:1], off s_load_b64 s[0:1], s[0:1], 0x0 s_waitcnt lgkmcnt(0) v_add_co_u32 v4, vcc_lo, s0, v4 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v5, vcc_lo s_lshl_b64 s[0:1], s[2:3], 3 s_mov_b64 s[2:3], s[6:7] .LBB0_3: global_load_b64 v[6:7], v[4:5], off v_add_co_u32 v4, vcc_lo, v4, s0 s_add_u32 s2, s2, -1 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v5, vcc_lo s_addc_u32 s3, s3, -1 s_delay_alu instid0(SALU_CYCLE_1) s_cmp_eq_u64 s[2:3], 0 s_waitcnt vmcnt(0) v_add_f64 v[2:3], v[6:7], v[2:3] global_store_b64 v[0:1], v[2:3], off s_cbranch_scc0 .LBB0_3 .LBB0_4: global_load_b64 v[2:3], v[0:1], off v_cvt_f64_u32_e32 v[4:5], s7 v_cvt_f64_u32_e32 v[6:7], s6 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_ldexp_f64 v[4:5], v[4:5], 32 v_add_f64 v[4:5], v[4:5], v[6:7] s_waitcnt vmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_div_scale_f64 v[6:7], null, v[4:5], v[4:5], v[2:3] v_rcp_f64_e32 v[8:9], v[6:7] s_waitcnt_depctr 0xfff v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_div_scale_f64 v[10:11], vcc_lo, v[2:3], v[4:5], v[2:3] v_mul_f64 v[12:13], v[10:11], v[8:9] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[6:7], -v[6:7], v[12:13], v[10:11] v_div_fmas_f64 v[6:7], v[6:7], v[8:9], v[12:13] s_delay_alu instid0(VALU_DEP_1) v_div_fixup_f64 v[2:3], v[6:7], v[4:5], v[2:3] global_store_b64 v[0:1], v[2:3], off .LBB0_5: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z12getMeanImagePKdPdmm .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 288 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z12getMeanImagePKdPdmm, .Lfunc_end0-_Z12getMeanImagePKdPdmm .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .offset: 16 .size: 8 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: hidden_block_count_x - .offset: 36 .size: 4 .value_kind: hidden_block_count_y - .offset: 40 .size: 4 .value_kind: hidden_block_count_z - .offset: 44 .size: 2 .value_kind: hidden_group_size_x - .offset: 46 .size: 2 .value_kind: hidden_group_size_y - .offset: 48 .size: 2 .value_kind: hidden_group_size_z - .offset: 50 .size: 2 .value_kind: hidden_remainder_x - .offset: 52 .size: 2 .value_kind: hidden_remainder_y - .offset: 54 .size: 2 .value_kind: hidden_remainder_z - .offset: 72 .size: 8 .value_kind: hidden_global_offset_x - .offset: 80 .size: 8 .value_kind: hidden_global_offset_y - .offset: 88 .size: 8 .value_kind: hidden_global_offset_z - .offset: 96 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 288 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z12getMeanImagePKdPdmm .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z12getMeanImagePKdPdmm.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include "includes.h" // helper for CUDA error handling __global__ void getMeanImage( const double* images, double* meanImage, std::size_t imageNum, std::size_t pixelNum ) { std::size_t col = blockIdx.x * blockDim.x + threadIdx.x; if(col >= pixelNum) { return; } meanImage[col] = 0.0; for(std::size_t row = 0; row < imageNum; ++row) { meanImage[col] += images[row*pixelNum + col]; } meanImage[col] /= imageNum; }
.text .file "getMeanImage.hip" .globl _Z27__device_stub__getMeanImagePKdPdmm # -- Begin function _Z27__device_stub__getMeanImagePKdPdmm .p2align 4, 0x90 .type _Z27__device_stub__getMeanImagePKdPdmm,@function _Z27__device_stub__getMeanImagePKdPdmm: # @_Z27__device_stub__getMeanImagePKdPdmm .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) movq %rcx, 48(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 48(%rsp), %rax movq %rax, 104(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z12getMeanImagePKdPdmm, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z27__device_stub__getMeanImagePKdPdmm, .Lfunc_end0-_Z27__device_stub__getMeanImagePKdPdmm .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12getMeanImagePKdPdmm, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z12getMeanImagePKdPdmm,@object # @_Z12getMeanImagePKdPdmm .section .rodata,"a",@progbits .globl _Z12getMeanImagePKdPdmm .p2align 3, 0x0 _Z12getMeanImagePKdPdmm: .quad _Z27__device_stub__getMeanImagePKdPdmm .size _Z12getMeanImagePKdPdmm, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z12getMeanImagePKdPdmm" .size .L__unnamed_1, 24 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z27__device_stub__getMeanImagePKdPdmm .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z12getMeanImagePKdPdmm .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z12getMeanImagePKdPdmm .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R4, SR_CTAID.X ; /* 0x0000000000047919 */ /* 0x000e280000002500 */ /*0020*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R4, R4, c[0x0][0x0], R5 ; /* 0x0000000004047a24 */ /* 0x001fca00078e0205 */ /*0040*/ ISETP.GE.U32.AND P0, PT, R4, c[0x0][0x178], PT ; /* 0x00005e0004007a0c */ /* 0x000fc80003f06070 */ /*0050*/ ISETP.GE.U32.AND.EX P0, PT, RZ, c[0x0][0x17c], PT, P0 ; /* 0x00005f00ff007a0c */ /* 0x000fda0003f06100 */ /*0060*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0070*/ IMAD.SHL.U32 R8, R4, 0x8, RZ ; /* 0x0000000804087824 */ /* 0x000fe200078e00ff */ /*0080*/ SHF.R.U32.HI R11, RZ, 0x1d, R4 ; /* 0x0000001dff0b7819 */ /* 0x000fe20000011604 */ /*0090*/ ULDC.64 UR10, c[0x0][0x118] ; /* 0x00004600000a7ab9 */ /* 0x000fe20000000a00 */ /*00a0*/ CS2R R12, SRZ ; /* 0x00000000000c7805 */ /* 0x000fe4000001ff00 */ /*00b0*/ IADD3 R2, P0, R8, c[0x0][0x168], RZ ; /* 0x00005a0008027a10 */ /* 0x000fc80007f1e0ff */ /*00c0*/ IADD3.X R3, R11, c[0x0][0x16c], RZ, P0, !PT ; /* 0x00005b000b037a10 */ /* 0x000fe400007fe4ff */ /*00d0*/ ISETP.NE.U32.AND P0, PT, RZ, c[0x0][0x170], PT ; /* 0x00005c00ff007a0c */ /* 0x000fc60003f05070 */ /*00e0*/ STG.E.64 [R2.64], RZ ; /* 0x000000ff02007986 */ /* 0x0001e2000c101b0a */ /*00f0*/ ISETP.NE.AND.EX P0, PT, RZ, c[0x0][0x174], PT, P0 ; /* 0x00005d00ff007a0c */ /* 0x000fda0003f05300 */ /*0100*/ @!P0 BRA 0x5e0 ; /* 0x000004d000008947 */ /* 0x000fea0003800000 */ /*0110*/ IMAD.MOV.U32 R7, RZ, RZ, 0x1 ; /* 0x00000001ff077424 */ /* 0x001fe200078e00ff */ /*0120*/ CS2R R12, SRZ ; /* 0x00000000000c7805 */ /* 0x000fe2000001ff00 */ /*0130*/ IMAD.MOV.U32 R9, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff097624 */ /* 0x000fe400078e00ff */ /*0140*/ IMAD.MOV.U32 R0, RZ, RZ, 0x3 ; /* 0x00000003ff007424 */ /* 0x000fe200078e00ff */ /*0150*/ IADD3 R7, P0, -R7, c[0x0][0x170], RZ ; /* 0x00005c0007077a10 */ /* 0x000fc80007f1e1ff */ /*0160*/ ISETP.GE.U32.AND P1, PT, R7, 0x3, PT ; /* 0x000000030700780c */ /* 0x000fe40003f26070 */ /*0170*/ IADD3.X R7, R9, -0x1, RZ, P0, !PT ; /* 0xffffffff09077810 */ /* 0x000fe200007fe4ff */ /*0180*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0190*/ LOP3.LUT R6, R0, c[0x0][0x170], RZ, 0xc0, !PT ; /* 0x00005c0000067a12 */ /* 0x000fe400078ec0ff */ /*01a0*/ ISETP.GE.U32.AND.EX P1, PT, R7, RZ, PT, P1 ; /* 0x000000ff0700720c */ /* 0x000fe20003f26110 */ /*01b0*/ IMAD.MOV.U32 R7, RZ, RZ, RZ ; /* 0x000000ffff077224 */ /* 0x000fe200078e00ff */ /*01c0*/ ISETP.NE.U32.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fc80003f05070 */ /*01d0*/ ISETP.NE.AND.EX P0, PT, RZ, RZ, PT, P0 ; /* 0x000000ffff00720c */ /* 0x000fce0003f05300 */ /*01e0*/ @!P1 BRA 0x460 ; /* 0x0000027000009947 */ /* 0x000fec0003800000 */ /*01f0*/ IADD3 R10, P2, R6, -c[0x0][0x170], RZ ; /* 0x80005c00060a7a10 */ /* 0x000fe20007f5e0ff */ /*0200*/ UMOV UR6, 0x8 ; /* 0x0000000800067882 */ /* 0x000fe20000000000 */ /*0210*/ IADD3 R22, P1, R8, c[0x0][0x160], RZ ; /* 0x0000580008167a10 */ /* 0x000fe20007f3e0ff */ /*0220*/ ULDC.64 UR8, c[0x0][0x178] ; /* 0x00005e0000087ab9 */ /* 0x000fe20000000a00 */ /*0230*/ IMAD.MOV.U32 R7, RZ, RZ, RZ ; /* 0x000000ffff077224 */ /* 0x000fe200078e00ff */ /*0240*/ UIMAD.WIDE.U32 UR4, UR6, UR8, URZ ; /* 0x00000008060472a5 */ /* 0x000fe2000f8e003f */ /*0250*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0260*/ UIMAD UR6, UR6, UR9, URZ ; /* 0x00000009060672a4 */ /* 0x000fe2000f8e023f */ /*0270*/ CS2R R12, SRZ ; /* 0x00000000000c7805 */ /* 0x000fe2000001ff00 */ /*0280*/ IADD3.X R23, R11, c[0x0][0x164], RZ, P1, !PT ; /* 0x000059000b177a10 */ /* 0x000fe20000ffe4ff */ /*0290*/ IMAD.X R8, RZ, RZ, ~c[0x0][0x174], P2 ; /* 0x80005d00ff087624 */ /* 0x000fe200010e06ff */ /*02a0*/ UIADD3 UR6, UR5, UR6, URZ ; /* 0x0000000605067290 */ /* 0x000fc6000fffe03f */ /*02b0*/ LDG.E.64 R14, [R22.64] ; /* 0x0000000a160e7981 */ /* 0x000ea2000c1e1b00 */ /*02c0*/ IADD3 R16, P1, R22, UR4, RZ ; /* 0x0000000416107c10 */ /* 0x000fc8000ff3e0ff */ /*02d0*/ IADD3.X R17, R23, UR6, RZ, P1, !PT ; /* 0x0000000617117c10 */ /* 0x000fe20008ffe4ff */ /*02e0*/ DADD R12, R14, R12 ; /* 0x000000000e0c7229 */ /* 0x006e0e000000000c */ /*02f0*/ STG.E.64 [R2.64], R12 ; /* 0x0000000c02007986 */ /* 0x0011e8000c101b0a */ /*0300*/ LDG.E.64 R14, [R16.64] ; /* 0x0000000a100e7981 */ /* 0x000ea2000c1e1b00 */ /*0310*/ IADD3 R24, P1, R16, UR4, RZ ; /* 0x0000000410187c10 */ /* 0x000fc8000ff3e0ff */ /*0320*/ IADD3.X R25, R17, UR6, RZ, P1, !PT ; /* 0x0000000611197c10 */ /* 0x000fe20008ffe4ff */ /*0330*/ DADD R18, R12, R14 ; /* 0x000000000c127229 */ /* 0x004e4e000000000e */ /*0340*/ STG.E.64 [R2.64], R18 ; /* 0x0000001202007986 */ /* 0x0023e8000c101b0a */ /*0350*/ LDG.E.64 R20, [R24.64] ; /* 0x0000000a18147981 */ /* 0x000ea2000c1e1b00 */ /*0360*/ IADD3 R14, P1, R24, UR4, RZ ; /* 0x00000004180e7c10 */ /* 0x000fc8000ff3e0ff */ /*0370*/ IADD3.X R15, R25, UR6, RZ, P1, !PT ; /* 0x00000006190f7c10 */ /* 0x000fe20008ffe4ff */ /*0380*/ DADD R20, R18, R20 ; /* 0x0000000012147229 */ /* 0x004e8e0000000014 */ /*0390*/ STG.E.64 [R2.64], R20 ; /* 0x0000001402007986 */ /* 0x0043e8000c101b0a */ /*03a0*/ LDG.E.64 R12, [R14.64] ; /* 0x0000000a0e0c7981 */ /* 0x001ea2000c1e1b00 */ /*03b0*/ IADD3 R7, P1, R7, 0x4, RZ ; /* 0x0000000407077810 */ /* 0x000fc80007f3e0ff */ /*03c0*/ IADD3 R11, P2, R7, R10, RZ ; /* 0x0000000a070b7210 */ /* 0x000fe20007f5e0ff */ /*03d0*/ IMAD.X R9, RZ, RZ, R9, P1 ; /* 0x000000ffff097224 */ /* 0x000fc600008e0609 */ /*03e0*/ ISETP.NE.U32.AND P1, PT, R11, RZ, PT ; /* 0x000000ff0b00720c */ /* 0x000fe20003f25070 */ /*03f0*/ IMAD.X R11, R9, 0x1, R8, P2 ; /* 0x00000001090b7824 */ /* 0x000fe200010e0608 */ /*0400*/ IADD3 R22, P2, R14, UR4, RZ ; /* 0x000000040e167c10 */ /* 0x000fc8000ff5e0ff */ /*0410*/ ISETP.NE.AND.EX P1, PT, R11, RZ, PT, P1 ; /* 0x000000ff0b00720c */ /* 0x000fe40003f25310 */ /*0420*/ IADD3.X R23, R15, UR6, RZ, P2, !PT ; /* 0x000000060f177c10 */ /* 0x000fe200097fe4ff */ /*0430*/ DADD R12, R20, R12 ; /* 0x00000000140c7229 */ /* 0x004e0e000000000c */ /*0440*/ STG.E.64 [R2.64], R12 ; /* 0x0000000c02007986 */ /* 0x0013e6000c101b0a */ /*0450*/ @P1 BRA 0x2b0 ; /* 0xfffffe5000001947 */ /* 0x000fea000383ffff */ /*0460*/ @!P0 BRA 0x5e0 ; /* 0x0000017000008947 */ /* 0x000fea0003800000 */ /*0470*/ IMAD R8, R9, c[0x0][0x178], RZ ; /* 0x00005e0009087a24 */ /* 0x000fe200078e02ff */ /*0480*/ IADD3 R6, P1, RZ, -R6, RZ ; /* 0x80000006ff067210 */ /* 0x000fe20007f3e0ff */ /*0490*/ IMAD.MOV.U32 R5, RZ, RZ, RZ ; /* 0x000000ffff057224 */ /* 0x000fe400078e00ff */ /*04a0*/ IMAD R9, R7.reuse, c[0x0][0x17c], R8 ; /* 0x00005f0007097a24 */ /* 0x040fe400078e0208 */ /*04b0*/ IMAD.WIDE.U32 R4, R7, c[0x0][0x178], R4 ; /* 0x00005e0007047a25 */ /* 0x000fc800078e0004 */ /*04c0*/ IMAD.MOV.U32 R10, RZ, RZ, c[0x0][0x178] ; /* 0x00005e00ff0a7624 */ /* 0x000fe200078e00ff */ /*04d0*/ LEA R7, P0, R4, c[0x0][0x160], 0x3 ; /* 0x0000580004077a11 */ /* 0x000fe200078018ff */ /*04e0*/ IMAD.IADD R5, R5, 0x1, R9 ; /* 0x0000000105057824 */ /* 0x000fc600078e0209 */ /*04f0*/ SHF.L.U64.HI R9, R10, R0, c[0x0][0x17c] ; /* 0x00005f000a097619 */ /* 0x000fe20000010200 */ /*0500*/ IMAD.X R0, RZ, RZ, -0x1, P1 ; /* 0xffffffffff007424 */ /* 0x000fe200008e06ff */ /*0510*/ LEA.HI.X R8, R4, c[0x0][0x164], R5, 0x3, P0 ; /* 0x0000590004087a11 */ /* 0x000fc600000f1c05 */ /*0520*/ IMAD.MOV.U32 R4, RZ, RZ, R7 ; /* 0x000000ffff047224 */ /* 0x000fe400078e0007 */ /*0530*/ IMAD.MOV.U32 R5, RZ, RZ, R8 ; /* 0x000000ffff057224 */ /* 0x000fcc00078e0008 */ /*0540*/ LDG.E.64 R4, [R4.64] ; /* 0x0000000a04047981 */ /* 0x000ea2000c1e1b00 */ /*0550*/ IADD3 R6, P0, R6, 0x1, RZ ; /* 0x0000000106067810 */ /* 0x000fe40007f1e0ff */ /*0560*/ LEA R7, P1, R10, R7, 0x3 ; /* 0x000000070a077211 */ /* 0x000fc600078218ff */ /*0570*/ IMAD.X R0, RZ, RZ, R0, P0 ; /* 0x000000ffff007224 */ /* 0x000fe200000e0600 */ /*0580*/ ISETP.NE.U32.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f05070 */ /*0590*/ IMAD.X R8, R8, 0x1, R9, P1 ; /* 0x0000000108087824 */ /* 0x000fc600008e0609 */ /*05a0*/ ISETP.NE.AND.EX P0, PT, R0, RZ, PT, P0 ; /* 0x000000ff0000720c */ /* 0x000fe20003f05300 */ /*05b0*/ DADD R12, R4, R12 ; /* 0x00000000040c7229 */ /* 0x007e0e000000000c */ /*05c0*/ STG.E.64 [R2.64], R12 ; /* 0x0000000c02007986 */ /* 0x0011ea000c101b0a */ /*05d0*/ @P0 BRA 0x520 ; /* 0xffffff4000000947 */ /* 0x000fea000383ffff */ /*05e0*/ I2F.F64.U64 R8, c[0x0][0x170] ; /* 0x00005c0000087b12 */ /* 0x001e220000301800 */ /*05f0*/ IMAD.MOV.U32 R4, RZ, RZ, 0x1 ; /* 0x00000001ff047424 */ /* 0x000fe200078e00ff */ /*0600*/ FSETP.GEU.AND P1, PT, |R13|, 6.5827683646048100446e-37, PT ; /* 0x036000000d00780b */ /* 0x000fe20003f2e200 */ /*0610*/ BSSY B0, 0x750 ; /* 0x0000013000007945 */ /* 0x000fea0003800000 */ /*0620*/ MUFU.RCP64H R5, R9 ; /* 0x0000000900057308 */ /* 0x001e240000001800 */ /*0630*/ DFMA R6, -R8, R4, 1 ; /* 0x3ff000000806742b */ /* 0x001e0c0000000104 */ /*0640*/ DFMA R6, R6, R6, R6 ; /* 0x000000060606722b */ /* 0x001e0c0000000006 */ /*0650*/ DFMA R6, R4, R6, R4 ; /* 0x000000060406722b */ /* 0x001e0c0000000004 */ /*0660*/ DFMA R4, -R8, R6, 1 ; /* 0x3ff000000804742b */ /* 0x001e0c0000000106 */ /*0670*/ DFMA R4, R6, R4, R6 ; /* 0x000000040604722b */ /* 0x001e0c0000000006 */ /*0680*/ DMUL R6, R4, R12 ; /* 0x0000000c04067228 */ /* 0x001e0c0000000000 */ /*0690*/ DFMA R10, -R8, R6, R12 ; /* 0x00000006080a722b */ /* 0x001e0c000000010c */ /*06a0*/ DFMA R4, R4, R10, R6 ; /* 0x0000000a0404722b */ /* 0x001e140000000006 */ /*06b0*/ FFMA R0, RZ, R9, R5 ; /* 0x00000009ff007223 */ /* 0x001fca0000000005 */ /*06c0*/ FSETP.GT.AND P0, PT, |R0|, 1.469367938527859385e-39, PT ; /* 0x001000000000780b */ /* 0x000fda0003f04200 */ /*06d0*/ @P0 BRA P1, 0x740 ; /* 0x0000006000000947 */ /* 0x000fea0000800000 */ /*06e0*/ IMAD.MOV.U32 R6, RZ, RZ, R12 ; /* 0x000000ffff067224 */ /* 0x000fe200078e000c */ /*06f0*/ MOV R0, 0x720 ; /* 0x0000072000007802 */ /* 0x000fe20000000f00 */ /*0700*/ IMAD.MOV.U32 R7, RZ, RZ, R13 ; /* 0x000000ffff077224 */ /* 0x000fe400078e000d */ /*0710*/ CALL.REL.NOINC 0x770 ; /* 0x0000005000007944 */ /* 0x002fea0003c00000 */ /*0720*/ IMAD.MOV.U32 R4, RZ, RZ, R12 ; /* 0x000000ffff047224 */ /* 0x000fe400078e000c */ /*0730*/ IMAD.MOV.U32 R5, RZ, RZ, R13 ; /* 0x000000ffff057224 */ /* 0x000fe400078e000d */ /*0740*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0750*/ STG.E.64 [R2.64], R4 ; /* 0x0000000402007986 */ /* 0x000fe2000c101b0a */ /*0760*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0770*/ FSETP.GEU.AND P0, PT, |R9|.reuse, 1.469367938527859385e-39, PT ; /* 0x001000000900780b */ /* 0x040fe20003f0e200 */ /*0780*/ IMAD.MOV.U32 R13, RZ, RZ, 0x1ca00000 ; /* 0x1ca00000ff0d7424 */ /* 0x000fe200078e00ff */ /*0790*/ LOP3.LUT R4, R9, 0x800fffff, RZ, 0xc0, !PT ; /* 0x800fffff09047812 */ /* 0x000fe200078ec0ff */ /*07a0*/ IMAD.MOV.U32 R16, RZ, RZ, 0x1 ; /* 0x00000001ff107424 */ /* 0x000fe200078e00ff */ /*07b0*/ FSETP.GEU.AND P2, PT, |R7|.reuse, 1.469367938527859385e-39, PT ; /* 0x001000000700780b */ /* 0x040fe20003f4e200 */ /*07c0*/ IMAD.MOV.U32 R10, RZ, RZ, R6 ; /* 0x000000ffff0a7224 */ /* 0x000fe200078e0006 */ /*07d0*/ LOP3.LUT R5, R4, 0x3ff00000, RZ, 0xfc, !PT ; /* 0x3ff0000004057812 */ /* 0x000fe200078efcff */ /*07e0*/ IMAD.MOV.U32 R4, RZ, RZ, R8 ; /* 0x000000ffff047224 */ /* 0x000fe200078e0008 */ /*07f0*/ LOP3.LUT R12, R7, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff00000070c7812 */ /* 0x000fe200078ec0ff */ /*0800*/ BSSY B1, 0xd10 ; /* 0x0000050000017945 */ /* 0x000fe20003800000 */ /*0810*/ LOP3.LUT R15, R9, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff00000090f7812 */ /* 0x000fc600078ec0ff */ /*0820*/ @!P0 DMUL R4, R8, 8.98846567431157953865e+307 ; /* 0x7fe0000008048828 */ /* 0x000e220000000000 */ /*0830*/ ISETP.GE.U32.AND P1, PT, R12, R15, PT ; /* 0x0000000f0c00720c */ /* 0x000fc60003f26070 */ /*0840*/ @!P2 LOP3.LUT R11, R9, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff00000090ba812 */ /* 0x000fe200078ec0ff */ /*0850*/ @!P2 IMAD.MOV.U32 R20, RZ, RZ, RZ ; /* 0x000000ffff14a224 */ /* 0x000fe200078e00ff */ /*0860*/ MUFU.RCP64H R17, R5 ; /* 0x0000000500117308 */ /* 0x001e240000001800 */ /*0870*/ @!P2 ISETP.GE.U32.AND P3, PT, R12, R11, PT ; /* 0x0000000b0c00a20c */ /* 0x000fe40003f66070 */ /*0880*/ SEL R11, R13.reuse, 0x63400000, !P1 ; /* 0x634000000d0b7807 */ /* 0x040fe40004800000 */ /*0890*/ @!P2 SEL R21, R13, 0x63400000, !P3 ; /* 0x634000000d15a807 */ /* 0x000fe40005800000 */ /*08a0*/ LOP3.LUT R11, R11, 0x800fffff, R7, 0xf8, !PT ; /* 0x800fffff0b0b7812 */ /* 0x000fc400078ef807 */ /*08b0*/ @!P2 LOP3.LUT R21, R21, 0x80000000, R7, 0xf8, !PT ; /* 0x800000001515a812 */ /* 0x000fc800078ef807 */ /*08c0*/ @!P2 LOP3.LUT R21, R21, 0x100000, RZ, 0xfc, !PT ; /* 0x001000001515a812 */ /* 0x000fe200078efcff */ /*08d0*/ DFMA R18, R16, -R4, 1 ; /* 0x3ff000001012742b */ /* 0x001e0a0000000804 */ /*08e0*/ @!P2 DFMA R10, R10, 2, -R20 ; /* 0x400000000a0aa82b */ /* 0x000fc80000000814 */ /*08f0*/ DFMA R18, R18, R18, R18 ; /* 0x000000121212722b */ /* 0x001e0c0000000012 */ /*0900*/ DFMA R18, R16, R18, R16 ; /* 0x000000121012722b */ /* 0x0010640000000010 */ /*0910*/ IMAD.MOV.U32 R16, RZ, RZ, R12 ; /* 0x000000ffff107224 */ /* 0x001fe200078e000c */ /*0920*/ @!P2 LOP3.LUT R16, R11, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff000000b10a812 */ /* 0x000fe200078ec0ff */ /*0930*/ IMAD.MOV.U32 R17, RZ, RZ, R15 ; /* 0x000000ffff117224 */ /* 0x000fe200078e000f */ /*0940*/ @!P0 LOP3.LUT R17, R5, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff0000005118812 */ /* 0x000fe200078ec0ff */ /*0950*/ DFMA R20, R18, -R4, 1 ; /* 0x3ff000001214742b */ /* 0x002e060000000804 */ /*0960*/ IADD3 R22, R17, -0x1, RZ ; /* 0xffffffff11167810 */ /* 0x000fc60007ffe0ff */ /*0970*/ DFMA R18, R18, R20, R18 ; /* 0x000000141212722b */ /* 0x0010640000000012 */ /*0980*/ IADD3 R20, R16, -0x1, RZ ; /* 0xffffffff10147810 */ /* 0x001fc80007ffe0ff */ /*0990*/ ISETP.GT.U32.AND P0, PT, R20, 0x7feffffe, PT ; /* 0x7feffffe1400780c */ /* 0x000fe20003f04070 */ /*09a0*/ DMUL R14, R18, R10 ; /* 0x0000000a120e7228 */ /* 0x002e060000000000 */ /*09b0*/ ISETP.GT.U32.OR P0, PT, R22, 0x7feffffe, P0 ; /* 0x7feffffe1600780c */ /* 0x000fc60000704470 */ /*09c0*/ DFMA R20, R14, -R4, R10 ; /* 0x800000040e14722b */ /* 0x001e0c000000000a */ /*09d0*/ DFMA R14, R18, R20, R14 ; /* 0x00000014120e722b */ /* 0x001048000000000e */ /*09e0*/ @P0 BRA 0xbb0 ; /* 0x000001c000000947 */ /* 0x000fea0003800000 */ /*09f0*/ LOP3.LUT R7, R9, 0x7ff00000, RZ, 0xc0, !PT ; /* 0x7ff0000009077812 */ /* 0x003fc800078ec0ff */ /*0a00*/ ISETP.GE.U32.AND P0, PT, R12.reuse, R7, PT ; /* 0x000000070c00720c */ /* 0x040fe20003f06070 */ /*0a10*/ IMAD.IADD R6, R12, 0x1, -R7 ; /* 0x000000010c067824 */ /* 0x000fc600078e0a07 */ /*0a20*/ SEL R13, R13, 0x63400000, !P0 ; /* 0x634000000d0d7807 */ /* 0x000fe40004000000 */ /*0a30*/ IMNMX R6, R6, -0x46a00000, !PT ; /* 0xb960000006067817 */ /* 0x000fc80007800200 */ /*0a40*/ IMNMX R6, R6, 0x46a00000, PT ; /* 0x46a0000006067817 */ /* 0x000fca0003800200 */ /*0a50*/ IMAD.IADD R16, R6, 0x1, -R13 ; /* 0x0000000106107824 */ /* 0x000fe400078e0a0d */ /*0a60*/ IMAD.MOV.U32 R6, RZ, RZ, RZ ; /* 0x000000ffff067224 */ /* 0x000fc600078e00ff */ /*0a70*/ IADD3 R7, R16, 0x7fe00000, RZ ; /* 0x7fe0000010077810 */ /* 0x000fcc0007ffe0ff */ /*0a80*/ DMUL R12, R14, R6 ; /* 0x000000060e0c7228 */ /* 0x000e140000000000 */ /*0a90*/ FSETP.GTU.AND P0, PT, |R13|, 1.469367938527859385e-39, PT ; /* 0x001000000d00780b */ /* 0x001fda0003f0c200 */ /*0aa0*/ @P0 BRA 0xd00 ; /* 0x0000025000000947 */ /* 0x000fea0003800000 */ /*0ab0*/ DFMA R4, R14, -R4, R10 ; /* 0x800000040e04722b */ /* 0x000e22000000000a */ /*0ac0*/ IMAD.MOV.U32 R6, RZ, RZ, RZ ; /* 0x000000ffff067224 */ /* 0x000fd200078e00ff */ /*0ad0*/ FSETP.NEU.AND P0, PT, R5.reuse, RZ, PT ; /* 0x000000ff0500720b */ /* 0x041fe40003f0d000 */ /*0ae0*/ LOP3.LUT R9, R5, 0x80000000, R9, 0x48, !PT ; /* 0x8000000005097812 */ /* 0x000fc800078e4809 */ /*0af0*/ LOP3.LUT R7, R9, R7, RZ, 0xfc, !PT ; /* 0x0000000709077212 */ /* 0x000fce00078efcff */ /*0b00*/ @!P0 BRA 0xd00 ; /* 0x000001f000008947 */ /* 0x000fea0003800000 */ /*0b10*/ IMAD.MOV R5, RZ, RZ, -R16 ; /* 0x000000ffff057224 */ /* 0x000fe200078e0a10 */ /*0b20*/ DMUL.RP R6, R14, R6 ; /* 0x000000060e067228 */ /* 0x000e220000008000 */ /*0b30*/ IMAD.MOV.U32 R4, RZ, RZ, RZ ; /* 0x000000ffff047224 */ /* 0x000fcc00078e00ff */ /*0b40*/ DFMA R4, R12, -R4, R14 ; /* 0x800000040c04722b */ /* 0x000e46000000000e */ /*0b50*/ LOP3.LUT R9, R7, R9, RZ, 0x3c, !PT ; /* 0x0000000907097212 */ /* 0x001fc600078e3cff */ /*0b60*/ IADD3 R4, -R16, -0x43300000, RZ ; /* 0xbcd0000010047810 */ /* 0x002fc80007ffe1ff */ /*0b70*/ FSETP.NEU.AND P0, PT, |R5|, R4, PT ; /* 0x000000040500720b */ /* 0x000fc80003f0d200 */ /*0b80*/ FSEL R12, R6, R12, !P0 ; /* 0x0000000c060c7208 */ /* 0x000fe40004000000 */ /*0b90*/ FSEL R13, R9, R13, !P0 ; /* 0x0000000d090d7208 */ /* 0x000fe20004000000 */ /*0ba0*/ BRA 0xd00 ; /* 0x0000015000007947 */ /* 0x000fea0003800000 */ /*0bb0*/ DSETP.NAN.AND P0, PT, R6, R6, PT ; /* 0x000000060600722a */ /* 0x003e1c0003f08000 */ /*0bc0*/ @P0 BRA 0xce0 ; /* 0x0000011000000947 */ /* 0x001fea0003800000 */ /*0bd0*/ DSETP.NAN.AND P0, PT, R8, R8, PT ; /* 0x000000080800722a */ /* 0x000e1c0003f08000 */ /*0be0*/ @P0 BRA 0xcb0 ; /* 0x000000c000000947 */ /* 0x001fea0003800000 */ /*0bf0*/ ISETP.NE.AND P0, PT, R16, R17, PT ; /* 0x000000111000720c */ /* 0x000fe20003f05270 */ /*0c00*/ IMAD.MOV.U32 R12, RZ, RZ, 0x0 ; /* 0x00000000ff0c7424 */ /* 0x000fe400078e00ff */ /*0c10*/ IMAD.MOV.U32 R13, RZ, RZ, -0x80000 ; /* 0xfff80000ff0d7424 */ /* 0x000fd400078e00ff */ /*0c20*/ @!P0 BRA 0xd00 ; /* 0x000000d000008947 */ /* 0x000fea0003800000 */ /*0c30*/ ISETP.NE.AND P0, PT, R16, 0x7ff00000, PT ; /* 0x7ff000001000780c */ /* 0x000fe40003f05270 */ /*0c40*/ LOP3.LUT R13, R7, 0x80000000, R9, 0x48, !PT ; /* 0x80000000070d7812 */ /* 0x000fe400078e4809 */ /*0c50*/ ISETP.EQ.OR P0, PT, R17, RZ, !P0 ; /* 0x000000ff1100720c */ /* 0x000fda0004702670 */ /*0c60*/ @P0 LOP3.LUT R4, R13, 0x7ff00000, RZ, 0xfc, !PT ; /* 0x7ff000000d040812 */ /* 0x000fe200078efcff */ /*0c70*/ @!P0 IMAD.MOV.U32 R12, RZ, RZ, RZ ; /* 0x000000ffff0c8224 */ /* 0x000fe400078e00ff */ /*0c80*/ @P0 IMAD.MOV.U32 R12, RZ, RZ, RZ ; /* 0x000000ffff0c0224 */ /* 0x000fe400078e00ff */ /*0c90*/ @P0 IMAD.MOV.U32 R13, RZ, RZ, R4 ; /* 0x000000ffff0d0224 */ /* 0x000fe200078e0004 */ /*0ca0*/ BRA 0xd00 ; /* 0x0000005000007947 */ /* 0x000fea0003800000 */ /*0cb0*/ LOP3.LUT R13, R9, 0x80000, RZ, 0xfc, !PT ; /* 0x00080000090d7812 */ /* 0x000fe200078efcff */ /*0cc0*/ IMAD.MOV.U32 R12, RZ, RZ, R8 ; /* 0x000000ffff0c7224 */ /* 0x000fe200078e0008 */ /*0cd0*/ BRA 0xd00 ; /* 0x0000002000007947 */ /* 0x000fea0003800000 */ /*0ce0*/ LOP3.LUT R13, R7, 0x80000, RZ, 0xfc, !PT ; /* 0x00080000070d7812 */ /* 0x000fe200078efcff */ /*0cf0*/ IMAD.MOV.U32 R12, RZ, RZ, R6 ; /* 0x000000ffff0c7224 */ /* 0x000fe400078e0006 */ /*0d00*/ BSYNC B1 ; /* 0x0000000000017941 */ /* 0x000fea0003800000 */ /*0d10*/ IMAD.MOV.U32 R4, RZ, RZ, R0 ; /* 0x000000ffff047224 */ /* 0x000fe400078e0000 */ /*0d20*/ IMAD.MOV.U32 R5, RZ, RZ, 0x0 ; /* 0x00000000ff057424 */ /* 0x000fc800078e00ff */ /*0d30*/ RET.REL.NODEC R4 0x0 ; /* 0xfffff2c004007950 */ /* 0x000fea0003c3ffff */ /*0d40*/ BRA 0xd40; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0d50*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0da0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0db0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0dc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0dd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0de0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0df0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z12getMeanImagePKdPdmm .globl _Z12getMeanImagePKdPdmm .p2align 8 .type _Z12getMeanImagePKdPdmm,@function _Z12getMeanImagePKdPdmm: s_clause 0x1 s_load_b32 s4, s[0:1], 0x2c s_load_b64 s[2:3], s[0:1], 0x18 s_waitcnt lgkmcnt(0) s_and_b32 s4, s4, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s4, v[0:1] v_mov_b32_e32 v2, 0 s_mov_b32 s4, exec_lo v_cmpx_gt_u64_e64 s[2:3], v[1:2] s_cbranch_execz .LBB0_5 s_load_b128 s[4:7], s[0:1], 0x8 v_lshlrev_b64 v[4:5], 3, v[1:2] v_mov_b32_e32 v2, 0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_mov_b32_e32 v3, v2 s_waitcnt lgkmcnt(0) v_add_co_u32 v0, vcc_lo, s4, v4 s_delay_alu instid0(VALU_DEP_4) v_add_co_ci_u32_e32 v1, vcc_lo, s5, v5, vcc_lo s_cmp_eq_u64 s[6:7], 0 global_store_b64 v[0:1], v[2:3], off s_cbranch_scc1 .LBB0_4 global_load_b64 v[2:3], v[0:1], off s_load_b64 s[0:1], s[0:1], 0x0 s_waitcnt lgkmcnt(0) v_add_co_u32 v4, vcc_lo, s0, v4 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v5, vcc_lo s_lshl_b64 s[0:1], s[2:3], 3 s_mov_b64 s[2:3], s[6:7] .LBB0_3: global_load_b64 v[6:7], v[4:5], off v_add_co_u32 v4, vcc_lo, v4, s0 s_add_u32 s2, s2, -1 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v5, vcc_lo s_addc_u32 s3, s3, -1 s_delay_alu instid0(SALU_CYCLE_1) s_cmp_eq_u64 s[2:3], 0 s_waitcnt vmcnt(0) v_add_f64 v[2:3], v[6:7], v[2:3] global_store_b64 v[0:1], v[2:3], off s_cbranch_scc0 .LBB0_3 .LBB0_4: global_load_b64 v[2:3], v[0:1], off v_cvt_f64_u32_e32 v[4:5], s7 v_cvt_f64_u32_e32 v[6:7], s6 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_ldexp_f64 v[4:5], v[4:5], 32 v_add_f64 v[4:5], v[4:5], v[6:7] s_waitcnt vmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_div_scale_f64 v[6:7], null, v[4:5], v[4:5], v[2:3] v_rcp_f64_e32 v[8:9], v[6:7] s_waitcnt_depctr 0xfff v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_div_scale_f64 v[10:11], vcc_lo, v[2:3], v[4:5], v[2:3] v_mul_f64 v[12:13], v[10:11], v[8:9] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[6:7], -v[6:7], v[12:13], v[10:11] v_div_fmas_f64 v[6:7], v[6:7], v[8:9], v[12:13] s_delay_alu instid0(VALU_DEP_1) v_div_fixup_f64 v[2:3], v[6:7], v[4:5], v[2:3] global_store_b64 v[0:1], v[2:3], off .LBB0_5: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z12getMeanImagePKdPdmm .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 288 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z12getMeanImagePKdPdmm, .Lfunc_end0-_Z12getMeanImagePKdPdmm .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .offset: 16 .size: 8 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: hidden_block_count_x - .offset: 36 .size: 4 .value_kind: hidden_block_count_y - .offset: 40 .size: 4 .value_kind: hidden_block_count_z - .offset: 44 .size: 2 .value_kind: hidden_group_size_x - .offset: 46 .size: 2 .value_kind: hidden_group_size_y - .offset: 48 .size: 2 .value_kind: hidden_group_size_z - .offset: 50 .size: 2 .value_kind: hidden_remainder_x - .offset: 52 .size: 2 .value_kind: hidden_remainder_y - .offset: 54 .size: 2 .value_kind: hidden_remainder_z - .offset: 72 .size: 8 .value_kind: hidden_global_offset_x - .offset: 80 .size: 8 .value_kind: hidden_global_offset_y - .offset: 88 .size: 8 .value_kind: hidden_global_offset_z - .offset: 96 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 288 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z12getMeanImagePKdPdmm .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z12getMeanImagePKdPdmm.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_0010e20e_00000000-6_getMeanImage.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm .type _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm, @function _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movq %rcx, (%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) movq %rsp, %rax movq %rax, 120(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 136(%rsp), %rax subq %fs:40, %rax jne .L8 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 168 pushq 40(%rsp) .cfi_def_cfa_offset 176 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z12getMeanImagePKdPdmm(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm, .-_Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm .globl _Z12getMeanImagePKdPdmm .type _Z12getMeanImagePKdPdmm, @function _Z12getMeanImagePKdPdmm: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z37__device_stub__Z12getMeanImagePKdPdmmPKdPdmm addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z12getMeanImagePKdPdmm, .-_Z12getMeanImagePKdPdmm .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z12getMeanImagePKdPdmm" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z12getMeanImagePKdPdmm(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "getMeanImage.hip" .globl _Z27__device_stub__getMeanImagePKdPdmm # -- Begin function _Z27__device_stub__getMeanImagePKdPdmm .p2align 4, 0x90 .type _Z27__device_stub__getMeanImagePKdPdmm,@function _Z27__device_stub__getMeanImagePKdPdmm: # @_Z27__device_stub__getMeanImagePKdPdmm .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) movq %rcx, 48(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 48(%rsp), %rax movq %rax, 104(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z12getMeanImagePKdPdmm, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z27__device_stub__getMeanImagePKdPdmm, .Lfunc_end0-_Z27__device_stub__getMeanImagePKdPdmm .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12getMeanImagePKdPdmm, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z12getMeanImagePKdPdmm,@object # @_Z12getMeanImagePKdPdmm .section .rodata,"a",@progbits .globl _Z12getMeanImagePKdPdmm .p2align 3, 0x0 _Z12getMeanImagePKdPdmm: .quad _Z27__device_stub__getMeanImagePKdPdmm .size _Z12getMeanImagePKdPdmm, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z12getMeanImagePKdPdmm" .size .L__unnamed_1, 24 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z27__device_stub__getMeanImagePKdPdmm .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z12getMeanImagePKdPdmm .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
/* * a simple test */ __shared__ float data1[32][32]; __shared__ float data2[32][32]; __shared__ float data3[32][32]; __device__ void mult(__shared__ float d1[32][32], __shared__ float d2[32][32], __shared__ float d3[32][32], int idx) { int i; for (i = 0; i < 31; i++) { d1[idx][i] = d2[idx+1][i-1] + d2[idx][i-1] + d2[idx-1][i-1] + d2[idx+1][i] + d2[idx][i] + d2[idx-1][i] + d2[idx+1][i+1] + d2[idx][i+1] + d2[idx-1][i+1]; } } __global__ void doit(int start, int end) { int i; for (i = start; i < end; i++) { mult(data1, data2, data3, i); } }
code for sm_80 Function : _Z4doitii .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0020*/ BRA 0x20; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0030*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0040*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0050*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0060*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0070*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0080*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0090*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
/* * a simple test */ __shared__ float data1[32][32]; __shared__ float data2[32][32]; __shared__ float data3[32][32]; __device__ void mult(__shared__ float d1[32][32], __shared__ float d2[32][32], __shared__ float d3[32][32], int idx) { int i; for (i = 0; i < 31; i++) { d1[idx][i] = d2[idx+1][i-1] + d2[idx][i-1] + d2[idx-1][i-1] + d2[idx+1][i] + d2[idx][i] + d2[idx-1][i] + d2[idx+1][i+1] + d2[idx][i+1] + d2[idx-1][i+1]; } } __global__ void doit(int start, int end) { int i; for (i = start; i < end; i++) { mult(data1, data2, data3, i); } }
.file "tmpxft_000688de_00000000-6_iv12.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2030: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2030: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z4multPA32_fS0_S0_i .type _Z4multPA32_fS0_S0_i, @function _Z4multPA32_fS0_S0_i: .LFB2027: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2027: .size _Z4multPA32_fS0_S0_i, .-_Z4multPA32_fS0_S0_i .globl _Z23__device_stub__Z4doitiiii .type _Z23__device_stub__Z4doitiiii, @function _Z23__device_stub__Z4doitiiii: .LFB2052: .cfi_startproc endbr64 subq $120, %rsp .cfi_def_cfa_offset 128 movl %edi, 12(%rsp) movl %esi, 8(%rsp) movq %fs:40, %rax movq %rax, 104(%rsp) xorl %eax, %eax leaq 12(%rsp), %rax movq %rax, 80(%rsp) leaq 8(%rsp), %rax movq %rax, 88(%rsp) movl $1, 32(%rsp) movl $1, 36(%rsp) movl $1, 40(%rsp) movl $1, 44(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) leaq 24(%rsp), %rcx leaq 16(%rsp), %rdx leaq 44(%rsp), %rsi leaq 32(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L9 .L5: movq 104(%rsp), %rax subq %fs:40, %rax jne .L10 addq $120, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L9: .cfi_restore_state pushq 24(%rsp) .cfi_def_cfa_offset 136 pushq 24(%rsp) .cfi_def_cfa_offset 144 leaq 96(%rsp), %r9 movq 60(%rsp), %rcx movl 68(%rsp), %r8d movq 48(%rsp), %rsi movl 56(%rsp), %edx leaq _Z4doitii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 128 jmp .L5 .L10: call __stack_chk_fail@PLT .cfi_endproc .LFE2052: .size _Z23__device_stub__Z4doitiiii, .-_Z23__device_stub__Z4doitiiii .globl _Z4doitii .type _Z4doitii, @function _Z4doitii: .LFB2053: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z23__device_stub__Z4doitiiii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2053: .size _Z4doitii, .-_Z4doitii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z4doitii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2055: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z4doitii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2055: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
/* * a simple test */ __shared__ float data1[32][32]; __shared__ float data2[32][32]; __shared__ float data3[32][32]; __device__ void mult(__shared__ float d1[32][32], __shared__ float d2[32][32], __shared__ float d3[32][32], int idx) { int i; for (i = 0; i < 31; i++) { d1[idx][i] = d2[idx+1][i-1] + d2[idx][i-1] + d2[idx-1][i-1] + d2[idx+1][i] + d2[idx][i] + d2[idx-1][i] + d2[idx+1][i+1] + d2[idx][i+1] + d2[idx-1][i+1]; } } __global__ void doit(int start, int end) { int i; for (i = start; i < end; i++) { mult(data1, data2, data3, i); } }
#include <hip/hip_runtime.h> /* * a simple test */ __shared__ float data1[32][32]; __shared__ float data2[32][32]; __shared__ float data3[32][32]; __device__ void mult(__shared__ float d1[32][32], __shared__ float d2[32][32], __shared__ float d3[32][32], int idx) { int i; for (i = 0; i < 31; i++) { d1[idx][i] = d2[idx+1][i-1] + d2[idx][i-1] + d2[idx-1][i-1] + d2[idx+1][i] + d2[idx][i] + d2[idx-1][i] + d2[idx+1][i+1] + d2[idx][i+1] + d2[idx-1][i+1]; } } __global__ void doit(int start, int end) { int i; for (i = start; i < end; i++) { mult(data1, data2, data3, i); } }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> /* * a simple test */ __shared__ float data1[32][32]; __shared__ float data2[32][32]; __shared__ float data3[32][32]; __device__ void mult(__shared__ float d1[32][32], __shared__ float d2[32][32], __shared__ float d3[32][32], int idx) { int i; for (i = 0; i < 31; i++) { d1[idx][i] = d2[idx+1][i-1] + d2[idx][i-1] + d2[idx-1][i-1] + d2[idx+1][i] + d2[idx][i] + d2[idx-1][i] + d2[idx+1][i+1] + d2[idx][i+1] + d2[idx-1][i+1]; } } __global__ void doit(int start, int end) { int i; for (i = start; i < end; i++) { mult(data1, data2, data3, i); } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z4doitii .globl _Z4doitii .p2align 8 .type _Z4doitii,@function _Z4doitii: s_load_b64 s[0:1], s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_cmp_ge_i32 s0, s1 s_cbranch_scc1 .LBB0_5 s_lshl_b32 s2, s0, 7 s_movk_i32 s5, 0xc00 s_set_inst_prefetch_distance 0x1 .p2align 6 .LBB0_2: s_mov_b32 s3, 0 .p2align 6 .LBB0_3: s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(VALU_DEP_1) s_add_i32 s4, s2, s3 s_add_i32 s3, s3, 4 v_mov_b32_e32 v8, s4 s_movk_i32 s4, 0x1000 s_cmpk_lg_i32 s3, 0x7c v_add_nc_u32_e32 v0, s4, v8 s_movk_i32 s4, 0xc00 v_add_nc_u32_e32 v2, s5, v8 v_add_nc_u32_e32 v4, s4, v8 s_movk_i32 s4, 0x1000 ds_load_2addr_b32 v[0:1], v0 offset0:31 offset1:32 ds_load_2addr_b32 v[2:3], v2 offset0:225 offset1:255 ds_load_2addr_b32 v[4:5], v4 offset0:223 offset1:224 v_add_nc_u32_e32 v6, s4, v8 ds_load_2addr_b32 v[6:7], v6 offset1:1 s_waitcnt lgkmcnt(2) v_add_f32_e32 v0, v0, v3 ds_load_b32 v3, v8 offset:4228 s_waitcnt lgkmcnt(2) v_add_f32_e32 v0, v0, v4 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_add_f32_e32 v0, v0, v1 s_waitcnt lgkmcnt(1) v_add_f32_e32 v0, v0, v6 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_add_f32_e32 v0, v0, v5 s_waitcnt lgkmcnt(0) v_add_f32_e32 v0, v0, v3 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_add_f32_e32 v0, v0, v7 v_add_f32_e32 v0, v0, v2 ds_store_b32 v8, v0 s_cbranch_scc1 .LBB0_3 s_add_i32 s0, s0, 1 s_addk_i32 s2, 0x80 s_cmp_lt_i32 s0, s1 s_cbranch_scc1 .LBB0_2 .LBB0_5: s_set_inst_prefetch_distance 0x2 s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z4doitii .amdhsa_group_segment_fixed_size 8192 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 8 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 9 .amdhsa_next_free_sgpr 6 .amdhsa_reserve_vcc 0 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z4doitii, .Lfunc_end0-_Z4doitii .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 4 .value_kind: by_value - .offset: 4 .size: 4 .value_kind: by_value .group_segment_fixed_size: 8192 .kernarg_segment_align: 4 .kernarg_segment_size: 8 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z4doitii .private_segment_fixed_size: 0 .sgpr_count: 6 .sgpr_spill_count: 0 .symbol: _Z4doitii.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 9 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> /* * a simple test */ __shared__ float data1[32][32]; __shared__ float data2[32][32]; __shared__ float data3[32][32]; __device__ void mult(__shared__ float d1[32][32], __shared__ float d2[32][32], __shared__ float d3[32][32], int idx) { int i; for (i = 0; i < 31; i++) { d1[idx][i] = d2[idx+1][i-1] + d2[idx][i-1] + d2[idx-1][i-1] + d2[idx+1][i] + d2[idx][i] + d2[idx-1][i] + d2[idx+1][i+1] + d2[idx][i+1] + d2[idx-1][i+1]; } } __global__ void doit(int start, int end) { int i; for (i = start; i < end; i++) { mult(data1, data2, data3, i); } }
.text .file "iv12.hip" .globl _Z19__device_stub__doitii # -- Begin function _Z19__device_stub__doitii .p2align 4, 0x90 .type _Z19__device_stub__doitii,@function _Z19__device_stub__doitii: # @_Z19__device_stub__doitii .cfi_startproc # %bb.0: subq $88, %rsp .cfi_def_cfa_offset 96 movl %edi, 12(%rsp) movl %esi, 8(%rsp) leaq 12(%rsp), %rax movq %rax, 64(%rsp) leaq 8(%rsp), %rax movq %rax, 72(%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d leaq 64(%rsp), %r9 movl $_Z4doitii, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $104, %rsp .cfi_adjust_cfa_offset -104 retq .Lfunc_end0: .size _Z19__device_stub__doitii, .Lfunc_end0-_Z19__device_stub__doitii .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z4doitii, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z4doitii,@object # @_Z4doitii .section .rodata,"a",@progbits .globl _Z4doitii .p2align 3, 0x0 _Z4doitii: .quad _Z19__device_stub__doitii .size _Z4doitii, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z4doitii" .size .L__unnamed_1, 10 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z19__device_stub__doitii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z4doitii .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z4doitii .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0020*/ BRA 0x20; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0030*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0040*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0050*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0060*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0070*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0080*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0090*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z4doitii .globl _Z4doitii .p2align 8 .type _Z4doitii,@function _Z4doitii: s_load_b64 s[0:1], s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_cmp_ge_i32 s0, s1 s_cbranch_scc1 .LBB0_5 s_lshl_b32 s2, s0, 7 s_movk_i32 s5, 0xc00 s_set_inst_prefetch_distance 0x1 .p2align 6 .LBB0_2: s_mov_b32 s3, 0 .p2align 6 .LBB0_3: s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(VALU_DEP_1) s_add_i32 s4, s2, s3 s_add_i32 s3, s3, 4 v_mov_b32_e32 v8, s4 s_movk_i32 s4, 0x1000 s_cmpk_lg_i32 s3, 0x7c v_add_nc_u32_e32 v0, s4, v8 s_movk_i32 s4, 0xc00 v_add_nc_u32_e32 v2, s5, v8 v_add_nc_u32_e32 v4, s4, v8 s_movk_i32 s4, 0x1000 ds_load_2addr_b32 v[0:1], v0 offset0:31 offset1:32 ds_load_2addr_b32 v[2:3], v2 offset0:225 offset1:255 ds_load_2addr_b32 v[4:5], v4 offset0:223 offset1:224 v_add_nc_u32_e32 v6, s4, v8 ds_load_2addr_b32 v[6:7], v6 offset1:1 s_waitcnt lgkmcnt(2) v_add_f32_e32 v0, v0, v3 ds_load_b32 v3, v8 offset:4228 s_waitcnt lgkmcnt(2) v_add_f32_e32 v0, v0, v4 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_add_f32_e32 v0, v0, v1 s_waitcnt lgkmcnt(1) v_add_f32_e32 v0, v0, v6 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_add_f32_e32 v0, v0, v5 s_waitcnt lgkmcnt(0) v_add_f32_e32 v0, v0, v3 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_add_f32_e32 v0, v0, v7 v_add_f32_e32 v0, v0, v2 ds_store_b32 v8, v0 s_cbranch_scc1 .LBB0_3 s_add_i32 s0, s0, 1 s_addk_i32 s2, 0x80 s_cmp_lt_i32 s0, s1 s_cbranch_scc1 .LBB0_2 .LBB0_5: s_set_inst_prefetch_distance 0x2 s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z4doitii .amdhsa_group_segment_fixed_size 8192 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 8 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 9 .amdhsa_next_free_sgpr 6 .amdhsa_reserve_vcc 0 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z4doitii, .Lfunc_end0-_Z4doitii .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 4 .value_kind: by_value - .offset: 4 .size: 4 .value_kind: by_value .group_segment_fixed_size: 8192 .kernarg_segment_align: 4 .kernarg_segment_size: 8 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z4doitii .private_segment_fixed_size: 0 .sgpr_count: 6 .sgpr_spill_count: 0 .symbol: _Z4doitii.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 9 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_000688de_00000000-6_iv12.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2030: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2030: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z4multPA32_fS0_S0_i .type _Z4multPA32_fS0_S0_i, @function _Z4multPA32_fS0_S0_i: .LFB2027: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2027: .size _Z4multPA32_fS0_S0_i, .-_Z4multPA32_fS0_S0_i .globl _Z23__device_stub__Z4doitiiii .type _Z23__device_stub__Z4doitiiii, @function _Z23__device_stub__Z4doitiiii: .LFB2052: .cfi_startproc endbr64 subq $120, %rsp .cfi_def_cfa_offset 128 movl %edi, 12(%rsp) movl %esi, 8(%rsp) movq %fs:40, %rax movq %rax, 104(%rsp) xorl %eax, %eax leaq 12(%rsp), %rax movq %rax, 80(%rsp) leaq 8(%rsp), %rax movq %rax, 88(%rsp) movl $1, 32(%rsp) movl $1, 36(%rsp) movl $1, 40(%rsp) movl $1, 44(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) leaq 24(%rsp), %rcx leaq 16(%rsp), %rdx leaq 44(%rsp), %rsi leaq 32(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L9 .L5: movq 104(%rsp), %rax subq %fs:40, %rax jne .L10 addq $120, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L9: .cfi_restore_state pushq 24(%rsp) .cfi_def_cfa_offset 136 pushq 24(%rsp) .cfi_def_cfa_offset 144 leaq 96(%rsp), %r9 movq 60(%rsp), %rcx movl 68(%rsp), %r8d movq 48(%rsp), %rsi movl 56(%rsp), %edx leaq _Z4doitii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 128 jmp .L5 .L10: call __stack_chk_fail@PLT .cfi_endproc .LFE2052: .size _Z23__device_stub__Z4doitiiii, .-_Z23__device_stub__Z4doitiiii .globl _Z4doitii .type _Z4doitii, @function _Z4doitii: .LFB2053: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z23__device_stub__Z4doitiiii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2053: .size _Z4doitii, .-_Z4doitii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z4doitii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2055: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z4doitii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2055: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "iv12.hip" .globl _Z19__device_stub__doitii # -- Begin function _Z19__device_stub__doitii .p2align 4, 0x90 .type _Z19__device_stub__doitii,@function _Z19__device_stub__doitii: # @_Z19__device_stub__doitii .cfi_startproc # %bb.0: subq $88, %rsp .cfi_def_cfa_offset 96 movl %edi, 12(%rsp) movl %esi, 8(%rsp) leaq 12(%rsp), %rax movq %rax, 64(%rsp) leaq 8(%rsp), %rax movq %rax, 72(%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d leaq 64(%rsp), %r9 movl $_Z4doitii, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $104, %rsp .cfi_adjust_cfa_offset -104 retq .Lfunc_end0: .size _Z19__device_stub__doitii, .Lfunc_end0-_Z19__device_stub__doitii .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z4doitii, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z4doitii,@object # @_Z4doitii .section .rodata,"a",@progbits .globl _Z4doitii .p2align 3, 0x0 _Z4doitii: .quad _Z19__device_stub__doitii .size _Z4doitii, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z4doitii" .size .L__unnamed_1, 10 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z19__device_stub__doitii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z4doitii .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include "includes.h" __global__ void InvertValuesKernel(float *input, float* outputs, int size) { int id = blockDim.x * blockIdx.y * gridDim.x + blockDim.x*blockIdx.x + threadIdx.x; if(id < size) { outputs[id] = 1.00f - input[id]; } }
code for sm_80 Function : _Z18InvertValuesKernelPfS_i .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R0, SR_CTAID.Y ; /* 0x0000000000007919 */ /* 0x000e280000002600 */ /*0020*/ S2R R3, SR_CTAID.X ; /* 0x0000000000037919 */ /* 0x000e280000002500 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD R0, R0, c[0x0][0xc], R3 ; /* 0x0000030000007a24 */ /* 0x001fc800078e0203 */ /*0050*/ IMAD R0, R0, c[0x0][0x0], R5 ; /* 0x0000000000007a24 */ /* 0x002fca00078e0205 */ /*0060*/ ISETP.GE.AND P0, PT, R0, c[0x0][0x170], PT ; /* 0x00005c0000007a0c */ /* 0x000fda0003f06270 */ /*0070*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0080*/ HFMA2.MMA R5, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff057435 */ /* 0x000fe200000001ff */ /*0090*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd20000000a00 */ /*00a0*/ IMAD.WIDE R2, R0, R5, c[0x0][0x160] ; /* 0x0000580000027625 */ /* 0x000fcc00078e0205 */ /*00b0*/ LDG.E R2, [R2.64] ; /* 0x0000000402027981 */ /* 0x000ea2000c1e1900 */ /*00c0*/ IMAD.WIDE R4, R0, R5, c[0x0][0x168] ; /* 0x00005a0000047625 */ /* 0x000fc800078e0205 */ /*00d0*/ FADD R7, -R2, 1 ; /* 0x3f80000002077421 */ /* 0x004fca0000000100 */ /*00e0*/ STG.E [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x000fe2000c101904 */ /*00f0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0100*/ BRA 0x100; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0110*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0120*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0130*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0140*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0150*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0160*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0170*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0180*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0190*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include "includes.h" __global__ void InvertValuesKernel(float *input, float* outputs, int size) { int id = blockDim.x * blockIdx.y * gridDim.x + blockDim.x*blockIdx.x + threadIdx.x; if(id < size) { outputs[id] = 1.00f - input[id]; } }
.file "tmpxft_001b6801_00000000-6_InvertValuesKernel.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i .type _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i, @function _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i: .LFB2051: .cfi_startproc endbr64 subq $136, %rsp .cfi_def_cfa_offset 144 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movl %edx, 12(%rsp) movq %fs:40, %rax movq %rax, 120(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 12(%rsp), %rax movq %rax, 112(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 120(%rsp), %rax subq %fs:40, %rax jne .L8 addq $136, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z18InvertValuesKernelPfS_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i, .-_Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i .globl _Z18InvertValuesKernelPfS_i .type _Z18InvertValuesKernelPfS_i, @function _Z18InvertValuesKernelPfS_i: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z18InvertValuesKernelPfS_i, .-_Z18InvertValuesKernelPfS_i .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z18InvertValuesKernelPfS_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z18InvertValuesKernelPfS_i(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include "includes.h" __global__ void InvertValuesKernel(float *input, float* outputs, int size) { int id = blockDim.x * blockIdx.y * gridDim.x + blockDim.x*blockIdx.x + threadIdx.x; if(id < size) { outputs[id] = 1.00f - input[id]; } }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void InvertValuesKernel(float *input, float* outputs, int size) { int id = blockDim.x * blockIdx.y * gridDim.x + blockDim.x*blockIdx.x + threadIdx.x; if(id < size) { outputs[id] = 1.00f - input[id]; } }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void InvertValuesKernel(float *input, float* outputs, int size) { int id = blockDim.x * blockIdx.y * gridDim.x + blockDim.x*blockIdx.x + threadIdx.x; if(id < size) { outputs[id] = 1.00f - input[id]; } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z18InvertValuesKernelPfS_i .globl _Z18InvertValuesKernelPfS_i .p2align 8 .type _Z18InvertValuesKernelPfS_i,@function _Z18InvertValuesKernelPfS_i: s_clause 0x2 s_load_b32 s2, s[0:1], 0x18 s_load_b32 s3, s[0:1], 0x24 s_load_b32 s4, s[0:1], 0x10 s_waitcnt lgkmcnt(0) s_mul_i32 s2, s2, s15 s_and_b32 s3, s3, 0xffff s_add_i32 s2, s2, s14 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s2, s3, v[0:1] s_mov_b32 s2, exec_lo v_cmpx_gt_i32_e64 s4, v1 s_cbranch_execz .LBB0_2 s_load_b128 s[0:3], s[0:1], 0x0 v_ashrrev_i32_e32 v2, 31, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[1:2] s_waitcnt lgkmcnt(0) v_add_co_u32 v2, vcc_lo, s0, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s1, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s2, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s3, v1, vcc_lo global_load_b32 v2, v[2:3], off s_waitcnt vmcnt(0) v_sub_f32_e32 v2, 1.0, v2 global_store_b32 v[0:1], v2, off .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z18InvertValuesKernelPfS_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 280 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 4 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z18InvertValuesKernelPfS_i, .Lfunc_end0-_Z18InvertValuesKernelPfS_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .offset: 16 .size: 4 .value_kind: by_value - .offset: 24 .size: 4 .value_kind: hidden_block_count_x - .offset: 28 .size: 4 .value_kind: hidden_block_count_y - .offset: 32 .size: 4 .value_kind: hidden_block_count_z - .offset: 36 .size: 2 .value_kind: hidden_group_size_x - .offset: 38 .size: 2 .value_kind: hidden_group_size_y - .offset: 40 .size: 2 .value_kind: hidden_group_size_z - .offset: 42 .size: 2 .value_kind: hidden_remainder_x - .offset: 44 .size: 2 .value_kind: hidden_remainder_y - .offset: 46 .size: 2 .value_kind: hidden_remainder_z - .offset: 64 .size: 8 .value_kind: hidden_global_offset_x - .offset: 72 .size: 8 .value_kind: hidden_global_offset_y - .offset: 80 .size: 8 .value_kind: hidden_global_offset_z - .offset: 88 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 280 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z18InvertValuesKernelPfS_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z18InvertValuesKernelPfS_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 4 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void InvertValuesKernel(float *input, float* outputs, int size) { int id = blockDim.x * blockIdx.y * gridDim.x + blockDim.x*blockIdx.x + threadIdx.x; if(id < size) { outputs[id] = 1.00f - input[id]; } }
.text .file "InvertValuesKernel.hip" .globl _Z33__device_stub__InvertValuesKernelPfS_i # -- Begin function _Z33__device_stub__InvertValuesKernelPfS_i .p2align 4, 0x90 .type _Z33__device_stub__InvertValuesKernelPfS_i,@function _Z33__device_stub__InvertValuesKernelPfS_i: # @_Z33__device_stub__InvertValuesKernelPfS_i .cfi_startproc # %bb.0: subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movl %edx, 12(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 12(%rsp), %rax movq %rax, 96(%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z18InvertValuesKernelPfS_i, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $120, %rsp .cfi_adjust_cfa_offset -120 retq .Lfunc_end0: .size _Z33__device_stub__InvertValuesKernelPfS_i, .Lfunc_end0-_Z33__device_stub__InvertValuesKernelPfS_i .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z18InvertValuesKernelPfS_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z18InvertValuesKernelPfS_i,@object # @_Z18InvertValuesKernelPfS_i .section .rodata,"a",@progbits .globl _Z18InvertValuesKernelPfS_i .p2align 3, 0x0 _Z18InvertValuesKernelPfS_i: .quad _Z33__device_stub__InvertValuesKernelPfS_i .size _Z18InvertValuesKernelPfS_i, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z18InvertValuesKernelPfS_i" .size .L__unnamed_1, 28 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z33__device_stub__InvertValuesKernelPfS_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z18InvertValuesKernelPfS_i .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z18InvertValuesKernelPfS_i .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R0, SR_CTAID.Y ; /* 0x0000000000007919 */ /* 0x000e280000002600 */ /*0020*/ S2R R3, SR_CTAID.X ; /* 0x0000000000037919 */ /* 0x000e280000002500 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD R0, R0, c[0x0][0xc], R3 ; /* 0x0000030000007a24 */ /* 0x001fc800078e0203 */ /*0050*/ IMAD R0, R0, c[0x0][0x0], R5 ; /* 0x0000000000007a24 */ /* 0x002fca00078e0205 */ /*0060*/ ISETP.GE.AND P0, PT, R0, c[0x0][0x170], PT ; /* 0x00005c0000007a0c */ /* 0x000fda0003f06270 */ /*0070*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0080*/ HFMA2.MMA R5, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff057435 */ /* 0x000fe200000001ff */ /*0090*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd20000000a00 */ /*00a0*/ IMAD.WIDE R2, R0, R5, c[0x0][0x160] ; /* 0x0000580000027625 */ /* 0x000fcc00078e0205 */ /*00b0*/ LDG.E R2, [R2.64] ; /* 0x0000000402027981 */ /* 0x000ea2000c1e1900 */ /*00c0*/ IMAD.WIDE R4, R0, R5, c[0x0][0x168] ; /* 0x00005a0000047625 */ /* 0x000fc800078e0205 */ /*00d0*/ FADD R7, -R2, 1 ; /* 0x3f80000002077421 */ /* 0x004fca0000000100 */ /*00e0*/ STG.E [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x000fe2000c101904 */ /*00f0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0100*/ BRA 0x100; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0110*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0120*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0130*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0140*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0150*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0160*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0170*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0180*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0190*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z18InvertValuesKernelPfS_i .globl _Z18InvertValuesKernelPfS_i .p2align 8 .type _Z18InvertValuesKernelPfS_i,@function _Z18InvertValuesKernelPfS_i: s_clause 0x2 s_load_b32 s2, s[0:1], 0x18 s_load_b32 s3, s[0:1], 0x24 s_load_b32 s4, s[0:1], 0x10 s_waitcnt lgkmcnt(0) s_mul_i32 s2, s2, s15 s_and_b32 s3, s3, 0xffff s_add_i32 s2, s2, s14 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s2, s3, v[0:1] s_mov_b32 s2, exec_lo v_cmpx_gt_i32_e64 s4, v1 s_cbranch_execz .LBB0_2 s_load_b128 s[0:3], s[0:1], 0x0 v_ashrrev_i32_e32 v2, 31, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[1:2] s_waitcnt lgkmcnt(0) v_add_co_u32 v2, vcc_lo, s0, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s1, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s2, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s3, v1, vcc_lo global_load_b32 v2, v[2:3], off s_waitcnt vmcnt(0) v_sub_f32_e32 v2, 1.0, v2 global_store_b32 v[0:1], v2, off .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z18InvertValuesKernelPfS_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 280 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 4 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z18InvertValuesKernelPfS_i, .Lfunc_end0-_Z18InvertValuesKernelPfS_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .offset: 16 .size: 4 .value_kind: by_value - .offset: 24 .size: 4 .value_kind: hidden_block_count_x - .offset: 28 .size: 4 .value_kind: hidden_block_count_y - .offset: 32 .size: 4 .value_kind: hidden_block_count_z - .offset: 36 .size: 2 .value_kind: hidden_group_size_x - .offset: 38 .size: 2 .value_kind: hidden_group_size_y - .offset: 40 .size: 2 .value_kind: hidden_group_size_z - .offset: 42 .size: 2 .value_kind: hidden_remainder_x - .offset: 44 .size: 2 .value_kind: hidden_remainder_y - .offset: 46 .size: 2 .value_kind: hidden_remainder_z - .offset: 64 .size: 8 .value_kind: hidden_global_offset_x - .offset: 72 .size: 8 .value_kind: hidden_global_offset_y - .offset: 80 .size: 8 .value_kind: hidden_global_offset_z - .offset: 88 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 280 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z18InvertValuesKernelPfS_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z18InvertValuesKernelPfS_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 4 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_001b6801_00000000-6_InvertValuesKernel.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i .type _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i, @function _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i: .LFB2051: .cfi_startproc endbr64 subq $136, %rsp .cfi_def_cfa_offset 144 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movl %edx, 12(%rsp) movq %fs:40, %rax movq %rax, 120(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 12(%rsp), %rax movq %rax, 112(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 120(%rsp), %rax subq %fs:40, %rax jne .L8 addq $136, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z18InvertValuesKernelPfS_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i, .-_Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i .globl _Z18InvertValuesKernelPfS_i .type _Z18InvertValuesKernelPfS_i, @function _Z18InvertValuesKernelPfS_i: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z41__device_stub__Z18InvertValuesKernelPfS_iPfS_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z18InvertValuesKernelPfS_i, .-_Z18InvertValuesKernelPfS_i .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z18InvertValuesKernelPfS_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z18InvertValuesKernelPfS_i(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "InvertValuesKernel.hip" .globl _Z33__device_stub__InvertValuesKernelPfS_i # -- Begin function _Z33__device_stub__InvertValuesKernelPfS_i .p2align 4, 0x90 .type _Z33__device_stub__InvertValuesKernelPfS_i,@function _Z33__device_stub__InvertValuesKernelPfS_i: # @_Z33__device_stub__InvertValuesKernelPfS_i .cfi_startproc # %bb.0: subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movl %edx, 12(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 12(%rsp), %rax movq %rax, 96(%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z18InvertValuesKernelPfS_i, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $120, %rsp .cfi_adjust_cfa_offset -120 retq .Lfunc_end0: .size _Z33__device_stub__InvertValuesKernelPfS_i, .Lfunc_end0-_Z33__device_stub__InvertValuesKernelPfS_i .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z18InvertValuesKernelPfS_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z18InvertValuesKernelPfS_i,@object # @_Z18InvertValuesKernelPfS_i .section .rodata,"a",@progbits .globl _Z18InvertValuesKernelPfS_i .p2align 3, 0x0 _Z18InvertValuesKernelPfS_i: .quad _Z33__device_stub__InvertValuesKernelPfS_i .size _Z18InvertValuesKernelPfS_i, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z18InvertValuesKernelPfS_i" .size .L__unnamed_1, 28 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z33__device_stub__InvertValuesKernelPfS_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z18InvertValuesKernelPfS_i .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include<stdio.h> #include<cuda.h> __global__ void find_unique_id(int* arr) { int bid = (blockIdx.z * gridDim.y * gridDim.x) + (blockIdx.y * gridDim.x) + blockIdx.x; int tid = (bid * blockDim.x * blockDim.y * blockDim.z) + (threadIdx.z * blockDim.y * blockDim.x) + (threadIdx.y * blockDim.x) + threadIdx.x; arr[tid] = tid; } int main() { dim3 grid(1,2,3); dim3 block(4,5,6); int threads = 1*2*3*4*5*6; int *arr, *darr; arr = (int*)malloc(threads*sizeof(int)); cudaMalloc(&darr, threads*sizeof(int)); find_unique_id<<<grid, block>>>(darr); cudaDeviceSynchronize(); cudaMemcpy(arr, darr, threads*sizeof(int), cudaMemcpyDeviceToHost); for(int i=0; i<threads; i++) printf("%d\n", arr[i]); printf("\n"); return 0; }
code for sm_80 Function : _Z14find_unique_idPi .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R0, SR_CTAID.Z ; /* 0x0000000000007919 */ /* 0x000e220000002700 */ /*0020*/ HFMA2.MMA R2, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff027435 */ /* 0x000fe200000001ff */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe40000000a00 */ /*0040*/ S2R R3, SR_CTAID.Y ; /* 0x0000000000037919 */ /* 0x000e280000002600 */ /*0050*/ S2R R5, SR_CTAID.X ; /* 0x0000000000057919 */ /* 0x000e680000002500 */ /*0060*/ S2R R7, SR_TID.Z ; /* 0x0000000000077919 */ /* 0x000ea80000002300 */ /*0070*/ S2R R9, SR_TID.Y ; /* 0x0000000000097919 */ /* 0x000ee80000002200 */ /*0080*/ S2R R11, SR_TID.X ; /* 0x00000000000b7919 */ /* 0x000f220000002100 */ /*0090*/ IMAD R0, R0, c[0x0][0x10], R3 ; /* 0x0000040000007a24 */ /* 0x001fc800078e0203 */ /*00a0*/ IMAD R0, R0, c[0x0][0xc], R5 ; /* 0x0000030000007a24 */ /* 0x002fc800078e0205 */ /*00b0*/ IMAD R0, R0, c[0x0][0x8], R7 ; /* 0x0000020000007a24 */ /* 0x004fc800078e0207 */ /*00c0*/ IMAD R0, R0, c[0x0][0x4], R9 ; /* 0x0000010000007a24 */ /* 0x008fc800078e0209 */ /*00d0*/ IMAD R5, R0, c[0x0][0x0], R11 ; /* 0x0000000000057a24 */ /* 0x010fc800078e020b */ /*00e0*/ IMAD.WIDE R2, R5, R2, c[0x0][0x160] ; /* 0x0000580005027625 */ /* 0x000fca00078e0202 */ /*00f0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe2000c101904 */ /*0100*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0110*/ BRA 0x110; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0120*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0130*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0140*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0150*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0160*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0170*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0180*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0190*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include<stdio.h> #include<cuda.h> __global__ void find_unique_id(int* arr) { int bid = (blockIdx.z * gridDim.y * gridDim.x) + (blockIdx.y * gridDim.x) + blockIdx.x; int tid = (bid * blockDim.x * blockDim.y * blockDim.z) + (threadIdx.z * blockDim.y * blockDim.x) + (threadIdx.y * blockDim.x) + threadIdx.x; arr[tid] = tid; } int main() { dim3 grid(1,2,3); dim3 block(4,5,6); int threads = 1*2*3*4*5*6; int *arr, *darr; arr = (int*)malloc(threads*sizeof(int)); cudaMalloc(&darr, threads*sizeof(int)); find_unique_id<<<grid, block>>>(darr); cudaDeviceSynchronize(); cudaMemcpy(arr, darr, threads*sizeof(int), cudaMemcpyDeviceToHost); for(int i=0; i<threads; i++) printf("%d\n", arr[i]); printf("\n"); return 0; }
.file "tmpxft_001062d8_00000000-6_unique_id.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2060: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2060: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z34__device_stub__Z14find_unique_idPiPi .type _Z34__device_stub__Z14find_unique_idPiPi, @function _Z34__device_stub__Z14find_unique_idPiPi: .LFB2082: .cfi_startproc endbr64 subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 8(%rsp) movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax leaq 8(%rsp), %rax movq %rax, 80(%rsp) movl $1, 32(%rsp) movl $1, 36(%rsp) movl $1, 40(%rsp) movl $1, 44(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) leaq 24(%rsp), %rcx leaq 16(%rsp), %rdx leaq 44(%rsp), %rsi leaq 32(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 88(%rsp), %rax subq %fs:40, %rax jne .L8 addq $104, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 24(%rsp) .cfi_def_cfa_offset 120 pushq 24(%rsp) .cfi_def_cfa_offset 128 leaq 96(%rsp), %r9 movq 60(%rsp), %rcx movl 68(%rsp), %r8d movq 48(%rsp), %rsi movl 56(%rsp), %edx leaq _Z14find_unique_idPi(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 112 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2082: .size _Z34__device_stub__Z14find_unique_idPiPi, .-_Z34__device_stub__Z14find_unique_idPiPi .globl _Z14find_unique_idPi .type _Z14find_unique_idPi, @function _Z14find_unique_idPi: .LFB2083: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z34__device_stub__Z14find_unique_idPiPi addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2083: .size _Z14find_unique_idPi, .-_Z14find_unique_idPi .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "%d\n" .LC1: .string "\n" .text .globl main .type main, @function main: .LFB2057: .cfi_startproc endbr64 pushq %r12 .cfi_def_cfa_offset 16 .cfi_offset 12, -16 pushq %rbp .cfi_def_cfa_offset 24 .cfi_offset 6, -24 pushq %rbx .cfi_def_cfa_offset 32 .cfi_offset 3, -32 subq $48, %rsp .cfi_def_cfa_offset 80 movq %fs:40, %rax movq %rax, 40(%rsp) xorl %eax, %eax movl $1, 16(%rsp) movl $2, 20(%rsp) movl $3, 24(%rsp) movl $4, 28(%rsp) movl $5, 32(%rsp) movl $6, 36(%rsp) movl $2880, %edi call malloc@PLT movq %rax, %rbp leaq 8(%rsp), %rdi movl $2880, %esi call cudaMalloc@PLT movl 36(%rsp), %ecx movl $0, %r9d movl $0, %r8d movq 28(%rsp), %rdx movq 16(%rsp), %rdi movl 24(%rsp), %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L17 .L12: call cudaDeviceSynchronize@PLT movl $2, %ecx movl $2880, %edx movq 8(%rsp), %rsi movq %rbp, %rdi call cudaMemcpy@PLT movq %rbp, %rbx addq $2880, %rbp leaq .LC0(%rip), %r12 .L13: movl (%rbx), %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L13 leaq .LC1(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 40(%rsp), %rax subq %fs:40, %rax jne .L18 movl $0, %eax addq $48, %rsp .cfi_remember_state .cfi_def_cfa_offset 32 popq %rbx .cfi_def_cfa_offset 24 popq %rbp .cfi_def_cfa_offset 16 popq %r12 .cfi_def_cfa_offset 8 ret .L17: .cfi_restore_state movq 8(%rsp), %rdi call _Z34__device_stub__Z14find_unique_idPiPi jmp .L12 .L18: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size main, .-main .section .rodata.str1.1 .LC2: .string "_Z14find_unique_idPi" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2085: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC2(%rip), %rdx movq %rdx, %rcx leaq _Z14find_unique_idPi(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include<stdio.h> #include<cuda.h> __global__ void find_unique_id(int* arr) { int bid = (blockIdx.z * gridDim.y * gridDim.x) + (blockIdx.y * gridDim.x) + blockIdx.x; int tid = (bid * blockDim.x * blockDim.y * blockDim.z) + (threadIdx.z * blockDim.y * blockDim.x) + (threadIdx.y * blockDim.x) + threadIdx.x; arr[tid] = tid; } int main() { dim3 grid(1,2,3); dim3 block(4,5,6); int threads = 1*2*3*4*5*6; int *arr, *darr; arr = (int*)malloc(threads*sizeof(int)); cudaMalloc(&darr, threads*sizeof(int)); find_unique_id<<<grid, block>>>(darr); cudaDeviceSynchronize(); cudaMemcpy(arr, darr, threads*sizeof(int), cudaMemcpyDeviceToHost); for(int i=0; i<threads; i++) printf("%d\n", arr[i]); printf("\n"); return 0; }
#include<stdio.h> #include<hip/hip_runtime.h> __global__ void find_unique_id(int* arr) { int bid = (blockIdx.z * gridDim.y * gridDim.x) + (blockIdx.y * gridDim.x) + blockIdx.x; int tid = (bid * blockDim.x * blockDim.y * blockDim.z) + (threadIdx.z * blockDim.y * blockDim.x) + (threadIdx.y * blockDim.x) + threadIdx.x; arr[tid] = tid; } int main() { dim3 grid(1,2,3); dim3 block(4,5,6); int threads = 1*2*3*4*5*6; int *arr, *darr; arr = (int*)malloc(threads*sizeof(int)); hipMalloc(&darr, threads*sizeof(int)); find_unique_id<<<grid, block>>>(darr); hipDeviceSynchronize(); hipMemcpy(arr, darr, threads*sizeof(int), hipMemcpyDeviceToHost); for(int i=0; i<threads; i++) printf("%d\n", arr[i]); printf("\n"); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include<stdio.h> #include<hip/hip_runtime.h> __global__ void find_unique_id(int* arr) { int bid = (blockIdx.z * gridDim.y * gridDim.x) + (blockIdx.y * gridDim.x) + blockIdx.x; int tid = (bid * blockDim.x * blockDim.y * blockDim.z) + (threadIdx.z * blockDim.y * blockDim.x) + (threadIdx.y * blockDim.x) + threadIdx.x; arr[tid] = tid; } int main() { dim3 grid(1,2,3); dim3 block(4,5,6); int threads = 1*2*3*4*5*6; int *arr, *darr; arr = (int*)malloc(threads*sizeof(int)); hipMalloc(&darr, threads*sizeof(int)); find_unique_id<<<grid, block>>>(darr); hipDeviceSynchronize(); hipMemcpy(arr, darr, threads*sizeof(int), hipMemcpyDeviceToHost); for(int i=0; i<threads; i++) printf("%d\n", arr[i]); printf("\n"); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z14find_unique_idPi .globl _Z14find_unique_idPi .p2align 8 .type _Z14find_unique_idPi,@function _Z14find_unique_idPi: s_clause 0x1 s_load_b128 s[4:7], s[0:1], 0x0 s_load_b64 s[0:1], s[0:1], 0x14 v_bfe_u32 v1, v0, 20, 10 s_waitcnt lgkmcnt(0) s_mul_i32 s2, s7, s15 s_and_b32 s1, s1, 0xffff s_add_i32 s2, s2, s14 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_mul_i32 s2, s2, s6 s_add_i32 s2, s2, s13 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(VALU_DEP_2) v_mad_u64_u32 v[2:3], null, s2, s1, v[1:2] v_bfe_u32 v1, v0, 10, 10 s_lshr_b32 s1, s0, 16 v_and_b32_e32 v0, 0x3ff, v0 s_and_b32 s0, s0, 0xffff v_mad_u64_u32 v[3:4], null, v2, s1, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, v3, s0, v[0:1] v_ashrrev_i32_e32 v2, 31, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[2:3], 2, v[1:2] v_add_co_u32 v2, vcc_lo, s4, v2 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s5, v3, vcc_lo global_store_b32 v[2:3], v1, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z14find_unique_idPi .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 264 .amdhsa_user_sgpr_count 13 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 1 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 2 .amdhsa_next_free_vgpr 5 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z14find_unique_idPi, .Lfunc_end0-_Z14find_unique_idPi .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .offset: 8 .size: 4 .value_kind: hidden_block_count_x - .offset: 12 .size: 4 .value_kind: hidden_block_count_y - .offset: 16 .size: 4 .value_kind: hidden_block_count_z - .offset: 20 .size: 2 .value_kind: hidden_group_size_x - .offset: 22 .size: 2 .value_kind: hidden_group_size_y - .offset: 24 .size: 2 .value_kind: hidden_group_size_z - .offset: 26 .size: 2 .value_kind: hidden_remainder_x - .offset: 28 .size: 2 .value_kind: hidden_remainder_y - .offset: 30 .size: 2 .value_kind: hidden_remainder_z - .offset: 48 .size: 8 .value_kind: hidden_global_offset_x - .offset: 56 .size: 8 .value_kind: hidden_global_offset_y - .offset: 64 .size: 8 .value_kind: hidden_global_offset_z - .offset: 72 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 264 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z14find_unique_idPi .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z14find_unique_idPi.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 5 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include<stdio.h> #include<hip/hip_runtime.h> __global__ void find_unique_id(int* arr) { int bid = (blockIdx.z * gridDim.y * gridDim.x) + (blockIdx.y * gridDim.x) + blockIdx.x; int tid = (bid * blockDim.x * blockDim.y * blockDim.z) + (threadIdx.z * blockDim.y * blockDim.x) + (threadIdx.y * blockDim.x) + threadIdx.x; arr[tid] = tid; } int main() { dim3 grid(1,2,3); dim3 block(4,5,6); int threads = 1*2*3*4*5*6; int *arr, *darr; arr = (int*)malloc(threads*sizeof(int)); hipMalloc(&darr, threads*sizeof(int)); find_unique_id<<<grid, block>>>(darr); hipDeviceSynchronize(); hipMemcpy(arr, darr, threads*sizeof(int), hipMemcpyDeviceToHost); for(int i=0; i<threads; i++) printf("%d\n", arr[i]); printf("\n"); return 0; }
.text .file "unique_id.hip" .globl _Z29__device_stub__find_unique_idPi # -- Begin function _Z29__device_stub__find_unique_idPi .p2align 4, 0x90 .type _Z29__device_stub__find_unique_idPi,@function _Z29__device_stub__find_unique_idPi: # @_Z29__device_stub__find_unique_idPi .cfi_startproc # %bb.0: subq $72, %rsp .cfi_def_cfa_offset 80 movq %rdi, 64(%rsp) leaq 64(%rsp), %rax movq %rax, (%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d movq %rsp, %r9 movl $_Z14find_unique_idPi, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $88, %rsp .cfi_adjust_cfa_offset -88 retq .Lfunc_end0: .size _Z29__device_stub__find_unique_idPi, .Lfunc_end0-_Z29__device_stub__find_unique_idPi .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %r14 .cfi_def_cfa_offset 16 pushq %rbx .cfi_def_cfa_offset 24 subq $88, %rsp .cfi_def_cfa_offset 112 .cfi_offset %rbx, -24 .cfi_offset %r14, -16 movl $2880, %edi # imm = 0xB40 callq malloc movq %rax, %rbx leaq 8(%rsp), %rdi movl $2880, %esi # imm = 0xB40 callq hipMalloc movabsq $8589934593, %rdi # imm = 0x200000001 movabsq $21474836484, %rdx # imm = 0x500000004 movl $3, %esi movl $6, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_2 # %bb.1: movq 8(%rsp), %rax movq %rax, 80(%rsp) leaq 80(%rsp), %rax movq %rax, 16(%rsp) leaq 64(%rsp), %rdi leaq 48(%rsp), %rsi leaq 40(%rsp), %rdx leaq 32(%rsp), %rcx callq __hipPopCallConfiguration movq 64(%rsp), %rsi movl 72(%rsp), %edx movq 48(%rsp), %rcx movl 56(%rsp), %r8d leaq 16(%rsp), %r9 movl $_Z14find_unique_idPi, %edi pushq 32(%rsp) .cfi_adjust_cfa_offset 8 pushq 48(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB1_2: callq hipDeviceSynchronize movq 8(%rsp), %rsi movl $2880, %edx # imm = 0xB40 movq %rbx, %rdi movl $2, %ecx callq hipMemcpy xorl %r14d, %r14d .p2align 4, 0x90 .LBB1_3: # =>This Inner Loop Header: Depth=1 movl (%rbx,%r14,4), %esi movl $.L.str, %edi xorl %eax, %eax callq printf incq %r14 cmpq $720, %r14 # imm = 0x2D0 jne .LBB1_3 # %bb.4: movl $10, %edi callq putchar@PLT xorl %eax, %eax addq $88, %rsp .cfi_def_cfa_offset 24 popq %rbx .cfi_def_cfa_offset 16 popq %r14 .cfi_def_cfa_offset 8 retq .Lfunc_end1: .size main, .Lfunc_end1-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB2_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB2_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z14find_unique_idPi, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end2: .size __hip_module_ctor, .Lfunc_end2-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB3_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB3_2: retq .Lfunc_end3: .size __hip_module_dtor, .Lfunc_end3-__hip_module_dtor .cfi_endproc # -- End function .type _Z14find_unique_idPi,@object # @_Z14find_unique_idPi .section .rodata,"a",@progbits .globl _Z14find_unique_idPi .p2align 3, 0x0 _Z14find_unique_idPi: .quad _Z29__device_stub__find_unique_idPi .size _Z14find_unique_idPi, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%d\n" .size .L.str, 4 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z14find_unique_idPi" .size .L__unnamed_1, 21 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z29__device_stub__find_unique_idPi .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z14find_unique_idPi .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z14find_unique_idPi .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R0, SR_CTAID.Z ; /* 0x0000000000007919 */ /* 0x000e220000002700 */ /*0020*/ HFMA2.MMA R2, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff027435 */ /* 0x000fe200000001ff */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe40000000a00 */ /*0040*/ S2R R3, SR_CTAID.Y ; /* 0x0000000000037919 */ /* 0x000e280000002600 */ /*0050*/ S2R R5, SR_CTAID.X ; /* 0x0000000000057919 */ /* 0x000e680000002500 */ /*0060*/ S2R R7, SR_TID.Z ; /* 0x0000000000077919 */ /* 0x000ea80000002300 */ /*0070*/ S2R R9, SR_TID.Y ; /* 0x0000000000097919 */ /* 0x000ee80000002200 */ /*0080*/ S2R R11, SR_TID.X ; /* 0x00000000000b7919 */ /* 0x000f220000002100 */ /*0090*/ IMAD R0, R0, c[0x0][0x10], R3 ; /* 0x0000040000007a24 */ /* 0x001fc800078e0203 */ /*00a0*/ IMAD R0, R0, c[0x0][0xc], R5 ; /* 0x0000030000007a24 */ /* 0x002fc800078e0205 */ /*00b0*/ IMAD R0, R0, c[0x0][0x8], R7 ; /* 0x0000020000007a24 */ /* 0x004fc800078e0207 */ /*00c0*/ IMAD R0, R0, c[0x0][0x4], R9 ; /* 0x0000010000007a24 */ /* 0x008fc800078e0209 */ /*00d0*/ IMAD R5, R0, c[0x0][0x0], R11 ; /* 0x0000000000057a24 */ /* 0x010fc800078e020b */ /*00e0*/ IMAD.WIDE R2, R5, R2, c[0x0][0x160] ; /* 0x0000580005027625 */ /* 0x000fca00078e0202 */ /*00f0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe2000c101904 */ /*0100*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0110*/ BRA 0x110; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0120*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0130*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0140*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0150*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0160*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0170*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0180*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0190*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z14find_unique_idPi .globl _Z14find_unique_idPi .p2align 8 .type _Z14find_unique_idPi,@function _Z14find_unique_idPi: s_clause 0x1 s_load_b128 s[4:7], s[0:1], 0x0 s_load_b64 s[0:1], s[0:1], 0x14 v_bfe_u32 v1, v0, 20, 10 s_waitcnt lgkmcnt(0) s_mul_i32 s2, s7, s15 s_and_b32 s1, s1, 0xffff s_add_i32 s2, s2, s14 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_mul_i32 s2, s2, s6 s_add_i32 s2, s2, s13 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(VALU_DEP_2) v_mad_u64_u32 v[2:3], null, s2, s1, v[1:2] v_bfe_u32 v1, v0, 10, 10 s_lshr_b32 s1, s0, 16 v_and_b32_e32 v0, 0x3ff, v0 s_and_b32 s0, s0, 0xffff v_mad_u64_u32 v[3:4], null, v2, s1, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, v3, s0, v[0:1] v_ashrrev_i32_e32 v2, 31, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[2:3], 2, v[1:2] v_add_co_u32 v2, vcc_lo, s4, v2 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s5, v3, vcc_lo global_store_b32 v[2:3], v1, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z14find_unique_idPi .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 264 .amdhsa_user_sgpr_count 13 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 1 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 2 .amdhsa_next_free_vgpr 5 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z14find_unique_idPi, .Lfunc_end0-_Z14find_unique_idPi .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .offset: 8 .size: 4 .value_kind: hidden_block_count_x - .offset: 12 .size: 4 .value_kind: hidden_block_count_y - .offset: 16 .size: 4 .value_kind: hidden_block_count_z - .offset: 20 .size: 2 .value_kind: hidden_group_size_x - .offset: 22 .size: 2 .value_kind: hidden_group_size_y - .offset: 24 .size: 2 .value_kind: hidden_group_size_z - .offset: 26 .size: 2 .value_kind: hidden_remainder_x - .offset: 28 .size: 2 .value_kind: hidden_remainder_y - .offset: 30 .size: 2 .value_kind: hidden_remainder_z - .offset: 48 .size: 8 .value_kind: hidden_global_offset_x - .offset: 56 .size: 8 .value_kind: hidden_global_offset_y - .offset: 64 .size: 8 .value_kind: hidden_global_offset_z - .offset: 72 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 264 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z14find_unique_idPi .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z14find_unique_idPi.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 5 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_001062d8_00000000-6_unique_id.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2060: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2060: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z34__device_stub__Z14find_unique_idPiPi .type _Z34__device_stub__Z14find_unique_idPiPi, @function _Z34__device_stub__Z14find_unique_idPiPi: .LFB2082: .cfi_startproc endbr64 subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 8(%rsp) movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax leaq 8(%rsp), %rax movq %rax, 80(%rsp) movl $1, 32(%rsp) movl $1, 36(%rsp) movl $1, 40(%rsp) movl $1, 44(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) leaq 24(%rsp), %rcx leaq 16(%rsp), %rdx leaq 44(%rsp), %rsi leaq 32(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 88(%rsp), %rax subq %fs:40, %rax jne .L8 addq $104, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 24(%rsp) .cfi_def_cfa_offset 120 pushq 24(%rsp) .cfi_def_cfa_offset 128 leaq 96(%rsp), %r9 movq 60(%rsp), %rcx movl 68(%rsp), %r8d movq 48(%rsp), %rsi movl 56(%rsp), %edx leaq _Z14find_unique_idPi(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 112 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2082: .size _Z34__device_stub__Z14find_unique_idPiPi, .-_Z34__device_stub__Z14find_unique_idPiPi .globl _Z14find_unique_idPi .type _Z14find_unique_idPi, @function _Z14find_unique_idPi: .LFB2083: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z34__device_stub__Z14find_unique_idPiPi addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2083: .size _Z14find_unique_idPi, .-_Z14find_unique_idPi .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "%d\n" .LC1: .string "\n" .text .globl main .type main, @function main: .LFB2057: .cfi_startproc endbr64 pushq %r12 .cfi_def_cfa_offset 16 .cfi_offset 12, -16 pushq %rbp .cfi_def_cfa_offset 24 .cfi_offset 6, -24 pushq %rbx .cfi_def_cfa_offset 32 .cfi_offset 3, -32 subq $48, %rsp .cfi_def_cfa_offset 80 movq %fs:40, %rax movq %rax, 40(%rsp) xorl %eax, %eax movl $1, 16(%rsp) movl $2, 20(%rsp) movl $3, 24(%rsp) movl $4, 28(%rsp) movl $5, 32(%rsp) movl $6, 36(%rsp) movl $2880, %edi call malloc@PLT movq %rax, %rbp leaq 8(%rsp), %rdi movl $2880, %esi call cudaMalloc@PLT movl 36(%rsp), %ecx movl $0, %r9d movl $0, %r8d movq 28(%rsp), %rdx movq 16(%rsp), %rdi movl 24(%rsp), %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L17 .L12: call cudaDeviceSynchronize@PLT movl $2, %ecx movl $2880, %edx movq 8(%rsp), %rsi movq %rbp, %rdi call cudaMemcpy@PLT movq %rbp, %rbx addq $2880, %rbp leaq .LC0(%rip), %r12 .L13: movl (%rbx), %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L13 leaq .LC1(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 40(%rsp), %rax subq %fs:40, %rax jne .L18 movl $0, %eax addq $48, %rsp .cfi_remember_state .cfi_def_cfa_offset 32 popq %rbx .cfi_def_cfa_offset 24 popq %rbp .cfi_def_cfa_offset 16 popq %r12 .cfi_def_cfa_offset 8 ret .L17: .cfi_restore_state movq 8(%rsp), %rdi call _Z34__device_stub__Z14find_unique_idPiPi jmp .L12 .L18: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size main, .-main .section .rodata.str1.1 .LC2: .string "_Z14find_unique_idPi" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2085: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC2(%rip), %rdx movq %rdx, %rcx leaq _Z14find_unique_idPi(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "unique_id.hip" .globl _Z29__device_stub__find_unique_idPi # -- Begin function _Z29__device_stub__find_unique_idPi .p2align 4, 0x90 .type _Z29__device_stub__find_unique_idPi,@function _Z29__device_stub__find_unique_idPi: # @_Z29__device_stub__find_unique_idPi .cfi_startproc # %bb.0: subq $72, %rsp .cfi_def_cfa_offset 80 movq %rdi, 64(%rsp) leaq 64(%rsp), %rax movq %rax, (%rsp) leaq 48(%rsp), %rdi leaq 32(%rsp), %rsi leaq 24(%rsp), %rdx leaq 16(%rsp), %rcx callq __hipPopCallConfiguration movq 48(%rsp), %rsi movl 56(%rsp), %edx movq 32(%rsp), %rcx movl 40(%rsp), %r8d movq %rsp, %r9 movl $_Z14find_unique_idPi, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $88, %rsp .cfi_adjust_cfa_offset -88 retq .Lfunc_end0: .size _Z29__device_stub__find_unique_idPi, .Lfunc_end0-_Z29__device_stub__find_unique_idPi .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %r14 .cfi_def_cfa_offset 16 pushq %rbx .cfi_def_cfa_offset 24 subq $88, %rsp .cfi_def_cfa_offset 112 .cfi_offset %rbx, -24 .cfi_offset %r14, -16 movl $2880, %edi # imm = 0xB40 callq malloc movq %rax, %rbx leaq 8(%rsp), %rdi movl $2880, %esi # imm = 0xB40 callq hipMalloc movabsq $8589934593, %rdi # imm = 0x200000001 movabsq $21474836484, %rdx # imm = 0x500000004 movl $3, %esi movl $6, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_2 # %bb.1: movq 8(%rsp), %rax movq %rax, 80(%rsp) leaq 80(%rsp), %rax movq %rax, 16(%rsp) leaq 64(%rsp), %rdi leaq 48(%rsp), %rsi leaq 40(%rsp), %rdx leaq 32(%rsp), %rcx callq __hipPopCallConfiguration movq 64(%rsp), %rsi movl 72(%rsp), %edx movq 48(%rsp), %rcx movl 56(%rsp), %r8d leaq 16(%rsp), %r9 movl $_Z14find_unique_idPi, %edi pushq 32(%rsp) .cfi_adjust_cfa_offset 8 pushq 48(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB1_2: callq hipDeviceSynchronize movq 8(%rsp), %rsi movl $2880, %edx # imm = 0xB40 movq %rbx, %rdi movl $2, %ecx callq hipMemcpy xorl %r14d, %r14d .p2align 4, 0x90 .LBB1_3: # =>This Inner Loop Header: Depth=1 movl (%rbx,%r14,4), %esi movl $.L.str, %edi xorl %eax, %eax callq printf incq %r14 cmpq $720, %r14 # imm = 0x2D0 jne .LBB1_3 # %bb.4: movl $10, %edi callq putchar@PLT xorl %eax, %eax addq $88, %rsp .cfi_def_cfa_offset 24 popq %rbx .cfi_def_cfa_offset 16 popq %r14 .cfi_def_cfa_offset 8 retq .Lfunc_end1: .size main, .Lfunc_end1-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB2_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB2_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z14find_unique_idPi, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end2: .size __hip_module_ctor, .Lfunc_end2-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB3_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB3_2: retq .Lfunc_end3: .size __hip_module_dtor, .Lfunc_end3-__hip_module_dtor .cfi_endproc # -- End function .type _Z14find_unique_idPi,@object # @_Z14find_unique_idPi .section .rodata,"a",@progbits .globl _Z14find_unique_idPi .p2align 3, 0x0 _Z14find_unique_idPi: .quad _Z29__device_stub__find_unique_idPi .size _Z14find_unique_idPi, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%d\n" .size .L.str, 4 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z14find_unique_idPi" .size .L__unnamed_1, 21 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z29__device_stub__find_unique_idPi .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z14find_unique_idPi .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include <stdio.h> #include <cuda.h> #include <stdlib.h> #define Tile_size 2 int numARows, numAColumns; int numBRows, numBColumns; int numCRows, numCColumns; __global__ void matrixMultiplyShared(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { __shared__ float sA[Tile_size][Tile_size]; // Tile size to store elements in shared memory __shared__ float sB[Tile_size][Tile_size]; int row = blockDim.y * blockIdx.y + threadIdx.y; int col = blockDim.x * blockIdx.x + threadIdx.x; float cvalue = 0.0; sA[threadIdx.y][threadIdx.x] = 0.0; sB[threadIdx.y][threadIdx.x] = 0.0; for (int k = 0; k < (((numAColumns - 1) / Tile_size) + 1); k++) { //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if ((row < numARows) && (threadIdx.x + (k * Tile_size)) < numAColumns) { sA[threadIdx.y][threadIdx.x] = A[(row * numAColumns) + threadIdx.x + (k * Tile_size)]; } else { sA[threadIdx.y][threadIdx.x] = 0.0; } //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if (col < numBColumns && (threadIdx.y + k * Tile_size) < numBRows) { sB[threadIdx.y][threadIdx.x] = B[(threadIdx.y + k * Tile_size) * numBColumns + col]; } else { sB[threadIdx.y][threadIdx.x] = 0.0; } __syncthreads(); for (int j = 0; j < Tile_size; ++j)//Multiplying Elements present in tile { cvalue += sA[threadIdx.y][j] * sB[j][threadIdx.x]; } } if (row < numCRows && col < numCColumns)//Saving Final result into Matrix C { C[row * numCColumns + col] = cvalue; } } void printMat(int row, int col, float *Mat) { for (int i = 0; i < row * col; i++) { printf("%f ", *(Mat + i)); if ((i % col) == 0 && i != 0) { printf("\n"); } } } void matMultiplyOnHost(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { for (int i = 0; i < numARows; i++) { for (int j = 0; j < numAColumns; j++) { C[i * numCColumns + j] = 0.0; for (int k = 0; k < numCColumns; k++) { C[i * numCColumns + j] += A[i * numAColumns + k] * B[k * numBColumns + j]; } } } } int main(int argc, char **argv) { float *hostA, *hostB, *hostC; float *hostComputedC; float *deviceA, *deviceB, *deviceC; printf("\nEnter Rows and Columns of A:"); scanf("%d %d", &numARows, &numAColumns); printf("\nEnter Rows and Columns of B:"); scanf("%d %d", &numBRows, &numBColumns); hostA = (float *) malloc(sizeof(float) * numARows * numAColumns); hostB = (float *) malloc(sizeof(float) * numBRows * numBColumns); int lower = 10.0, upper = 20.0; for (int i = 0; i < numARows * numAColumns; i++) hostA[i] = (rand() % (upper - lower + 1)) + lower; for (int i = 0; i < numBRows * numBColumns; i++) hostB[i] = (rand() % (upper - lower + 1)) + lower; printf("\nMatrix A Values:\n"); printMat(numARows, numAColumns, hostA); printf("\n\nMatrix B Values:\n"); printMat(numBRows, numBColumns, hostB); numCRows = numARows; numCColumns = numBColumns; hostC = (float *) malloc(sizeof(float) * numCRows * numCColumns); hostComputedC = (float *) malloc(sizeof(float) * numCRows * numCColumns); // Allocating GPU memory cudaMalloc((void **) &deviceA, sizeof(float) * numARows * numAColumns); cudaMalloc((void **) &deviceB, sizeof(float) * numBRows * numBColumns); cudaMalloc((void **) &deviceC, sizeof(float) * numCRows * numCColumns); // Copy memory to the GPU cudaMemcpy(deviceA, hostA, sizeof(float) * numARows * numAColumns, cudaMemcpyHostToDevice); cudaMemcpy(deviceB, hostB, sizeof(float) * numBRows * numBColumns, cudaMemcpyHostToDevice); // Initialize the grid and block dimensions dim3 dimGrid((numCColumns / Tile_size) + 1, (numCRows / Tile_size) + 1, 1);//Number of Blocks required dim3 dimBlock(Tile_size, Tile_size, 1);//Number of threads in each block matrixMultiplyShared<<<dimGrid, dimBlock>>>(deviceA, deviceB, deviceC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); cudaError_t err1 = cudaPeekAtLastError();//To capture last error in function call cudaDeviceSynchronize();//To synchronize the device // Copy the results in GPU memory back to the CPU cudaMemcpy(hostC, deviceC, sizeof(float) * numCRows * numCColumns, cudaMemcpyDeviceToHost); printf("\nMatrix C From Device\n"); printMat(numCRows, numCColumns, hostC);//Function Call matMultiplyOnHost(hostA, hostB, hostComputedC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); printf("\nMatrix C From Host\n"); printMat(numCRows, numCColumns, hostComputedC); printf("\n\n"); for (int i = 0; i < numCColumns * numCRows; i++)//Compare both the result matrices 1. MatrixMultiplyonHost 2. MatrixMultiplyonDevice { if (hostComputedC[i] != hostC[i]) { printf("Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n", i / numCColumns, i % numCColumns, hostComputedC[i], hostC[i]); break; } } printf("\nNumber of Blocks Created:%d \n", ((numCColumns / Tile_size) + 1) * ((numCColumns / Tile_size) + 1)); printf("\nNumber of Threads Per Block: %d \n", (Tile_size * Tile_size)); // Free the GPU memory cudaFree(deviceA); cudaFree(deviceB); cudaFree(deviceC); //Free the Pointer Memory free(hostA); free(hostB); free(hostC); free(hostComputedC); return 0; }
code for sm_80 Function : _Z20matrixMultiplySharedPfS_S_iiiiii .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R0, SR_TID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002200 */ /*0020*/ ISETP.LE.AND P0, PT, RZ, c[0x0][0x17c], PT ; /* 0x00005f00ff007a0c */ /* 0x000fe20003f03270 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fe200078e00ff */ /*0050*/ S2R R2, SR_TID.X ; /* 0x0000000000027919 */ /* 0x000e680000002100 */ /*0060*/ S2R R5, SR_CTAID.Y ; /* 0x0000000000057919 */ /* 0x000ea80000002600 */ /*0070*/ S2R R7, SR_CTAID.X ; /* 0x0000000000077919 */ /* 0x000ee20000002500 */ /*0080*/ IMAD.SHL.U32 R3, R0, 0x8, RZ ; /* 0x0000000800037824 */ /* 0x001fca00078e00ff */ /*0090*/ LEA R3, R2, R3, 0x2 ; /* 0x0000000302037211 */ /* 0x002fe200078e10ff */ /*00a0*/ IMAD R5, R5, c[0x0][0x4], R0 ; /* 0x0000010005057a24 */ /* 0x004fc800078e0200 */ /*00b0*/ STS [R3], RZ ; /* 0x000000ff03007388 */ /* 0x0001e20000000800 */ /*00c0*/ IMAD R6, R7, c[0x0][0x0], R2 ; /* 0x0000000007067a24 */ /* 0x008fc600078e0202 */ /*00d0*/ STS [R3+0x10], RZ ; /* 0x000010ff03007388 */ /* 0x0001e20000000800 */ /*00e0*/ @!P0 BRA 0xab0 ; /* 0x000009c000008947 */ /* 0x000fea0003800000 */ /*00f0*/ IMAD.MOV.U32 R8, RZ, RZ, c[0x0][0x17c] ; /* 0x00005f00ff087624 */ /* 0x000fe200078e00ff */ /*0100*/ HFMA2.MMA R9, -RZ, RZ, 0, 0 ; /* 0x00000000ff097435 */ /* 0x000fe200000001ff */ /*0110*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fc600078e00ff */ /*0120*/ IADD3 R4, R8, -0x1, RZ ; /* 0xffffffff08047810 */ /* 0x000fc80007ffe0ff */ /*0130*/ LEA.HI R4, R4, R4, RZ, 0x1 ; /* 0x0000000404047211 */ /* 0x000fc800078f08ff */ /*0140*/ SHF.R.S32.HI R4, RZ, 0x1, R4 ; /* 0x00000001ff047819 */ /* 0x000fc80000011404 */ /*0150*/ IMNMX R4, RZ, R4, !PT ; /* 0x00000004ff047217 */ /* 0x000fc80007800200 */ /*0160*/ ISETP.GE.U32.AND P0, PT, R4.reuse, 0x3, PT ; /* 0x000000030400780c */ /* 0x040fe40003f06070 */ /*0170*/ IADD3 R7, R4, 0x1, RZ ; /* 0x0000000104077810 */ /* 0x000fc80007ffe0ff */ /*0180*/ LOP3.LUT R7, R7, 0x3, RZ, 0xc0, !PT ; /* 0x0000000307077812 */ /* 0x000fc800078ec0ff */ /*0190*/ ISETP.NE.AND P1, PT, R7, RZ, PT ; /* 0x000000ff0700720c */ /* 0x000fc60003f25270 */ /*01a0*/ @!P0 BRA 0x880 ; /* 0x000006d000008947 */ /* 0x000ff40003800000 */ /*01b0*/ IMAD R21, R5.reuse, c[0x0][0x17c], R2 ; /* 0x00005f0005157a24 */ /* 0x040fe200078e0202 */ /*01c0*/ IADD3 R11, R0.reuse, 0x6, RZ ; /* 0x00000006000b7810 */ /* 0x040fe20007ffe0ff */ /*01d0*/ IMAD R8, R5.reuse, R8, 0x2 ; /* 0x0000000205087424 */ /* 0x040fe200078e0208 */ /*01e0*/ IADD3 R23, R0.reuse, 0x4, RZ ; /* 0x0000000400177810 */ /* 0x040fe20007ffe0ff */ /*01f0*/ IMAD R22, R0.reuse, c[0x0][0x184], R6 ; /* 0x0000610000167a24 */ /* 0x040fe200078e0206 */ /*0200*/ IADD3 R13, R0, 0x2, RZ ; /* 0x00000002000d7810 */ /* 0x000fe20007ffe0ff */ /*0210*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0220*/ IADD3 R10, -R4, -0x1, R7 ; /* 0xffffffff040a7810 */ /* 0x000fe20007ffe107 */ /*0230*/ IMAD.MOV.U32 R20, RZ, RZ, R2 ; /* 0x000000ffff147224 */ /* 0x000fe200078e0002 */ /*0240*/ ISETP.GE.AND P2, PT, R5, c[0x0][0x178], PT ; /* 0x00005e0005007a0c */ /* 0x000fe20003f46270 */ /*0250*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fe200078e00ff */ /*0260*/ MOV R4, R0 ; /* 0x0000000000047202 */ /* 0x000fe20000000f00 */ /*0270*/ IMAD R11, R11, c[0x0][0x184], R6.reuse ; /* 0x000061000b0b7a24 */ /* 0x100fe200078e0206 */ /*0280*/ IADD3 R21, R21, 0x6, RZ ; /* 0x0000000615157810 */ /* 0x000fe20007ffe0ff */ /*0290*/ IMAD R23, R23, c[0x0][0x184], R6 ; /* 0x0000610017177a24 */ /* 0x000fc400078e0206 */ /*02a0*/ IMAD R24, R13, c[0x0][0x184], R6 ; /* 0x000061000d187a24 */ /* 0x000fe400078e0206 */ /*02b0*/ ISETP.GE.AND P0, PT, R6, c[0x0][0x184], PT ; /* 0x0000610006007a0c */ /* 0x000fe20003f06270 */ /*02c0*/ HFMA2.MMA R16, -RZ, RZ, 0, 0 ; /* 0x00000000ff107435 */ /* 0x000fe200000001ff */ /*02d0*/ ISETP.GE.U32.OR P3, PT, R20, c[0x0][0x17c], P2 ; /* 0x00005f0014007a0c */ /* 0x000fe20001766470 */ /*02e0*/ IMAD.MOV.U32 R28, RZ, RZ, RZ ; /* 0x000000ffff1c7224 */ /* 0x000fe200078e00ff */ /*02f0*/ ISETP.GE.U32.OR P4, PT, R4, c[0x0][0x180], P0 ; /* 0x0000600004007a0c */ /* 0x000fd60000786470 */ /*0300*/ @!P3 MOV R15, 0x4 ; /* 0x00000004000fb802 */ /* 0x000fe20000000f00 */ /*0310*/ @!P3 IMAD R14, R5, c[0x0][0x17c], R20 ; /* 0x00005f00050eba24 */ /* 0x000fe400078e0214 */ /*0320*/ @!P4 IMAD.MOV.U32 R13, RZ, RZ, 0x4 ; /* 0x00000004ff0dc424 */ /* 0x000fe400078e00ff */ /*0330*/ @!P3 IMAD.WIDE.U32 R14, R14, R15, c[0x0][0x160] ; /* 0x000058000e0eb625 */ /* 0x000fc800078e000f */ /*0340*/ @!P4 IMAD.WIDE.U32 R12, R22, R13, c[0x0][0x168] ; /* 0x00005a00160cc625 */ /* 0x000fe200078e000d */ /*0350*/ @!P3 LDG.E R28, [R14.64] ; /* 0x000000040e1cb981 */ /* 0x000ea8000c1e1900 */ /*0360*/ @!P4 LDG.E R16, [R12.64] ; /* 0x000000040c10c981 */ /* 0x0002e2000c1e1900 */ /*0370*/ IADD3 R17, R20, 0x2, RZ ; /* 0x0000000214117810 */ /* 0x000fe20007ffe0ff */ /*0380*/ CS2R R26, SRZ ; /* 0x00000000001a7805 */ /* 0x000fc6000001ff00 */ /*0390*/ ISETP.GE.U32.OR P3, PT, R17, c[0x0][0x17c], P2 ; /* 0x00005f0011007a0c */ /* 0x000fe40001766470 */ /*03a0*/ IADD3 R17, R4, 0x2, RZ ; /* 0x0000000204117810 */ /* 0x000fc80007ffe0ff */ /*03b0*/ ISETP.GE.U32.OR P4, PT, R17, c[0x0][0x180], P0 ; /* 0x0000600011007a0c */ /* 0x000fce0000786470 */ /*03c0*/ @!P3 IMAD.IADD R18, R8, 0x1, R20 ; /* 0x000000010812b824 */ /* 0x000fe400078e0214 */ /*03d0*/ @!P3 IMAD.MOV.U32 R19, RZ, RZ, 0x4 ; /* 0x00000004ff13b424 */ /* 0x000fc800078e00ff */ /*03e0*/ @!P4 MOV R17, 0x4 ; /* 0x000000040011c802 */ /* 0x000fe20000000f00 */ /*03f0*/ @!P3 IMAD.WIDE.U32 R18, R18, R19, c[0x0][0x160] ; /* 0x000058001212b625 */ /* 0x000fc800078e0013 */ /*0400*/ @!P4 IMAD.WIDE.U32 R12, R24, R17, c[0x0][0x168] ; /* 0x00005a00180cc625 */ /* 0x002fe200078e0011 */ /*0410*/ STS [R3], R28 ; /* 0x0000001c03007388 */ /* 0x004fe80000000800 */ /*0420*/ STS [R3+0x10], R16 ; /* 0x0000101003007388 */ /* 0x0083e80000000800 */ /*0430*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0440*/ @!P3 LDG.E R26, [R18.64] ; /* 0x00000004121ab981 */ /* 0x000ea8000c1e1900 */ /*0450*/ @!P4 LDG.E R27, [R12.64] ; /* 0x000000040c1bc981 */ /* 0x000722000c1e1900 */ /*0460*/ IADD3 R14, R20, 0x4, RZ ; /* 0x00000004140e7810 */ /* 0x000fc60007ffe0ff */ /*0470*/ LDS R28, [R2.X4+0x18] ; /* 0x00001800021c7984 */ /* 0x000fe20000004800 */ /*0480*/ ISETP.GE.U32.OR P3, PT, R14, c[0x0][0x17c], P2 ; /* 0x00005f000e007a0c */ /* 0x000fc60001766470 */ /*0490*/ LDS.64 R14, [R0.X8] ; /* 0x00000000000e7984 */ /* 0x000fe80000008a00 */ /*04a0*/ LDS R12, [R2.X4+0x10] ; /* 0x00001000020c7984 */ /* 0x008eec0000004800 */ /*04b0*/ @!P3 IADD3 R17, R21, -0x2, RZ ; /* 0xfffffffe1511b810 */ /* 0x000fe20007ffe0ff */ /*04c0*/ @!P3 IMAD.MOV.U32 R29, RZ, RZ, 0x4 ; /* 0x00000004ff1db424 */ /* 0x000fc800078e00ff */ /*04d0*/ @!P3 IMAD.WIDE.U32 R16, R17, R29, c[0x0][0x160] ; /* 0x000058001110b625 */ /* 0x002fe200078e001d */ /*04e0*/ HFMA2.MMA R29, -RZ, RZ, 0, 0 ; /* 0x00000000ff1d7435 */ /* 0x000fe200000001ff */ /*04f0*/ IADD3 R13, R4, 0x4, RZ ; /* 0x00000004040d7810 */ /* 0x000fe20007ffe0ff */ /*0500*/ STS [R3], R26 ; /* 0x0000001a03007388 */ /* 0x004fe80000000800 */ /*0510*/ STS [R3+0x10], R27 ; /* 0x0000101b03007388 */ /* 0x0107e80000000800 */ /*0520*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0530*/ @!P3 LDG.E R29, [R16.64] ; /* 0x00000004101db981 */ /* 0x0002a2000c1e1900 */ /*0540*/ ISETP.GE.U32.OR P3, PT, R13, c[0x0][0x180], P0 ; /* 0x000060000d007a0c */ /* 0x000fe20000766470 */ /*0550*/ FFMA R27, R12, R14, R25 ; /* 0x0000000e0c1b7223 */ /* 0x008fc40000000019 */ /*0560*/ LDS R14, [R2.X4+0x10] ; /* 0x00001000020e7984 */ /* 0x000fe80000004800 */ /*0570*/ LDS R26, [R2.X4+0x18] ; /* 0x00001800021a7984 */ /* 0x000fe20000004800 */ /*0580*/ MOV R16, RZ ; /* 0x000000ff00107202 */ /* 0x002fc60000000f00 */ /*0590*/ LDS.64 R12, [R0.X8] ; /* 0x00000000000c7984 */ /* 0x000e640000008a00 */ /*05a0*/ @!P3 IMAD.MOV.U32 R18, RZ, RZ, 0x4 ; /* 0x00000004ff12b424 */ /* 0x000fc800078e00ff */ /*05b0*/ @!P3 IMAD.WIDE.U32 R18, R23, R18, c[0x0][0x168] ; /* 0x00005a001712b625 */ /* 0x000fca00078e0012 */ /*05c0*/ @!P3 LDG.E R16, [R18.64] ; /* 0x000000041210b981 */ /* 0x000722000c1e1900 */ /*05d0*/ IADD3 R17, R20, 0x6, RZ ; /* 0x0000000614117810 */ /* 0x000fc80007ffe0ff */ /*05e0*/ ISETP.GE.U32.OR P3, PT, R17, c[0x0][0x17c], P2 ; /* 0x00005f0011007a0c */ /* 0x000fe20001766470 */ /*05f0*/ HFMA2.MMA R25, -RZ, RZ, 0, 0 ; /* 0x00000000ff197435 */ /* 0x000fd800000001ff */ /*0600*/ @!P3 IMAD.MOV.U32 R17, RZ, RZ, 0x4 ; /* 0x00000004ff11b424 */ /* 0x000fe200078e00ff */ /*0610*/ MOV R18, RZ ; /* 0x000000ff00127202 */ /* 0x008fe20000000f00 */ /*0620*/ STS [R3], R29 ; /* 0x0000001d03007388 */ /* 0x0045e40000000800 */ /*0630*/ IADD3 R29, R4, 0x6, RZ ; /* 0x00000006041d7810 */ /* 0x004fc80007ffe0ff */ /*0640*/ ISETP.GE.U32.OR P0, PT, R29, c[0x0][0x180], P0 ; /* 0x000060001d007a0c */ /* 0x000fe20000706470 */ /*0650*/ STS [R3+0x10], R16 ; /* 0x0000101003007388 */ /* 0x0105d80000000800 */ /*0660*/ @!P0 IMAD.MOV.U32 R19, RZ, RZ, 0x4 ; /* 0x00000004ff138424 */ /* 0x000fe400078e00ff */ /*0670*/ @!P3 IMAD.WIDE.U32 R16, R21, R17, c[0x0][0x160] ; /* 0x000058001510b625 */ /* 0x004fe200078e0011 */ /*0680*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0690*/ @!P3 LDG.E R25, [R16.64] ; /* 0x000000041019b981 */ /* 0x0004e4000c1e1900 */ /*06a0*/ @!P0 IMAD.WIDE.U32 R16, R11, R19, c[0x0][0x168] ; /* 0x00005a000b108625 */ /* 0x004fca00078e0013 */ /*06b0*/ @!P0 LDG.E R18, [R16.64] ; /* 0x0000000410128981 */ /* 0x000ea2000c1e1900 */ /*06c0*/ FFMA R15, R28, R15, R27 ; /* 0x0000000f1c0f7223 */ /* 0x000fc6000000001b */ /*06d0*/ LDS R27, [R2.X4+0x10] ; /* 0x00001000021b7984 */ /* 0x000fe20000004800 */ /*06e0*/ FFMA R28, R14, R12, R15 ; /* 0x0000000c0e1c7223 */ /* 0x002fc6000000000f */ /*06f0*/ LDS R12, [R2.X4+0x18] ; /* 0x00001800020c7984 */ /* 0x000fe80000004800 */ /*0700*/ LDS.64 R14, [R0.X8] ; /* 0x00000000000e7984 */ /* 0x000e620000008a00 */ /*0710*/ FFMA R13, R26, R13, R28 ; /* 0x0000000d1a0d7223 */ /* 0x000fe2000000001c */ /*0720*/ IADD3 R9, R9, 0x4, RZ ; /* 0x0000000409097810 */ /* 0x000fe40007ffe0ff */ /*0730*/ IADD3 R4, R4, 0x8, RZ ; /* 0x0000000804047810 */ /* 0x000fe40007ffe0ff */ /*0740*/ IADD3 R20, R20, 0x8, RZ ; /* 0x0000000814147810 */ /* 0x000fc40007ffe0ff */ /*0750*/ IADD3 R21, R21, 0x8, RZ ; /* 0x0000000815157810 */ /* 0x000fe20007ffe0ff */ /*0760*/ FFMA R14, R27, R14, R13 ; /* 0x0000000e1b0e7223 */ /* 0x002fc8000000000d */ /*0770*/ FFMA R14, R12, R15, R14 ; /* 0x0000000f0c0e7223 */ /* 0x000fe4000000000e */ /*0780*/ IMAD.IADD R12, R10, 0x1, R9 ; /* 0x000000010a0c7824 */ /* 0x000fca00078e0209 */ /*0790*/ ISETP.NE.AND P0, PT, R12, RZ, PT ; /* 0x000000ff0c00720c */ /* 0x000fe20003f05270 */ /*07a0*/ HFMA2.MMA R13, -RZ, RZ, 0, 4.76837158203125e-07 ; /* 0x00000008ff0d7435 */ /* 0x000fd400000001ff */ /*07b0*/ IMAD R11, R13.reuse, c[0x0][0x184], R11 ; /* 0x000061000d0b7a24 */ /* 0x040fe400078e020b */ /*07c0*/ IMAD R23, R13.reuse, c[0x0][0x184], R23 ; /* 0x000061000d177a24 */ /* 0x040fe400078e0217 */ /*07d0*/ IMAD R24, R13.reuse, c[0x0][0x184], R24 ; /* 0x000061000d187a24 */ /* 0x040fe400078e0218 */ /*07e0*/ IMAD R22, R13, c[0x0][0x184], R22 ; /* 0x000061000d167a24 */ /* 0x000fe200078e0216 */ /*07f0*/ STS [R3], R25 ; /* 0x0000001903007388 */ /* 0x008fe80000000800 */ /*0800*/ STS [R3+0x10], R18 ; /* 0x0000101203007388 */ /* 0x004fe80000000800 */ /*0810*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0820*/ LDS R29, [R2.X4+0x10] ; /* 0x00001000021d7984 */ /* 0x000fe80000004800 */ /*0830*/ LDS.64 R18, [R0.X8] ; /* 0x0000000000127984 */ /* 0x000e680000008a00 */ /*0840*/ LDS R26, [R2.X4+0x18] ; /* 0x00001800021a7984 */ /* 0x000ea20000004800 */ /*0850*/ FFMA R18, R29, R18, R14 ; /* 0x000000121d127223 */ /* 0x002fc8000000000e */ /*0860*/ FFMA R25, R26, R19, R18 ; /* 0x000000131a197223 */ /* 0x004fe20000000012 */ /*0870*/ @P0 BRA 0x2b0 ; /* 0xfffffa3000000947 */ /* 0x000fea000383ffff */ /*0880*/ @!P1 BRA 0xab0 ; /* 0x0000022000009947 */ /* 0x000fea0003800000 */ /*0890*/ IMAD R4, R5, c[0x0][0x17c], R2 ; /* 0x00005f0005047a24 */ /* 0x000fe200078e0202 */ /*08a0*/ IADD3 R16, -R7, RZ, RZ ; /* 0x000000ff07107210 */ /* 0x000fe20007ffe1ff */ /*08b0*/ IMAD R15, R9.reuse, 0x2, R0 ; /* 0x00000002090f7824 */ /* 0x040fe200078e0200 */ /*08c0*/ LEA R14, R9.reuse, R2, 0x1 ; /* 0x00000002090e7211 */ /* 0x040fe200078e08ff */ /*08d0*/ IMAD R7, R9, 0x2, R4 ; /* 0x0000000209077824 */ /* 0x000fe400078e0204 */ /*08e0*/ IMAD R4, R15, c[0x0][0x184], R6 ; /* 0x000061000f047a24 */ /* 0x000fe400078e0206 */ /*08f0*/ ISETP.GE.U32.AND P1, PT, R15, c[0x0][0x180], PT ; /* 0x000060000f007a0c */ /* 0x000fe20003f26070 */ /*0900*/ HFMA2.MMA R18, -RZ, RZ, 0, 0 ; /* 0x00000000ff127435 */ /* 0x000fe200000001ff */ /*0910*/ ISETP.GE.U32.AND P0, PT, R14, c[0x0][0x17c], PT ; /* 0x00005f000e007a0c */ /* 0x000fe20003f06070 */ /*0920*/ IMAD.MOV.U32 R20, RZ, RZ, RZ ; /* 0x000000ffff147224 */ /* 0x000fe200078e00ff */ /*0930*/ ISETP.GE.OR P1, PT, R6, c[0x0][0x184], P1 ; /* 0x0000610006007a0c */ /* 0x000fc40000f26670 */ /*0940*/ ISETP.GE.OR P0, PT, R5, c[0x0][0x178], P0 ; /* 0x00005e0005007a0c */ /* 0x000fd60000706670 */ /*0950*/ @!P1 MOV R11, 0x4 ; /* 0x00000004000b9802 */ /* 0x000fe40000000f00 */ /*0960*/ @!P0 IMAD.MOV.U32 R8, RZ, RZ, 0x4 ; /* 0x00000004ff088424 */ /* 0x000fc600078e00ff */ /*0970*/ @!P1 IMAD.WIDE.U32 R10, R4, R11, c[0x0][0x168] ; /* 0x00005a00040a9625 */ /* 0x000fc800078e000b */ /*0980*/ @!P0 IMAD.WIDE.U32 R8, R7, R8, c[0x0][0x160] ; /* 0x0000580007088625 */ /* 0x000fe200078e0008 */ /*0990*/ @!P1 LDG.E R18, [R10.64] ; /* 0x000000040a129981 */ /* 0x000ea8000c1e1900 */ /*09a0*/ @!P0 LDG.E R20, [R8.64] ; /* 0x0000000408148981 */ /* 0x000ee2000c1e1900 */ /*09b0*/ IADD3 R16, R16, 0x1, RZ ; /* 0x0000000110107810 */ /* 0x000fc80007ffe0ff */ /*09c0*/ ISETP.NE.AND P0, PT, R16, RZ, PT ; /* 0x000000ff1000720c */ /* 0x000fe20003f05270 */ /*09d0*/ IMAD.MOV.U32 R21, RZ, RZ, 0x2 ; /* 0x00000002ff157424 */ /* 0x000fe200078e00ff */ /*09e0*/ IADD3 R15, R15, 0x2, RZ ; /* 0x000000020f0f7810 */ /* 0x000fe40007ffe0ff */ /*09f0*/ IADD3 R14, R14, 0x2, RZ ; /* 0x000000020e0e7810 */ /* 0x000fe20007ffe0ff */ /*0a00*/ IMAD R4, R21, c[0x0][0x184], R4 ; /* 0x0000610015047a24 */ /* 0x000fe200078e0204 */ /*0a10*/ IADD3 R7, R7, 0x2, RZ ; /* 0x0000000207077810 */ /* 0x000fe20007ffe0ff */ /*0a20*/ STS [R3+0x10], R18 ; /* 0x0000101203007388 */ /* 0x004fe80000000800 */ /*0a30*/ STS [R3], R20 ; /* 0x0000001403007388 */ /* 0x008fe80000000800 */ /*0a40*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0a50*/ LDS R17, [R2.X4+0x10] ; /* 0x0000100002117984 */ /* 0x000fe80000004800 */ /*0a60*/ LDS.64 R12, [R0.X8] ; /* 0x00000000000c7984 */ /* 0x000e680000008a00 */ /*0a70*/ LDS R19, [R2.X4+0x18] ; /* 0x0000180002137984 */ /* 0x000ea20000004800 */ /*0a80*/ FFMA R12, R17, R12, R25 ; /* 0x0000000c110c7223 */ /* 0x002fc80000000019 */ /*0a90*/ FFMA R25, R19, R13, R12 ; /* 0x0000000d13197223 */ /* 0x004fe2000000000c */ /*0aa0*/ @P0 BRA 0x8f0 ; /* 0xfffffe4000000947 */ /* 0x000fea000383ffff */ /*0ab0*/ ISETP.GE.AND P0, PT, R6, c[0x0][0x18c], PT ; /* 0x0000630006007a0c */ /* 0x000fc80003f06270 */ /*0ac0*/ ISETP.GE.OR P0, PT, R5, c[0x0][0x188], P0 ; /* 0x0000620005007a0c */ /* 0x000fda0000706670 */ /*0ad0*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0ae0*/ MOV R3, 0x4 ; /* 0x0000000400037802 */ /* 0x001fe20000000f00 */ /*0af0*/ IMAD R2, R5, c[0x0][0x18c], R6 ; /* 0x0000630005027a24 */ /* 0x000fc800078e0206 */ /*0b00*/ IMAD.WIDE R2, R2, R3, c[0x0][0x170] ; /* 0x00005c0002027625 */ /* 0x000fca00078e0203 */ /*0b10*/ STG.E [R2.64], R25 ; /* 0x0000001902007986 */ /* 0x000fe2000c101904 */ /*0b20*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0b30*/ BRA 0xb30; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0b40*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b50*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ba0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0be0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <stdio.h> #include <cuda.h> #include <stdlib.h> #define Tile_size 2 int numARows, numAColumns; int numBRows, numBColumns; int numCRows, numCColumns; __global__ void matrixMultiplyShared(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { __shared__ float sA[Tile_size][Tile_size]; // Tile size to store elements in shared memory __shared__ float sB[Tile_size][Tile_size]; int row = blockDim.y * blockIdx.y + threadIdx.y; int col = blockDim.x * blockIdx.x + threadIdx.x; float cvalue = 0.0; sA[threadIdx.y][threadIdx.x] = 0.0; sB[threadIdx.y][threadIdx.x] = 0.0; for (int k = 0; k < (((numAColumns - 1) / Tile_size) + 1); k++) { //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if ((row < numARows) && (threadIdx.x + (k * Tile_size)) < numAColumns) { sA[threadIdx.y][threadIdx.x] = A[(row * numAColumns) + threadIdx.x + (k * Tile_size)]; } else { sA[threadIdx.y][threadIdx.x] = 0.0; } //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if (col < numBColumns && (threadIdx.y + k * Tile_size) < numBRows) { sB[threadIdx.y][threadIdx.x] = B[(threadIdx.y + k * Tile_size) * numBColumns + col]; } else { sB[threadIdx.y][threadIdx.x] = 0.0; } __syncthreads(); for (int j = 0; j < Tile_size; ++j)//Multiplying Elements present in tile { cvalue += sA[threadIdx.y][j] * sB[j][threadIdx.x]; } } if (row < numCRows && col < numCColumns)//Saving Final result into Matrix C { C[row * numCColumns + col] = cvalue; } } void printMat(int row, int col, float *Mat) { for (int i = 0; i < row * col; i++) { printf("%f ", *(Mat + i)); if ((i % col) == 0 && i != 0) { printf("\n"); } } } void matMultiplyOnHost(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { for (int i = 0; i < numARows; i++) { for (int j = 0; j < numAColumns; j++) { C[i * numCColumns + j] = 0.0; for (int k = 0; k < numCColumns; k++) { C[i * numCColumns + j] += A[i * numAColumns + k] * B[k * numBColumns + j]; } } } } int main(int argc, char **argv) { float *hostA, *hostB, *hostC; float *hostComputedC; float *deviceA, *deviceB, *deviceC; printf("\nEnter Rows and Columns of A:"); scanf("%d %d", &numARows, &numAColumns); printf("\nEnter Rows and Columns of B:"); scanf("%d %d", &numBRows, &numBColumns); hostA = (float *) malloc(sizeof(float) * numARows * numAColumns); hostB = (float *) malloc(sizeof(float) * numBRows * numBColumns); int lower = 10.0, upper = 20.0; for (int i = 0; i < numARows * numAColumns; i++) hostA[i] = (rand() % (upper - lower + 1)) + lower; for (int i = 0; i < numBRows * numBColumns; i++) hostB[i] = (rand() % (upper - lower + 1)) + lower; printf("\nMatrix A Values:\n"); printMat(numARows, numAColumns, hostA); printf("\n\nMatrix B Values:\n"); printMat(numBRows, numBColumns, hostB); numCRows = numARows; numCColumns = numBColumns; hostC = (float *) malloc(sizeof(float) * numCRows * numCColumns); hostComputedC = (float *) malloc(sizeof(float) * numCRows * numCColumns); // Allocating GPU memory cudaMalloc((void **) &deviceA, sizeof(float) * numARows * numAColumns); cudaMalloc((void **) &deviceB, sizeof(float) * numBRows * numBColumns); cudaMalloc((void **) &deviceC, sizeof(float) * numCRows * numCColumns); // Copy memory to the GPU cudaMemcpy(deviceA, hostA, sizeof(float) * numARows * numAColumns, cudaMemcpyHostToDevice); cudaMemcpy(deviceB, hostB, sizeof(float) * numBRows * numBColumns, cudaMemcpyHostToDevice); // Initialize the grid and block dimensions dim3 dimGrid((numCColumns / Tile_size) + 1, (numCRows / Tile_size) + 1, 1);//Number of Blocks required dim3 dimBlock(Tile_size, Tile_size, 1);//Number of threads in each block matrixMultiplyShared<<<dimGrid, dimBlock>>>(deviceA, deviceB, deviceC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); cudaError_t err1 = cudaPeekAtLastError();//To capture last error in function call cudaDeviceSynchronize();//To synchronize the device // Copy the results in GPU memory back to the CPU cudaMemcpy(hostC, deviceC, sizeof(float) * numCRows * numCColumns, cudaMemcpyDeviceToHost); printf("\nMatrix C From Device\n"); printMat(numCRows, numCColumns, hostC);//Function Call matMultiplyOnHost(hostA, hostB, hostComputedC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); printf("\nMatrix C From Host\n"); printMat(numCRows, numCColumns, hostComputedC); printf("\n\n"); for (int i = 0; i < numCColumns * numCRows; i++)//Compare both the result matrices 1. MatrixMultiplyonHost 2. MatrixMultiplyonDevice { if (hostComputedC[i] != hostC[i]) { printf("Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n", i / numCColumns, i % numCColumns, hostComputedC[i], hostC[i]); break; } } printf("\nNumber of Blocks Created:%d \n", ((numCColumns / Tile_size) + 1) * ((numCColumns / Tile_size) + 1)); printf("\nNumber of Threads Per Block: %d \n", (Tile_size * Tile_size)); // Free the GPU memory cudaFree(deviceA); cudaFree(deviceB); cudaFree(deviceC); //Free the Pointer Memory free(hostA); free(hostB); free(hostC); free(hostComputedC); return 0; }
.file "tmpxft_000ceb69_00000000-6_mat_mul_shared_memory.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2062: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2062: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "%f " .LC1: .string "\n" .text .globl _Z8printMatiiPf .type _Z8printMatiiPf, @function _Z8printMatiiPf: .LFB2057: .cfi_startproc endbr64 imull %esi, %edi testl %edi, %edi jle .L9 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $8, %rsp .cfi_def_cfa_offset 64 movl %esi, %ebp movq %rdx, %r13 movslq %edi, %r12 movl $0, %ebx leaq .LC0(%rip), %r14 leaq .LC1(%rip), %r15 jmp .L6 .L5: addq $1, %rbx cmpq %r12, %rbx je .L12 .L6: pxor %xmm0, %xmm0 cvtss2sd 0(%r13,%rbx,4), %xmm0 movq %r14, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movl %ebx, %eax cltd idivl %ebp testl %edx, %edx jne .L5 testl %ebx, %ebx je .L5 movq %r15, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT jmp .L5 .L12: addq $8, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .L9: .cfi_restore 3 .cfi_restore 6 .cfi_restore 12 .cfi_restore 13 .cfi_restore 14 .cfi_restore 15 ret .cfi_endproc .LFE2057: .size _Z8printMatiiPf, .-_Z8printMatiiPf .globl _Z17matMultiplyOnHostPfS_S_iiiiii .type _Z17matMultiplyOnHostPfS_S_iiiiii, @function _Z17matMultiplyOnHostPfS_S_iiiiii: .LFB2058: .cfi_startproc endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 movq %rsi, -8(%rsp) movl %ecx, -12(%rsp) movl 72(%rsp), %ebx testl %ecx, %ecx jle .L13 movq %rdi, %rbp movq %rdx, %r15 movl %r8d, %r11d movslq 56(%rsp), %rdi salq $2, %rdi movl $0, %r13d movl $0, %ecx movl $0, %edx movslq %ebx, %r14 movq %r15, %r8 movq %r14, %rsi jmp .L15 .L19: movslq %ecx, %rax leaq (%r8,%rax,4), %r10 movq -8(%rsp), %r14 movslq %r13d, %rax leaq 0(%rbp,%rax,4), %r15 addq %rsi, %rax leaq 0(%rbp,%rax,4), %r9 movl $0, %r12d movl %edx, -20(%rsp) movl %ecx, -16(%rsp) .L18: movq %r10, %rcx movl $0x00000000, (%r10) testl %ebx, %ebx jle .L16 movq %r14, %rdx movq %r15, %rax .L17: movss (%rax), %xmm0 mulss (%rdx), %xmm0 addss (%rcx), %xmm0 movss %xmm0, (%rcx) addq $4, %rax addq %rdi, %rdx cmpq %r9, %rax jne .L17 .L16: addl $1, %r12d addq $4, %r10 addq $4, %r14 cmpl %r12d, %r11d jne .L18 movl -20(%rsp), %edx movl -16(%rsp), %ecx .L20: addl $1, %edx addl %ebx, %ecx addl %r11d, %r13d cmpl %edx, -12(%rsp) je .L13 .L15: testl %r11d, %r11d jg .L19 jmp .L20 .L13: popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2058: .size _Z17matMultiplyOnHostPfS_S_iiiiii, .-_Z17matMultiplyOnHostPfS_S_iiiiii .globl _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii .type _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii, @function _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii: .LFB2084: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 20(%rsp), %rax movq %rax, 136(%rsp) leaq 16(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) leaq 224(%rsp), %rax movq %rax, 176(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L28 .L24: movq 184(%rsp), %rax subq %fs:40, %rax jne .L29 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L28: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z20matrixMultiplySharedPfS_S_iiiiii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L24 .L29: call __stack_chk_fail@PLT .cfi_endproc .LFE2084: .size _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii, .-_Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii .globl _Z20matrixMultiplySharedPfS_S_iiiiii .type _Z20matrixMultiplySharedPfS_S_iiiiii, @function _Z20matrixMultiplySharedPfS_S_iiiiii: .LFB2085: .cfi_startproc endbr64 subq $16, %rsp .cfi_def_cfa_offset 24 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 40 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 48 call _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii addq $40, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _Z20matrixMultiplySharedPfS_S_iiiiii, .-_Z20matrixMultiplySharedPfS_S_iiiiii .section .rodata.str1.1 .LC3: .string "\nEnter Rows and Columns of A:" .LC4: .string "%d %d" .LC5: .string "\nEnter Rows and Columns of B:" .LC6: .string "\nMatrix A Values:\n" .LC7: .string "\n\nMatrix B Values:\n" .LC8: .string "\nMatrix C From Device\n" .LC9: .string "\nMatrix C From Host\n" .LC10: .string "\n\n" .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC11: .string "Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n" .align 8 .LC12: .string "\nNumber of Blocks Created:%d \n" .align 8 .LC13: .string "\nNumber of Threads Per Block: %d \n" .text .globl main .type main, @function main: .LFB2059: .cfi_startproc endbr64 pushq %r14 .cfi_def_cfa_offset 16 .cfi_offset 14, -16 pushq %r13 .cfi_def_cfa_offset 24 .cfi_offset 13, -24 pushq %r12 .cfi_def_cfa_offset 32 .cfi_offset 12, -32 pushq %rbp .cfi_def_cfa_offset 40 .cfi_offset 6, -40 pushq %rbx .cfi_def_cfa_offset 48 .cfi_offset 3, -48 subq $64, %rsp .cfi_def_cfa_offset 112 movq %fs:40, %rax movq %rax, 56(%rsp) xorl %eax, %eax leaq .LC3(%rip), %rsi movl $2, %edi call __printf_chk@PLT leaq numAColumns(%rip), %rdx leaq numARows(%rip), %rsi leaq .LC4(%rip), %rbx movq %rbx, %rdi movl $0, %eax call __isoc23_scanf@PLT leaq .LC5(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq numBColumns(%rip), %rdx leaq numBRows(%rip), %rsi movq %rbx, %rdi movl $0, %eax call __isoc23_scanf@PLT movl numARows(%rip), %ebx movl numAColumns(%rip), %r13d movslq %ebx, %rdi movslq %r13d, %rax imulq %rax, %rdi salq $2, %rdi call malloc@PLT movq %rax, %r12 movslq numBRows(%rip), %rdi movslq numBColumns(%rip), %rax imulq %rax, %rdi salq $2, %rdi call malloc@PLT movq %rax, %rbp imull %r13d, %ebx testl %ebx, %ebx jle .L33 movl $0, %ebx .L34: call rand@PLT movslq %eax, %rdx imulq $780903145, %rdx, %rdx sarq $33, %rdx movl %eax, %ecx sarl $31, %ecx subl %ecx, %edx leal (%rdx,%rdx,4), %ecx leal (%rdx,%rcx,2), %edx subl %edx, %eax addl $10, %eax pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, (%r12,%rbx,4) addq $1, %rbx movl numARows(%rip), %eax imull numAColumns(%rip), %eax cmpl %ebx, %eax jg .L34 .L33: movl numBRows(%rip), %eax imull numBColumns(%rip), %eax testl %eax, %eax jle .L35 movl $0, %ebx .L36: call rand@PLT movslq %eax, %rdx imulq $780903145, %rdx, %rdx sarq $33, %rdx movl %eax, %ecx sarl $31, %ecx subl %ecx, %edx leal (%rdx,%rdx,4), %ecx leal (%rdx,%rcx,2), %edx subl %edx, %eax addl $10, %eax pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, 0(%rbp,%rbx,4) addq $1, %rbx movl numBRows(%rip), %eax imull numBColumns(%rip), %eax cmpl %ebx, %eax jg .L36 .L35: leaq .LC6(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %r12, %rdx movl numAColumns(%rip), %esi movl numARows(%rip), %edi call _Z8printMatiiPf leaq .LC7(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %rbp, %rdx movl numBColumns(%rip), %esi movl numBRows(%rip), %edi call _Z8printMatiiPf movl numARows(%rip), %r14d movl %r14d, numCRows(%rip) movl numBColumns(%rip), %ebx movl %ebx, numCColumns(%rip) movslq %r14d, %r14 movslq %ebx, %rbx imulq %r14, %rbx salq $2, %rbx movq %rbx, %rdi call malloc@PLT movq %rax, %r13 movq %rbx, %rdi call malloc@PLT movq %rax, %rbx movslq numAColumns(%rip), %rsi imulq %r14, %rsi salq $2, %rsi leaq 8(%rsp), %rdi call cudaMalloc@PLT movslq numBRows(%rip), %rsi movslq numBColumns(%rip), %rax imulq %rax, %rsi salq $2, %rsi leaq 16(%rsp), %rdi call cudaMalloc@PLT movslq numCRows(%rip), %rsi movslq numCColumns(%rip), %rax imulq %rax, %rsi salq $2, %rsi leaq 24(%rsp), %rdi call cudaMalloc@PLT movslq numARows(%rip), %rdx movslq numAColumns(%rip), %rax imulq %rax, %rdx salq $2, %rdx movl $1, %ecx movq %r12, %rsi movq 8(%rsp), %rdi call cudaMemcpy@PLT movslq numBRows(%rip), %rdx movslq numBColumns(%rip), %rax imulq %rax, %rdx salq $2, %rdx movl $1, %ecx movq %rbp, %rsi movq 16(%rsp), %rdi call cudaMemcpy@PLT movl numCRows(%rip), %edx movl %edx, %eax shrl $31, %eax addl %edx, %eax sarl %eax addl $1, %eax movl numCColumns(%rip), %ecx movl %ecx, %edx shrl $31, %edx addl %ecx, %edx sarl %edx addl $1, %edx movl %edx, 32(%rsp) movl %eax, 36(%rsp) movl $2, 44(%rsp) movl $2, 48(%rsp) movl $0, %r9d movl $0, %r8d movq 44(%rsp), %rdx movl $1, %ecx movq 32(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L48 .L37: call cudaPeekAtLastError@PLT call cudaDeviceSynchronize@PLT movslq numCRows(%rip), %rdx movslq numCColumns(%rip), %rax imulq %rax, %rdx salq $2, %rdx movl $2, %ecx movq 24(%rsp), %rsi movq %r13, %rdi call cudaMemcpy@PLT leaq .LC8(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %r13, %rdx movl numCColumns(%rip), %esi movl numCRows(%rip), %edi call _Z8printMatiiPf subq $8, %rsp .cfi_def_cfa_offset 120 movl numCColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 128 movl numCRows(%rip), %eax pushq %rax .cfi_def_cfa_offset 136 movl numBColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 144 movl numBRows(%rip), %r9d movl numAColumns(%rip), %r8d movl numARows(%rip), %ecx movq %rbx, %rdx movq %rbp, %rsi movq %r12, %rdi call _Z17matMultiplyOnHostPfS_S_iiiiii addq $32, %rsp .cfi_def_cfa_offset 112 leaq .LC9(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %rbx, %rdx movl numCColumns(%rip), %esi movl numCRows(%rip), %edi call _Z8printMatiiPf leaq .LC10(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movl numCColumns(%rip), %esi movl %esi, %edx imull numCRows(%rip), %edx testl %edx, %edx jle .L38 movslq %edx, %rdx movl $0, %eax .L41: movss (%rbx,%rax,4), %xmm0 movss 0(%r13,%rax,4), %xmm1 ucomiss %xmm1, %xmm0 jp .L43 jne .L43 addq $1, %rax cmpq %rax, %rdx jne .L41 jmp .L38 .L48: subq $8, %rsp .cfi_def_cfa_offset 120 movl numCColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 128 movl numCRows(%rip), %eax pushq %rax .cfi_def_cfa_offset 136 movl numBColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 144 movl numBRows(%rip), %r9d movl numAColumns(%rip), %r8d movl numARows(%rip), %ecx movq 56(%rsp), %rdx movq 48(%rsp), %rsi movq 40(%rsp), %rdi call _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii addq $32, %rsp .cfi_def_cfa_offset 112 jmp .L37 .L43: cvtss2sd %xmm0, %xmm0 cltd idivl %esi movl %edx, %ecx cvtss2sd %xmm1, %xmm1 movl %eax, %edx leaq .LC11(%rip), %rsi movl $2, %edi movl $2, %eax call __printf_chk@PLT .L38: movl numCColumns(%rip), %eax movl %eax, %edx shrl $31, %edx addl %eax, %edx sarl %edx addl $1, %edx imull %edx, %edx leaq .LC12(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movl $4, %edx leaq .LC13(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 8(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call cudaFree@PLT movq 24(%rsp), %rdi call cudaFree@PLT movq %r12, %rdi call free@PLT movq %rbp, %rdi call free@PLT movq %r13, %rdi call free@PLT movq %rbx, %rdi call free@PLT movq 56(%rsp), %rax subq %fs:40, %rax jne .L49 movl $0, %eax addq $64, %rsp .cfi_remember_state .cfi_def_cfa_offset 48 popq %rbx .cfi_def_cfa_offset 40 popq %rbp .cfi_def_cfa_offset 32 popq %r12 .cfi_def_cfa_offset 24 popq %r13 .cfi_def_cfa_offset 16 popq %r14 .cfi_def_cfa_offset 8 ret .L49: .cfi_restore_state call __stack_chk_fail@PLT .cfi_endproc .LFE2059: .size main, .-main .section .rodata.str1.8 .align 8 .LC14: .string "_Z20matrixMultiplySharedPfS_S_iiiiii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2087: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC14(%rip), %rdx movq %rdx, %rcx leaq _Z20matrixMultiplySharedPfS_S_iiiiii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2087: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .globl numCColumns .bss .align 4 .type numCColumns, @object .size numCColumns, 4 numCColumns: .zero 4 .globl numCRows .align 4 .type numCRows, @object .size numCRows, 4 numCRows: .zero 4 .globl numBColumns .align 4 .type numBColumns, @object .size numBColumns, 4 numBColumns: .zero 4 .globl numBRows .align 4 .type numBRows, @object .size numBRows, 4 numBRows: .zero 4 .globl numAColumns .align 4 .type numAColumns, @object .size numAColumns, 4 numAColumns: .zero 4 .globl numARows .align 4 .type numARows, @object .size numARows, 4 numARows: .zero 4 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include <stdio.h> #include <cuda.h> #include <stdlib.h> #define Tile_size 2 int numARows, numAColumns; int numBRows, numBColumns; int numCRows, numCColumns; __global__ void matrixMultiplyShared(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { __shared__ float sA[Tile_size][Tile_size]; // Tile size to store elements in shared memory __shared__ float sB[Tile_size][Tile_size]; int row = blockDim.y * blockIdx.y + threadIdx.y; int col = blockDim.x * blockIdx.x + threadIdx.x; float cvalue = 0.0; sA[threadIdx.y][threadIdx.x] = 0.0; sB[threadIdx.y][threadIdx.x] = 0.0; for (int k = 0; k < (((numAColumns - 1) / Tile_size) + 1); k++) { //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if ((row < numARows) && (threadIdx.x + (k * Tile_size)) < numAColumns) { sA[threadIdx.y][threadIdx.x] = A[(row * numAColumns) + threadIdx.x + (k * Tile_size)]; } else { sA[threadIdx.y][threadIdx.x] = 0.0; } //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if (col < numBColumns && (threadIdx.y + k * Tile_size) < numBRows) { sB[threadIdx.y][threadIdx.x] = B[(threadIdx.y + k * Tile_size) * numBColumns + col]; } else { sB[threadIdx.y][threadIdx.x] = 0.0; } __syncthreads(); for (int j = 0; j < Tile_size; ++j)//Multiplying Elements present in tile { cvalue += sA[threadIdx.y][j] * sB[j][threadIdx.x]; } } if (row < numCRows && col < numCColumns)//Saving Final result into Matrix C { C[row * numCColumns + col] = cvalue; } } void printMat(int row, int col, float *Mat) { for (int i = 0; i < row * col; i++) { printf("%f ", *(Mat + i)); if ((i % col) == 0 && i != 0) { printf("\n"); } } } void matMultiplyOnHost(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { for (int i = 0; i < numARows; i++) { for (int j = 0; j < numAColumns; j++) { C[i * numCColumns + j] = 0.0; for (int k = 0; k < numCColumns; k++) { C[i * numCColumns + j] += A[i * numAColumns + k] * B[k * numBColumns + j]; } } } } int main(int argc, char **argv) { float *hostA, *hostB, *hostC; float *hostComputedC; float *deviceA, *deviceB, *deviceC; printf("\nEnter Rows and Columns of A:"); scanf("%d %d", &numARows, &numAColumns); printf("\nEnter Rows and Columns of B:"); scanf("%d %d", &numBRows, &numBColumns); hostA = (float *) malloc(sizeof(float) * numARows * numAColumns); hostB = (float *) malloc(sizeof(float) * numBRows * numBColumns); int lower = 10.0, upper = 20.0; for (int i = 0; i < numARows * numAColumns; i++) hostA[i] = (rand() % (upper - lower + 1)) + lower; for (int i = 0; i < numBRows * numBColumns; i++) hostB[i] = (rand() % (upper - lower + 1)) + lower; printf("\nMatrix A Values:\n"); printMat(numARows, numAColumns, hostA); printf("\n\nMatrix B Values:\n"); printMat(numBRows, numBColumns, hostB); numCRows = numARows; numCColumns = numBColumns; hostC = (float *) malloc(sizeof(float) * numCRows * numCColumns); hostComputedC = (float *) malloc(sizeof(float) * numCRows * numCColumns); // Allocating GPU memory cudaMalloc((void **) &deviceA, sizeof(float) * numARows * numAColumns); cudaMalloc((void **) &deviceB, sizeof(float) * numBRows * numBColumns); cudaMalloc((void **) &deviceC, sizeof(float) * numCRows * numCColumns); // Copy memory to the GPU cudaMemcpy(deviceA, hostA, sizeof(float) * numARows * numAColumns, cudaMemcpyHostToDevice); cudaMemcpy(deviceB, hostB, sizeof(float) * numBRows * numBColumns, cudaMemcpyHostToDevice); // Initialize the grid and block dimensions dim3 dimGrid((numCColumns / Tile_size) + 1, (numCRows / Tile_size) + 1, 1);//Number of Blocks required dim3 dimBlock(Tile_size, Tile_size, 1);//Number of threads in each block matrixMultiplyShared<<<dimGrid, dimBlock>>>(deviceA, deviceB, deviceC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); cudaError_t err1 = cudaPeekAtLastError();//To capture last error in function call cudaDeviceSynchronize();//To synchronize the device // Copy the results in GPU memory back to the CPU cudaMemcpy(hostC, deviceC, sizeof(float) * numCRows * numCColumns, cudaMemcpyDeviceToHost); printf("\nMatrix C From Device\n"); printMat(numCRows, numCColumns, hostC);//Function Call matMultiplyOnHost(hostA, hostB, hostComputedC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); printf("\nMatrix C From Host\n"); printMat(numCRows, numCColumns, hostComputedC); printf("\n\n"); for (int i = 0; i < numCColumns * numCRows; i++)//Compare both the result matrices 1. MatrixMultiplyonHost 2. MatrixMultiplyonDevice { if (hostComputedC[i] != hostC[i]) { printf("Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n", i / numCColumns, i % numCColumns, hostComputedC[i], hostC[i]); break; } } printf("\nNumber of Blocks Created:%d \n", ((numCColumns / Tile_size) + 1) * ((numCColumns / Tile_size) + 1)); printf("\nNumber of Threads Per Block: %d \n", (Tile_size * Tile_size)); // Free the GPU memory cudaFree(deviceA); cudaFree(deviceB); cudaFree(deviceC); //Free the Pointer Memory free(hostA); free(hostB); free(hostC); free(hostComputedC); return 0; }
#include <stdio.h> #include <hip/hip_runtime.h> #include <stdlib.h> #define Tile_size 2 int numARows, numAColumns; int numBRows, numBColumns; int numCRows, numCColumns; __global__ void matrixMultiplyShared(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { __shared__ float sA[Tile_size][Tile_size]; // Tile size to store elements in shared memory __shared__ float sB[Tile_size][Tile_size]; int row = blockDim.y * blockIdx.y + threadIdx.y; int col = blockDim.x * blockIdx.x + threadIdx.x; float cvalue = 0.0; sA[threadIdx.y][threadIdx.x] = 0.0; sB[threadIdx.y][threadIdx.x] = 0.0; for (int k = 0; k < (((numAColumns - 1) / Tile_size) + 1); k++) { //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if ((row < numARows) && (threadIdx.x + (k * Tile_size)) < numAColumns) { sA[threadIdx.y][threadIdx.x] = A[(row * numAColumns) + threadIdx.x + (k * Tile_size)]; } else { sA[threadIdx.y][threadIdx.x] = 0.0; } //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if (col < numBColumns && (threadIdx.y + k * Tile_size) < numBRows) { sB[threadIdx.y][threadIdx.x] = B[(threadIdx.y + k * Tile_size) * numBColumns + col]; } else { sB[threadIdx.y][threadIdx.x] = 0.0; } __syncthreads(); for (int j = 0; j < Tile_size; ++j)//Multiplying Elements present in tile { cvalue += sA[threadIdx.y][j] * sB[j][threadIdx.x]; } } if (row < numCRows && col < numCColumns)//Saving Final result into Matrix C { C[row * numCColumns + col] = cvalue; } } void printMat(int row, int col, float *Mat) { for (int i = 0; i < row * col; i++) { printf("%f ", *(Mat + i)); if ((i % col) == 0 && i != 0) { printf("\n"); } } } void matMultiplyOnHost(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { for (int i = 0; i < numARows; i++) { for (int j = 0; j < numAColumns; j++) { C[i * numCColumns + j] = 0.0; for (int k = 0; k < numCColumns; k++) { C[i * numCColumns + j] += A[i * numAColumns + k] * B[k * numBColumns + j]; } } } } int main(int argc, char **argv) { float *hostA, *hostB, *hostC; float *hostComputedC; float *deviceA, *deviceB, *deviceC; printf("\nEnter Rows and Columns of A:"); scanf("%d %d", &numARows, &numAColumns); printf("\nEnter Rows and Columns of B:"); scanf("%d %d", &numBRows, &numBColumns); hostA = (float *) malloc(sizeof(float) * numARows * numAColumns); hostB = (float *) malloc(sizeof(float) * numBRows * numBColumns); int lower = 10.0, upper = 20.0; for (int i = 0; i < numARows * numAColumns; i++) hostA[i] = (rand() % (upper - lower + 1)) + lower; for (int i = 0; i < numBRows * numBColumns; i++) hostB[i] = (rand() % (upper - lower + 1)) + lower; printf("\nMatrix A Values:\n"); printMat(numARows, numAColumns, hostA); printf("\n\nMatrix B Values:\n"); printMat(numBRows, numBColumns, hostB); numCRows = numARows; numCColumns = numBColumns; hostC = (float *) malloc(sizeof(float) * numCRows * numCColumns); hostComputedC = (float *) malloc(sizeof(float) * numCRows * numCColumns); // Allocating GPU memory hipMalloc((void **) &deviceA, sizeof(float) * numARows * numAColumns); hipMalloc((void **) &deviceB, sizeof(float) * numBRows * numBColumns); hipMalloc((void **) &deviceC, sizeof(float) * numCRows * numCColumns); // Copy memory to the GPU hipMemcpy(deviceA, hostA, sizeof(float) * numARows * numAColumns, hipMemcpyHostToDevice); hipMemcpy(deviceB, hostB, sizeof(float) * numBRows * numBColumns, hipMemcpyHostToDevice); // Initialize the grid and block dimensions dim3 dimGrid((numCColumns / Tile_size) + 1, (numCRows / Tile_size) + 1, 1);//Number of Blocks required dim3 dimBlock(Tile_size, Tile_size, 1);//Number of threads in each block matrixMultiplyShared<<<dimGrid, dimBlock>>>(deviceA, deviceB, deviceC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); hipError_t err1 = hipPeekAtLastError();//To capture last error in function call hipDeviceSynchronize();//To synchronize the device // Copy the results in GPU memory back to the CPU hipMemcpy(hostC, deviceC, sizeof(float) * numCRows * numCColumns, hipMemcpyDeviceToHost); printf("\nMatrix C From Device\n"); printMat(numCRows, numCColumns, hostC);//Function Call matMultiplyOnHost(hostA, hostB, hostComputedC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); printf("\nMatrix C From Host\n"); printMat(numCRows, numCColumns, hostComputedC); printf("\n\n"); for (int i = 0; i < numCColumns * numCRows; i++)//Compare both the result matrices 1. MatrixMultiplyonHost 2. MatrixMultiplyonDevice { if (hostComputedC[i] != hostC[i]) { printf("Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n", i / numCColumns, i % numCColumns, hostComputedC[i], hostC[i]); break; } } printf("\nNumber of Blocks Created:%d \n", ((numCColumns / Tile_size) + 1) * ((numCColumns / Tile_size) + 1)); printf("\nNumber of Threads Per Block: %d \n", (Tile_size * Tile_size)); // Free the GPU memory hipFree(deviceA); hipFree(deviceB); hipFree(deviceC); //Free the Pointer Memory free(hostA); free(hostB); free(hostC); free(hostComputedC); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <stdio.h> #include <hip/hip_runtime.h> #include <stdlib.h> #define Tile_size 2 int numARows, numAColumns; int numBRows, numBColumns; int numCRows, numCColumns; __global__ void matrixMultiplyShared(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { __shared__ float sA[Tile_size][Tile_size]; // Tile size to store elements in shared memory __shared__ float sB[Tile_size][Tile_size]; int row = blockDim.y * blockIdx.y + threadIdx.y; int col = blockDim.x * blockIdx.x + threadIdx.x; float cvalue = 0.0; sA[threadIdx.y][threadIdx.x] = 0.0; sB[threadIdx.y][threadIdx.x] = 0.0; for (int k = 0; k < (((numAColumns - 1) / Tile_size) + 1); k++) { //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if ((row < numARows) && (threadIdx.x + (k * Tile_size)) < numAColumns) { sA[threadIdx.y][threadIdx.x] = A[(row * numAColumns) + threadIdx.x + (k * Tile_size)]; } else { sA[threadIdx.y][threadIdx.x] = 0.0; } //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if (col < numBColumns && (threadIdx.y + k * Tile_size) < numBRows) { sB[threadIdx.y][threadIdx.x] = B[(threadIdx.y + k * Tile_size) * numBColumns + col]; } else { sB[threadIdx.y][threadIdx.x] = 0.0; } __syncthreads(); for (int j = 0; j < Tile_size; ++j)//Multiplying Elements present in tile { cvalue += sA[threadIdx.y][j] * sB[j][threadIdx.x]; } } if (row < numCRows && col < numCColumns)//Saving Final result into Matrix C { C[row * numCColumns + col] = cvalue; } } void printMat(int row, int col, float *Mat) { for (int i = 0; i < row * col; i++) { printf("%f ", *(Mat + i)); if ((i % col) == 0 && i != 0) { printf("\n"); } } } void matMultiplyOnHost(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { for (int i = 0; i < numARows; i++) { for (int j = 0; j < numAColumns; j++) { C[i * numCColumns + j] = 0.0; for (int k = 0; k < numCColumns; k++) { C[i * numCColumns + j] += A[i * numAColumns + k] * B[k * numBColumns + j]; } } } } int main(int argc, char **argv) { float *hostA, *hostB, *hostC; float *hostComputedC; float *deviceA, *deviceB, *deviceC; printf("\nEnter Rows and Columns of A:"); scanf("%d %d", &numARows, &numAColumns); printf("\nEnter Rows and Columns of B:"); scanf("%d %d", &numBRows, &numBColumns); hostA = (float *) malloc(sizeof(float) * numARows * numAColumns); hostB = (float *) malloc(sizeof(float) * numBRows * numBColumns); int lower = 10.0, upper = 20.0; for (int i = 0; i < numARows * numAColumns; i++) hostA[i] = (rand() % (upper - lower + 1)) + lower; for (int i = 0; i < numBRows * numBColumns; i++) hostB[i] = (rand() % (upper - lower + 1)) + lower; printf("\nMatrix A Values:\n"); printMat(numARows, numAColumns, hostA); printf("\n\nMatrix B Values:\n"); printMat(numBRows, numBColumns, hostB); numCRows = numARows; numCColumns = numBColumns; hostC = (float *) malloc(sizeof(float) * numCRows * numCColumns); hostComputedC = (float *) malloc(sizeof(float) * numCRows * numCColumns); // Allocating GPU memory hipMalloc((void **) &deviceA, sizeof(float) * numARows * numAColumns); hipMalloc((void **) &deviceB, sizeof(float) * numBRows * numBColumns); hipMalloc((void **) &deviceC, sizeof(float) * numCRows * numCColumns); // Copy memory to the GPU hipMemcpy(deviceA, hostA, sizeof(float) * numARows * numAColumns, hipMemcpyHostToDevice); hipMemcpy(deviceB, hostB, sizeof(float) * numBRows * numBColumns, hipMemcpyHostToDevice); // Initialize the grid and block dimensions dim3 dimGrid((numCColumns / Tile_size) + 1, (numCRows / Tile_size) + 1, 1);//Number of Blocks required dim3 dimBlock(Tile_size, Tile_size, 1);//Number of threads in each block matrixMultiplyShared<<<dimGrid, dimBlock>>>(deviceA, deviceB, deviceC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); hipError_t err1 = hipPeekAtLastError();//To capture last error in function call hipDeviceSynchronize();//To synchronize the device // Copy the results in GPU memory back to the CPU hipMemcpy(hostC, deviceC, sizeof(float) * numCRows * numCColumns, hipMemcpyDeviceToHost); printf("\nMatrix C From Device\n"); printMat(numCRows, numCColumns, hostC);//Function Call matMultiplyOnHost(hostA, hostB, hostComputedC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); printf("\nMatrix C From Host\n"); printMat(numCRows, numCColumns, hostComputedC); printf("\n\n"); for (int i = 0; i < numCColumns * numCRows; i++)//Compare both the result matrices 1. MatrixMultiplyonHost 2. MatrixMultiplyonDevice { if (hostComputedC[i] != hostC[i]) { printf("Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n", i / numCColumns, i % numCColumns, hostComputedC[i], hostC[i]); break; } } printf("\nNumber of Blocks Created:%d \n", ((numCColumns / Tile_size) + 1) * ((numCColumns / Tile_size) + 1)); printf("\nNumber of Threads Per Block: %d \n", (Tile_size * Tile_size)); // Free the GPU memory hipFree(deviceA); hipFree(deviceB); hipFree(deviceC); //Free the Pointer Memory free(hostA); free(hostB); free(hostC); free(hostComputedC); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z20matrixMultiplySharedPfS_S_iiiiii .globl _Z20matrixMultiplySharedPfS_S_iiiiii .p2align 8 .type _Z20matrixMultiplySharedPfS_S_iiiiii,@function _Z20matrixMultiplySharedPfS_S_iiiiii: s_clause 0x1 s_load_b32 s2, s[0:1], 0x3c s_load_b32 s10, s[0:1], 0x1c v_dual_mov_b32 v8, 0 :: v_dual_and_b32 v3, 0x3ff, v0 v_bfe_u32 v4, v0, 10, 10 s_mov_b32 s11, 0 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_3) | instid1(VALU_DEP_1) v_lshlrev_b32_e32 v5, 2, v3 s_waitcnt lgkmcnt(0) s_lshr_b32 s3, s2, 16 s_and_b32 s2, s2, 0xffff v_mad_u64_u32 v[0:1], null, s15, s3, v[4:5] v_mad_u64_u32 v[1:2], null, s14, s2, v[3:4] v_lshl_add_u32 v2, v4, 3, v5 v_mov_b32_e32 v5, 0 s_cmp_lt_i32 s10, 0 ds_store_2addr_b32 v2, v5, v5 offset1:4 s_cbranch_scc1 .LBB0_14 s_clause 0x2 s_load_b32 s2, s[0:1], 0x18 s_load_b64 s[8:9], s[0:1], 0x20 s_load_b128 s[4:7], s[0:1], 0x0 s_add_i32 s3, s10, -1 v_mad_u64_u32 v[5:6], null, v0, s10, v[3:4] s_lshr_b32 s12, s3, 31 v_dual_mov_b32 v8, 0 :: v_dual_add_nc_u32 v9, 16, v2 s_add_i32 s3, s3, s12 v_dual_mov_b32 v7, 0 :: v_dual_lshlrev_b32 v10, 3, v4 v_lshl_add_u32 v11, v3, 2, 16 s_ashr_i32 s3, s3, 1 s_delay_alu instid0(SALU_CYCLE_1) s_max_i32 s12, s3, 0 s_waitcnt lgkmcnt(0) v_cmp_gt_i32_e32 vcc_lo, s2, v0 v_cmp_gt_i32_e64 s2, s9, v1 .LBB0_2: v_mov_b32_e32 v6, 0 s_and_saveexec_b32 s13, vcc_lo s_cbranch_execz .LBB0_6 s_lshl_b32 s15, s11, 1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_add_nc_u32_e32 v6, s15, v3 v_cmp_gt_u32_e64 s3, s10, v6 v_mov_b32_e32 v6, 0 s_delay_alu instid0(VALU_DEP_2) s_and_saveexec_b32 s14, s3 s_cbranch_execz .LBB0_5 v_add_nc_u32_e32 v6, s15, v5 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[12:13], 2, v[6:7] v_add_co_u32 v12, s3, s4, v12 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v13, s3, s5, v13, s3 global_load_b32 v6, v[12:13], off .LBB0_5: s_or_b32 exec_lo, exec_lo, s14 .LBB0_6: s_delay_alu instid0(SALU_CYCLE_1) s_or_b32 exec_lo, exec_lo, s13 v_mov_b32_e32 v12, 0 s_waitcnt vmcnt(0) ds_store_b32 v2, v6 s_and_saveexec_b32 s13, s2 s_cbranch_execz .LBB0_10 v_lshl_add_u32 v6, s11, 1, v4 v_mov_b32_e32 v12, 0 s_mov_b32 s14, exec_lo s_delay_alu instid0(VALU_DEP_2) v_cmpx_gt_u32_e64 s8, v6 s_cbranch_execz .LBB0_9 v_mad_u64_u32 v[12:13], null, v6, s9, v[1:2] v_mov_b32_e32 v13, v7 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[12:13], 2, v[12:13] v_add_co_u32 v12, s3, s6, v12 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v13, s3, s7, v13, s3 global_load_b32 v12, v[12:13], off .LBB0_9: s_or_b32 exec_lo, exec_lo, s14 .LBB0_10: s_delay_alu instid0(SALU_CYCLE_1) s_or_b32 exec_lo, exec_lo, s13 v_mov_b32_e32 v6, v11 s_mov_b32 s3, 0 s_waitcnt vmcnt(0) ds_store_b32 v9, v12 s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv .LBB0_11: v_add_nc_u32_e32 v12, s3, v10 s_add_i32 s3, s3, 4 ds_load_b32 v13, v6 ds_load_b32 v12, v12 v_add_nc_u32_e32 v6, 8, v6 s_cmp_lg_u32 s3, 4 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v8, v12, v13 s_cbranch_scc0 .LBB0_11 s_add_i32 s3, s11, 1 s_cmp_eq_u32 s11, s12 s_cbranch_scc1 .LBB0_14 s_mov_b32 s11, s3 s_branch .LBB0_2 .LBB0_14: s_load_b64 s[2:3], s[0:1], 0x28 s_waitcnt lgkmcnt(0) v_cmp_gt_i32_e32 vcc_lo, s2, v0 v_cmp_gt_i32_e64 s2, s3, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_and_b32 s2, vcc_lo, s2 s_and_saveexec_b32 s4, s2 s_cbranch_execz .LBB0_16 s_load_b64 s[0:1], s[0:1], 0x10 v_mad_u64_u32 v[2:3], null, v0, s3, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[0:1], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v0, vcc_lo, s0, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo global_store_b32 v[0:1], v8, off .LBB0_16: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z20matrixMultiplySharedPfS_S_iiiiii .amdhsa_group_segment_fixed_size 32 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 304 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z20matrixMultiplySharedPfS_S_iiiiii, .Lfunc_end0-_Z20matrixMultiplySharedPfS_S_iiiiii .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value - .offset: 28 .size: 4 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: by_value - .offset: 36 .size: 4 .value_kind: by_value - .offset: 40 .size: 4 .value_kind: by_value - .offset: 44 .size: 4 .value_kind: by_value - .offset: 48 .size: 4 .value_kind: hidden_block_count_x - .offset: 52 .size: 4 .value_kind: hidden_block_count_y - .offset: 56 .size: 4 .value_kind: hidden_block_count_z - .offset: 60 .size: 2 .value_kind: hidden_group_size_x - .offset: 62 .size: 2 .value_kind: hidden_group_size_y - .offset: 64 .size: 2 .value_kind: hidden_group_size_z - .offset: 66 .size: 2 .value_kind: hidden_remainder_x - .offset: 68 .size: 2 .value_kind: hidden_remainder_y - .offset: 70 .size: 2 .value_kind: hidden_remainder_z - .offset: 88 .size: 8 .value_kind: hidden_global_offset_x - .offset: 96 .size: 8 .value_kind: hidden_global_offset_y - .offset: 104 .size: 8 .value_kind: hidden_global_offset_z - .offset: 112 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 32 .kernarg_segment_align: 8 .kernarg_segment_size: 304 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z20matrixMultiplySharedPfS_S_iiiiii .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z20matrixMultiplySharedPfS_S_iiiiii.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <stdio.h> #include <hip/hip_runtime.h> #include <stdlib.h> #define Tile_size 2 int numARows, numAColumns; int numBRows, numBColumns; int numCRows, numCColumns; __global__ void matrixMultiplyShared(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { __shared__ float sA[Tile_size][Tile_size]; // Tile size to store elements in shared memory __shared__ float sB[Tile_size][Tile_size]; int row = blockDim.y * blockIdx.y + threadIdx.y; int col = blockDim.x * blockIdx.x + threadIdx.x; float cvalue = 0.0; sA[threadIdx.y][threadIdx.x] = 0.0; sB[threadIdx.y][threadIdx.x] = 0.0; for (int k = 0; k < (((numAColumns - 1) / Tile_size) + 1); k++) { //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if ((row < numARows) && (threadIdx.x + (k * Tile_size)) < numAColumns) { sA[threadIdx.y][threadIdx.x] = A[(row * numAColumns) + threadIdx.x + (k * Tile_size)]; } else { sA[threadIdx.y][threadIdx.x] = 0.0; } //Copy Data to Tile from Matrix (Global Memory to Shared Memory) if (col < numBColumns && (threadIdx.y + k * Tile_size) < numBRows) { sB[threadIdx.y][threadIdx.x] = B[(threadIdx.y + k * Tile_size) * numBColumns + col]; } else { sB[threadIdx.y][threadIdx.x] = 0.0; } __syncthreads(); for (int j = 0; j < Tile_size; ++j)//Multiplying Elements present in tile { cvalue += sA[threadIdx.y][j] * sB[j][threadIdx.x]; } } if (row < numCRows && col < numCColumns)//Saving Final result into Matrix C { C[row * numCColumns + col] = cvalue; } } void printMat(int row, int col, float *Mat) { for (int i = 0; i < row * col; i++) { printf("%f ", *(Mat + i)); if ((i % col) == 0 && i != 0) { printf("\n"); } } } void matMultiplyOnHost(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) { for (int i = 0; i < numARows; i++) { for (int j = 0; j < numAColumns; j++) { C[i * numCColumns + j] = 0.0; for (int k = 0; k < numCColumns; k++) { C[i * numCColumns + j] += A[i * numAColumns + k] * B[k * numBColumns + j]; } } } } int main(int argc, char **argv) { float *hostA, *hostB, *hostC; float *hostComputedC; float *deviceA, *deviceB, *deviceC; printf("\nEnter Rows and Columns of A:"); scanf("%d %d", &numARows, &numAColumns); printf("\nEnter Rows and Columns of B:"); scanf("%d %d", &numBRows, &numBColumns); hostA = (float *) malloc(sizeof(float) * numARows * numAColumns); hostB = (float *) malloc(sizeof(float) * numBRows * numBColumns); int lower = 10.0, upper = 20.0; for (int i = 0; i < numARows * numAColumns; i++) hostA[i] = (rand() % (upper - lower + 1)) + lower; for (int i = 0; i < numBRows * numBColumns; i++) hostB[i] = (rand() % (upper - lower + 1)) + lower; printf("\nMatrix A Values:\n"); printMat(numARows, numAColumns, hostA); printf("\n\nMatrix B Values:\n"); printMat(numBRows, numBColumns, hostB); numCRows = numARows; numCColumns = numBColumns; hostC = (float *) malloc(sizeof(float) * numCRows * numCColumns); hostComputedC = (float *) malloc(sizeof(float) * numCRows * numCColumns); // Allocating GPU memory hipMalloc((void **) &deviceA, sizeof(float) * numARows * numAColumns); hipMalloc((void **) &deviceB, sizeof(float) * numBRows * numBColumns); hipMalloc((void **) &deviceC, sizeof(float) * numCRows * numCColumns); // Copy memory to the GPU hipMemcpy(deviceA, hostA, sizeof(float) * numARows * numAColumns, hipMemcpyHostToDevice); hipMemcpy(deviceB, hostB, sizeof(float) * numBRows * numBColumns, hipMemcpyHostToDevice); // Initialize the grid and block dimensions dim3 dimGrid((numCColumns / Tile_size) + 1, (numCRows / Tile_size) + 1, 1);//Number of Blocks required dim3 dimBlock(Tile_size, Tile_size, 1);//Number of threads in each block matrixMultiplyShared<<<dimGrid, dimBlock>>>(deviceA, deviceB, deviceC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); hipError_t err1 = hipPeekAtLastError();//To capture last error in function call hipDeviceSynchronize();//To synchronize the device // Copy the results in GPU memory back to the CPU hipMemcpy(hostC, deviceC, sizeof(float) * numCRows * numCColumns, hipMemcpyDeviceToHost); printf("\nMatrix C From Device\n"); printMat(numCRows, numCColumns, hostC);//Function Call matMultiplyOnHost(hostA, hostB, hostComputedC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns); printf("\nMatrix C From Host\n"); printMat(numCRows, numCColumns, hostComputedC); printf("\n\n"); for (int i = 0; i < numCColumns * numCRows; i++)//Compare both the result matrices 1. MatrixMultiplyonHost 2. MatrixMultiplyonDevice { if (hostComputedC[i] != hostC[i]) { printf("Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n", i / numCColumns, i % numCColumns, hostComputedC[i], hostC[i]); break; } } printf("\nNumber of Blocks Created:%d \n", ((numCColumns / Tile_size) + 1) * ((numCColumns / Tile_size) + 1)); printf("\nNumber of Threads Per Block: %d \n", (Tile_size * Tile_size)); // Free the GPU memory hipFree(deviceA); hipFree(deviceB); hipFree(deviceC); //Free the Pointer Memory free(hostA); free(hostB); free(hostC); free(hostComputedC); return 0; }
.text .file "mat_mul_shared_memory.hip" .globl _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii # -- Begin function _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .p2align 4, 0x90 .type _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii,@function _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii: # @_Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 20(%rsp), %rax movq %rax, 120(%rsp) leaq 16(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 192(%rsp), %rax movq %rax, 160(%rsp) leaq 56(%rsp), %rdi leaq 40(%rsp), %rsi leaq 32(%rsp), %rdx leaq 24(%rsp), %rcx callq __hipPopCallConfiguration movq 56(%rsp), %rsi movl 64(%rsp), %edx movq 40(%rsp), %rcx movl 48(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z20matrixMultiplySharedPfS_S_iiiiii, %edi pushq 24(%rsp) .cfi_adjust_cfa_offset 8 pushq 40(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end0: .size _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii, .Lfunc_end0-_Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .cfi_endproc # -- End function .globl _Z8printMatiiPf # -- Begin function _Z8printMatiiPf .p2align 4, 0x90 .type _Z8printMatiiPf,@function _Z8printMatiiPf: # @_Z8printMatiiPf .cfi_startproc # %bb.0: imull %esi, %edi testl %edi, %edi jle .LBB1_7 # %bb.1: # %.lr.ph.preheader pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %rbx .cfi_def_cfa_offset 40 pushq %rax .cfi_def_cfa_offset 48 .cfi_offset %rbx, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movq %rdx, %rbx movl %esi, %ebp movl %edi, %r15d xorl %r14d, %r14d jmp .LBB1_2 .p2align 4, 0x90 .LBB1_5: # in Loop: Header=BB1_2 Depth=1 incq %r14 cmpq %r14, %r15 je .LBB1_6 .LBB1_2: # %.lr.ph # =>This Inner Loop Header: Depth=1 movss (%rbx,%r14,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r14d, %eax cltd idivl %ebp testq %r14, %r14 je .LBB1_5 # %bb.3: # %.lr.ph # in Loop: Header=BB1_2 Depth=1 testl %edx, %edx jne .LBB1_5 # %bb.4: # in Loop: Header=BB1_2 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB1_5 .LBB1_6: addq $8, %rsp .cfi_def_cfa_offset 40 popq %rbx .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 .cfi_restore %rbx .cfi_restore %r14 .cfi_restore %r15 .cfi_restore %rbp .LBB1_7: # %._crit_edge retq .Lfunc_end1: .size _Z8printMatiiPf, .Lfunc_end1-_Z8printMatiiPf .cfi_endproc # -- End function .globl _Z17matMultiplyOnHostPfS_S_iiiiii # -- Begin function _Z17matMultiplyOnHostPfS_S_iiiiii .p2align 4, 0x90 .type _Z17matMultiplyOnHostPfS_S_iiiiii,@function _Z17matMultiplyOnHostPfS_S_iiiiii: # @_Z17matMultiplyOnHostPfS_S_iiiiii .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movq %rdx, -8(%rsp) # 8-byte Spill movq %rsi, -16(%rsp) # 8-byte Spill testl %ecx, %ecx jle .LBB2_9 # %bb.1: # %.preheader.lr.ph movl 72(%rsp), %eax movslq 56(%rsp), %r9 movslq %eax, %r10 movslq %r8d, %r11 movl %ecx, %ecx movl %r11d, %ebx movl %r10d, %r14d shlq $2, %r11 shlq $2, %r9 xorl %r15d, %r15d jmp .LBB2_2 .p2align 4, 0x90 .LBB2_8: # %._crit_edge30 # in Loop: Header=BB2_2 Depth=1 incq %r15 addq %r11, %rdi cmpq %rcx, %r15 je .LBB2_9 .LBB2_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB2_4 Depth 2 # Child Loop BB2_6 Depth 3 testl %r8d, %r8d jle .LBB2_8 # %bb.3: # %.lr.ph29 # in Loop: Header=BB2_2 Depth=1 movq %r15, %rdx imulq %r10, %rdx movq -8(%rsp), %rsi # 8-byte Reload leaq (%rsi,%rdx,4), %r12 movq -16(%rsp), %rsi # 8-byte Reload xorl %ebp, %ebp jmp .LBB2_4 .p2align 4, 0x90 .LBB2_7: # %._crit_edge # in Loop: Header=BB2_4 Depth=2 incq %rbp addq $4, %rsi cmpq %rbx, %rbp je .LBB2_8 .LBB2_4: # Parent Loop BB2_2 Depth=1 # => This Loop Header: Depth=2 # Child Loop BB2_6 Depth 3 movl $0, (%r12,%rbp,4) testl %eax, %eax jle .LBB2_7 # %bb.5: # %.lr.ph # in Loop: Header=BB2_4 Depth=2 movss (%r12,%rbp,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movq %rsi, %r13 xorl %edx, %edx .p2align 4, 0x90 .LBB2_6: # Parent Loop BB2_2 Depth=1 # Parent Loop BB2_4 Depth=2 # => This Inner Loop Header: Depth=3 movss (%rdi,%rdx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero mulss (%r13), %xmm1 addss %xmm1, %xmm0 movss %xmm0, (%r12,%rbp,4) incq %rdx addq %r9, %r13 cmpq %rdx, %r14 jne .LBB2_6 jmp .LBB2_7 .LBB2_9: # %._crit_edge32 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end2: .size _Z17matMultiplyOnHostPfS_S_iiiiii, .Lfunc_end2-_Z17matMultiplyOnHostPfS_S_iiiiii .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 subq $216, %rsp .cfi_def_cfa_offset 272 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movl $.L.str.2, %edi xorl %eax, %eax callq printf movl $.L.str.3, %edi movl $numARows, %esi movl $numAColumns, %edx xorl %eax, %eax callq __isoc23_scanf movl $.L.str.4, %edi xorl %eax, %eax callq printf movl $.L.str.3, %edi movl $numBRows, %esi movl $numBColumns, %edx xorl %eax, %eax callq __isoc23_scanf movslq numARows(%rip), %rax movslq numAColumns(%rip), %rdi imulq %rax, %rdi shlq $2, %rdi callq malloc movq %rax, %r13 movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rdi imulq %rax, %rdi shlq $2, %rdi callq malloc movq %rax, %r14 movl numAColumns(%rip), %eax imull numARows(%rip), %eax testl %eax, %eax jle .LBB3_3 # %bb.1: # %.lr.ph.preheader xorl %ebx, %ebx .p2align 4, 0x90 .LBB3_2: # %.lr.ph # =>This Inner Loop Header: Depth=1 callq rand cltq imulq $780903145, %rax, %rcx # imm = 0x2E8BA2E9 movq %rcx, %rdx shrq $63, %rdx sarq $33, %rcx addl %edx, %ecx leal (%rcx,%rcx,4), %edx leal (%rcx,%rdx,2), %ecx negl %ecx addl %ecx, %eax addl $10, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r13,%rbx,4) incq %rbx movslq numARows(%rip), %rax movslq numAColumns(%rip), %rcx imulq %rax, %rcx cmpq %rcx, %rbx jl .LBB3_2 .LBB3_3: # %.preheader movl numBColumns(%rip), %eax imull numBRows(%rip), %eax testl %eax, %eax jle .LBB3_6 # %bb.4: # %.lr.ph95.preheader xorl %ebx, %ebx .p2align 4, 0x90 .LBB3_5: # %.lr.ph95 # =>This Inner Loop Header: Depth=1 callq rand cltq imulq $780903145, %rax, %rcx # imm = 0x2E8BA2E9 movq %rcx, %rdx shrq $63, %rdx sarq $33, %rcx addl %edx, %ecx leal (%rcx,%rcx,4), %edx leal (%rcx,%rdx,2), %ecx negl %ecx addl %ecx, %eax addl $10, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r14,%rbx,4) incq %rbx movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rcx imulq %rax, %rcx cmpq %rcx, %rbx jl .LBB3_5 .LBB3_6: # %._crit_edge movl $.Lstr, %edi callq puts@PLT movl numAColumns(%rip), %ebp movl numARows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_12 # %bb.7: # %.lr.ph.preheader.i movl %eax, %r12d xorl %r15d, %r15d jmp .LBB3_8 .p2align 4, 0x90 .LBB3_11: # in Loop: Header=BB3_8 Depth=1 incq %r15 cmpq %r15, %r12 je .LBB3_12 .LBB3_8: # %.lr.ph.i # =>This Inner Loop Header: Depth=1 movss (%r13,%r15,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r15d, %eax cltd idivl %ebp testq %r15, %r15 je .LBB3_11 # %bb.9: # %.lr.ph.i # in Loop: Header=BB3_8 Depth=1 testl %edx, %edx jne .LBB3_11 # %bb.10: # in Loop: Header=BB3_8 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_11 .LBB3_12: # %_Z8printMatiiPf.exit movl $.Lstr.1, %edi callq puts@PLT movl numBColumns(%rip), %ebp movl numBRows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_18 # %bb.13: # %.lr.ph.preheader.i54 movl %eax, %r12d xorl %r15d, %r15d jmp .LBB3_14 .p2align 4, 0x90 .LBB3_17: # in Loop: Header=BB3_14 Depth=1 incq %r15 cmpq %r15, %r12 je .LBB3_18 .LBB3_14: # %.lr.ph.i56 # =>This Inner Loop Header: Depth=1 movss (%r14,%r15,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r15d, %eax cltd idivl %ebp testq %r15, %r15 je .LBB3_17 # %bb.15: # %.lr.ph.i56 # in Loop: Header=BB3_14 Depth=1 testl %edx, %edx jne .LBB3_17 # %bb.16: # in Loop: Header=BB3_14 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_17 .LBB3_18: # %_Z8printMatiiPf.exit62 movslq numARows(%rip), %rbx movl %ebx, numCRows(%rip) movslq numBColumns(%rip), %r12 movl %r12d, numCColumns(%rip) shlq $2, %rbx imulq %rbx, %r12 movq %r12, %rdi callq malloc movq %rax, %r15 movq %r12, %rdi callq malloc movq %rax, %r12 movslq numAColumns(%rip), %rsi imulq %rbx, %rsi leaq 24(%rsp), %rdi callq hipMalloc movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rsi imulq %rax, %rsi shlq $2, %rsi leaq 16(%rsp), %rdi callq hipMalloc movslq numCRows(%rip), %rax movslq numCColumns(%rip), %rsi imulq %rax, %rsi shlq $2, %rsi leaq 8(%rsp), %rdi callq hipMalloc movq 24(%rsp), %rdi movslq numARows(%rip), %rax movslq numAColumns(%rip), %rdx imulq %rax, %rdx shlq $2, %rdx movq %r13, 56(%rsp) # 8-byte Spill movq %r13, %rsi movl $1, %ecx callq hipMemcpy movq 16(%rsp), %rdi movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rdx imulq %rax, %rdx shlq $2, %rdx movq %r14, 64(%rsp) # 8-byte Spill movq %r14, %rsi movl $1, %ecx callq hipMemcpy movl numCColumns(%rip), %eax movl %eax, %ecx shrl $31, %ecx addl %eax, %ecx sarl %ecx incl %ecx movl numCRows(%rip), %eax movl %eax, %edi shrl $31, %edi addl %eax, %edi sarl %edi incl %edi shlq $32, %rdi orq %rcx, %rdi movabsq $8589934594, %rdx # imm = 0x200000002 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB3_20 # %bb.19: movq 24(%rsp), %rax movq 16(%rsp), %rcx movq 8(%rsp), %rdx movl numARows(%rip), %esi movl numAColumns(%rip), %edi movl numBRows(%rip), %r8d movl numBColumns(%rip), %r9d movl numCRows(%rip), %r10d movl numCColumns(%rip), %r11d movq %rax, 136(%rsp) movq %rcx, 128(%rsp) movq %rdx, 120(%rsp) movl %esi, 52(%rsp) movl %edi, 48(%rsp) movl %r8d, 44(%rsp) movl %r9d, 40(%rsp) movl %r10d, 36(%rsp) movl %r11d, 32(%rsp) leaq 136(%rsp), %rax movq %rax, 144(%rsp) leaq 128(%rsp), %rax movq %rax, 152(%rsp) leaq 120(%rsp), %rax movq %rax, 160(%rsp) leaq 52(%rsp), %rax movq %rax, 168(%rsp) leaq 48(%rsp), %rax movq %rax, 176(%rsp) leaq 44(%rsp), %rax movq %rax, 184(%rsp) leaq 40(%rsp), %rax movq %rax, 192(%rsp) leaq 36(%rsp), %rax movq %rax, 200(%rsp) leaq 32(%rsp), %rax movq %rax, 208(%rsp) leaq 104(%rsp), %rdi leaq 88(%rsp), %rsi leaq 80(%rsp), %rdx leaq 72(%rsp), %rcx callq __hipPopCallConfiguration movq 104(%rsp), %rsi movl 112(%rsp), %edx movq 88(%rsp), %rcx movl 96(%rsp), %r8d leaq 144(%rsp), %r9 movl $_Z20matrixMultiplySharedPfS_S_iiiiii, %edi pushq 72(%rsp) .cfi_adjust_cfa_offset 8 pushq 88(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB3_20: callq hipPeekAtLastError callq hipDeviceSynchronize movq 8(%rsp), %rsi movslq numCRows(%rip), %rax movslq numCColumns(%rip), %rdx imulq %rax, %rdx shlq $2, %rdx movq %r15, %r14 movq %r15, %rdi movl $2, %ecx callq hipMemcpy movl $.Lstr.2, %edi callq puts@PLT movl numCColumns(%rip), %ebp movl numCRows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_26 # %bb.21: # %.lr.ph.preheader.i63 movl %eax, %ebx xorl %r13d, %r13d jmp .LBB3_22 .p2align 4, 0x90 .LBB3_25: # in Loop: Header=BB3_22 Depth=1 incq %r13 cmpq %r13, %rbx je .LBB3_26 .LBB3_22: # %.lr.ph.i65 # =>This Inner Loop Header: Depth=1 movss (%r14,%r13,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r13d, %eax cltd idivl %ebp testq %r13, %r13 je .LBB3_25 # %bb.23: # %.lr.ph.i65 # in Loop: Header=BB3_22 Depth=1 testl %edx, %edx jne .LBB3_25 # %bb.24: # in Loop: Header=BB3_22 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_25 .LBB3_26: # %_Z8printMatiiPf.exit71 movl numARows(%rip), %eax testl %eax, %eax jle .LBB3_36 # %bb.27: # %.preheader.lr.ph.i movl numCColumns(%rip), %ecx movl numAColumns(%rip), %edx movslq numBColumns(%rip), %rsi movslq %ecx, %rdi movslq %edx, %r8 shlq $2, %r8 shlq $2, %rsi xorl %r9d, %r9d movq 56(%rsp), %r10 # 8-byte Reload jmp .LBB3_28 .p2align 4, 0x90 .LBB3_35: # %._crit_edge30.i # in Loop: Header=BB3_28 Depth=1 incq %r9 addq %r8, %r10 cmpq %rax, %r9 je .LBB3_36 .LBB3_28: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB3_30 Depth 2 # Child Loop BB3_32 Depth 3 testl %edx, %edx jle .LBB3_35 # %bb.29: # %.lr.ph29.i # in Loop: Header=BB3_28 Depth=1 movq %r9, %r11 imulq %rdi, %r11 leaq (%r12,%r11,4), %r11 movq 64(%rsp), %r13 # 8-byte Reload xorl %ebp, %ebp jmp .LBB3_30 .p2align 4, 0x90 .LBB3_34: # %._crit_edge.i # in Loop: Header=BB3_30 Depth=2 incq %rbp addq $4, %r13 cmpq %rdx, %rbp je .LBB3_35 .LBB3_30: # Parent Loop BB3_28 Depth=1 # => This Loop Header: Depth=2 # Child Loop BB3_32 Depth 3 movl $0, (%r11,%rbp,4) testl %ecx, %ecx jle .LBB3_34 # %bb.31: # %.lr.ph.i72 # in Loop: Header=BB3_30 Depth=2 xorps %xmm0, %xmm0 movq %r13, %rbx xorl %r15d, %r15d .p2align 4, 0x90 .LBB3_32: # Parent Loop BB3_28 Depth=1 # Parent Loop BB3_30 Depth=2 # => This Inner Loop Header: Depth=3 movss (%r10,%r15,4), %xmm1 # xmm1 = mem[0],zero,zero,zero mulss (%rbx), %xmm1 addss %xmm1, %xmm0 incq %r15 addq %rsi, %rbx cmpq %r15, %rcx jne .LBB3_32 # %bb.33: # %._crit_edge.i.loopexit # in Loop: Header=BB3_30 Depth=2 movss %xmm0, (%r11,%rbp,4) jmp .LBB3_34 .LBB3_36: # %_Z17matMultiplyOnHostPfS_S_iiiiii.exit movl $.Lstr.3, %edi callq puts@PLT movl numCColumns(%rip), %ebp movl numCRows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_42 # %bb.37: # %.lr.ph.preheader.i78 movl %eax, %ebx xorl %r13d, %r13d jmp .LBB3_38 .p2align 4, 0x90 .LBB3_41: # in Loop: Header=BB3_38 Depth=1 incq %r13 cmpq %r13, %rbx je .LBB3_42 .LBB3_38: # %.lr.ph.i80 # =>This Inner Loop Header: Depth=1 movss (%r12,%r13,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r13d, %eax cltd idivl %ebp testq %r13, %r13 je .LBB3_41 # %bb.39: # %.lr.ph.i80 # in Loop: Header=BB3_38 Depth=1 testl %edx, %edx jne .LBB3_41 # %bb.40: # in Loop: Header=BB3_38 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_41 .LBB3_42: # %_Z8printMatiiPf.exit86 movl $.Lstr.4, %edi callq puts@PLT movl numCColumns(%rip), %ecx movl numCRows(%rip), %eax imull %ecx, %eax testl %eax, %eax movq 56(%rsp), %rbx # 8-byte Reload movq %r14, %r15 movq 64(%rsp), %r14 # 8-byte Reload jle .LBB3_47 # %bb.43: # %.lr.ph101.preheader movl %eax, %edx xorl %eax, %eax .p2align 4, 0x90 .LBB3_44: # %.lr.ph101 # =>This Inner Loop Header: Depth=1 movss (%r12,%rax,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss (%r15,%rax,4), %xmm1 # xmm1 = mem[0],zero,zero,zero ucomiss %xmm1, %xmm0 jne .LBB3_45 jp .LBB3_45 # %bb.46: # in Loop: Header=BB3_44 Depth=1 incq %rax cmpq %rax, %rdx jne .LBB3_44 jmp .LBB3_47 .LBB3_45: # kill: def $eax killed $eax killed $rax cltd idivl %ecx cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 movl $.L.str.10, %edi movl %eax, %esi movb $2, %al callq printf .LBB3_47: # %.loopexit movl numCColumns(%rip), %eax movl %eax, %esi shrl $31, %esi addl %eax, %esi sarl %esi incl %esi imull %esi, %esi movl $.L.str.11, %edi xorl %eax, %eax callq printf movl $.L.str.12, %edi movl $4, %esi xorl %eax, %eax callq printf movq 24(%rsp), %rdi callq hipFree movq 16(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree movq %rbx, %rdi callq free movq %r14, %rdi callq free movq %r15, %rdi callq free movq %r12, %rdi callq free xorl %eax, %eax addq $216, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end3: .size main, .Lfunc_end3-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB4_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB4_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z20matrixMultiplySharedPfS_S_iiiiii, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end4: .size __hip_module_ctor, .Lfunc_end4-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB5_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB5_2: retq .Lfunc_end5: .size __hip_module_dtor, .Lfunc_end5-__hip_module_dtor .cfi_endproc # -- End function .type numARows,@object # @numARows .bss .globl numARows .p2align 2, 0x0 numARows: .long 0 # 0x0 .size numARows, 4 .type numAColumns,@object # @numAColumns .globl numAColumns .p2align 2, 0x0 numAColumns: .long 0 # 0x0 .size numAColumns, 4 .type numBRows,@object # @numBRows .globl numBRows .p2align 2, 0x0 numBRows: .long 0 # 0x0 .size numBRows, 4 .type numBColumns,@object # @numBColumns .globl numBColumns .p2align 2, 0x0 numBColumns: .long 0 # 0x0 .size numBColumns, 4 .type numCRows,@object # @numCRows .globl numCRows .p2align 2, 0x0 numCRows: .long 0 # 0x0 .size numCRows, 4 .type numCColumns,@object # @numCColumns .globl numCColumns .p2align 2, 0x0 numCColumns: .long 0 # 0x0 .size numCColumns, 4 .type _Z20matrixMultiplySharedPfS_S_iiiiii,@object # @_Z20matrixMultiplySharedPfS_S_iiiiii .section .rodata,"a",@progbits .globl _Z20matrixMultiplySharedPfS_S_iiiiii .p2align 3, 0x0 _Z20matrixMultiplySharedPfS_S_iiiiii: .quad _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .size _Z20matrixMultiplySharedPfS_S_iiiiii, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%f " .size .L.str, 5 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "\nEnter Rows and Columns of A:" .size .L.str.2, 30 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "%d %d" .size .L.str.3, 6 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "\nEnter Rows and Columns of B:" .size .L.str.4, 30 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n" .size .L.str.10, 65 .type .L.str.11,@object # @.str.11 .L.str.11: .asciz "\nNumber of Blocks Created:%d \n" .size .L.str.11, 31 .type .L.str.12,@object # @.str.12 .L.str.12: .asciz "\nNumber of Threads Per Block: %d \n" .size .L.str.12, 35 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z20matrixMultiplySharedPfS_S_iiiiii" .size .L__unnamed_1, 37 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "\nMatrix A Values:" .size .Lstr, 18 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "\n\nMatrix B Values:" .size .Lstr.1, 19 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "\nMatrix C From Device" .size .Lstr.2, 22 .type .Lstr.3,@object # @str.3 .Lstr.3: .asciz "\nMatrix C From Host" .size .Lstr.3, 20 .type .Lstr.4,@object # @str.4 .Lstr.4: .asciz "\n" .size .Lstr.4, 2 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym numARows .addrsig_sym numAColumns .addrsig_sym numBRows .addrsig_sym numBColumns .addrsig_sym _Z20matrixMultiplySharedPfS_S_iiiiii .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z20matrixMultiplySharedPfS_S_iiiiii .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R0, SR_TID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002200 */ /*0020*/ ISETP.LE.AND P0, PT, RZ, c[0x0][0x17c], PT ; /* 0x00005f00ff007a0c */ /* 0x000fe20003f03270 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fe200078e00ff */ /*0050*/ S2R R2, SR_TID.X ; /* 0x0000000000027919 */ /* 0x000e680000002100 */ /*0060*/ S2R R5, SR_CTAID.Y ; /* 0x0000000000057919 */ /* 0x000ea80000002600 */ /*0070*/ S2R R7, SR_CTAID.X ; /* 0x0000000000077919 */ /* 0x000ee20000002500 */ /*0080*/ IMAD.SHL.U32 R3, R0, 0x8, RZ ; /* 0x0000000800037824 */ /* 0x001fca00078e00ff */ /*0090*/ LEA R3, R2, R3, 0x2 ; /* 0x0000000302037211 */ /* 0x002fe200078e10ff */ /*00a0*/ IMAD R5, R5, c[0x0][0x4], R0 ; /* 0x0000010005057a24 */ /* 0x004fc800078e0200 */ /*00b0*/ STS [R3], RZ ; /* 0x000000ff03007388 */ /* 0x0001e20000000800 */ /*00c0*/ IMAD R6, R7, c[0x0][0x0], R2 ; /* 0x0000000007067a24 */ /* 0x008fc600078e0202 */ /*00d0*/ STS [R3+0x10], RZ ; /* 0x000010ff03007388 */ /* 0x0001e20000000800 */ /*00e0*/ @!P0 BRA 0xab0 ; /* 0x000009c000008947 */ /* 0x000fea0003800000 */ /*00f0*/ IMAD.MOV.U32 R8, RZ, RZ, c[0x0][0x17c] ; /* 0x00005f00ff087624 */ /* 0x000fe200078e00ff */ /*0100*/ HFMA2.MMA R9, -RZ, RZ, 0, 0 ; /* 0x00000000ff097435 */ /* 0x000fe200000001ff */ /*0110*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fc600078e00ff */ /*0120*/ IADD3 R4, R8, -0x1, RZ ; /* 0xffffffff08047810 */ /* 0x000fc80007ffe0ff */ /*0130*/ LEA.HI R4, R4, R4, RZ, 0x1 ; /* 0x0000000404047211 */ /* 0x000fc800078f08ff */ /*0140*/ SHF.R.S32.HI R4, RZ, 0x1, R4 ; /* 0x00000001ff047819 */ /* 0x000fc80000011404 */ /*0150*/ IMNMX R4, RZ, R4, !PT ; /* 0x00000004ff047217 */ /* 0x000fc80007800200 */ /*0160*/ ISETP.GE.U32.AND P0, PT, R4.reuse, 0x3, PT ; /* 0x000000030400780c */ /* 0x040fe40003f06070 */ /*0170*/ IADD3 R7, R4, 0x1, RZ ; /* 0x0000000104077810 */ /* 0x000fc80007ffe0ff */ /*0180*/ LOP3.LUT R7, R7, 0x3, RZ, 0xc0, !PT ; /* 0x0000000307077812 */ /* 0x000fc800078ec0ff */ /*0190*/ ISETP.NE.AND P1, PT, R7, RZ, PT ; /* 0x000000ff0700720c */ /* 0x000fc60003f25270 */ /*01a0*/ @!P0 BRA 0x880 ; /* 0x000006d000008947 */ /* 0x000ff40003800000 */ /*01b0*/ IMAD R21, R5.reuse, c[0x0][0x17c], R2 ; /* 0x00005f0005157a24 */ /* 0x040fe200078e0202 */ /*01c0*/ IADD3 R11, R0.reuse, 0x6, RZ ; /* 0x00000006000b7810 */ /* 0x040fe20007ffe0ff */ /*01d0*/ IMAD R8, R5.reuse, R8, 0x2 ; /* 0x0000000205087424 */ /* 0x040fe200078e0208 */ /*01e0*/ IADD3 R23, R0.reuse, 0x4, RZ ; /* 0x0000000400177810 */ /* 0x040fe20007ffe0ff */ /*01f0*/ IMAD R22, R0.reuse, c[0x0][0x184], R6 ; /* 0x0000610000167a24 */ /* 0x040fe200078e0206 */ /*0200*/ IADD3 R13, R0, 0x2, RZ ; /* 0x00000002000d7810 */ /* 0x000fe20007ffe0ff */ /*0210*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0220*/ IADD3 R10, -R4, -0x1, R7 ; /* 0xffffffff040a7810 */ /* 0x000fe20007ffe107 */ /*0230*/ IMAD.MOV.U32 R20, RZ, RZ, R2 ; /* 0x000000ffff147224 */ /* 0x000fe200078e0002 */ /*0240*/ ISETP.GE.AND P2, PT, R5, c[0x0][0x178], PT ; /* 0x00005e0005007a0c */ /* 0x000fe20003f46270 */ /*0250*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fe200078e00ff */ /*0260*/ MOV R4, R0 ; /* 0x0000000000047202 */ /* 0x000fe20000000f00 */ /*0270*/ IMAD R11, R11, c[0x0][0x184], R6.reuse ; /* 0x000061000b0b7a24 */ /* 0x100fe200078e0206 */ /*0280*/ IADD3 R21, R21, 0x6, RZ ; /* 0x0000000615157810 */ /* 0x000fe20007ffe0ff */ /*0290*/ IMAD R23, R23, c[0x0][0x184], R6 ; /* 0x0000610017177a24 */ /* 0x000fc400078e0206 */ /*02a0*/ IMAD R24, R13, c[0x0][0x184], R6 ; /* 0x000061000d187a24 */ /* 0x000fe400078e0206 */ /*02b0*/ ISETP.GE.AND P0, PT, R6, c[0x0][0x184], PT ; /* 0x0000610006007a0c */ /* 0x000fe20003f06270 */ /*02c0*/ HFMA2.MMA R16, -RZ, RZ, 0, 0 ; /* 0x00000000ff107435 */ /* 0x000fe200000001ff */ /*02d0*/ ISETP.GE.U32.OR P3, PT, R20, c[0x0][0x17c], P2 ; /* 0x00005f0014007a0c */ /* 0x000fe20001766470 */ /*02e0*/ IMAD.MOV.U32 R28, RZ, RZ, RZ ; /* 0x000000ffff1c7224 */ /* 0x000fe200078e00ff */ /*02f0*/ ISETP.GE.U32.OR P4, PT, R4, c[0x0][0x180], P0 ; /* 0x0000600004007a0c */ /* 0x000fd60000786470 */ /*0300*/ @!P3 MOV R15, 0x4 ; /* 0x00000004000fb802 */ /* 0x000fe20000000f00 */ /*0310*/ @!P3 IMAD R14, R5, c[0x0][0x17c], R20 ; /* 0x00005f00050eba24 */ /* 0x000fe400078e0214 */ /*0320*/ @!P4 IMAD.MOV.U32 R13, RZ, RZ, 0x4 ; /* 0x00000004ff0dc424 */ /* 0x000fe400078e00ff */ /*0330*/ @!P3 IMAD.WIDE.U32 R14, R14, R15, c[0x0][0x160] ; /* 0x000058000e0eb625 */ /* 0x000fc800078e000f */ /*0340*/ @!P4 IMAD.WIDE.U32 R12, R22, R13, c[0x0][0x168] ; /* 0x00005a00160cc625 */ /* 0x000fe200078e000d */ /*0350*/ @!P3 LDG.E R28, [R14.64] ; /* 0x000000040e1cb981 */ /* 0x000ea8000c1e1900 */ /*0360*/ @!P4 LDG.E R16, [R12.64] ; /* 0x000000040c10c981 */ /* 0x0002e2000c1e1900 */ /*0370*/ IADD3 R17, R20, 0x2, RZ ; /* 0x0000000214117810 */ /* 0x000fe20007ffe0ff */ /*0380*/ CS2R R26, SRZ ; /* 0x00000000001a7805 */ /* 0x000fc6000001ff00 */ /*0390*/ ISETP.GE.U32.OR P3, PT, R17, c[0x0][0x17c], P2 ; /* 0x00005f0011007a0c */ /* 0x000fe40001766470 */ /*03a0*/ IADD3 R17, R4, 0x2, RZ ; /* 0x0000000204117810 */ /* 0x000fc80007ffe0ff */ /*03b0*/ ISETP.GE.U32.OR P4, PT, R17, c[0x0][0x180], P0 ; /* 0x0000600011007a0c */ /* 0x000fce0000786470 */ /*03c0*/ @!P3 IMAD.IADD R18, R8, 0x1, R20 ; /* 0x000000010812b824 */ /* 0x000fe400078e0214 */ /*03d0*/ @!P3 IMAD.MOV.U32 R19, RZ, RZ, 0x4 ; /* 0x00000004ff13b424 */ /* 0x000fc800078e00ff */ /*03e0*/ @!P4 MOV R17, 0x4 ; /* 0x000000040011c802 */ /* 0x000fe20000000f00 */ /*03f0*/ @!P3 IMAD.WIDE.U32 R18, R18, R19, c[0x0][0x160] ; /* 0x000058001212b625 */ /* 0x000fc800078e0013 */ /*0400*/ @!P4 IMAD.WIDE.U32 R12, R24, R17, c[0x0][0x168] ; /* 0x00005a00180cc625 */ /* 0x002fe200078e0011 */ /*0410*/ STS [R3], R28 ; /* 0x0000001c03007388 */ /* 0x004fe80000000800 */ /*0420*/ STS [R3+0x10], R16 ; /* 0x0000101003007388 */ /* 0x0083e80000000800 */ /*0430*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0440*/ @!P3 LDG.E R26, [R18.64] ; /* 0x00000004121ab981 */ /* 0x000ea8000c1e1900 */ /*0450*/ @!P4 LDG.E R27, [R12.64] ; /* 0x000000040c1bc981 */ /* 0x000722000c1e1900 */ /*0460*/ IADD3 R14, R20, 0x4, RZ ; /* 0x00000004140e7810 */ /* 0x000fc60007ffe0ff */ /*0470*/ LDS R28, [R2.X4+0x18] ; /* 0x00001800021c7984 */ /* 0x000fe20000004800 */ /*0480*/ ISETP.GE.U32.OR P3, PT, R14, c[0x0][0x17c], P2 ; /* 0x00005f000e007a0c */ /* 0x000fc60001766470 */ /*0490*/ LDS.64 R14, [R0.X8] ; /* 0x00000000000e7984 */ /* 0x000fe80000008a00 */ /*04a0*/ LDS R12, [R2.X4+0x10] ; /* 0x00001000020c7984 */ /* 0x008eec0000004800 */ /*04b0*/ @!P3 IADD3 R17, R21, -0x2, RZ ; /* 0xfffffffe1511b810 */ /* 0x000fe20007ffe0ff */ /*04c0*/ @!P3 IMAD.MOV.U32 R29, RZ, RZ, 0x4 ; /* 0x00000004ff1db424 */ /* 0x000fc800078e00ff */ /*04d0*/ @!P3 IMAD.WIDE.U32 R16, R17, R29, c[0x0][0x160] ; /* 0x000058001110b625 */ /* 0x002fe200078e001d */ /*04e0*/ HFMA2.MMA R29, -RZ, RZ, 0, 0 ; /* 0x00000000ff1d7435 */ /* 0x000fe200000001ff */ /*04f0*/ IADD3 R13, R4, 0x4, RZ ; /* 0x00000004040d7810 */ /* 0x000fe20007ffe0ff */ /*0500*/ STS [R3], R26 ; /* 0x0000001a03007388 */ /* 0x004fe80000000800 */ /*0510*/ STS [R3+0x10], R27 ; /* 0x0000101b03007388 */ /* 0x0107e80000000800 */ /*0520*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0530*/ @!P3 LDG.E R29, [R16.64] ; /* 0x00000004101db981 */ /* 0x0002a2000c1e1900 */ /*0540*/ ISETP.GE.U32.OR P3, PT, R13, c[0x0][0x180], P0 ; /* 0x000060000d007a0c */ /* 0x000fe20000766470 */ /*0550*/ FFMA R27, R12, R14, R25 ; /* 0x0000000e0c1b7223 */ /* 0x008fc40000000019 */ /*0560*/ LDS R14, [R2.X4+0x10] ; /* 0x00001000020e7984 */ /* 0x000fe80000004800 */ /*0570*/ LDS R26, [R2.X4+0x18] ; /* 0x00001800021a7984 */ /* 0x000fe20000004800 */ /*0580*/ MOV R16, RZ ; /* 0x000000ff00107202 */ /* 0x002fc60000000f00 */ /*0590*/ LDS.64 R12, [R0.X8] ; /* 0x00000000000c7984 */ /* 0x000e640000008a00 */ /*05a0*/ @!P3 IMAD.MOV.U32 R18, RZ, RZ, 0x4 ; /* 0x00000004ff12b424 */ /* 0x000fc800078e00ff */ /*05b0*/ @!P3 IMAD.WIDE.U32 R18, R23, R18, c[0x0][0x168] ; /* 0x00005a001712b625 */ /* 0x000fca00078e0012 */ /*05c0*/ @!P3 LDG.E R16, [R18.64] ; /* 0x000000041210b981 */ /* 0x000722000c1e1900 */ /*05d0*/ IADD3 R17, R20, 0x6, RZ ; /* 0x0000000614117810 */ /* 0x000fc80007ffe0ff */ /*05e0*/ ISETP.GE.U32.OR P3, PT, R17, c[0x0][0x17c], P2 ; /* 0x00005f0011007a0c */ /* 0x000fe20001766470 */ /*05f0*/ HFMA2.MMA R25, -RZ, RZ, 0, 0 ; /* 0x00000000ff197435 */ /* 0x000fd800000001ff */ /*0600*/ @!P3 IMAD.MOV.U32 R17, RZ, RZ, 0x4 ; /* 0x00000004ff11b424 */ /* 0x000fe200078e00ff */ /*0610*/ MOV R18, RZ ; /* 0x000000ff00127202 */ /* 0x008fe20000000f00 */ /*0620*/ STS [R3], R29 ; /* 0x0000001d03007388 */ /* 0x0045e40000000800 */ /*0630*/ IADD3 R29, R4, 0x6, RZ ; /* 0x00000006041d7810 */ /* 0x004fc80007ffe0ff */ /*0640*/ ISETP.GE.U32.OR P0, PT, R29, c[0x0][0x180], P0 ; /* 0x000060001d007a0c */ /* 0x000fe20000706470 */ /*0650*/ STS [R3+0x10], R16 ; /* 0x0000101003007388 */ /* 0x0105d80000000800 */ /*0660*/ @!P0 IMAD.MOV.U32 R19, RZ, RZ, 0x4 ; /* 0x00000004ff138424 */ /* 0x000fe400078e00ff */ /*0670*/ @!P3 IMAD.WIDE.U32 R16, R21, R17, c[0x0][0x160] ; /* 0x000058001510b625 */ /* 0x004fe200078e0011 */ /*0680*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0690*/ @!P3 LDG.E R25, [R16.64] ; /* 0x000000041019b981 */ /* 0x0004e4000c1e1900 */ /*06a0*/ @!P0 IMAD.WIDE.U32 R16, R11, R19, c[0x0][0x168] ; /* 0x00005a000b108625 */ /* 0x004fca00078e0013 */ /*06b0*/ @!P0 LDG.E R18, [R16.64] ; /* 0x0000000410128981 */ /* 0x000ea2000c1e1900 */ /*06c0*/ FFMA R15, R28, R15, R27 ; /* 0x0000000f1c0f7223 */ /* 0x000fc6000000001b */ /*06d0*/ LDS R27, [R2.X4+0x10] ; /* 0x00001000021b7984 */ /* 0x000fe20000004800 */ /*06e0*/ FFMA R28, R14, R12, R15 ; /* 0x0000000c0e1c7223 */ /* 0x002fc6000000000f */ /*06f0*/ LDS R12, [R2.X4+0x18] ; /* 0x00001800020c7984 */ /* 0x000fe80000004800 */ /*0700*/ LDS.64 R14, [R0.X8] ; /* 0x00000000000e7984 */ /* 0x000e620000008a00 */ /*0710*/ FFMA R13, R26, R13, R28 ; /* 0x0000000d1a0d7223 */ /* 0x000fe2000000001c */ /*0720*/ IADD3 R9, R9, 0x4, RZ ; /* 0x0000000409097810 */ /* 0x000fe40007ffe0ff */ /*0730*/ IADD3 R4, R4, 0x8, RZ ; /* 0x0000000804047810 */ /* 0x000fe40007ffe0ff */ /*0740*/ IADD3 R20, R20, 0x8, RZ ; /* 0x0000000814147810 */ /* 0x000fc40007ffe0ff */ /*0750*/ IADD3 R21, R21, 0x8, RZ ; /* 0x0000000815157810 */ /* 0x000fe20007ffe0ff */ /*0760*/ FFMA R14, R27, R14, R13 ; /* 0x0000000e1b0e7223 */ /* 0x002fc8000000000d */ /*0770*/ FFMA R14, R12, R15, R14 ; /* 0x0000000f0c0e7223 */ /* 0x000fe4000000000e */ /*0780*/ IMAD.IADD R12, R10, 0x1, R9 ; /* 0x000000010a0c7824 */ /* 0x000fca00078e0209 */ /*0790*/ ISETP.NE.AND P0, PT, R12, RZ, PT ; /* 0x000000ff0c00720c */ /* 0x000fe20003f05270 */ /*07a0*/ HFMA2.MMA R13, -RZ, RZ, 0, 4.76837158203125e-07 ; /* 0x00000008ff0d7435 */ /* 0x000fd400000001ff */ /*07b0*/ IMAD R11, R13.reuse, c[0x0][0x184], R11 ; /* 0x000061000d0b7a24 */ /* 0x040fe400078e020b */ /*07c0*/ IMAD R23, R13.reuse, c[0x0][0x184], R23 ; /* 0x000061000d177a24 */ /* 0x040fe400078e0217 */ /*07d0*/ IMAD R24, R13.reuse, c[0x0][0x184], R24 ; /* 0x000061000d187a24 */ /* 0x040fe400078e0218 */ /*07e0*/ IMAD R22, R13, c[0x0][0x184], R22 ; /* 0x000061000d167a24 */ /* 0x000fe200078e0216 */ /*07f0*/ STS [R3], R25 ; /* 0x0000001903007388 */ /* 0x008fe80000000800 */ /*0800*/ STS [R3+0x10], R18 ; /* 0x0000101203007388 */ /* 0x004fe80000000800 */ /*0810*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0820*/ LDS R29, [R2.X4+0x10] ; /* 0x00001000021d7984 */ /* 0x000fe80000004800 */ /*0830*/ LDS.64 R18, [R0.X8] ; /* 0x0000000000127984 */ /* 0x000e680000008a00 */ /*0840*/ LDS R26, [R2.X4+0x18] ; /* 0x00001800021a7984 */ /* 0x000ea20000004800 */ /*0850*/ FFMA R18, R29, R18, R14 ; /* 0x000000121d127223 */ /* 0x002fc8000000000e */ /*0860*/ FFMA R25, R26, R19, R18 ; /* 0x000000131a197223 */ /* 0x004fe20000000012 */ /*0870*/ @P0 BRA 0x2b0 ; /* 0xfffffa3000000947 */ /* 0x000fea000383ffff */ /*0880*/ @!P1 BRA 0xab0 ; /* 0x0000022000009947 */ /* 0x000fea0003800000 */ /*0890*/ IMAD R4, R5, c[0x0][0x17c], R2 ; /* 0x00005f0005047a24 */ /* 0x000fe200078e0202 */ /*08a0*/ IADD3 R16, -R7, RZ, RZ ; /* 0x000000ff07107210 */ /* 0x000fe20007ffe1ff */ /*08b0*/ IMAD R15, R9.reuse, 0x2, R0 ; /* 0x00000002090f7824 */ /* 0x040fe200078e0200 */ /*08c0*/ LEA R14, R9.reuse, R2, 0x1 ; /* 0x00000002090e7211 */ /* 0x040fe200078e08ff */ /*08d0*/ IMAD R7, R9, 0x2, R4 ; /* 0x0000000209077824 */ /* 0x000fe400078e0204 */ /*08e0*/ IMAD R4, R15, c[0x0][0x184], R6 ; /* 0x000061000f047a24 */ /* 0x000fe400078e0206 */ /*08f0*/ ISETP.GE.U32.AND P1, PT, R15, c[0x0][0x180], PT ; /* 0x000060000f007a0c */ /* 0x000fe20003f26070 */ /*0900*/ HFMA2.MMA R18, -RZ, RZ, 0, 0 ; /* 0x00000000ff127435 */ /* 0x000fe200000001ff */ /*0910*/ ISETP.GE.U32.AND P0, PT, R14, c[0x0][0x17c], PT ; /* 0x00005f000e007a0c */ /* 0x000fe20003f06070 */ /*0920*/ IMAD.MOV.U32 R20, RZ, RZ, RZ ; /* 0x000000ffff147224 */ /* 0x000fe200078e00ff */ /*0930*/ ISETP.GE.OR P1, PT, R6, c[0x0][0x184], P1 ; /* 0x0000610006007a0c */ /* 0x000fc40000f26670 */ /*0940*/ ISETP.GE.OR P0, PT, R5, c[0x0][0x178], P0 ; /* 0x00005e0005007a0c */ /* 0x000fd60000706670 */ /*0950*/ @!P1 MOV R11, 0x4 ; /* 0x00000004000b9802 */ /* 0x000fe40000000f00 */ /*0960*/ @!P0 IMAD.MOV.U32 R8, RZ, RZ, 0x4 ; /* 0x00000004ff088424 */ /* 0x000fc600078e00ff */ /*0970*/ @!P1 IMAD.WIDE.U32 R10, R4, R11, c[0x0][0x168] ; /* 0x00005a00040a9625 */ /* 0x000fc800078e000b */ /*0980*/ @!P0 IMAD.WIDE.U32 R8, R7, R8, c[0x0][0x160] ; /* 0x0000580007088625 */ /* 0x000fe200078e0008 */ /*0990*/ @!P1 LDG.E R18, [R10.64] ; /* 0x000000040a129981 */ /* 0x000ea8000c1e1900 */ /*09a0*/ @!P0 LDG.E R20, [R8.64] ; /* 0x0000000408148981 */ /* 0x000ee2000c1e1900 */ /*09b0*/ IADD3 R16, R16, 0x1, RZ ; /* 0x0000000110107810 */ /* 0x000fc80007ffe0ff */ /*09c0*/ ISETP.NE.AND P0, PT, R16, RZ, PT ; /* 0x000000ff1000720c */ /* 0x000fe20003f05270 */ /*09d0*/ IMAD.MOV.U32 R21, RZ, RZ, 0x2 ; /* 0x00000002ff157424 */ /* 0x000fe200078e00ff */ /*09e0*/ IADD3 R15, R15, 0x2, RZ ; /* 0x000000020f0f7810 */ /* 0x000fe40007ffe0ff */ /*09f0*/ IADD3 R14, R14, 0x2, RZ ; /* 0x000000020e0e7810 */ /* 0x000fe20007ffe0ff */ /*0a00*/ IMAD R4, R21, c[0x0][0x184], R4 ; /* 0x0000610015047a24 */ /* 0x000fe200078e0204 */ /*0a10*/ IADD3 R7, R7, 0x2, RZ ; /* 0x0000000207077810 */ /* 0x000fe20007ffe0ff */ /*0a20*/ STS [R3+0x10], R18 ; /* 0x0000101203007388 */ /* 0x004fe80000000800 */ /*0a30*/ STS [R3], R20 ; /* 0x0000001403007388 */ /* 0x008fe80000000800 */ /*0a40*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0a50*/ LDS R17, [R2.X4+0x10] ; /* 0x0000100002117984 */ /* 0x000fe80000004800 */ /*0a60*/ LDS.64 R12, [R0.X8] ; /* 0x00000000000c7984 */ /* 0x000e680000008a00 */ /*0a70*/ LDS R19, [R2.X4+0x18] ; /* 0x0000180002137984 */ /* 0x000ea20000004800 */ /*0a80*/ FFMA R12, R17, R12, R25 ; /* 0x0000000c110c7223 */ /* 0x002fc80000000019 */ /*0a90*/ FFMA R25, R19, R13, R12 ; /* 0x0000000d13197223 */ /* 0x004fe2000000000c */ /*0aa0*/ @P0 BRA 0x8f0 ; /* 0xfffffe4000000947 */ /* 0x000fea000383ffff */ /*0ab0*/ ISETP.GE.AND P0, PT, R6, c[0x0][0x18c], PT ; /* 0x0000630006007a0c */ /* 0x000fc80003f06270 */ /*0ac0*/ ISETP.GE.OR P0, PT, R5, c[0x0][0x188], P0 ; /* 0x0000620005007a0c */ /* 0x000fda0000706670 */ /*0ad0*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0ae0*/ MOV R3, 0x4 ; /* 0x0000000400037802 */ /* 0x001fe20000000f00 */ /*0af0*/ IMAD R2, R5, c[0x0][0x18c], R6 ; /* 0x0000630005027a24 */ /* 0x000fc800078e0206 */ /*0b00*/ IMAD.WIDE R2, R2, R3, c[0x0][0x170] ; /* 0x00005c0002027625 */ /* 0x000fca00078e0203 */ /*0b10*/ STG.E [R2.64], R25 ; /* 0x0000001902007986 */ /* 0x000fe2000c101904 */ /*0b20*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0b30*/ BRA 0xb30; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0b40*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b50*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0b90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ba0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0be0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z20matrixMultiplySharedPfS_S_iiiiii .globl _Z20matrixMultiplySharedPfS_S_iiiiii .p2align 8 .type _Z20matrixMultiplySharedPfS_S_iiiiii,@function _Z20matrixMultiplySharedPfS_S_iiiiii: s_clause 0x1 s_load_b32 s2, s[0:1], 0x3c s_load_b32 s10, s[0:1], 0x1c v_dual_mov_b32 v8, 0 :: v_dual_and_b32 v3, 0x3ff, v0 v_bfe_u32 v4, v0, 10, 10 s_mov_b32 s11, 0 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_3) | instid1(VALU_DEP_1) v_lshlrev_b32_e32 v5, 2, v3 s_waitcnt lgkmcnt(0) s_lshr_b32 s3, s2, 16 s_and_b32 s2, s2, 0xffff v_mad_u64_u32 v[0:1], null, s15, s3, v[4:5] v_mad_u64_u32 v[1:2], null, s14, s2, v[3:4] v_lshl_add_u32 v2, v4, 3, v5 v_mov_b32_e32 v5, 0 s_cmp_lt_i32 s10, 0 ds_store_2addr_b32 v2, v5, v5 offset1:4 s_cbranch_scc1 .LBB0_14 s_clause 0x2 s_load_b32 s2, s[0:1], 0x18 s_load_b64 s[8:9], s[0:1], 0x20 s_load_b128 s[4:7], s[0:1], 0x0 s_add_i32 s3, s10, -1 v_mad_u64_u32 v[5:6], null, v0, s10, v[3:4] s_lshr_b32 s12, s3, 31 v_dual_mov_b32 v8, 0 :: v_dual_add_nc_u32 v9, 16, v2 s_add_i32 s3, s3, s12 v_dual_mov_b32 v7, 0 :: v_dual_lshlrev_b32 v10, 3, v4 v_lshl_add_u32 v11, v3, 2, 16 s_ashr_i32 s3, s3, 1 s_delay_alu instid0(SALU_CYCLE_1) s_max_i32 s12, s3, 0 s_waitcnt lgkmcnt(0) v_cmp_gt_i32_e32 vcc_lo, s2, v0 v_cmp_gt_i32_e64 s2, s9, v1 .LBB0_2: v_mov_b32_e32 v6, 0 s_and_saveexec_b32 s13, vcc_lo s_cbranch_execz .LBB0_6 s_lshl_b32 s15, s11, 1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_add_nc_u32_e32 v6, s15, v3 v_cmp_gt_u32_e64 s3, s10, v6 v_mov_b32_e32 v6, 0 s_delay_alu instid0(VALU_DEP_2) s_and_saveexec_b32 s14, s3 s_cbranch_execz .LBB0_5 v_add_nc_u32_e32 v6, s15, v5 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[12:13], 2, v[6:7] v_add_co_u32 v12, s3, s4, v12 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v13, s3, s5, v13, s3 global_load_b32 v6, v[12:13], off .LBB0_5: s_or_b32 exec_lo, exec_lo, s14 .LBB0_6: s_delay_alu instid0(SALU_CYCLE_1) s_or_b32 exec_lo, exec_lo, s13 v_mov_b32_e32 v12, 0 s_waitcnt vmcnt(0) ds_store_b32 v2, v6 s_and_saveexec_b32 s13, s2 s_cbranch_execz .LBB0_10 v_lshl_add_u32 v6, s11, 1, v4 v_mov_b32_e32 v12, 0 s_mov_b32 s14, exec_lo s_delay_alu instid0(VALU_DEP_2) v_cmpx_gt_u32_e64 s8, v6 s_cbranch_execz .LBB0_9 v_mad_u64_u32 v[12:13], null, v6, s9, v[1:2] v_mov_b32_e32 v13, v7 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[12:13], 2, v[12:13] v_add_co_u32 v12, s3, s6, v12 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v13, s3, s7, v13, s3 global_load_b32 v12, v[12:13], off .LBB0_9: s_or_b32 exec_lo, exec_lo, s14 .LBB0_10: s_delay_alu instid0(SALU_CYCLE_1) s_or_b32 exec_lo, exec_lo, s13 v_mov_b32_e32 v6, v11 s_mov_b32 s3, 0 s_waitcnt vmcnt(0) ds_store_b32 v9, v12 s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv .LBB0_11: v_add_nc_u32_e32 v12, s3, v10 s_add_i32 s3, s3, 4 ds_load_b32 v13, v6 ds_load_b32 v12, v12 v_add_nc_u32_e32 v6, 8, v6 s_cmp_lg_u32 s3, 4 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v8, v12, v13 s_cbranch_scc0 .LBB0_11 s_add_i32 s3, s11, 1 s_cmp_eq_u32 s11, s12 s_cbranch_scc1 .LBB0_14 s_mov_b32 s11, s3 s_branch .LBB0_2 .LBB0_14: s_load_b64 s[2:3], s[0:1], 0x28 s_waitcnt lgkmcnt(0) v_cmp_gt_i32_e32 vcc_lo, s2, v0 v_cmp_gt_i32_e64 s2, s3, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_and_b32 s2, vcc_lo, s2 s_and_saveexec_b32 s4, s2 s_cbranch_execz .LBB0_16 s_load_b64 s[0:1], s[0:1], 0x10 v_mad_u64_u32 v[2:3], null, v0, s3, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[0:1], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v0, vcc_lo, s0, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo global_store_b32 v[0:1], v8, off .LBB0_16: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z20matrixMultiplySharedPfS_S_iiiiii .amdhsa_group_segment_fixed_size 32 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 304 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z20matrixMultiplySharedPfS_S_iiiiii, .Lfunc_end0-_Z20matrixMultiplySharedPfS_S_iiiiii .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value - .offset: 28 .size: 4 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: by_value - .offset: 36 .size: 4 .value_kind: by_value - .offset: 40 .size: 4 .value_kind: by_value - .offset: 44 .size: 4 .value_kind: by_value - .offset: 48 .size: 4 .value_kind: hidden_block_count_x - .offset: 52 .size: 4 .value_kind: hidden_block_count_y - .offset: 56 .size: 4 .value_kind: hidden_block_count_z - .offset: 60 .size: 2 .value_kind: hidden_group_size_x - .offset: 62 .size: 2 .value_kind: hidden_group_size_y - .offset: 64 .size: 2 .value_kind: hidden_group_size_z - .offset: 66 .size: 2 .value_kind: hidden_remainder_x - .offset: 68 .size: 2 .value_kind: hidden_remainder_y - .offset: 70 .size: 2 .value_kind: hidden_remainder_z - .offset: 88 .size: 8 .value_kind: hidden_global_offset_x - .offset: 96 .size: 8 .value_kind: hidden_global_offset_y - .offset: 104 .size: 8 .value_kind: hidden_global_offset_z - .offset: 112 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 32 .kernarg_segment_align: 8 .kernarg_segment_size: 304 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z20matrixMultiplySharedPfS_S_iiiiii .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z20matrixMultiplySharedPfS_S_iiiiii.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_000ceb69_00000000-6_mat_mul_shared_memory.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2062: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2062: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "%f " .LC1: .string "\n" .text .globl _Z8printMatiiPf .type _Z8printMatiiPf, @function _Z8printMatiiPf: .LFB2057: .cfi_startproc endbr64 imull %esi, %edi testl %edi, %edi jle .L9 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $8, %rsp .cfi_def_cfa_offset 64 movl %esi, %ebp movq %rdx, %r13 movslq %edi, %r12 movl $0, %ebx leaq .LC0(%rip), %r14 leaq .LC1(%rip), %r15 jmp .L6 .L5: addq $1, %rbx cmpq %r12, %rbx je .L12 .L6: pxor %xmm0, %xmm0 cvtss2sd 0(%r13,%rbx,4), %xmm0 movq %r14, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movl %ebx, %eax cltd idivl %ebp testl %edx, %edx jne .L5 testl %ebx, %ebx je .L5 movq %r15, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT jmp .L5 .L12: addq $8, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .L9: .cfi_restore 3 .cfi_restore 6 .cfi_restore 12 .cfi_restore 13 .cfi_restore 14 .cfi_restore 15 ret .cfi_endproc .LFE2057: .size _Z8printMatiiPf, .-_Z8printMatiiPf .globl _Z17matMultiplyOnHostPfS_S_iiiiii .type _Z17matMultiplyOnHostPfS_S_iiiiii, @function _Z17matMultiplyOnHostPfS_S_iiiiii: .LFB2058: .cfi_startproc endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 movq %rsi, -8(%rsp) movl %ecx, -12(%rsp) movl 72(%rsp), %ebx testl %ecx, %ecx jle .L13 movq %rdi, %rbp movq %rdx, %r15 movl %r8d, %r11d movslq 56(%rsp), %rdi salq $2, %rdi movl $0, %r13d movl $0, %ecx movl $0, %edx movslq %ebx, %r14 movq %r15, %r8 movq %r14, %rsi jmp .L15 .L19: movslq %ecx, %rax leaq (%r8,%rax,4), %r10 movq -8(%rsp), %r14 movslq %r13d, %rax leaq 0(%rbp,%rax,4), %r15 addq %rsi, %rax leaq 0(%rbp,%rax,4), %r9 movl $0, %r12d movl %edx, -20(%rsp) movl %ecx, -16(%rsp) .L18: movq %r10, %rcx movl $0x00000000, (%r10) testl %ebx, %ebx jle .L16 movq %r14, %rdx movq %r15, %rax .L17: movss (%rax), %xmm0 mulss (%rdx), %xmm0 addss (%rcx), %xmm0 movss %xmm0, (%rcx) addq $4, %rax addq %rdi, %rdx cmpq %r9, %rax jne .L17 .L16: addl $1, %r12d addq $4, %r10 addq $4, %r14 cmpl %r12d, %r11d jne .L18 movl -20(%rsp), %edx movl -16(%rsp), %ecx .L20: addl $1, %edx addl %ebx, %ecx addl %r11d, %r13d cmpl %edx, -12(%rsp) je .L13 .L15: testl %r11d, %r11d jg .L19 jmp .L20 .L13: popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2058: .size _Z17matMultiplyOnHostPfS_S_iiiiii, .-_Z17matMultiplyOnHostPfS_S_iiiiii .globl _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii .type _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii, @function _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii: .LFB2084: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 20(%rsp), %rax movq %rax, 136(%rsp) leaq 16(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) leaq 224(%rsp), %rax movq %rax, 176(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L28 .L24: movq 184(%rsp), %rax subq %fs:40, %rax jne .L29 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L28: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z20matrixMultiplySharedPfS_S_iiiiii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L24 .L29: call __stack_chk_fail@PLT .cfi_endproc .LFE2084: .size _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii, .-_Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii .globl _Z20matrixMultiplySharedPfS_S_iiiiii .type _Z20matrixMultiplySharedPfS_S_iiiiii, @function _Z20matrixMultiplySharedPfS_S_iiiiii: .LFB2085: .cfi_startproc endbr64 subq $16, %rsp .cfi_def_cfa_offset 24 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 40 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 48 call _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii addq $40, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _Z20matrixMultiplySharedPfS_S_iiiiii, .-_Z20matrixMultiplySharedPfS_S_iiiiii .section .rodata.str1.1 .LC3: .string "\nEnter Rows and Columns of A:" .LC4: .string "%d %d" .LC5: .string "\nEnter Rows and Columns of B:" .LC6: .string "\nMatrix A Values:\n" .LC7: .string "\n\nMatrix B Values:\n" .LC8: .string "\nMatrix C From Device\n" .LC9: .string "\nMatrix C From Host\n" .LC10: .string "\n\n" .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC11: .string "Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n" .align 8 .LC12: .string "\nNumber of Blocks Created:%d \n" .align 8 .LC13: .string "\nNumber of Threads Per Block: %d \n" .text .globl main .type main, @function main: .LFB2059: .cfi_startproc endbr64 pushq %r14 .cfi_def_cfa_offset 16 .cfi_offset 14, -16 pushq %r13 .cfi_def_cfa_offset 24 .cfi_offset 13, -24 pushq %r12 .cfi_def_cfa_offset 32 .cfi_offset 12, -32 pushq %rbp .cfi_def_cfa_offset 40 .cfi_offset 6, -40 pushq %rbx .cfi_def_cfa_offset 48 .cfi_offset 3, -48 subq $64, %rsp .cfi_def_cfa_offset 112 movq %fs:40, %rax movq %rax, 56(%rsp) xorl %eax, %eax leaq .LC3(%rip), %rsi movl $2, %edi call __printf_chk@PLT leaq numAColumns(%rip), %rdx leaq numARows(%rip), %rsi leaq .LC4(%rip), %rbx movq %rbx, %rdi movl $0, %eax call __isoc23_scanf@PLT leaq .LC5(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq numBColumns(%rip), %rdx leaq numBRows(%rip), %rsi movq %rbx, %rdi movl $0, %eax call __isoc23_scanf@PLT movl numARows(%rip), %ebx movl numAColumns(%rip), %r13d movslq %ebx, %rdi movslq %r13d, %rax imulq %rax, %rdi salq $2, %rdi call malloc@PLT movq %rax, %r12 movslq numBRows(%rip), %rdi movslq numBColumns(%rip), %rax imulq %rax, %rdi salq $2, %rdi call malloc@PLT movq %rax, %rbp imull %r13d, %ebx testl %ebx, %ebx jle .L33 movl $0, %ebx .L34: call rand@PLT movslq %eax, %rdx imulq $780903145, %rdx, %rdx sarq $33, %rdx movl %eax, %ecx sarl $31, %ecx subl %ecx, %edx leal (%rdx,%rdx,4), %ecx leal (%rdx,%rcx,2), %edx subl %edx, %eax addl $10, %eax pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, (%r12,%rbx,4) addq $1, %rbx movl numARows(%rip), %eax imull numAColumns(%rip), %eax cmpl %ebx, %eax jg .L34 .L33: movl numBRows(%rip), %eax imull numBColumns(%rip), %eax testl %eax, %eax jle .L35 movl $0, %ebx .L36: call rand@PLT movslq %eax, %rdx imulq $780903145, %rdx, %rdx sarq $33, %rdx movl %eax, %ecx sarl $31, %ecx subl %ecx, %edx leal (%rdx,%rdx,4), %ecx leal (%rdx,%rcx,2), %edx subl %edx, %eax addl $10, %eax pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, 0(%rbp,%rbx,4) addq $1, %rbx movl numBRows(%rip), %eax imull numBColumns(%rip), %eax cmpl %ebx, %eax jg .L36 .L35: leaq .LC6(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %r12, %rdx movl numAColumns(%rip), %esi movl numARows(%rip), %edi call _Z8printMatiiPf leaq .LC7(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %rbp, %rdx movl numBColumns(%rip), %esi movl numBRows(%rip), %edi call _Z8printMatiiPf movl numARows(%rip), %r14d movl %r14d, numCRows(%rip) movl numBColumns(%rip), %ebx movl %ebx, numCColumns(%rip) movslq %r14d, %r14 movslq %ebx, %rbx imulq %r14, %rbx salq $2, %rbx movq %rbx, %rdi call malloc@PLT movq %rax, %r13 movq %rbx, %rdi call malloc@PLT movq %rax, %rbx movslq numAColumns(%rip), %rsi imulq %r14, %rsi salq $2, %rsi leaq 8(%rsp), %rdi call cudaMalloc@PLT movslq numBRows(%rip), %rsi movslq numBColumns(%rip), %rax imulq %rax, %rsi salq $2, %rsi leaq 16(%rsp), %rdi call cudaMalloc@PLT movslq numCRows(%rip), %rsi movslq numCColumns(%rip), %rax imulq %rax, %rsi salq $2, %rsi leaq 24(%rsp), %rdi call cudaMalloc@PLT movslq numARows(%rip), %rdx movslq numAColumns(%rip), %rax imulq %rax, %rdx salq $2, %rdx movl $1, %ecx movq %r12, %rsi movq 8(%rsp), %rdi call cudaMemcpy@PLT movslq numBRows(%rip), %rdx movslq numBColumns(%rip), %rax imulq %rax, %rdx salq $2, %rdx movl $1, %ecx movq %rbp, %rsi movq 16(%rsp), %rdi call cudaMemcpy@PLT movl numCRows(%rip), %edx movl %edx, %eax shrl $31, %eax addl %edx, %eax sarl %eax addl $1, %eax movl numCColumns(%rip), %ecx movl %ecx, %edx shrl $31, %edx addl %ecx, %edx sarl %edx addl $1, %edx movl %edx, 32(%rsp) movl %eax, 36(%rsp) movl $2, 44(%rsp) movl $2, 48(%rsp) movl $0, %r9d movl $0, %r8d movq 44(%rsp), %rdx movl $1, %ecx movq 32(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L48 .L37: call cudaPeekAtLastError@PLT call cudaDeviceSynchronize@PLT movslq numCRows(%rip), %rdx movslq numCColumns(%rip), %rax imulq %rax, %rdx salq $2, %rdx movl $2, %ecx movq 24(%rsp), %rsi movq %r13, %rdi call cudaMemcpy@PLT leaq .LC8(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %r13, %rdx movl numCColumns(%rip), %esi movl numCRows(%rip), %edi call _Z8printMatiiPf subq $8, %rsp .cfi_def_cfa_offset 120 movl numCColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 128 movl numCRows(%rip), %eax pushq %rax .cfi_def_cfa_offset 136 movl numBColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 144 movl numBRows(%rip), %r9d movl numAColumns(%rip), %r8d movl numARows(%rip), %ecx movq %rbx, %rdx movq %rbp, %rsi movq %r12, %rdi call _Z17matMultiplyOnHostPfS_S_iiiiii addq $32, %rsp .cfi_def_cfa_offset 112 leaq .LC9(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq %rbx, %rdx movl numCColumns(%rip), %esi movl numCRows(%rip), %edi call _Z8printMatiiPf leaq .LC10(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movl numCColumns(%rip), %esi movl %esi, %edx imull numCRows(%rip), %edx testl %edx, %edx jle .L38 movslq %edx, %rdx movl $0, %eax .L41: movss (%rbx,%rax,4), %xmm0 movss 0(%r13,%rax,4), %xmm1 ucomiss %xmm1, %xmm0 jp .L43 jne .L43 addq $1, %rax cmpq %rax, %rdx jne .L41 jmp .L38 .L48: subq $8, %rsp .cfi_def_cfa_offset 120 movl numCColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 128 movl numCRows(%rip), %eax pushq %rax .cfi_def_cfa_offset 136 movl numBColumns(%rip), %eax pushq %rax .cfi_def_cfa_offset 144 movl numBRows(%rip), %r9d movl numAColumns(%rip), %r8d movl numARows(%rip), %ecx movq 56(%rsp), %rdx movq 48(%rsp), %rsi movq 40(%rsp), %rdi call _Z50__device_stub__Z20matrixMultiplySharedPfS_S_iiiiiiPfS_S_iiiiii addq $32, %rsp .cfi_def_cfa_offset 112 jmp .L37 .L43: cvtss2sd %xmm0, %xmm0 cltd idivl %esi movl %edx, %ecx cvtss2sd %xmm1, %xmm1 movl %eax, %edx leaq .LC11(%rip), %rsi movl $2, %edi movl $2, %eax call __printf_chk@PLT .L38: movl numCColumns(%rip), %eax movl %eax, %edx shrl $31, %edx addl %eax, %edx sarl %edx addl $1, %edx imull %edx, %edx leaq .LC12(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movl $4, %edx leaq .LC13(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 8(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call cudaFree@PLT movq 24(%rsp), %rdi call cudaFree@PLT movq %r12, %rdi call free@PLT movq %rbp, %rdi call free@PLT movq %r13, %rdi call free@PLT movq %rbx, %rdi call free@PLT movq 56(%rsp), %rax subq %fs:40, %rax jne .L49 movl $0, %eax addq $64, %rsp .cfi_remember_state .cfi_def_cfa_offset 48 popq %rbx .cfi_def_cfa_offset 40 popq %rbp .cfi_def_cfa_offset 32 popq %r12 .cfi_def_cfa_offset 24 popq %r13 .cfi_def_cfa_offset 16 popq %r14 .cfi_def_cfa_offset 8 ret .L49: .cfi_restore_state call __stack_chk_fail@PLT .cfi_endproc .LFE2059: .size main, .-main .section .rodata.str1.8 .align 8 .LC14: .string "_Z20matrixMultiplySharedPfS_S_iiiiii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2087: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC14(%rip), %rdx movq %rdx, %rcx leaq _Z20matrixMultiplySharedPfS_S_iiiiii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2087: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .globl numCColumns .bss .align 4 .type numCColumns, @object .size numCColumns, 4 numCColumns: .zero 4 .globl numCRows .align 4 .type numCRows, @object .size numCRows, 4 numCRows: .zero 4 .globl numBColumns .align 4 .type numBColumns, @object .size numBColumns, 4 numBColumns: .zero 4 .globl numBRows .align 4 .type numBRows, @object .size numBRows, 4 numBRows: .zero 4 .globl numAColumns .align 4 .type numAColumns, @object .size numAColumns, 4 numAColumns: .zero 4 .globl numARows .align 4 .type numARows, @object .size numARows, 4 numARows: .zero 4 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "mat_mul_shared_memory.hip" .globl _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii # -- Begin function _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .p2align 4, 0x90 .type _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii,@function _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii: # @_Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 20(%rsp), %rax movq %rax, 120(%rsp) leaq 16(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 192(%rsp), %rax movq %rax, 160(%rsp) leaq 56(%rsp), %rdi leaq 40(%rsp), %rsi leaq 32(%rsp), %rdx leaq 24(%rsp), %rcx callq __hipPopCallConfiguration movq 56(%rsp), %rsi movl 64(%rsp), %edx movq 40(%rsp), %rcx movl 48(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z20matrixMultiplySharedPfS_S_iiiiii, %edi pushq 24(%rsp) .cfi_adjust_cfa_offset 8 pushq 40(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end0: .size _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii, .Lfunc_end0-_Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .cfi_endproc # -- End function .globl _Z8printMatiiPf # -- Begin function _Z8printMatiiPf .p2align 4, 0x90 .type _Z8printMatiiPf,@function _Z8printMatiiPf: # @_Z8printMatiiPf .cfi_startproc # %bb.0: imull %esi, %edi testl %edi, %edi jle .LBB1_7 # %bb.1: # %.lr.ph.preheader pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %rbx .cfi_def_cfa_offset 40 pushq %rax .cfi_def_cfa_offset 48 .cfi_offset %rbx, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movq %rdx, %rbx movl %esi, %ebp movl %edi, %r15d xorl %r14d, %r14d jmp .LBB1_2 .p2align 4, 0x90 .LBB1_5: # in Loop: Header=BB1_2 Depth=1 incq %r14 cmpq %r14, %r15 je .LBB1_6 .LBB1_2: # %.lr.ph # =>This Inner Loop Header: Depth=1 movss (%rbx,%r14,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r14d, %eax cltd idivl %ebp testq %r14, %r14 je .LBB1_5 # %bb.3: # %.lr.ph # in Loop: Header=BB1_2 Depth=1 testl %edx, %edx jne .LBB1_5 # %bb.4: # in Loop: Header=BB1_2 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB1_5 .LBB1_6: addq $8, %rsp .cfi_def_cfa_offset 40 popq %rbx .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 .cfi_restore %rbx .cfi_restore %r14 .cfi_restore %r15 .cfi_restore %rbp .LBB1_7: # %._crit_edge retq .Lfunc_end1: .size _Z8printMatiiPf, .Lfunc_end1-_Z8printMatiiPf .cfi_endproc # -- End function .globl _Z17matMultiplyOnHostPfS_S_iiiiii # -- Begin function _Z17matMultiplyOnHostPfS_S_iiiiii .p2align 4, 0x90 .type _Z17matMultiplyOnHostPfS_S_iiiiii,@function _Z17matMultiplyOnHostPfS_S_iiiiii: # @_Z17matMultiplyOnHostPfS_S_iiiiii .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movq %rdx, -8(%rsp) # 8-byte Spill movq %rsi, -16(%rsp) # 8-byte Spill testl %ecx, %ecx jle .LBB2_9 # %bb.1: # %.preheader.lr.ph movl 72(%rsp), %eax movslq 56(%rsp), %r9 movslq %eax, %r10 movslq %r8d, %r11 movl %ecx, %ecx movl %r11d, %ebx movl %r10d, %r14d shlq $2, %r11 shlq $2, %r9 xorl %r15d, %r15d jmp .LBB2_2 .p2align 4, 0x90 .LBB2_8: # %._crit_edge30 # in Loop: Header=BB2_2 Depth=1 incq %r15 addq %r11, %rdi cmpq %rcx, %r15 je .LBB2_9 .LBB2_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB2_4 Depth 2 # Child Loop BB2_6 Depth 3 testl %r8d, %r8d jle .LBB2_8 # %bb.3: # %.lr.ph29 # in Loop: Header=BB2_2 Depth=1 movq %r15, %rdx imulq %r10, %rdx movq -8(%rsp), %rsi # 8-byte Reload leaq (%rsi,%rdx,4), %r12 movq -16(%rsp), %rsi # 8-byte Reload xorl %ebp, %ebp jmp .LBB2_4 .p2align 4, 0x90 .LBB2_7: # %._crit_edge # in Loop: Header=BB2_4 Depth=2 incq %rbp addq $4, %rsi cmpq %rbx, %rbp je .LBB2_8 .LBB2_4: # Parent Loop BB2_2 Depth=1 # => This Loop Header: Depth=2 # Child Loop BB2_6 Depth 3 movl $0, (%r12,%rbp,4) testl %eax, %eax jle .LBB2_7 # %bb.5: # %.lr.ph # in Loop: Header=BB2_4 Depth=2 movss (%r12,%rbp,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movq %rsi, %r13 xorl %edx, %edx .p2align 4, 0x90 .LBB2_6: # Parent Loop BB2_2 Depth=1 # Parent Loop BB2_4 Depth=2 # => This Inner Loop Header: Depth=3 movss (%rdi,%rdx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero mulss (%r13), %xmm1 addss %xmm1, %xmm0 movss %xmm0, (%r12,%rbp,4) incq %rdx addq %r9, %r13 cmpq %rdx, %r14 jne .LBB2_6 jmp .LBB2_7 .LBB2_9: # %._crit_edge32 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end2: .size _Z17matMultiplyOnHostPfS_S_iiiiii, .Lfunc_end2-_Z17matMultiplyOnHostPfS_S_iiiiii .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 subq $216, %rsp .cfi_def_cfa_offset 272 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movl $.L.str.2, %edi xorl %eax, %eax callq printf movl $.L.str.3, %edi movl $numARows, %esi movl $numAColumns, %edx xorl %eax, %eax callq __isoc23_scanf movl $.L.str.4, %edi xorl %eax, %eax callq printf movl $.L.str.3, %edi movl $numBRows, %esi movl $numBColumns, %edx xorl %eax, %eax callq __isoc23_scanf movslq numARows(%rip), %rax movslq numAColumns(%rip), %rdi imulq %rax, %rdi shlq $2, %rdi callq malloc movq %rax, %r13 movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rdi imulq %rax, %rdi shlq $2, %rdi callq malloc movq %rax, %r14 movl numAColumns(%rip), %eax imull numARows(%rip), %eax testl %eax, %eax jle .LBB3_3 # %bb.1: # %.lr.ph.preheader xorl %ebx, %ebx .p2align 4, 0x90 .LBB3_2: # %.lr.ph # =>This Inner Loop Header: Depth=1 callq rand cltq imulq $780903145, %rax, %rcx # imm = 0x2E8BA2E9 movq %rcx, %rdx shrq $63, %rdx sarq $33, %rcx addl %edx, %ecx leal (%rcx,%rcx,4), %edx leal (%rcx,%rdx,2), %ecx negl %ecx addl %ecx, %eax addl $10, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r13,%rbx,4) incq %rbx movslq numARows(%rip), %rax movslq numAColumns(%rip), %rcx imulq %rax, %rcx cmpq %rcx, %rbx jl .LBB3_2 .LBB3_3: # %.preheader movl numBColumns(%rip), %eax imull numBRows(%rip), %eax testl %eax, %eax jle .LBB3_6 # %bb.4: # %.lr.ph95.preheader xorl %ebx, %ebx .p2align 4, 0x90 .LBB3_5: # %.lr.ph95 # =>This Inner Loop Header: Depth=1 callq rand cltq imulq $780903145, %rax, %rcx # imm = 0x2E8BA2E9 movq %rcx, %rdx shrq $63, %rdx sarq $33, %rcx addl %edx, %ecx leal (%rcx,%rcx,4), %edx leal (%rcx,%rdx,2), %ecx negl %ecx addl %ecx, %eax addl $10, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r14,%rbx,4) incq %rbx movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rcx imulq %rax, %rcx cmpq %rcx, %rbx jl .LBB3_5 .LBB3_6: # %._crit_edge movl $.Lstr, %edi callq puts@PLT movl numAColumns(%rip), %ebp movl numARows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_12 # %bb.7: # %.lr.ph.preheader.i movl %eax, %r12d xorl %r15d, %r15d jmp .LBB3_8 .p2align 4, 0x90 .LBB3_11: # in Loop: Header=BB3_8 Depth=1 incq %r15 cmpq %r15, %r12 je .LBB3_12 .LBB3_8: # %.lr.ph.i # =>This Inner Loop Header: Depth=1 movss (%r13,%r15,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r15d, %eax cltd idivl %ebp testq %r15, %r15 je .LBB3_11 # %bb.9: # %.lr.ph.i # in Loop: Header=BB3_8 Depth=1 testl %edx, %edx jne .LBB3_11 # %bb.10: # in Loop: Header=BB3_8 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_11 .LBB3_12: # %_Z8printMatiiPf.exit movl $.Lstr.1, %edi callq puts@PLT movl numBColumns(%rip), %ebp movl numBRows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_18 # %bb.13: # %.lr.ph.preheader.i54 movl %eax, %r12d xorl %r15d, %r15d jmp .LBB3_14 .p2align 4, 0x90 .LBB3_17: # in Loop: Header=BB3_14 Depth=1 incq %r15 cmpq %r15, %r12 je .LBB3_18 .LBB3_14: # %.lr.ph.i56 # =>This Inner Loop Header: Depth=1 movss (%r14,%r15,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r15d, %eax cltd idivl %ebp testq %r15, %r15 je .LBB3_17 # %bb.15: # %.lr.ph.i56 # in Loop: Header=BB3_14 Depth=1 testl %edx, %edx jne .LBB3_17 # %bb.16: # in Loop: Header=BB3_14 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_17 .LBB3_18: # %_Z8printMatiiPf.exit62 movslq numARows(%rip), %rbx movl %ebx, numCRows(%rip) movslq numBColumns(%rip), %r12 movl %r12d, numCColumns(%rip) shlq $2, %rbx imulq %rbx, %r12 movq %r12, %rdi callq malloc movq %rax, %r15 movq %r12, %rdi callq malloc movq %rax, %r12 movslq numAColumns(%rip), %rsi imulq %rbx, %rsi leaq 24(%rsp), %rdi callq hipMalloc movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rsi imulq %rax, %rsi shlq $2, %rsi leaq 16(%rsp), %rdi callq hipMalloc movslq numCRows(%rip), %rax movslq numCColumns(%rip), %rsi imulq %rax, %rsi shlq $2, %rsi leaq 8(%rsp), %rdi callq hipMalloc movq 24(%rsp), %rdi movslq numARows(%rip), %rax movslq numAColumns(%rip), %rdx imulq %rax, %rdx shlq $2, %rdx movq %r13, 56(%rsp) # 8-byte Spill movq %r13, %rsi movl $1, %ecx callq hipMemcpy movq 16(%rsp), %rdi movslq numBRows(%rip), %rax movslq numBColumns(%rip), %rdx imulq %rax, %rdx shlq $2, %rdx movq %r14, 64(%rsp) # 8-byte Spill movq %r14, %rsi movl $1, %ecx callq hipMemcpy movl numCColumns(%rip), %eax movl %eax, %ecx shrl $31, %ecx addl %eax, %ecx sarl %ecx incl %ecx movl numCRows(%rip), %eax movl %eax, %edi shrl $31, %edi addl %eax, %edi sarl %edi incl %edi shlq $32, %rdi orq %rcx, %rdi movabsq $8589934594, %rdx # imm = 0x200000002 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB3_20 # %bb.19: movq 24(%rsp), %rax movq 16(%rsp), %rcx movq 8(%rsp), %rdx movl numARows(%rip), %esi movl numAColumns(%rip), %edi movl numBRows(%rip), %r8d movl numBColumns(%rip), %r9d movl numCRows(%rip), %r10d movl numCColumns(%rip), %r11d movq %rax, 136(%rsp) movq %rcx, 128(%rsp) movq %rdx, 120(%rsp) movl %esi, 52(%rsp) movl %edi, 48(%rsp) movl %r8d, 44(%rsp) movl %r9d, 40(%rsp) movl %r10d, 36(%rsp) movl %r11d, 32(%rsp) leaq 136(%rsp), %rax movq %rax, 144(%rsp) leaq 128(%rsp), %rax movq %rax, 152(%rsp) leaq 120(%rsp), %rax movq %rax, 160(%rsp) leaq 52(%rsp), %rax movq %rax, 168(%rsp) leaq 48(%rsp), %rax movq %rax, 176(%rsp) leaq 44(%rsp), %rax movq %rax, 184(%rsp) leaq 40(%rsp), %rax movq %rax, 192(%rsp) leaq 36(%rsp), %rax movq %rax, 200(%rsp) leaq 32(%rsp), %rax movq %rax, 208(%rsp) leaq 104(%rsp), %rdi leaq 88(%rsp), %rsi leaq 80(%rsp), %rdx leaq 72(%rsp), %rcx callq __hipPopCallConfiguration movq 104(%rsp), %rsi movl 112(%rsp), %edx movq 88(%rsp), %rcx movl 96(%rsp), %r8d leaq 144(%rsp), %r9 movl $_Z20matrixMultiplySharedPfS_S_iiiiii, %edi pushq 72(%rsp) .cfi_adjust_cfa_offset 8 pushq 88(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB3_20: callq hipPeekAtLastError callq hipDeviceSynchronize movq 8(%rsp), %rsi movslq numCRows(%rip), %rax movslq numCColumns(%rip), %rdx imulq %rax, %rdx shlq $2, %rdx movq %r15, %r14 movq %r15, %rdi movl $2, %ecx callq hipMemcpy movl $.Lstr.2, %edi callq puts@PLT movl numCColumns(%rip), %ebp movl numCRows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_26 # %bb.21: # %.lr.ph.preheader.i63 movl %eax, %ebx xorl %r13d, %r13d jmp .LBB3_22 .p2align 4, 0x90 .LBB3_25: # in Loop: Header=BB3_22 Depth=1 incq %r13 cmpq %r13, %rbx je .LBB3_26 .LBB3_22: # %.lr.ph.i65 # =>This Inner Loop Header: Depth=1 movss (%r14,%r13,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r13d, %eax cltd idivl %ebp testq %r13, %r13 je .LBB3_25 # %bb.23: # %.lr.ph.i65 # in Loop: Header=BB3_22 Depth=1 testl %edx, %edx jne .LBB3_25 # %bb.24: # in Loop: Header=BB3_22 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_25 .LBB3_26: # %_Z8printMatiiPf.exit71 movl numARows(%rip), %eax testl %eax, %eax jle .LBB3_36 # %bb.27: # %.preheader.lr.ph.i movl numCColumns(%rip), %ecx movl numAColumns(%rip), %edx movslq numBColumns(%rip), %rsi movslq %ecx, %rdi movslq %edx, %r8 shlq $2, %r8 shlq $2, %rsi xorl %r9d, %r9d movq 56(%rsp), %r10 # 8-byte Reload jmp .LBB3_28 .p2align 4, 0x90 .LBB3_35: # %._crit_edge30.i # in Loop: Header=BB3_28 Depth=1 incq %r9 addq %r8, %r10 cmpq %rax, %r9 je .LBB3_36 .LBB3_28: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB3_30 Depth 2 # Child Loop BB3_32 Depth 3 testl %edx, %edx jle .LBB3_35 # %bb.29: # %.lr.ph29.i # in Loop: Header=BB3_28 Depth=1 movq %r9, %r11 imulq %rdi, %r11 leaq (%r12,%r11,4), %r11 movq 64(%rsp), %r13 # 8-byte Reload xorl %ebp, %ebp jmp .LBB3_30 .p2align 4, 0x90 .LBB3_34: # %._crit_edge.i # in Loop: Header=BB3_30 Depth=2 incq %rbp addq $4, %r13 cmpq %rdx, %rbp je .LBB3_35 .LBB3_30: # Parent Loop BB3_28 Depth=1 # => This Loop Header: Depth=2 # Child Loop BB3_32 Depth 3 movl $0, (%r11,%rbp,4) testl %ecx, %ecx jle .LBB3_34 # %bb.31: # %.lr.ph.i72 # in Loop: Header=BB3_30 Depth=2 xorps %xmm0, %xmm0 movq %r13, %rbx xorl %r15d, %r15d .p2align 4, 0x90 .LBB3_32: # Parent Loop BB3_28 Depth=1 # Parent Loop BB3_30 Depth=2 # => This Inner Loop Header: Depth=3 movss (%r10,%r15,4), %xmm1 # xmm1 = mem[0],zero,zero,zero mulss (%rbx), %xmm1 addss %xmm1, %xmm0 incq %r15 addq %rsi, %rbx cmpq %r15, %rcx jne .LBB3_32 # %bb.33: # %._crit_edge.i.loopexit # in Loop: Header=BB3_30 Depth=2 movss %xmm0, (%r11,%rbp,4) jmp .LBB3_34 .LBB3_36: # %_Z17matMultiplyOnHostPfS_S_iiiiii.exit movl $.Lstr.3, %edi callq puts@PLT movl numCColumns(%rip), %ebp movl numCRows(%rip), %eax imull %ebp, %eax testl %eax, %eax jle .LBB3_42 # %bb.37: # %.lr.ph.preheader.i78 movl %eax, %ebx xorl %r13d, %r13d jmp .LBB3_38 .p2align 4, 0x90 .LBB3_41: # in Loop: Header=BB3_38 Depth=1 incq %r13 cmpq %r13, %rbx je .LBB3_42 .LBB3_38: # %.lr.ph.i80 # =>This Inner Loop Header: Depth=1 movss (%r12,%r13,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str, %edi movb $1, %al callq printf movl %r13d, %eax cltd idivl %ebp testq %r13, %r13 je .LBB3_41 # %bb.39: # %.lr.ph.i80 # in Loop: Header=BB3_38 Depth=1 testl %edx, %edx jne .LBB3_41 # %bb.40: # in Loop: Header=BB3_38 Depth=1 movl $10, %edi callq putchar@PLT jmp .LBB3_41 .LBB3_42: # %_Z8printMatiiPf.exit86 movl $.Lstr.4, %edi callq puts@PLT movl numCColumns(%rip), %ecx movl numCRows(%rip), %eax imull %ecx, %eax testl %eax, %eax movq 56(%rsp), %rbx # 8-byte Reload movq %r14, %r15 movq 64(%rsp), %r14 # 8-byte Reload jle .LBB3_47 # %bb.43: # %.lr.ph101.preheader movl %eax, %edx xorl %eax, %eax .p2align 4, 0x90 .LBB3_44: # %.lr.ph101 # =>This Inner Loop Header: Depth=1 movss (%r12,%rax,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss (%r15,%rax,4), %xmm1 # xmm1 = mem[0],zero,zero,zero ucomiss %xmm1, %xmm0 jne .LBB3_45 jp .LBB3_45 # %bb.46: # in Loop: Header=BB3_44 Depth=1 incq %rax cmpq %rax, %rdx jne .LBB3_44 jmp .LBB3_47 .LBB3_45: # kill: def $eax killed $eax killed $rax cltd idivl %ecx cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 movl $.L.str.10, %edi movl %eax, %esi movb $2, %al callq printf .LBB3_47: # %.loopexit movl numCColumns(%rip), %eax movl %eax, %esi shrl $31, %esi addl %eax, %esi sarl %esi incl %esi imull %esi, %esi movl $.L.str.11, %edi xorl %eax, %eax callq printf movl $.L.str.12, %edi movl $4, %esi xorl %eax, %eax callq printf movq 24(%rsp), %rdi callq hipFree movq 16(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree movq %rbx, %rdi callq free movq %r14, %rdi callq free movq %r15, %rdi callq free movq %r12, %rdi callq free xorl %eax, %eax addq $216, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end3: .size main, .Lfunc_end3-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB4_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB4_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z20matrixMultiplySharedPfS_S_iiiiii, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end4: .size __hip_module_ctor, .Lfunc_end4-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB5_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB5_2: retq .Lfunc_end5: .size __hip_module_dtor, .Lfunc_end5-__hip_module_dtor .cfi_endproc # -- End function .type numARows,@object # @numARows .bss .globl numARows .p2align 2, 0x0 numARows: .long 0 # 0x0 .size numARows, 4 .type numAColumns,@object # @numAColumns .globl numAColumns .p2align 2, 0x0 numAColumns: .long 0 # 0x0 .size numAColumns, 4 .type numBRows,@object # @numBRows .globl numBRows .p2align 2, 0x0 numBRows: .long 0 # 0x0 .size numBRows, 4 .type numBColumns,@object # @numBColumns .globl numBColumns .p2align 2, 0x0 numBColumns: .long 0 # 0x0 .size numBColumns, 4 .type numCRows,@object # @numCRows .globl numCRows .p2align 2, 0x0 numCRows: .long 0 # 0x0 .size numCRows, 4 .type numCColumns,@object # @numCColumns .globl numCColumns .p2align 2, 0x0 numCColumns: .long 0 # 0x0 .size numCColumns, 4 .type _Z20matrixMultiplySharedPfS_S_iiiiii,@object # @_Z20matrixMultiplySharedPfS_S_iiiiii .section .rodata,"a",@progbits .globl _Z20matrixMultiplySharedPfS_S_iiiiii .p2align 3, 0x0 _Z20matrixMultiplySharedPfS_S_iiiiii: .quad _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .size _Z20matrixMultiplySharedPfS_S_iiiiii, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%f " .size .L.str, 5 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "\nEnter Rows and Columns of A:" .size .L.str.2, 30 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "%d %d" .size .L.str.3, 6 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "\nEnter Rows and Columns of B:" .size .L.str.4, 30 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "Mismatch at Row = %d Col = %d hostComputed[] = %f --device[] %f\n" .size .L.str.10, 65 .type .L.str.11,@object # @.str.11 .L.str.11: .asciz "\nNumber of Blocks Created:%d \n" .size .L.str.11, 31 .type .L.str.12,@object # @.str.12 .L.str.12: .asciz "\nNumber of Threads Per Block: %d \n" .size .L.str.12, 35 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z20matrixMultiplySharedPfS_S_iiiiii" .size .L__unnamed_1, 37 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "\nMatrix A Values:" .size .Lstr, 18 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "\n\nMatrix B Values:" .size .Lstr.1, 19 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "\nMatrix C From Device" .size .Lstr.2, 22 .type .Lstr.3,@object # @str.3 .Lstr.3: .asciz "\nMatrix C From Host" .size .Lstr.3, 20 .type .Lstr.4,@object # @str.4 .Lstr.4: .asciz "\n" .size .Lstr.4, 2 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z35__device_stub__matrixMultiplySharedPfS_S_iiiiii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym numARows .addrsig_sym numAColumns .addrsig_sym numBRows .addrsig_sym numBColumns .addrsig_sym _Z20matrixMultiplySharedPfS_S_iiiiii .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
/*** This script is an example of usign CUDA Thrust library. ***/ #include <thrust/host_vector.h> #include <thrust/device_vector.h> #include <iostream> using namespace std; int main(void) { thrust::host_vector<int> v; v.push_back(1); v.push_back(2); v.push_back(3); v.push_back(4); for (int i = 0; i < v.size(); i++) cout << "v[" << i << "] == " << v[i] << endl; thrust::device_vector<int> v_gpu = v; v_gpu.push_back(5); for (int i = 0; i < v_gpu.size(); i++) std::cout << "v_gpu[" << i << "] == " << v_gpu[i] << std::endl; return 0; }
code for sm_80 Function : _ZN3cub17CUB_200700_800_NS6detail8for_each13static_kernelINS2_12policy_hub_t12policy_350_tElN6thrust20THRUST_200700_800_NS8cuda_cub6__fill7functorINS7_6detail15normal_iteratorINS7_10device_ptrIiEEEEiEEEEvT0_T1_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD.WIDE.U32 R2, R2, 0x200, RZ ; /* 0x0000020002027825 */ /* 0x001fca00078e00ff */ /*0050*/ IADD3 R4, P1, -R2.reuse, c[0x0][0x160], RZ ; /* 0x0000580002047a10 */ /* 0x040fe40007f3e1ff */ /*0060*/ IADD3 R0, P2, R2, R5, RZ ; /* 0x0000000502007210 */ /* 0x002fe40007f5e0ff */ /*0070*/ ISETP.GT.U32.AND P0, PT, R4, 0x1ff, PT ; /* 0x000001ff0400780c */ /* 0x000fe40003f04070 */ /*0080*/ IADD3.X R6, ~R3, c[0x0][0x164], RZ, P1, !PT ; /* 0x0000590003067a10 */ /* 0x000fe20000ffe5ff */ /*0090*/ IMAD.X R3, RZ, RZ, R3, P2 ; /* 0x000000ffff037224 */ /* 0x000fe200010e0603 */ /*00a0*/ LEA R2, P1, R0, c[0x0][0x168], 0x2 ; /* 0x00005a0000027a11 */ /* 0x000fe400078210ff */ /*00b0*/ ISETP.GT.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fc40003f04300 */ /*00c0*/ LEA.HI.X R3, R0, c[0x0][0x16c], R3, 0x2, P1 ; /* 0x00005b0000037a11 */ /* 0x000fd600008f1403 */ /*00d0*/ @P0 BRA 0x1a0 ; /* 0x000000c000000947 */ /* 0x000fea0003800000 */ /*00e0*/ ISETP.GT.U32.AND P0, PT, R4, R5, PT ; /* 0x000000050400720c */ /* 0x000fe40003f04070 */ /*00f0*/ SHF.R.S32.HI R6, RZ, 0x1f, R4 ; /* 0x0000001fff067819 */ /* 0x000fe40000011404 */ /*0100*/ IADD3 R0, R5, 0x100, RZ ; /* 0x0000010005007810 */ /* 0x000fe40007ffe0ff */ /*0110*/ ISETP.GT.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04300 */ /*0120*/ @P0 IMAD.MOV.U32 R7, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff070624 */ /* 0x000fca00078e00ff */ /*0130*/ @P0 STG.E [R2.64], R7 ; /* 0x0000000702000986 */ /* 0x0001e2000c101904 */ /*0140*/ ISETP.GT.U32.AND P0, PT, R4, R0, PT ; /* 0x000000000400720c */ /* 0x000fc80003f04070 */ /*0150*/ ISETP.GT.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04300 */ /*0160*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0170*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x001fca00078e00ff */ /*0180*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*0190*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01a0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x000fca00078e00ff */ /*01b0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe8000c101904 */ /*01c0*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*01d0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01e0*/ BRA 0x1e0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0200*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0210*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0220*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0230*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0240*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0250*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0260*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0270*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ .......... Function : _ZN3cub17CUB_200700_800_NS6detail8for_each13static_kernelINS2_12policy_hub_t12policy_350_tEmN6thrust20THRUST_200700_800_NS8cuda_cub20__uninitialized_fill7functorINS7_10device_ptrIiEEiEEEEvT0_T1_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD.WIDE.U32 R2, R2, 0x200, RZ ; /* 0x0000020002027825 */ /* 0x001fca00078e00ff */ /*0050*/ IADD3 R4, P1, -R2.reuse, c[0x0][0x160], RZ ; /* 0x0000580002047a10 */ /* 0x040fe40007f3e1ff */ /*0060*/ IADD3 R0, P2, R2, R5, RZ ; /* 0x0000000502007210 */ /* 0x002fe40007f5e0ff */ /*0070*/ ISETP.GT.U32.AND P0, PT, R4, 0x1ff, PT ; /* 0x000001ff0400780c */ /* 0x000fe40003f04070 */ /*0080*/ IADD3.X R6, ~R3, c[0x0][0x164], RZ, P1, !PT ; /* 0x0000590003067a10 */ /* 0x000fe20000ffe5ff */ /*0090*/ IMAD.X R3, RZ, RZ, R3, P2 ; /* 0x000000ffff037224 */ /* 0x000fe200010e0603 */ /*00a0*/ LEA R2, P1, R0, c[0x0][0x168], 0x2 ; /* 0x00005a0000027a11 */ /* 0x000fe400078210ff */ /*00b0*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fc40003f04100 */ /*00c0*/ LEA.HI.X R3, R0, c[0x0][0x16c], R3, 0x2, P1 ; /* 0x00005b0000037a11 */ /* 0x000fd600008f1403 */ /*00d0*/ @P0 BRA 0x1a0 ; /* 0x000000c000000947 */ /* 0x000fea0003800000 */ /*00e0*/ ISETP.GT.U32.AND P0, PT, R4, R5, PT ; /* 0x000000050400720c */ /* 0x000fe40003f04070 */ /*00f0*/ SHF.R.S32.HI R6, RZ, 0x1f, R4 ; /* 0x0000001fff067819 */ /* 0x000fe40000011404 */ /*0100*/ IADD3 R0, R5, 0x100, RZ ; /* 0x0000010005007810 */ /* 0x000fe40007ffe0ff */ /*0110*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0120*/ @P0 IMAD.MOV.U32 R7, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff070624 */ /* 0x000fca00078e00ff */ /*0130*/ @P0 STG.E [R2.64], R7 ; /* 0x0000000702000986 */ /* 0x0001e2000c101904 */ /*0140*/ ISETP.GT.U32.AND P0, PT, R4, R0, PT ; /* 0x000000000400720c */ /* 0x000fc80003f04070 */ /*0150*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0160*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0170*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x001fca00078e00ff */ /*0180*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*0190*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01a0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x000fca00078e00ff */ /*01b0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe8000c101904 */ /*01c0*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*01d0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01e0*/ BRA 0x1e0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0200*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0210*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0220*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0230*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0240*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0250*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0260*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0270*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ .......... Function : _ZN3cub17CUB_200700_800_NS6detail8for_each13static_kernelINS2_12policy_hub_t12policy_350_tEmN6thrust20THRUST_200700_800_NS8cuda_cub6__fill7functorINS7_6detail15normal_iteratorINS7_10device_ptrIiEEEEiEEEEvT0_T1_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD.WIDE.U32 R2, R2, 0x200, RZ ; /* 0x0000020002027825 */ /* 0x001fca00078e00ff */ /*0050*/ IADD3 R4, P1, -R2.reuse, c[0x0][0x160], RZ ; /* 0x0000580002047a10 */ /* 0x040fe40007f3e1ff */ /*0060*/ IADD3 R0, P2, R2, R5, RZ ; /* 0x0000000502007210 */ /* 0x002fe40007f5e0ff */ /*0070*/ ISETP.GT.U32.AND P0, PT, R4, 0x1ff, PT ; /* 0x000001ff0400780c */ /* 0x000fe40003f04070 */ /*0080*/ IADD3.X R6, ~R3, c[0x0][0x164], RZ, P1, !PT ; /* 0x0000590003067a10 */ /* 0x000fe20000ffe5ff */ /*0090*/ IMAD.X R3, RZ, RZ, R3, P2 ; /* 0x000000ffff037224 */ /* 0x000fe200010e0603 */ /*00a0*/ LEA R2, P1, R0, c[0x0][0x168], 0x2 ; /* 0x00005a0000027a11 */ /* 0x000fe400078210ff */ /*00b0*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fc40003f04100 */ /*00c0*/ LEA.HI.X R3, R0, c[0x0][0x16c], R3, 0x2, P1 ; /* 0x00005b0000037a11 */ /* 0x000fd600008f1403 */ /*00d0*/ @P0 BRA 0x1a0 ; /* 0x000000c000000947 */ /* 0x000fea0003800000 */ /*00e0*/ ISETP.GT.U32.AND P0, PT, R4, R5, PT ; /* 0x000000050400720c */ /* 0x000fe40003f04070 */ /*00f0*/ SHF.R.S32.HI R6, RZ, 0x1f, R4 ; /* 0x0000001fff067819 */ /* 0x000fe40000011404 */ /*0100*/ IADD3 R0, R5, 0x100, RZ ; /* 0x0000010005007810 */ /* 0x000fe40007ffe0ff */ /*0110*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0120*/ @P0 IMAD.MOV.U32 R7, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff070624 */ /* 0x000fca00078e00ff */ /*0130*/ @P0 STG.E [R2.64], R7 ; /* 0x0000000702000986 */ /* 0x0001e2000c101904 */ /*0140*/ ISETP.GT.U32.AND P0, PT, R4, R0, PT ; /* 0x000000000400720c */ /* 0x000fc80003f04070 */ /*0150*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0160*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0170*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x001fca00078e00ff */ /*0180*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*0190*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01a0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x000fca00078e00ff */ /*01b0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe8000c101904 */ /*01c0*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*01d0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01e0*/ BRA 0x1e0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0200*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0210*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0220*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0230*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0240*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0250*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0260*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0270*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ .......... Function : _ZN3cub17CUB_200700_800_NS11EmptyKernelIvEEvv .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0020*/ BRA 0x20; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0030*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0040*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0050*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0060*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0070*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0080*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0090*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
/*** This script is an example of usign CUDA Thrust library. ***/ #include <thrust/host_vector.h> #include <thrust/device_vector.h> #include <iostream> using namespace std; int main(void) { thrust::host_vector<int> v; v.push_back(1); v.push_back(2); v.push_back(3); v.push_back(4); for (int i = 0; i < v.size(); i++) cout << "v[" << i << "] == " << v[i] << endl; thrust::device_vector<int> v_gpu = v; v_gpu.push_back(5); for (int i = 0; i < v_gpu.size(); i++) std::cout << "v_gpu[" << i << "] == " << v_gpu[i] << std::endl; return 0; }
/*** This script is an example of usign CUDA Thrust library. ***/ #include <hip/hip_runtime.h> #include <thrust/host_vector.h> #include <thrust/device_vector.h> #include <iostream> using namespace std; int main(void) { thrust::host_vector<int> v; v.push_back(1); v.push_back(2); v.push_back(3); v.push_back(4); for (int i = 0; i < v.size(); i++) cout << "v[" << i << "] == " << v[i] << endl; thrust::device_vector<int> v_gpu = v; v_gpu.push_back(5); for (int i = 0; i < v_gpu.size(); i++) std::cout << "v_gpu[" << i << "] == " << v_gpu[i] << std::endl; return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
/*** This script is an example of usign CUDA Thrust library. ***/ #include <hip/hip_runtime.h> #include <thrust/host_vector.h> #include <thrust/device_vector.h> #include <iostream> using namespace std; int main(void) { thrust::host_vector<int> v; v.push_back(1); v.push_back(2); v.push_back(3); v.push_back(4); for (int i = 0; i < v.size(); i++) cout << "v[" << i << "] == " << v[i] << endl; thrust::device_vector<int> v_gpu = v; v_gpu.push_back(5); for (int i = 0; i < v_gpu.size(); i++) std::cout << "v_gpu[" << i << "] == " << v_gpu[i] << std::endl; return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,comdat .protected _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .globl _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .p2align 8 .type _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,@function _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_: s_load_b128 s[4:7], s[0:1], 0x18 s_lshl_b32 s2, s15, 8 s_waitcnt lgkmcnt(0) s_add_u32 s2, s2, s6 s_addc_u32 s3, 0, s7 s_sub_u32 s4, s4, s2 s_subb_u32 s5, s5, s3 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_i64_e64 s5, 0x100, s[4:5] s_and_b32 s5, s5, exec_lo s_cselect_b32 s4, s4, 0x100 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1) v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_cmpk_eq_i32 s4, 0x100 s_cselect_b32 s4, -1, 0 s_or_b32 s4, s4, vcc_lo s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s5, s4 s_cbranch_execz .LBB0_2 s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x8 s_load_b32 s6, s[0:1], 0x10 v_lshlrev_b32_e32 v0, 2, v0 s_lshl_b64 s[0:1], s[2:3], 2 s_waitcnt lgkmcnt(0) s_add_u32 s0, s4, s0 s_addc_u32 s1, s5, s1 v_add_co_u32 v0, s0, s0, v0 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v1, null, s1, 0, s0 v_mov_b32_e32 v2, s6 flat_store_b32 v[0:1], v2 .LBB0_2: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 40 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 3 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,comdat .Lfunc_end0: .size _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_, .Lfunc_end0-_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .section .AMDGPU.csdata,"",@progbits .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,comdat .protected _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .globl _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .p2align 8 .type _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,@function _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_: s_load_b128 s[4:7], s[0:1], 0x10 s_lshl_b32 s2, s15, 8 s_waitcnt lgkmcnt(0) s_add_u32 s2, s2, s6 s_addc_u32 s3, 0, s7 s_sub_u32 s4, s4, s2 s_subb_u32 s5, s5, s3 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_u64_e64 s5, 0x100, s[4:5] s_and_b32 s5, s5, exec_lo s_cselect_b32 s4, s4, 0x100 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1) v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_cmpk_eq_i32 s4, 0x100 s_cselect_b32 s4, -1, 0 s_or_b32 s4, s4, vcc_lo s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s5, s4 s_cbranch_execz .LBB1_2 s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x0 s_load_b32 s6, s[0:1], 0x8 v_lshlrev_b32_e32 v0, 2, v0 s_lshl_b64 s[0:1], s[2:3], 2 s_waitcnt lgkmcnt(0) s_add_u32 s0, s4, s0 s_addc_u32 s1, s5, s1 v_add_co_u32 v0, s0, s0, v0 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v1, null, s1, 0, s0 v_mov_b32_e32 v2, s6 flat_store_b32 v[0:1], v2 .LBB1_2: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 32 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 3 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,comdat .Lfunc_end1: .size _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_, .Lfunc_end1-_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .section .AMDGPU.csdata,"",@progbits .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,comdat .protected _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .globl _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .p2align 8 .type _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,@function _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_: s_load_b128 s[4:7], s[0:1], 0x18 s_lshl_b32 s2, s15, 8 s_waitcnt lgkmcnt(0) s_add_u32 s2, s2, s6 s_addc_u32 s3, 0, s7 s_sub_u32 s4, s4, s2 s_subb_u32 s5, s5, s3 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_i64_e64 s5, 0x100, s[4:5] s_and_b32 s5, s5, exec_lo s_cselect_b32 s4, s4, 0x100 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1) v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_cmpk_eq_i32 s4, 0x100 s_cselect_b32 s4, -1, 0 s_or_b32 s4, s4, vcc_lo s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s5, s4 s_cbranch_execz .LBB2_2 s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x8 s_load_b32 s6, s[0:1], 0x10 v_lshlrev_b32_e32 v0, 2, v0 s_lshl_b64 s[0:1], s[2:3], 2 s_waitcnt lgkmcnt(0) s_add_u32 s0, s4, s0 s_addc_u32 s1, s5, s1 v_add_co_u32 v0, s0, s0, v0 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v1, null, s1, 0, s0 v_mov_b32_e32 v2, s6 flat_store_b32 v[0:1], v2 .LBB2_2: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 40 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 3 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,comdat .Lfunc_end2: .size _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_, .Lfunc_end2-_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 24 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value - .offset: 32 .size: 8 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 40 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 256 .name: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 3 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 - .args: - .offset: 0 .size: 16 .value_kind: by_value - .offset: 16 .size: 8 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 32 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 256 .name: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 3 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 - .args: - .offset: 0 .size: 24 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value - .offset: 32 .size: 8 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 40 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 256 .name: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 3 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _ZN3cub17CUB_200700_800_NS6detail8for_each13static_kernelINS2_12policy_hub_t12policy_350_tElN6thrust20THRUST_200700_800_NS8cuda_cub6__fill7functorINS7_6detail15normal_iteratorINS7_10device_ptrIiEEEEiEEEEvT0_T1_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD.WIDE.U32 R2, R2, 0x200, RZ ; /* 0x0000020002027825 */ /* 0x001fca00078e00ff */ /*0050*/ IADD3 R4, P1, -R2.reuse, c[0x0][0x160], RZ ; /* 0x0000580002047a10 */ /* 0x040fe40007f3e1ff */ /*0060*/ IADD3 R0, P2, R2, R5, RZ ; /* 0x0000000502007210 */ /* 0x002fe40007f5e0ff */ /*0070*/ ISETP.GT.U32.AND P0, PT, R4, 0x1ff, PT ; /* 0x000001ff0400780c */ /* 0x000fe40003f04070 */ /*0080*/ IADD3.X R6, ~R3, c[0x0][0x164], RZ, P1, !PT ; /* 0x0000590003067a10 */ /* 0x000fe20000ffe5ff */ /*0090*/ IMAD.X R3, RZ, RZ, R3, P2 ; /* 0x000000ffff037224 */ /* 0x000fe200010e0603 */ /*00a0*/ LEA R2, P1, R0, c[0x0][0x168], 0x2 ; /* 0x00005a0000027a11 */ /* 0x000fe400078210ff */ /*00b0*/ ISETP.GT.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fc40003f04300 */ /*00c0*/ LEA.HI.X R3, R0, c[0x0][0x16c], R3, 0x2, P1 ; /* 0x00005b0000037a11 */ /* 0x000fd600008f1403 */ /*00d0*/ @P0 BRA 0x1a0 ; /* 0x000000c000000947 */ /* 0x000fea0003800000 */ /*00e0*/ ISETP.GT.U32.AND P0, PT, R4, R5, PT ; /* 0x000000050400720c */ /* 0x000fe40003f04070 */ /*00f0*/ SHF.R.S32.HI R6, RZ, 0x1f, R4 ; /* 0x0000001fff067819 */ /* 0x000fe40000011404 */ /*0100*/ IADD3 R0, R5, 0x100, RZ ; /* 0x0000010005007810 */ /* 0x000fe40007ffe0ff */ /*0110*/ ISETP.GT.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04300 */ /*0120*/ @P0 IMAD.MOV.U32 R7, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff070624 */ /* 0x000fca00078e00ff */ /*0130*/ @P0 STG.E [R2.64], R7 ; /* 0x0000000702000986 */ /* 0x0001e2000c101904 */ /*0140*/ ISETP.GT.U32.AND P0, PT, R4, R0, PT ; /* 0x000000000400720c */ /* 0x000fc80003f04070 */ /*0150*/ ISETP.GT.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04300 */ /*0160*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0170*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x001fca00078e00ff */ /*0180*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*0190*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01a0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x000fca00078e00ff */ /*01b0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe8000c101904 */ /*01c0*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*01d0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01e0*/ BRA 0x1e0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0200*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0210*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0220*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0230*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0240*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0250*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0260*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0270*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ .......... Function : _ZN3cub17CUB_200700_800_NS6detail8for_each13static_kernelINS2_12policy_hub_t12policy_350_tEmN6thrust20THRUST_200700_800_NS8cuda_cub20__uninitialized_fill7functorINS7_10device_ptrIiEEiEEEEvT0_T1_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD.WIDE.U32 R2, R2, 0x200, RZ ; /* 0x0000020002027825 */ /* 0x001fca00078e00ff */ /*0050*/ IADD3 R4, P1, -R2.reuse, c[0x0][0x160], RZ ; /* 0x0000580002047a10 */ /* 0x040fe40007f3e1ff */ /*0060*/ IADD3 R0, P2, R2, R5, RZ ; /* 0x0000000502007210 */ /* 0x002fe40007f5e0ff */ /*0070*/ ISETP.GT.U32.AND P0, PT, R4, 0x1ff, PT ; /* 0x000001ff0400780c */ /* 0x000fe40003f04070 */ /*0080*/ IADD3.X R6, ~R3, c[0x0][0x164], RZ, P1, !PT ; /* 0x0000590003067a10 */ /* 0x000fe20000ffe5ff */ /*0090*/ IMAD.X R3, RZ, RZ, R3, P2 ; /* 0x000000ffff037224 */ /* 0x000fe200010e0603 */ /*00a0*/ LEA R2, P1, R0, c[0x0][0x168], 0x2 ; /* 0x00005a0000027a11 */ /* 0x000fe400078210ff */ /*00b0*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fc40003f04100 */ /*00c0*/ LEA.HI.X R3, R0, c[0x0][0x16c], R3, 0x2, P1 ; /* 0x00005b0000037a11 */ /* 0x000fd600008f1403 */ /*00d0*/ @P0 BRA 0x1a0 ; /* 0x000000c000000947 */ /* 0x000fea0003800000 */ /*00e0*/ ISETP.GT.U32.AND P0, PT, R4, R5, PT ; /* 0x000000050400720c */ /* 0x000fe40003f04070 */ /*00f0*/ SHF.R.S32.HI R6, RZ, 0x1f, R4 ; /* 0x0000001fff067819 */ /* 0x000fe40000011404 */ /*0100*/ IADD3 R0, R5, 0x100, RZ ; /* 0x0000010005007810 */ /* 0x000fe40007ffe0ff */ /*0110*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0120*/ @P0 IMAD.MOV.U32 R7, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff070624 */ /* 0x000fca00078e00ff */ /*0130*/ @P0 STG.E [R2.64], R7 ; /* 0x0000000702000986 */ /* 0x0001e2000c101904 */ /*0140*/ ISETP.GT.U32.AND P0, PT, R4, R0, PT ; /* 0x000000000400720c */ /* 0x000fc80003f04070 */ /*0150*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0160*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0170*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x001fca00078e00ff */ /*0180*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*0190*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01a0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x000fca00078e00ff */ /*01b0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe8000c101904 */ /*01c0*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*01d0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01e0*/ BRA 0x1e0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0200*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0210*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0220*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0230*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0240*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0250*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0260*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0270*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ .......... Function : _ZN3cub17CUB_200700_800_NS6detail8for_each13static_kernelINS2_12policy_hub_t12policy_350_tEmN6thrust20THRUST_200700_800_NS8cuda_cub6__fill7functorINS7_6detail15normal_iteratorINS7_10device_ptrIiEEEEiEEEEvT0_T1_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0040*/ IMAD.WIDE.U32 R2, R2, 0x200, RZ ; /* 0x0000020002027825 */ /* 0x001fca00078e00ff */ /*0050*/ IADD3 R4, P1, -R2.reuse, c[0x0][0x160], RZ ; /* 0x0000580002047a10 */ /* 0x040fe40007f3e1ff */ /*0060*/ IADD3 R0, P2, R2, R5, RZ ; /* 0x0000000502007210 */ /* 0x002fe40007f5e0ff */ /*0070*/ ISETP.GT.U32.AND P0, PT, R4, 0x1ff, PT ; /* 0x000001ff0400780c */ /* 0x000fe40003f04070 */ /*0080*/ IADD3.X R6, ~R3, c[0x0][0x164], RZ, P1, !PT ; /* 0x0000590003067a10 */ /* 0x000fe20000ffe5ff */ /*0090*/ IMAD.X R3, RZ, RZ, R3, P2 ; /* 0x000000ffff037224 */ /* 0x000fe200010e0603 */ /*00a0*/ LEA R2, P1, R0, c[0x0][0x168], 0x2 ; /* 0x00005a0000027a11 */ /* 0x000fe400078210ff */ /*00b0*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fc40003f04100 */ /*00c0*/ LEA.HI.X R3, R0, c[0x0][0x16c], R3, 0x2, P1 ; /* 0x00005b0000037a11 */ /* 0x000fd600008f1403 */ /*00d0*/ @P0 BRA 0x1a0 ; /* 0x000000c000000947 */ /* 0x000fea0003800000 */ /*00e0*/ ISETP.GT.U32.AND P0, PT, R4, R5, PT ; /* 0x000000050400720c */ /* 0x000fe40003f04070 */ /*00f0*/ SHF.R.S32.HI R6, RZ, 0x1f, R4 ; /* 0x0000001fff067819 */ /* 0x000fe40000011404 */ /*0100*/ IADD3 R0, R5, 0x100, RZ ; /* 0x0000010005007810 */ /* 0x000fe40007ffe0ff */ /*0110*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0120*/ @P0 IMAD.MOV.U32 R7, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff070624 */ /* 0x000fca00078e00ff */ /*0130*/ @P0 STG.E [R2.64], R7 ; /* 0x0000000702000986 */ /* 0x0001e2000c101904 */ /*0140*/ ISETP.GT.U32.AND P0, PT, R4, R0, PT ; /* 0x000000000400720c */ /* 0x000fc80003f04070 */ /*0150*/ ISETP.GT.U32.AND.EX P0, PT, R6, RZ, PT, P0 ; /* 0x000000ff0600720c */ /* 0x000fda0003f04100 */ /*0160*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0170*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x001fca00078e00ff */ /*0180*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*0190*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01a0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff057624 */ /* 0x000fca00078e00ff */ /*01b0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe8000c101904 */ /*01c0*/ STG.E [R2.64+0x400], R5 ; /* 0x0004000502007986 */ /* 0x000fe2000c101904 */ /*01d0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*01e0*/ BRA 0x1e0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*01f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0200*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0210*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0220*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0230*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0240*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0250*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0260*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0270*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ .......... Function : _ZN3cub17CUB_200700_800_NS11EmptyKernelIvEEvv .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0020*/ BRA 0x20; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0030*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0040*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0050*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0060*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0070*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0080*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0090*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*00f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,comdat .protected _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .globl _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .p2align 8 .type _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,@function _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_: s_load_b128 s[4:7], s[0:1], 0x18 s_lshl_b32 s2, s15, 8 s_waitcnt lgkmcnt(0) s_add_u32 s2, s2, s6 s_addc_u32 s3, 0, s7 s_sub_u32 s4, s4, s2 s_subb_u32 s5, s5, s3 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_i64_e64 s5, 0x100, s[4:5] s_and_b32 s5, s5, exec_lo s_cselect_b32 s4, s4, 0x100 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1) v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_cmpk_eq_i32 s4, 0x100 s_cselect_b32 s4, -1, 0 s_or_b32 s4, s4, vcc_lo s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s5, s4 s_cbranch_execz .LBB0_2 s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x8 s_load_b32 s6, s[0:1], 0x10 v_lshlrev_b32_e32 v0, 2, v0 s_lshl_b64 s[0:1], s[2:3], 2 s_waitcnt lgkmcnt(0) s_add_u32 s0, s4, s0 s_addc_u32 s1, s5, s1 v_add_co_u32 v0, s0, s0, v0 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v1, null, s1, 0, s0 v_mov_b32_e32 v2, s6 flat_store_b32 v[0:1], v2 .LBB0_2: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 40 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 3 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_,comdat .Lfunc_end0: .size _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_, .Lfunc_end0-_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .section .AMDGPU.csdata,"",@progbits .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,comdat .protected _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .globl _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .p2align 8 .type _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,@function _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_: s_load_b128 s[4:7], s[0:1], 0x10 s_lshl_b32 s2, s15, 8 s_waitcnt lgkmcnt(0) s_add_u32 s2, s2, s6 s_addc_u32 s3, 0, s7 s_sub_u32 s4, s4, s2 s_subb_u32 s5, s5, s3 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_u64_e64 s5, 0x100, s[4:5] s_and_b32 s5, s5, exec_lo s_cselect_b32 s4, s4, 0x100 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1) v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_cmpk_eq_i32 s4, 0x100 s_cselect_b32 s4, -1, 0 s_or_b32 s4, s4, vcc_lo s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s5, s4 s_cbranch_execz .LBB1_2 s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x0 s_load_b32 s6, s[0:1], 0x8 v_lshlrev_b32_e32 v0, 2, v0 s_lshl_b64 s[0:1], s[2:3], 2 s_waitcnt lgkmcnt(0) s_add_u32 s0, s4, s0 s_addc_u32 s1, s5, s1 v_add_co_u32 v0, s0, s0, v0 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v1, null, s1, 0, s0 v_mov_b32_e32 v2, s6 flat_store_b32 v[0:1], v2 .LBB1_2: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 32 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 3 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_,comdat .Lfunc_end1: .size _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_, .Lfunc_end1-_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .section .AMDGPU.csdata,"",@progbits .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,comdat .protected _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .globl _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .p2align 8 .type _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,@function _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_: s_load_b128 s[4:7], s[0:1], 0x18 s_lshl_b32 s2, s15, 8 s_waitcnt lgkmcnt(0) s_add_u32 s2, s2, s6 s_addc_u32 s3, 0, s7 s_sub_u32 s4, s4, s2 s_subb_u32 s5, s5, s3 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_i64_e64 s5, 0x100, s[4:5] s_and_b32 s5, s5, exec_lo s_cselect_b32 s4, s4, 0x100 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(SALU_CYCLE_1) v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_cmpk_eq_i32 s4, 0x100 s_cselect_b32 s4, -1, 0 s_or_b32 s4, s4, vcc_lo s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s5, s4 s_cbranch_execz .LBB2_2 s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x8 s_load_b32 s6, s[0:1], 0x10 v_lshlrev_b32_e32 v0, 2, v0 s_lshl_b64 s[0:1], s[2:3], 2 s_waitcnt lgkmcnt(0) s_add_u32 s0, s4, s0 s_addc_u32 s1, s5, s1 v_add_co_u32 v0, s0, s0, v0 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v1, null, s1, 0, s0 v_mov_b32_e32 v2, s6 flat_store_b32 v[0:1], v2 .LBB2_2: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 40 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 3 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .section .text._ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,"axG",@progbits,_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_,comdat .Lfunc_end2: .size _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_, .Lfunc_end2-_ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 24 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value - .offset: 32 .size: 8 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 40 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 256 .name: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorImNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_miEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUlmE_NS3_21always_true_predicateEEExLj1EEEvSG_SL_SL_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 3 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 - .args: - .offset: 0 .size: 16 .value_kind: by_value - .offset: 16 .size: 8 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 32 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 256 .name: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_20__uninitialized_fill7functorINS_10device_ptrIiEEiEEmLj1EEEvT0_T1_S9_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 3 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 - .args: - .offset: 0 .size: 24 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value - .offset: 32 .size: 8 .value_kind: by_value .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 40 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 256 .name: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _ZN6thrust11hip_rocprim14__parallel_for6kernelILj256ENS0_11__transform17unary_transform_fINS_17counting_iteratorIlNS_11use_defaultES6_S6_EENS_6detail15normal_iteratorINS_10device_ptrIiEEEENS3_14no_stencil_tagEZNS0_6fill_nINS0_3tagESC_liEET0_RNS0_16execution_policyIT_EESG_T1_RKT2_EUllE_NS3_21always_true_predicateEEElLj1EEEvSG_SL_SL_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 3 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include <stdio.h> #include <iostream> #include <fstream> #include <string> #include <cuda_runtime.h> #include <cstdlib> #include <sstream> #include <iomanip> using namespace std; //input, filter and output considering maximum dimension possible double a[1000][1000], h[10][10], c[1000][1000]; /** * CUDA Kernel Device code * * Computes the 2D convolution of input b with filter h and output c */ __global__ void convolution2D(double *a_1, double *h_1, double *c_1, int rows_a, int columns_a, int rows_h, int columns_h) { int idx = blockDim.x * blockIdx.x + threadIdx.x; int idy = blockDim.y * blockIdx.y + threadIdx.y; double sum; if((idx < (columns_h+columns_a - 1)) && (idy < (rows_h + rows_a - 1))) { sum = 0; for(int k=0; k< rows_h; k++) { for(int j=0; j< columns_h; j++) { if ((idy - k) >= 0 && (idx - j) >= 0 && (idy - k) < rows_a && (idx - j) < columns_a) { sum += a_1[((idy - k)*columns_a) + (idx - j)] * h_1[k*columns_h + j]; } } } c_1[idy*(columns_a + columns_h - 1) + idx] = sum; __syncthreads(); } } /** * Host main routine */ int main(int argc, char** argv) { // Error code to check return values for CUDA calls cudaError_t err = cudaSuccess; int columns_a, columns_h, rows_a, rows_h, size_a, size_h; columns_a = columns_h = 0; size_a = size_h = rows_a = rows_h = 0; char *ip_file; string line; double input; ip_file = argv[1]; int i, j, k; ifstream file(ip_file); if(file.is_open()) { i=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { a[i][j] = input; size_a++; j++; } i++; rows_a++; } k=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { h[k][j] = input; j++; size_h++; } k++; rows_h++; } } file.close(); columns_a = size_a/rows_a; columns_h = size_h/rows_h; int op_size = ((rows_a+rows_h-1)*(columns_a+columns_h-1)); size_t size_ax = size_a*sizeof(double); size_t size_hx = size_h*sizeof(double); size_t size_cx = op_size*sizeof(double); // Allocate the host input vector a double *h_a; cudaMallocManaged(&h_a, size_ax); // Allocate the host input vector h double *h_h; cudaMallocManaged(&h_h, size_hx); // Allocate the host output vector c double *h_c; cudaMallocManaged(&h_c, size_cx); for (int i = 0; i < rows_a; i++) { for(int j=0;j< columns_a; j++) { h_a[i*columns_a + j] = a[i][j]; } } for (int i = 0; i < rows_h; i++) { for(int j=0;j< columns_h; j++) { h_h[i*columns_h + j] = h[i][j]; } } // Launch the CUDA Kernel dim3 blocksPerGrid(((rows_a + rows_h - 2) / 32) + 1, ((columns_h + columns_a - 2) / 32) + 1, 1); dim3 threadsPerBlock(32,32,1); convolution2D<<<blocksPerGrid, threadsPerBlock>>>(h_a, h_h, h_c, rows_a, columns_a, rows_h, columns_h); err = cudaGetLastError(); if (err != cudaSuccess) { fprintf(stderr, "Failed to launch convolution2D kernel (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaDeviceSynchronize(); if (err != cudaSuccess) { fprintf(stderr, "Failed to synchronize (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } ///////////////////////////////////////////////////////////////////////////////////////////////////////// for (int i = 0; i < rows_a + rows_h - 1; i++) { for (int j = 0; j < columns_a + columns_h - 1; j++) { cout << fixed << setprecision(3) << h_c[(i*(columns_h+columns_a-1))+j] << " "; } cout << endl; } // Free device global memory err = cudaFree(h_a); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector a (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaFree(h_h); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector h (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaFree(h_c); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector c (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } // Reset the device and exit // cudaDeviceReset causes the driver to clean up all state. While // not mandatory in normal operation, it is good practice. It is also // needed to ensure correct operation when the application is being // profiled. Calling cudaDeviceReset causes all profile data to be // flushed before the application exits err = cudaDeviceReset(); if (err != cudaSuccess) { fprintf(stderr, "Failed to deinitialize the device! error=%s\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } printf("Done\n"); return 0; }
code for sm_80 Function : _Z13convolution2DPdS_S_iiii .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R0, SR_CTAID.X ; /* 0x0000000000007919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR6, c[0x0][0x178] ; /* 0x00005e0000067ab9 */ /* 0x000fe40000000a00 */ /*0030*/ ULDC.64 UR8, c[0x0][0x180] ; /* 0x0000600000087ab9 */ /* 0x000fe20000000a00 */ /*0040*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e220000002100 */ /*0050*/ UIADD3 UR4, UR7, UR9, URZ ; /* 0x0000000907047290 */ /* 0x000fe4000fffe03f */ /*0060*/ UIADD3 UR5, UR6, UR8, URZ ; /* 0x0000000806057290 */ /* 0x000fe2000fffe03f */ /*0070*/ S2R R21, SR_CTAID.Y ; /* 0x0000000000157919 */ /* 0x000e620000002600 */ /*0080*/ UIADD3 UR4, UR4, -0x1, URZ ; /* 0xffffffff04047890 */ /* 0x000fe4000fffe03f */ /*0090*/ UIADD3 UR5, UR5, -0x1, URZ ; /* 0xffffffff05057890 */ /* 0x000fe2000fffe03f */ /*00a0*/ S2R R2, SR_TID.Y ; /* 0x0000000000027919 */ /* 0x000e620000002200 */ /*00b0*/ IMAD R0, R0, c[0x0][0x0], R3 ; /* 0x0000000000007a24 */ /* 0x001fca00078e0203 */ /*00c0*/ ISETP.GE.AND P0, PT, R0, UR4, PT ; /* 0x0000000400007c0c */ /* 0x000fe2000bf06270 */ /*00d0*/ IMAD R21, R21, c[0x0][0x4], R2 ; /* 0x0000010015157a24 */ /* 0x002fca00078e0202 */ /*00e0*/ ISETP.GE.OR P0, PT, R21, UR5, P0 ; /* 0x0000000515007c0c */ /* 0x000fda0008706670 */ /*00f0*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0100*/ IMAD.MOV.U32 R2, RZ, RZ, 0x1 ; /* 0x00000001ff027424 */ /* 0x000fe200078e00ff */ /*0110*/ ULDC.64 UR6, c[0x0][0x118] ; /* 0x0000460000067ab9 */ /* 0x000fc80000000a00 */ /*0120*/ ISETP.LE.AND P0, PT, R2, c[0x0][0x184], PT ; /* 0x0000610002007a0c */ /* 0x000fc80003f03270 */ /*0130*/ ISETP.GT.OR P0, PT, R2, c[0x0][0x180], !P0 ; /* 0x0000600002007a0c */ /* 0x000fe40004704670 */ /*0140*/ CS2R R2, SRZ ; /* 0x0000000000027805 */ /* 0x000fd6000001ff00 */ /*0150*/ @P0 BRA 0x7a0 ; /* 0x0000064000000947 */ /* 0x000fea0003800000 */ /*0160*/ IMAD.MOV.U32 R2, RZ, RZ, 0x1 ; /* 0x00000001ff027424 */ /* 0x000fe400078e00ff */ /*0170*/ IMAD.MOV.U32 R20, RZ, RZ, c[0x0][0x184] ; /* 0x00006100ff147624 */ /* 0x000fe400078e00ff */ /*0180*/ IMAD.MOV.U32 R22, RZ, RZ, RZ ; /* 0x000000ffff167224 */ /* 0x000fe200078e00ff */ /*0190*/ IADD3 R2, -R2, c[0x0][0x184], RZ ; /* 0x0000610002027a10 */ /* 0x000fe40007ffe1ff */ /*01a0*/ LOP3.LUT R20, R20, 0x3, RZ, 0xc0, !PT ; /* 0x0000000314147812 */ /* 0x000fe400078ec0ff */ /*01b0*/ ISETP.GE.U32.AND P3, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe40003f66070 */ /*01c0*/ CS2R R2, SRZ ; /* 0x0000000000027805 */ /* 0x000fe2000001ff00 */ /*01d0*/ IADD3 R23, -R20, c[0x0][0x184], RZ ; /* 0x0000610014177a10 */ /* 0x000fc40007ffe1ff */ /*01e0*/ IMAD.MOV.U32 R26, RZ, RZ, R22.reuse ; /* 0x000000ffff1a7224 */ /* 0x100fe200078e0016 */ /*01f0*/ ISETP.NE.AND P4, PT, R20, RZ, PT ; /* 0x000000ff1400720c */ /* 0x000fe20003f85270 */ /*0200*/ IMAD.IADD R24, R21, 0x1, -R22 ; /* 0x0000000115187824 */ /* 0x000fe200078e0a16 */ /*0210*/ IADD3 R22, R22, 0x1, RZ ; /* 0x0000000116167810 */ /* 0x000fe20007ffe0ff */ /*0220*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fc600078e00ff */ /*0230*/ ISETP.GE.AND P5, PT, R22, c[0x0][0x180], PT ; /* 0x0000600016007a0c */ /* 0x000fe20003fa6270 */ /*0240*/ @!P3 BRA 0x4f0 ; /* 0x000002a00000b947 */ /* 0x00ffee0003800000 */ /*0250*/ ISETP.GE.AND P6, PT, R24, RZ, PT ; /* 0x000000ff1800720c */ /* 0x000fe20003fc6270 */ /*0260*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fe400078e00ff */ /*0270*/ IMAD.MOV.U32 R27, RZ, RZ, R23 ; /* 0x000000ffff1b7224 */ /* 0x000fc600078e0017 */ /*0280*/ ISETP.LT.OR P0, PT, R0.reuse, R25, !P6 ; /* 0x000000190000720c */ /* 0x040fe20007701670 */ /*0290*/ IMAD.IADD R5, R0, 0x1, -R25 ; /* 0x0000000100057824 */ /* 0x000fe400078e0a19 */ /*02a0*/ IMAD.MOV.U32 R17, RZ, RZ, 0x8 ; /* 0x00000008ff117424 */ /* 0x000fe200078e00ff */ /*02b0*/ ISETP.GE.OR P0, PT, R24.reuse, c[0x0][0x178], P0 ; /* 0x00005e0018007a0c */ /* 0x040fe20000706670 */ /*02c0*/ IMAD R12, R24, c[0x0][0x17c], R5 ; /* 0x00005f00180c7a24 */ /* 0x000fe400078e0205 */ /*02d0*/ IMAD R16, R26, c[0x0][0x184], R25 ; /* 0x000061001a107a24 */ /* 0x000fe200078e0219 */ /*02e0*/ ISETP.GE.OR P2, PT, R5, c[0x0][0x17c], P0 ; /* 0x00005f0005007a0c */ /* 0x000fe20000746670 */ /*02f0*/ IMAD.WIDE R12, R12, R17, c[0x0][0x160] ; /* 0x000058000c0c7625 */ /* 0x000fc800078e0211 */ /*0300*/ IMAD.WIDE R16, R16, R17, c[0x0][0x168] ; /* 0x00005a0010107625 */ /* 0x000fd000078e0211 */ /*0310*/ @!P2 LDG.E.64 R6, [R12.64] ; /* 0x000000060c06a981 */ /* 0x000ea8000c1e1b00 */ /*0320*/ @!P2 LDG.E.64 R4, [R16.64] ; /* 0x000000061004a981 */ /* 0x000ea2000c1e1b00 */ /*0330*/ ISETP.LE.OR P0, PT, R0.reuse, R25, !P6 ; /* 0x000000190000720c */ /* 0x040fe40007703670 */ /*0340*/ IADD3 R8, R0, -0x1, -R25 ; /* 0xffffffff00087810 */ /* 0x000fe40007ffe819 */ /*0350*/ IADD3 R9, R25, 0x2, RZ ; /* 0x0000000219097810 */ /* 0x000fe40007ffe0ff */ /*0360*/ ISETP.GE.OR P0, PT, R24, c[0x0][0x178], P0 ; /* 0x00005e0018007a0c */ /* 0x000fc40000706670 */ /*0370*/ ISETP.LT.OR P1, PT, R0.reuse, R9, !P6 ; /* 0x000000090000720c */ /* 0x040fe20007721670 */ /*0380*/ IMAD.IADD R9, R0, 0x1, -R9 ; /* 0x0000000100097824 */ /* 0x000fe200078e0a09 */ /*0390*/ ISETP.GE.OR P0, PT, R8, c[0x0][0x17c], P0 ; /* 0x00005f0008007a0c */ /* 0x000fe40000706670 */ /*03a0*/ IADD3 R11, R25, 0x3, RZ ; /* 0x00000003190b7810 */ /* 0x000fe40007ffe0ff */ /*03b0*/ ISETP.GE.OR P1, PT, R24, c[0x0][0x178], P1 ; /* 0x00005e0018007a0c */ /* 0x000fc80000f26670 */ /*03c0*/ ISETP.GE.OR P1, PT, R9, c[0x0][0x17c], P1 ; /* 0x00005f0009007a0c */ /* 0x000fda0000f26670 */ /*03d0*/ @!P1 LDG.E.64 R8, [R16.64+0x10] ; /* 0x0000100610089981 */ /* 0x000ee2000c1e1b00 */ /*03e0*/ @!P2 DFMA R2, R6, R4, R2 ; /* 0x000000040602a22b */ /* 0x0070620000000002 */ /*03f0*/ ISETP.LT.OR P2, PT, R0.reuse, R11, !P6 ; /* 0x0000000b0000720c */ /* 0x040fe20007741670 */ /*0400*/ IMAD.IADD R11, R0, 0x1, -R11 ; /* 0x00000001000b7824 */ /* 0x000fe200078e0a0b */ /*0410*/ @!P0 LDG.E.64 R4, [R16.64+0x8] ; /* 0x0000080610048981 */ /* 0x001e64000c1e1b00 */ /*0420*/ ISETP.GE.OR P2, PT, R24, c[0x0][0x178], P2 ; /* 0x00005e0018007a0c */ /* 0x000fe40001746670 */ /*0430*/ @!P0 LDG.E.64 R6, [R12.64+-0x8] ; /* 0xfffff8060c068981 */ /* 0x000e64000c1e1b00 */ /*0440*/ ISETP.GE.OR P2, PT, R11, c[0x0][0x17c], P2 ; /* 0x00005f000b007a0c */ /* 0x000fc40001746670 */ /*0450*/ @!P1 LDG.E.64 R10, [R12.64+-0x10] ; /* 0xfffff0060c0a9981 */ /* 0x000ef6000c1e1b00 */ /*0460*/ @!P2 LDG.E.64 R14, [R16.64+0x18] ; /* 0x00001806100ea981 */ /* 0x000ea8000c1e1b00 */ /*0470*/ @!P2 LDG.E.64 R18, [R12.64+-0x18] ; /* 0xffffe8060c12a981 */ /* 0x000ea2000c1e1b00 */ /*0480*/ IADD3 R27, R27, -0x4, RZ ; /* 0xfffffffc1b1b7810 */ /* 0x000fc40007ffe0ff */ /*0490*/ IADD3 R25, R25, 0x4, RZ ; /* 0x0000000419197810 */ /* 0x000fe20007ffe0ff */ /*04a0*/ @!P0 DFMA R2, R4, R6, R2 ; /* 0x000000060402822b */ /* 0x002ee20000000002 */ /*04b0*/ ISETP.NE.AND P0, PT, R27, RZ, PT ; /* 0x000000ff1b00720c */ /* 0x000fca0003f05270 */ /*04c0*/ @!P1 DFMA R2, R8, R10, R2 ; /* 0x0000000a0802922b */ /* 0x008e8c0000000002 */ /*04d0*/ @!P2 DFMA R2, R14, R18, R2 ; /* 0x000000120e02a22b */ /* 0x0040640000000002 */ /*04e0*/ @P0 BRA 0x280 ; /* 0xfffffd9000000947 */ /* 0x000fea000383ffff */ /*04f0*/ @!P4 BRA 0x790 ; /* 0x000002900000c947 */ /* 0x000fea0003800000 */ /*0500*/ ISETP.GE.AND P0, PT, R24, RZ, PT ; /* 0x000000ff1800720c */ /* 0x000fe20003f06270 */ /*0510*/ IMAD.IADD R5, R0.reuse, 0x1, -R25.reuse ; /* 0x0000000100057824 */ /* 0x140fe200078e0a19 */ /*0520*/ BSSY B0, 0x610 ; /* 0x000000e000007945 */ /* 0x000fe20003800000 */ /*0530*/ IMAD.MOV.U32 R4, RZ, RZ, 0x8 ; /* 0x00000008ff047424 */ /* 0x000fe200078e00ff */ /*0540*/ ISETP.LT.OR P1, PT, R0, R25, !P0 ; /* 0x000000190000720c */ /* 0x000fe20004721670 */ /*0550*/ IMAD R6, R26, c[0x0][0x184], R25 ; /* 0x000061001a067a24 */ /* 0x000fe200078e0219 */ /*0560*/ ISETP.NE.AND P2, PT, R20, 0x1, PT ; /* 0x000000011400780c */ /* 0x000fe40003f45270 */ /*0570*/ ISETP.GE.OR P1, PT, R24, c[0x0][0x178], P1 ; /* 0x00005e0018007a0c */ /* 0x000fe20000f26670 */ /*0580*/ IMAD.WIDE R6, R6, R4, c[0x0][0x168] ; /* 0x00005a0006067625 */ /* 0x000fc600078e0204 */ /*0590*/ ISETP.GE.OR P1, PT, R5, c[0x0][0x17c], P1 ; /* 0x00005f0005007a0c */ /* 0x000fe20000f26670 */ /*05a0*/ IMAD R5, R24, c[0x0][0x17c], R5 ; /* 0x00005f0018057a24 */ /* 0x000fc800078e0205 */ /*05b0*/ IMAD.WIDE R4, R5, R4, c[0x0][0x160] ; /* 0x0000580005047625 */ /* 0x000fd000078e0204 */ /*05c0*/ @P1 BRA 0x600 ; /* 0x0000003000001947 */ /* 0x000fea0003800000 */ /*05d0*/ LDG.E.64 R8, [R4.64] ; /* 0x0000000604087981 */ /* 0x000ea8000c1e1b00 */ /*05e0*/ LDG.E.64 R10, [R6.64] ; /* 0x00000006060a7981 */ /* 0x000ea4000c1e1b00 */ /*05f0*/ DFMA R2, R8, R10, R2 ; /* 0x0000000a0802722b */ /* 0x00628c0000000002 */ /*0600*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0610*/ @!P2 BRA 0x790 ; /* 0x000001700000a947 */ /* 0x000fea0003800000 */ /*0620*/ ISETP.LE.OR P1, PT, R0.reuse, R25, !P0 ; /* 0x000000190000720c */ /* 0x040fe20004723670 */ /*0630*/ BSSY B0, 0x6d0 ; /* 0x0000009000007945 */ /* 0x000fe20003800000 */ /*0640*/ IADD3 R8, R0, -0x1, -R25 ; /* 0xffffffff00087810 */ /* 0x002fe40007ffe819 */ /*0650*/ ISETP.GE.OR P1, PT, R24, c[0x0][0x178], P1 ; /* 0x00005e0018007a0c */ /* 0x000fc40000f26670 */ /*0660*/ ISETP.NE.AND P2, PT, R20, 0x2, PT ; /* 0x000000021400780c */ /* 0x000fe40003f45270 */ /*0670*/ ISETP.GE.OR P1, PT, R8, c[0x0][0x17c], P1 ; /* 0x00005f0008007a0c */ /* 0x000fda0000f26670 */ /*0680*/ @P1 BRA 0x6c0 ; /* 0x0000003000001947 */ /* 0x000fea0003800000 */ /*0690*/ LDG.E.64 R8, [R6.64+0x8] ; /* 0x0000080606087981 */ /* 0x000ee8000c1e1b00 */ /*06a0*/ LDG.E.64 R10, [R4.64+-0x8] ; /* 0xfffff806040a7981 */ /* 0x000ee4000c1e1b00 */ /*06b0*/ DFMA R2, R8, R10, R2 ; /* 0x0000000a0802722b */ /* 0x00c28c0000000002 */ /*06c0*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*06d0*/ @!P2 BRA 0x790 ; /* 0x000000b00000a947 */ /* 0x000fea0003800000 */ /*06e0*/ IADD3 R25, R25, 0x2, RZ ; /* 0x0000000219197810 */ /* 0x000fe20007ffe0ff */ /*06f0*/ BSSY B0, 0x790 ; /* 0x0000009000007945 */ /* 0x000fe60003800000 */ /*0700*/ ISETP.LT.OR P0, PT, R0.reuse, R25, !P0 ; /* 0x000000190000720c */ /* 0x040fe20004701670 */ /*0710*/ IMAD.IADD R25, R0, 0x1, -R25 ; /* 0x0000000100197824 */ /* 0x000fc600078e0a19 */ /*0720*/ ISETP.GE.OR P0, PT, R24, c[0x0][0x178], P0 ; /* 0x00005e0018007a0c */ /* 0x000fc80000706670 */ /*0730*/ ISETP.GE.OR P0, PT, R25, c[0x0][0x17c], P0 ; /* 0x00005f0019007a0c */ /* 0x000fda0000706670 */ /*0740*/ @P0 BRA 0x780 ; /* 0x0000003000000947 */ /* 0x000fea0003800000 */ /*0750*/ LDG.E.64 R6, [R6.64+0x10] ; /* 0x0000100606067981 */ /* 0x000ee8000c1e1b00 */ /*0760*/ LDG.E.64 R4, [R4.64+-0x10] ; /* 0xfffff00604047981 */ /* 0x000ee4000c1e1b00 */ /*0770*/ DFMA R2, R6, R4, R2 ; /* 0x000000040602722b */ /* 0x00c4cc0000000002 */ /*0780*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0790*/ @!P5 BRA 0x1e0 ; /* 0xfffffa400000d947 */ /* 0x000fea000383ffff */ /*07a0*/ IMAD.MOV.U32 R5, RZ, RZ, 0x8 ; /* 0x00000008ff057424 */ /* 0x004fe400078e00ff */ /*07b0*/ IMAD R4, R21, UR4, R0 ; /* 0x0000000415047c24 */ /* 0x000fc8000f8e0200 */ /*07c0*/ IMAD.WIDE R4, R4, R5, c[0x0][0x170] ; /* 0x00005c0004047625 */ /* 0x000fca00078e0205 */ /*07d0*/ STG.E.64 [R4.64], R2 ; /* 0x0000000204007986 */ /* 0x00afe8000c101b06 */ /*07e0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*07f0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0800*/ BRA 0x800; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0810*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0820*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0830*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0840*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0850*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0860*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0870*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0880*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0890*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <stdio.h> #include <iostream> #include <fstream> #include <string> #include <cuda_runtime.h> #include <cstdlib> #include <sstream> #include <iomanip> using namespace std; //input, filter and output considering maximum dimension possible double a[1000][1000], h[10][10], c[1000][1000]; /** * CUDA Kernel Device code * * Computes the 2D convolution of input b with filter h and output c */ __global__ void convolution2D(double *a_1, double *h_1, double *c_1, int rows_a, int columns_a, int rows_h, int columns_h) { int idx = blockDim.x * blockIdx.x + threadIdx.x; int idy = blockDim.y * blockIdx.y + threadIdx.y; double sum; if((idx < (columns_h+columns_a - 1)) && (idy < (rows_h + rows_a - 1))) { sum = 0; for(int k=0; k< rows_h; k++) { for(int j=0; j< columns_h; j++) { if ((idy - k) >= 0 && (idx - j) >= 0 && (idy - k) < rows_a && (idx - j) < columns_a) { sum += a_1[((idy - k)*columns_a) + (idx - j)] * h_1[k*columns_h + j]; } } } c_1[idy*(columns_a + columns_h - 1) + idx] = sum; __syncthreads(); } } /** * Host main routine */ int main(int argc, char** argv) { // Error code to check return values for CUDA calls cudaError_t err = cudaSuccess; int columns_a, columns_h, rows_a, rows_h, size_a, size_h; columns_a = columns_h = 0; size_a = size_h = rows_a = rows_h = 0; char *ip_file; string line; double input; ip_file = argv[1]; int i, j, k; ifstream file(ip_file); if(file.is_open()) { i=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { a[i][j] = input; size_a++; j++; } i++; rows_a++; } k=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { h[k][j] = input; j++; size_h++; } k++; rows_h++; } } file.close(); columns_a = size_a/rows_a; columns_h = size_h/rows_h; int op_size = ((rows_a+rows_h-1)*(columns_a+columns_h-1)); size_t size_ax = size_a*sizeof(double); size_t size_hx = size_h*sizeof(double); size_t size_cx = op_size*sizeof(double); // Allocate the host input vector a double *h_a; cudaMallocManaged(&h_a, size_ax); // Allocate the host input vector h double *h_h; cudaMallocManaged(&h_h, size_hx); // Allocate the host output vector c double *h_c; cudaMallocManaged(&h_c, size_cx); for (int i = 0; i < rows_a; i++) { for(int j=0;j< columns_a; j++) { h_a[i*columns_a + j] = a[i][j]; } } for (int i = 0; i < rows_h; i++) { for(int j=0;j< columns_h; j++) { h_h[i*columns_h + j] = h[i][j]; } } // Launch the CUDA Kernel dim3 blocksPerGrid(((rows_a + rows_h - 2) / 32) + 1, ((columns_h + columns_a - 2) / 32) + 1, 1); dim3 threadsPerBlock(32,32,1); convolution2D<<<blocksPerGrid, threadsPerBlock>>>(h_a, h_h, h_c, rows_a, columns_a, rows_h, columns_h); err = cudaGetLastError(); if (err != cudaSuccess) { fprintf(stderr, "Failed to launch convolution2D kernel (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaDeviceSynchronize(); if (err != cudaSuccess) { fprintf(stderr, "Failed to synchronize (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } ///////////////////////////////////////////////////////////////////////////////////////////////////////// for (int i = 0; i < rows_a + rows_h - 1; i++) { for (int j = 0; j < columns_a + columns_h - 1; j++) { cout << fixed << setprecision(3) << h_c[(i*(columns_h+columns_a-1))+j] << " "; } cout << endl; } // Free device global memory err = cudaFree(h_a); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector a (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaFree(h_h); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector h (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaFree(h_c); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector c (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } // Reset the device and exit // cudaDeviceReset causes the driver to clean up all state. While // not mandatory in normal operation, it is good practice. It is also // needed to ensure correct operation when the application is being // profiled. Calling cudaDeviceReset causes all profile data to be // flushed before the application exits err = cudaDeviceReset(); if (err != cudaSuccess) { fprintf(stderr, "Failed to deinitialize the device! error=%s\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } printf("Done\n"); return 0; }
.file "tmpxft_001387e5_00000000-6_exec2.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB4043: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE4043: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z41__device_stub__Z13convolution2DPdS_S_iiiiPdS_S_iiii .type _Z41__device_stub__Z13convolution2DPdS_S_iiiiPdS_S_iiii, @function _Z41__device_stub__Z13convolution2DPdS_S_iiiiPdS_S_iiii: .LFB4065: .cfi_startproc endbr64 subq $184, %rsp .cfi_def_cfa_offset 192 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) movq %fs:40, %rax movq %rax, 168(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 20(%rsp), %rax movq %rax, 136(%rsp) leaq 16(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%rsp) leaq 192(%rsp), %rax movq %rax, 160(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 168(%rsp), %rax subq %fs:40, %rax jne .L8 addq $184, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 200 pushq 56(%rsp) .cfi_def_cfa_offset 208 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z13convolution2DPdS_S_iiii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 192 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE4065: .size _Z41__device_stub__Z13convolution2DPdS_S_iiiiPdS_S_iiii, .-_Z41__device_stub__Z13convolution2DPdS_S_iiiiPdS_S_iiii .globl _Z13convolution2DPdS_S_iiii .type _Z13convolution2DPdS_S_iiii, @function _Z13convolution2DPdS_S_iiii: .LFB4066: .cfi_startproc endbr64 subq $16, %rsp .cfi_def_cfa_offset 24 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 call _Z41__device_stub__Z13convolution2DPdS_S_iiiiPdS_S_iiii addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE4066: .size _Z13convolution2DPdS_S_iiii, .-_Z13convolution2DPdS_S_iiii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z13convolution2DPdS_S_iiii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB4068: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z13convolution2DPdS_S_iiii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE4068: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .text._ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED2Ev,"axG",@progbits,_ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED5Ev,comdat .align 2 .weak _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED2Ev .type _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED2Ev, @function _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED2Ev: .LFB4400: .cfi_startproc endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 movq %rdi, %rbx leaq 16+_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, (%rdi) movq 72(%rdi), %rdi leaq 88(%rbx), %rax cmpq %rax, %rdi je .L14 movq 88(%rbx), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L14: leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, (%rbx) leaq 56(%rbx), %rdi call _ZNSt6localeD1Ev@PLT popq %rbx .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE4400: .size _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED2Ev, .-_ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED2Ev .weak _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED1Ev .set _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED1Ev,_ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED2Ev .section .text._ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED0Ev,"axG",@progbits,_ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED5Ev,comdat .align 2 .weak _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED0Ev .type _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED0Ev, @function _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED0Ev: .LFB4402: .cfi_startproc endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 movq %rdi, %rbx leaq 16+_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, (%rdi) movq 72(%rdi), %rdi leaq 88(%rbx), %rax cmpq %rax, %rdi je .L17 movq 88(%rbx), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L17: leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, (%rbx) leaq 56(%rbx), %rdi call _ZNSt6localeD1Ev@PLT movl $104, %esi movq %rbx, %rdi call _ZdlPvm@PLT popq %rbx .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE4402: .size _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED0Ev, .-_ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED0Ev .section .text._ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag,"axG",@progbits,_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag,comdat .align 2 .weak _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag .type _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag, @function _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag: .LFB4495: .cfi_startproc endbr64 pushq %r12 .cfi_def_cfa_offset 16 .cfi_offset 12, -16 pushq %rbp .cfi_def_cfa_offset 24 .cfi_offset 6, -24 pushq %rbx .cfi_def_cfa_offset 32 .cfi_offset 3, -32 subq $16, %rsp .cfi_def_cfa_offset 48 movq %rdi, %rbx movq %rsi, %r12 movq %fs:40, %rax movq %rax, 8(%rsp) xorl %eax, %eax subq %rsi, %rdx movq %rdx, %rbp movq %rdx, (%rsp) cmpq $15, %rdx ja .L26 movq (%rdi), %rdi cmpq $1, %rdx jne .L22 movzbl (%rsi), %eax movb %al, (%rdi) .L23: movq (%rsp), %rax movq %rax, 8(%rbx) movq (%rbx), %rdx movb $0, (%rdx,%rax) movq 8(%rsp), %rax subq %fs:40, %rax jne .L27 addq $16, %rsp .cfi_remember_state .cfi_def_cfa_offset 32 popq %rbx .cfi_def_cfa_offset 24 popq %rbp .cfi_def_cfa_offset 16 popq %r12 .cfi_def_cfa_offset 8 ret .L26: .cfi_restore_state movq %rsp, %rsi movl $0, %edx call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE9_M_createERmm@PLT movq %rax, %rdi movq %rax, (%rbx) movq (%rsp), %rax movq %rax, 16(%rbx) .L21: movq %rbp, %rdx movq %r12, %rsi call memcpy@PLT jmp .L23 .L22: testq %rdx, %rdx je .L23 jmp .L21 .L27: call __stack_chk_fail@PLT .cfi_endproc .LFE4495: .size _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag, .-_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC1: .string "basic_string: construction from null is not valid" .align 8 .LC2: .string "Failed to launch convolution2D kernel (error code %s)!\n" .align 8 .LC3: .string "Failed to synchronize (error code %s)!\n" .section .rodata.str1.1 .LC4: .string " " .section .rodata.str1.8 .align 8 .LC5: .string "Failed to free device vector a (error code %s)!\n" .align 8 .LC6: .string "Failed to free device vector h (error code %s)!\n" .align 8 .LC7: .string "Failed to free device vector c (error code %s)!\n" .align 8 .LC8: .string "Failed to deinitialize the device! error=%s\n" .section .rodata.str1.1 .LC9: .string "Done\n" .text .globl main .type main, @function main: .LFB4039: .cfi_startproc .cfi_personality 0x9b,DW.ref.__gxx_personality_v0 .cfi_lsda 0x1b,.LLSDA4039 endbr64 pushq %rbp .cfi_def_cfa_offset 16 .cfi_offset 6, -16 movq %rsp, %rbp .cfi_def_cfa_register 6 pushq %r15 pushq %r14 pushq %r13 pushq %r12 pushq %rbx subq $1080, %rsp .cfi_offset 15, -24 .cfi_offset 14, -32 .cfi_offset 13, -40 .cfi_offset 12, -48 .cfi_offset 3, -56 movq %fs:40, %rax movq %rax, -56(%rbp) xorl %eax, %eax leaq -992(%rbp), %rax movq %rax, -1008(%rbp) movq $0, -1000(%rbp) movb $0, -992(%rbp) movq 8(%rsi), %rsi leaq -576(%rbp), %rdi movl $8, %edx .LEHB0: call _ZNSt14basic_ifstreamIcSt11char_traitsIcEEC1EPKcSt13_Ios_Openmode@PLT .LEHE0: leaq -456(%rbp), %rdi call _ZNKSt12__basic_fileIcE7is_openEv@PLT testb %al, %al je .L103 movq -576(%rbp), %rax movq -24(%rax), %rax movq -336(%rbp,%rax), %rbx testq %rbx, %rbx je .L30 leaq a(%rip), %rax movq %rax, -1088(%rbp) movl $0, -1080(%rbp) movl $0, -1092(%rbp) jmp .L34 .L141: movq -1088(%rbp), %rbx leaq -1064(%rbp), %r12 jmp .L49 .L134: movq (%rax), %rdx movq -24(%rdx), %rdx testb $5, 32(%rax,%rdx) jne .L133 movsd -1064(%rbp), %xmm0 movsd %xmm0, (%rbx) addl $1, -1080(%rbp) addq $8, %rbx .L49: leaq -976(%rbp), %rdi movq %r12, %rsi .LEHB1: call _ZNSi10_M_extractIdEERSiRT_@PLT .LEHE1: jmp .L134 .L133: addl $1, -1092(%rbp) leaq 24+_ZTVNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -976(%rbp) leaq 80(%rax), %rax movq %rax, -848(%rbp) leaq -40(%rax), %rax movq %rax, -960(%rbp) leaq 16+_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -952(%rbp) movq -880(%rbp), %rdi leaq -864(%rbp), %rax cmpq %rax, %rdi je .L33 movq -864(%rbp), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L33: leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, -952(%rbp) leaq -896(%rbp), %rdi call _ZNSt6localeD1Ev@PLT movq %r14, -976(%rbp) movq -24(%r14), %rax movq 48+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -976(%rbp,%rax) movq %r15, -960(%rbp) movq -24(%r15), %rax movq 40+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -960(%rbp,%rax) movq %r13, -976(%rbp) movq -24(%r13), %rax movq 24+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rbx movq %rbx, -976(%rbp,%rax) movq $0, -968(%rbp) leaq 16+_ZTVSt9basic_iosIcSt11char_traitsIcEE(%rip), %rax movq %rax, -848(%rbp) leaq -848(%rbp), %rdi call _ZNSt8ios_baseD2Ev@PLT movq -576(%rbp), %rax movq -24(%rax), %rax movq -336(%rbp,%rax), %rbx addq $8000, -1088(%rbp) testq %rbx, %rbx je .L30 .L34: cmpb $0, 56(%rbx) je .L36 movzbl 67(%rbx), %edx .L37: movsbl %dl, %edx leaq -1008(%rbp), %rsi leaq -576(%rbp), %rdi .LEHB2: call _ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EES4_@PLT jmp .L135 .L30: movq -56(%rbp), %rax subq %fs:40, %rax jne .L136 call _ZSt16__throw_bad_castv@PLT .L136: call __stack_chk_fail@PLT .L36: movq %rbx, %rdi call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq (%rbx), %rax movl $10, %esi movq %rbx, %rdi call *48(%rax) .LEHE2: movl %eax, %edx jmp .L37 .L135: movq (%rax), %rdx movq -24(%rdx), %rdx testb $5, 32(%rax,%rdx) jne .L38 cmpq $0, -1000(%rbp) je .L38 leaq -976(%rbp), %rbx leaq -848(%rbp), %rdi call _ZNSt8ios_baseC2Ev@PLT leaq 16+_ZTVSt9basic_iosIcSt11char_traitsIcEE(%rip), %rax movq %rax, -848(%rbp) movq $0, -632(%rbp) movb $0, -624(%rbp) movb $0, -623(%rbp) movq $0, -616(%rbp) movq $0, -608(%rbp) movq $0, -600(%rbp) movq $0, -592(%rbp) movq 16+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %r13 movq %r13, -976(%rbp) movq -24(%r13), %rax movq 24+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -976(%rbp,%rax) movq $0, -968(%rbp) movq -976(%rbp), %rax addq -24(%rax), %rbx movq %rbx, %rdi movl $0, %esi .LEHB3: call _ZNSt9basic_iosIcSt11char_traitsIcEE4initEPSt15basic_streambufIcS1_E@PLT .LEHE3: movq 32+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %r15 movq %r15, -960(%rbp) movq -24(%r15), %rax movq 40+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rbx movq %rbx, -960(%rbp,%rax) movq -960(%rbp), %rax movq -24(%rax), %rax leaq -960(%rbp,%rax), %rdi movl $0, %esi .LEHB4: call _ZNSt9basic_iosIcSt11char_traitsIcEE4initEPSt15basic_streambufIcS1_E@PLT .LEHE4: movq 8+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %r14 movq %r14, -976(%rbp) movq -24(%r14), %rax movq 48+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -976(%rbp,%rax) leaq 24+_ZTVNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -976(%rbp) leaq 80(%rax), %rax movq %rax, -848(%rbp) leaq -40(%rax), %rax movq %rax, -960(%rbp) leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, -952(%rbp) movq $0, -944(%rbp) movq $0, -936(%rbp) movq $0, -928(%rbp) movq $0, -920(%rbp) movq $0, -912(%rbp) movq $0, -904(%rbp) leaq -896(%rbp), %rdi call _ZNSt6localeC1Ev@PLT leaq 16+_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -952(%rbp) movl $0, -888(%rbp) movq -1000(%rbp), %rdx movq -1008(%rbp), %rsi leaq -864(%rbp), %rax movq %rax, -880(%rbp) testq %rsi, %rsi jne .L41 testq %rdx, %rdx jne .L137 .L41: addq %rsi, %rdx leaq -880(%rbp), %rdi .LEHB5: call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag jmp .L138 .L112: endbr64 movq %rax, %rbx movq %r13, -976(%rbp) movq -24(%r13), %rax movq 24+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -976(%rbp,%rax) movq $0, -968(%rbp) .L43: leaq 16+_ZTVSt9basic_iosIcSt11char_traitsIcEE(%rip), %rax movq %rax, -848(%rbp) leaq -848(%rbp), %rdi call _ZNSt8ios_baseD2Ev@PLT .L51: leaq -576(%rbp), %rdi call _ZNSt14basic_ifstreamIcSt11char_traitsIcEED1Ev@PLT .L100: leaq -1008(%rbp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq -56(%rbp), %rax subq %fs:40, %rax je .L101 call __stack_chk_fail@PLT .L137: movq -56(%rbp), %rax subq %fs:40, %rax jne .L139 leaq .LC1(%rip), %rdi call _ZSt19__throw_logic_errorPKc@PLT .LEHE5: .L113: endbr64 movq %rax, %rbx jmp .L47 .L139: call __stack_chk_fail@PLT .L138: movl $24, -888(%rbp) leaq -952(%rbp), %rdi movl $0, %ecx movl $0, %edx movq -880(%rbp), %rsi .LEHB6: call _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEE7_M_syncEPcmm@PLT .LEHE6: jmp .L140 .L114: endbr64 movq %rax, %rbx leaq -880(%rbp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT .L47: leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, -952(%rbp) leaq -896(%rbp), %rdi call _ZNSt6localeD1Ev@PLT .L48: leaq -976(%rbp), %rdi leaq 8+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi call _ZNSdD2Ev@PLT jmp .L43 .L140: leaq -952(%rbp), %rsi leaq -848(%rbp), %rdi .LEHB7: call _ZNSt9basic_iosIcSt11char_traitsIcEE4initEPSt15basic_streambufIcS1_E@PLT .LEHE7: jmp .L141 .L111: endbr64 movq %rax, %rbx leaq -952(%rbp), %rdi call _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED1Ev jmp .L48 .L110: endbr64 movq %rax, %rbx jmp .L43 .L150: movq -1088(%rbp), %rbx leaq -1064(%rbp), %r13 jmp .L70 .L143: movq (%rax), %rdx movq -24(%rdx), %rdx testb $5, 32(%rax,%rdx) jne .L142 movsd -1064(%rbp), %xmm0 movsd %xmm0, (%rbx) addl $1, %r12d addq $8, %rbx .L70: leaq -976(%rbp), %rdi movq %r13, %rsi .LEHB8: call _ZNSi10_M_extractIdEERSiRT_@PLT .LEHE8: jmp .L143 .L142: addl $1, -1096(%rbp) leaq 24+_ZTVNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -976(%rbp) leaq 80(%rax), %rax movq %rax, -848(%rbp) leaq -40(%rax), %rax movq %rax, -960(%rbp) leaq 16+_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -952(%rbp) movq -880(%rbp), %rdi leaq -864(%rbp), %rax cmpq %rax, %rdi je .L53 movq -864(%rbp), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L53: leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, -952(%rbp) leaq -896(%rbp), %rdi call _ZNSt6localeD1Ev@PLT movq 8+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -976(%rbp) movq -24(%rax), %rax movq 48+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rbx movq %rbx, -976(%rbp,%rax) movq %r15, -960(%rbp) movq -24(%r15), %rax movq 40+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -960(%rbp,%rax) movq %r14, -976(%rbp) movq -24(%r14), %rax movq 24+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rbx movq %rbx, -976(%rbp,%rax) movq $0, -968(%rbp) leaq 16+_ZTVSt9basic_iosIcSt11char_traitsIcEE(%rip), %rax movq %rax, -848(%rbp) leaq -848(%rbp), %rdi call _ZNSt8ios_baseD2Ev@PLT movq -576(%rbp), %rax movq -24(%rax), %rax movq -336(%rbp,%rax), %rbx addq $80, -1088(%rbp) testq %rbx, %rbx je .L54 .L55: cmpb $0, 56(%rbx) je .L58 movzbl 67(%rbx), %edx .L59: movsbl %dl, %edx leaq -1008(%rbp), %rsi leaq -576(%rbp), %rdi .LEHB9: call _ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EES4_@PLT jmp .L144 .L38: movq -576(%rbp), %rax movq -24(%rax), %rax movq -336(%rbp,%rax), %rbx testq %rbx, %rbx je .L54 leaq h(%rip), %rax movq %rax, -1088(%rbp) movl $0, -1096(%rbp) movl $0, %r12d jmp .L55 .L54: movq -56(%rbp), %rax subq %fs:40, %rax jne .L145 call _ZSt16__throw_bad_castv@PLT .L145: call __stack_chk_fail@PLT .L58: movq %rbx, %rdi call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq (%rbx), %rax movl $10, %esi movq %rbx, %rdi call *48(%rax) .LEHE9: movl %eax, %edx jmp .L59 .L144: movq (%rax), %rdx movq -24(%rdx), %rdx testb $5, 32(%rax,%rdx) jne .L29 cmpq $0, -1000(%rbp) je .L29 leaq -976(%rbp), %rbx leaq -848(%rbp), %rdi call _ZNSt8ios_baseC2Ev@PLT leaq 16+_ZTVSt9basic_iosIcSt11char_traitsIcEE(%rip), %rax movq %rax, -848(%rbp) movq $0, -632(%rbp) movb $0, -624(%rbp) movb $0, -623(%rbp) movq $0, -616(%rbp) movq $0, -608(%rbp) movq $0, -600(%rbp) movq $0, -592(%rbp) movq 16+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %r14 movq %r14, -976(%rbp) movq -24(%r14), %rax movq 24+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -976(%rbp,%rax) movq $0, -968(%rbp) movq -976(%rbp), %rax addq -24(%rax), %rbx movq %rbx, %rdi movl $0, %esi .LEHB10: call _ZNSt9basic_iosIcSt11char_traitsIcEE4initEPSt15basic_streambufIcS1_E@PLT .LEHE10: movq 32+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %r15 movq %r15, -960(%rbp) movq -24(%r15), %rax movq 40+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rbx movq %rbx, -960(%rbp,%rax) movq -960(%rbp), %rax movq -24(%rax), %rax leaq -960(%rbp,%rax), %rdi movl $0, %esi .LEHB11: call _ZNSt9basic_iosIcSt11char_traitsIcEE4initEPSt15basic_streambufIcS1_E@PLT .LEHE11: movq 8+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -976(%rbp) movq -24(%rax), %rax movq 48+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -976(%rbp,%rax) leaq 24+_ZTVNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -976(%rbp) leaq 80(%rax), %rax movq %rax, -848(%rbp) leaq -40(%rax), %rax movq %rax, -960(%rbp) leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, -952(%rbp) movq $0, -944(%rbp) movq $0, -936(%rbp) movq $0, -928(%rbp) movq $0, -920(%rbp) movq $0, -912(%rbp) movq $0, -904(%rbp) leaq -896(%rbp), %rdi call _ZNSt6localeC1Ev@PLT leaq 16+_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, -952(%rbp) movl $0, -888(%rbp) movq -1000(%rbp), %rdx movq -1008(%rbp), %rsi leaq -864(%rbp), %rax movq %rax, -880(%rbp) testq %rsi, %rsi jne .L62 testq %rdx, %rdx jne .L146 .L62: addq %rsi, %rdx leaq -880(%rbp), %rdi .LEHB12: call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPKcEEvT_S8_St20forward_iterator_tag jmp .L147 .L117: endbr64 movq %rax, %rbx movq %r14, -976(%rbp) movq -24(%r14), %rax movq 24+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi movq %rsi, -976(%rbp,%rax) movq $0, -968(%rbp) .L64: leaq 16+_ZTVSt9basic_iosIcSt11char_traitsIcEE(%rip), %rax movq %rax, -848(%rbp) leaq -848(%rbp), %rdi call _ZNSt8ios_baseD2Ev@PLT jmp .L51 .L146: movq -56(%rbp), %rax subq %fs:40, %rax jne .L148 leaq .LC1(%rip), %rdi call _ZSt19__throw_logic_errorPKc@PLT .LEHE12: .L118: endbr64 movq %rax, %rbx jmp .L68 .L148: call __stack_chk_fail@PLT .L147: movl $24, -888(%rbp) leaq -952(%rbp), %rdi movl $0, %ecx movl $0, %edx movq -880(%rbp), %rsi .LEHB13: call _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEE7_M_syncEPcmm@PLT .LEHE13: jmp .L149 .L119: endbr64 movq %rax, %rbx leaq -880(%rbp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT .L68: leaq 16+_ZTVSt15basic_streambufIcSt11char_traitsIcEE(%rip), %rax movq %rax, -952(%rbp) leaq -896(%rbp), %rdi call _ZNSt6localeD1Ev@PLT .L69: leaq -976(%rbp), %rdi leaq 8+_ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rsi call _ZNSdD2Ev@PLT jmp .L64 .L149: leaq -952(%rbp), %rsi leaq -848(%rbp), %rdi .LEHB14: call _ZNSt9basic_iosIcSt11char_traitsIcEE4initEPSt15basic_streambufIcS1_E@PLT .LEHE14: jmp .L150 .L116: endbr64 movq %rax, %rbx leaq -952(%rbp), %rdi call _ZNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEED1Ev jmp .L69 .L115: endbr64 movq %rax, %rbx jmp .L64 .L103: movl $0, %r12d movl $0, -1080(%rbp) movl $0, -1096(%rbp) movl $0, -1092(%rbp) .L29: leaq -576(%rbp), %rdi .LEHB15: call _ZNSt14basic_ifstreamIcSt11char_traitsIcEE5closeEv@PLT movl -1080(%rbp), %eax movl -1092(%rbp), %ebx cltd idivl %ebx movl %eax, %r13d movl %r12d, %eax movl -1096(%rbp), %esi cltd idivl %esi movl %eax, %r14d leal (%rbx,%rsi), %r15d leal -1(%r15), %edi movl %edi, -1100(%rbp) leal 0(%r13,%rax), %ebx leal -1(%rbx), %edx movl %edx, -1088(%rbp) movslq %r12d, %rax leaq 0(,%rax,8), %rsi movq %rsi, -1112(%rbp) imull %edx, %edi movslq %edi, %r12 salq $3, %r12 movslq -1080(%rbp), %rsi salq $3, %rsi leaq -1056(%rbp), %rdi movl $1, %edx call cudaMallocManaged@PLT leaq -1048(%rbp), %rdi movl $1, %edx movq -1112(%rbp), %rsi call cudaMallocManaged@PLT leaq -1040(%rbp), %rdi movl $1, %edx movq %r12, %rsi call cudaMallocManaged@PLT movl -1092(%rbp), %eax testl %eax, %eax jle .L72 movslq %eax, %r9 imulq $8000, %r9, %r9 movl $0, %edi movl $0, %r8d movslq %r13d, %r11 leaq a(%rip), %r10 jmp .L73 .L76: addl %r13d, %r8d addq $8000, %rdi cmpq %rdi, %r9 je .L72 .L73: testl %r13d, %r13d jle .L76 movslq %r8d, %rdx leaq 0(,%rdx,8), %rax leaq (%r11,%rdx), %rsi salq $3, %rsi negq %rdx leaq (%rdi,%rdx,8), %rcx addq %r10, %rcx .L74: movsd (%rcx,%rax), %xmm0 movq -1056(%rbp), %rdx movsd %xmm0, (%rdx,%rax) addq $8, %rax cmpq %rsi, %rax jne .L74 jmp .L76 .L72: movl -1096(%rbp), %eax testl %eax, %eax jle .L77 cltq leaq (%rax,%rax,4), %r9 salq $4, %r9 movl $0, %edi movl $0, %r8d movslq %r14d, %r11 leaq h(%rip), %r10 jmp .L78 .L81: addl %r14d, %r8d addq $80, %rdi cmpq %rdi, %r9 je .L77 .L78: testl %r14d, %r14d jle .L81 movslq %r8d, %rdx leaq 0(,%rdx,8), %rax leaq (%r11,%rdx), %rsi salq $3, %rsi negq %rdx leaq (%rdi,%rdx,8), %rcx addq %r10, %rcx .L79: movsd (%rcx,%rax), %xmm0 movq -1048(%rbp), %rdx movsd %xmm0, (%rdx,%rax) addq $8, %rax cmpq %rsi, %rax jne .L79 jmp .L81 .L77: leal 29(%r15), %eax movl %r15d, %edx subl $2, %edx cmovns %edx, %eax sarl $5, %eax addl $1, %eax movl %eax, -1032(%rbp) leal 29(%rbx), %eax movl %ebx, %edx subl $2, %edx cmovns %edx, %eax sarl $5, %eax addl $1, %eax movl %eax, -1028(%rbp) movl $1, -1024(%rbp) movl $32, -1020(%rbp) movl $32, -1016(%rbp) movl $1, -1012(%rbp) movl $0, %r9d movl $0, %r8d movq -1020(%rbp), %rdx movl $1, %ecx movq -1032(%rbp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax jne .L82 subq $8, %rsp pushq %r14 movl -1096(%rbp), %r9d movl %r13d, %r8d movl -1092(%rbp), %ecx movq -1040(%rbp), %rdx movq -1048(%rbp), %rsi movq -1056(%rbp), %rdi .cfi_escape 0x2e,0x10 call _Z41__device_stub__Z13convolution2DPdS_S_iiiiPdS_S_iiii addq $16, %rsp .L82: .cfi_escape 0x2e,0 call cudaGetLastError@PLT testl %eax, %eax jne .L151 call cudaDeviceSynchronize@PLT jmp .L152 .L151: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC2(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $1, %edi call exit@PLT .L152: movl %eax, %r14d testl %eax, %eax jne .L84 cmpl $0, -1100(%rbp) jle .L85 leal -1(%r15), %eax movl %eax, -1092(%rbp) movl $0, %r15d leal -1(%rbx), %eax movl %eax, -1096(%rbp) leaq _ZSt4cout(%rip), %rbx leaq .LC4(%rip), %r13 jmp .L86 .L84: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC3(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $1, %edi call exit@PLT .L153: movq %rax, %rdi movl $1, %edx movq %r13, %rsi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT addq $8, -1080(%rbp) movq -1080(%rbp), %rcx cmpq %rcx, %r12 je .L93 .L87: movq (%rbx), %rdx movq %rbx, %rcx addq -24(%rdx), %rcx movl 24(%rcx), %eax andl $-261, %eax orl $4, %eax movl %eax, 24(%rcx) movq -24(%rdx), %rax movq $3, 8(%rbx,%rax) movq -1040(%rbp), %rax movq -1080(%rbp), %rcx movsd (%rax,%rcx), %xmm0 movq %rbx, %rdi call _ZNSo9_M_insertIdEERSoT_@PLT jmp .L153 .L93: movq (%rbx), %rax movq -24(%rax), %rax movq 240(%rbx,%rax), %rax movq %rax, %rdi testq %rax, %rax je .L154 cmpb $0, 56(%rax) je .L90 movzbl 67(%rax), %esi .L91: movsbl %sil, %esi movq %rbx, %rdi call _ZNSo3putEc@PLT jmp .L155 .L154: movq -56(%rbp), %rax subq %fs:40, %rax jne .L156 call _ZSt16__throw_bad_castv@PLT .L108: endbr64 movq %rax, %rbx jmp .L51 .L156: call __stack_chk_fail@PLT .L90: movq %rax, %r12 call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq %r12, %rdi movq (%r12), %rax movl $10, %esi call *48(%rax) movl %eax, %esi jmp .L91 .L155: movq %rax, %rdi call _ZNSo5flushEv@PLT addl $1, %r15d movl -1088(%rbp), %eax addl %eax, %r14d movl -1092(%rbp), %eax cmpl %eax, %r15d je .L85 .L86: cmpl $0, -1088(%rbp) jle .L93 movslq %r14d, %rax leaq 0(,%rax,8), %rsi movq %rsi, -1080(%rbp) movl -1096(%rbp), %r12d addq %rax, %r12 salq $3, %r12 jmp .L87 .L85: movq -1056(%rbp), %rdi call cudaFree@PLT testl %eax, %eax jne .L157 movq -1048(%rbp), %rdi call cudaFree@PLT jmp .L158 .L157: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC5(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $1, %edi call exit@PLT .L158: testl %eax, %eax jne .L159 movq -1040(%rbp), %rdi call cudaFree@PLT jmp .L160 .L159: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC6(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $1, %edi call exit@PLT .L160: testl %eax, %eax jne .L161 call cudaDeviceReset@PLT jmp .L162 .L161: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC7(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $1, %edi call exit@PLT .L162: testl %eax, %eax jne .L163 leaq .LC9(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT jmp .L164 .L163: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC8(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT .LEHE15: movl $1, %edi call exit@PLT .L164: leaq -576(%rbp), %rdi call _ZNSt14basic_ifstreamIcSt11char_traitsIcEED1Ev@PLT leaq -1008(%rbp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq -56(%rbp), %rax subq %fs:40, %rax jne .L165 movl $0, %eax leaq -40(%rbp), %rsp popq %rbx popq %r12 popq %r13 popq %r14 popq %r15 popq %rbp .cfi_remember_state .cfi_def_cfa 7, 8 ret .L107: .cfi_restore_state endbr64 movq %rax, %rbx leaq -976(%rbp), %rdi call _ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEED1Ev@PLT jmp .L51 .L109: endbr64 movq %rax, %rbx leaq -976(%rbp), %rdi call _ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEED1Ev@PLT jmp .L51 .L106: endbr64 movq %rax, %rbx jmp .L100 .L101: movq %rbx, %rdi .LEHB16: call _Unwind_Resume@PLT .LEHE16: .L165: call __stack_chk_fail@PLT .cfi_endproc .LFE4039: .globl __gxx_personality_v0 .section .gcc_except_table,"a",@progbits .LLSDA4039: .byte 0xff .byte 0xff .byte 0x1 .uleb128 .LLSDACSE4039-.LLSDACSB4039 .LLSDACSB4039: .uleb128 .LEHB0-.LFB4039 .uleb128 .LEHE0-.LEHB0 .uleb128 .L106-.LFB4039 .uleb128 0 .uleb128 .LEHB1-.LFB4039 .uleb128 .LEHE1-.LEHB1 .uleb128 .L107-.LFB4039 .uleb128 0 .uleb128 .LEHB2-.LFB4039 .uleb128 .LEHE2-.LEHB2 .uleb128 .L108-.LFB4039 .uleb128 0 .uleb128 .LEHB3-.LFB4039 .uleb128 .LEHE3-.LEHB3 .uleb128 .L110-.LFB4039 .uleb128 0 .uleb128 .LEHB4-.LFB4039 .uleb128 .LEHE4-.LEHB4 .uleb128 .L112-.LFB4039 .uleb128 0 .uleb128 .LEHB5-.LFB4039 .uleb128 .LEHE5-.LEHB5 .uleb128 .L113-.LFB4039 .uleb128 0 .uleb128 .LEHB6-.LFB4039 .uleb128 .LEHE6-.LEHB6 .uleb128 .L114-.LFB4039 .uleb128 0 .uleb128 .LEHB7-.LFB4039 .uleb128 .LEHE7-.LEHB7 .uleb128 .L111-.LFB4039 .uleb128 0 .uleb128 .LEHB8-.LFB4039 .uleb128 .LEHE8-.LEHB8 .uleb128 .L109-.LFB4039 .uleb128 0 .uleb128 .LEHB9-.LFB4039 .uleb128 .LEHE9-.LEHB9 .uleb128 .L108-.LFB4039 .uleb128 0 .uleb128 .LEHB10-.LFB4039 .uleb128 .LEHE10-.LEHB10 .uleb128 .L115-.LFB4039 .uleb128 0 .uleb128 .LEHB11-.LFB4039 .uleb128 .LEHE11-.LEHB11 .uleb128 .L117-.LFB4039 .uleb128 0 .uleb128 .LEHB12-.LFB4039 .uleb128 .LEHE12-.LEHB12 .uleb128 .L118-.LFB4039 .uleb128 0 .uleb128 .LEHB13-.LFB4039 .uleb128 .LEHE13-.LEHB13 .uleb128 .L119-.LFB4039 .uleb128 0 .uleb128 .LEHB14-.LFB4039 .uleb128 .LEHE14-.LEHB14 .uleb128 .L116-.LFB4039 .uleb128 0 .uleb128 .LEHB15-.LFB4039 .uleb128 .LEHE15-.LEHB15 .uleb128 .L108-.LFB4039 .uleb128 0 .uleb128 .LEHB16-.LFB4039 .uleb128 .LEHE16-.LEHB16 .uleb128 0 .uleb128 0 .LLSDACSE4039: .text .size main, .-main .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .globl c .bss .align 32 .type c, @object .size c, 8000000 c: .zero 8000000 .globl h .align 32 .type h, @object .size h, 800 h: .zero 800 .globl a .align 32 .type a, @object .size a, 8000000 a: .zero 8000000 .hidden DW.ref.__gxx_personality_v0 .weak DW.ref.__gxx_personality_v0 .section .data.rel.local.DW.ref.__gxx_personality_v0,"awG",@progbits,DW.ref.__gxx_personality_v0,comdat .align 8 .type DW.ref.__gxx_personality_v0, @object .size DW.ref.__gxx_personality_v0, 8 DW.ref.__gxx_personality_v0: .quad __gxx_personality_v0 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include <stdio.h> #include <iostream> #include <fstream> #include <string> #include <cuda_runtime.h> #include <cstdlib> #include <sstream> #include <iomanip> using namespace std; //input, filter and output considering maximum dimension possible double a[1000][1000], h[10][10], c[1000][1000]; /** * CUDA Kernel Device code * * Computes the 2D convolution of input b with filter h and output c */ __global__ void convolution2D(double *a_1, double *h_1, double *c_1, int rows_a, int columns_a, int rows_h, int columns_h) { int idx = blockDim.x * blockIdx.x + threadIdx.x; int idy = blockDim.y * blockIdx.y + threadIdx.y; double sum; if((idx < (columns_h+columns_a - 1)) && (idy < (rows_h + rows_a - 1))) { sum = 0; for(int k=0; k< rows_h; k++) { for(int j=0; j< columns_h; j++) { if ((idy - k) >= 0 && (idx - j) >= 0 && (idy - k) < rows_a && (idx - j) < columns_a) { sum += a_1[((idy - k)*columns_a) + (idx - j)] * h_1[k*columns_h + j]; } } } c_1[idy*(columns_a + columns_h - 1) + idx] = sum; __syncthreads(); } } /** * Host main routine */ int main(int argc, char** argv) { // Error code to check return values for CUDA calls cudaError_t err = cudaSuccess; int columns_a, columns_h, rows_a, rows_h, size_a, size_h; columns_a = columns_h = 0; size_a = size_h = rows_a = rows_h = 0; char *ip_file; string line; double input; ip_file = argv[1]; int i, j, k; ifstream file(ip_file); if(file.is_open()) { i=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { a[i][j] = input; size_a++; j++; } i++; rows_a++; } k=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { h[k][j] = input; j++; size_h++; } k++; rows_h++; } } file.close(); columns_a = size_a/rows_a; columns_h = size_h/rows_h; int op_size = ((rows_a+rows_h-1)*(columns_a+columns_h-1)); size_t size_ax = size_a*sizeof(double); size_t size_hx = size_h*sizeof(double); size_t size_cx = op_size*sizeof(double); // Allocate the host input vector a double *h_a; cudaMallocManaged(&h_a, size_ax); // Allocate the host input vector h double *h_h; cudaMallocManaged(&h_h, size_hx); // Allocate the host output vector c double *h_c; cudaMallocManaged(&h_c, size_cx); for (int i = 0; i < rows_a; i++) { for(int j=0;j< columns_a; j++) { h_a[i*columns_a + j] = a[i][j]; } } for (int i = 0; i < rows_h; i++) { for(int j=0;j< columns_h; j++) { h_h[i*columns_h + j] = h[i][j]; } } // Launch the CUDA Kernel dim3 blocksPerGrid(((rows_a + rows_h - 2) / 32) + 1, ((columns_h + columns_a - 2) / 32) + 1, 1); dim3 threadsPerBlock(32,32,1); convolution2D<<<blocksPerGrid, threadsPerBlock>>>(h_a, h_h, h_c, rows_a, columns_a, rows_h, columns_h); err = cudaGetLastError(); if (err != cudaSuccess) { fprintf(stderr, "Failed to launch convolution2D kernel (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaDeviceSynchronize(); if (err != cudaSuccess) { fprintf(stderr, "Failed to synchronize (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } ///////////////////////////////////////////////////////////////////////////////////////////////////////// for (int i = 0; i < rows_a + rows_h - 1; i++) { for (int j = 0; j < columns_a + columns_h - 1; j++) { cout << fixed << setprecision(3) << h_c[(i*(columns_h+columns_a-1))+j] << " "; } cout << endl; } // Free device global memory err = cudaFree(h_a); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector a (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaFree(h_h); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector h (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } err = cudaFree(h_c); if (err != cudaSuccess) { fprintf(stderr, "Failed to free device vector c (error code %s)!\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } // Reset the device and exit // cudaDeviceReset causes the driver to clean up all state. While // not mandatory in normal operation, it is good practice. It is also // needed to ensure correct operation when the application is being // profiled. Calling cudaDeviceReset causes all profile data to be // flushed before the application exits err = cudaDeviceReset(); if (err != cudaSuccess) { fprintf(stderr, "Failed to deinitialize the device! error=%s\n", cudaGetErrorString(err)); exit(EXIT_FAILURE); } printf("Done\n"); return 0; }
#include <stdio.h> #include <iostream> #include <fstream> #include <string> #include <hip/hip_runtime.h> #include <cstdlib> #include <sstream> #include <iomanip> using namespace std; //input, filter and output considering maximum dimension possible double a[1000][1000], h[10][10], c[1000][1000]; /** * CUDA Kernel Device code * * Computes the 2D convolution of input b with filter h and output c */ __global__ void convolution2D(double *a_1, double *h_1, double *c_1, int rows_a, int columns_a, int rows_h, int columns_h) { int idx = blockDim.x * blockIdx.x + threadIdx.x; int idy = blockDim.y * blockIdx.y + threadIdx.y; double sum; if((idx < (columns_h+columns_a - 1)) && (idy < (rows_h + rows_a - 1))) { sum = 0; for(int k=0; k< rows_h; k++) { for(int j=0; j< columns_h; j++) { if ((idy - k) >= 0 && (idx - j) >= 0 && (idy - k) < rows_a && (idx - j) < columns_a) { sum += a_1[((idy - k)*columns_a) + (idx - j)] * h_1[k*columns_h + j]; } } } c_1[idy*(columns_a + columns_h - 1) + idx] = sum; __syncthreads(); } } /** * Host main routine */ int main(int argc, char** argv) { // Error code to check return values for CUDA calls hipError_t err = hipSuccess; int columns_a, columns_h, rows_a, rows_h, size_a, size_h; columns_a = columns_h = 0; size_a = size_h = rows_a = rows_h = 0; char *ip_file; string line; double input; ip_file = argv[1]; int i, j, k; ifstream file(ip_file); if(file.is_open()) { i=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { a[i][j] = input; size_a++; j++; } i++; rows_a++; } k=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { h[k][j] = input; j++; size_h++; } k++; rows_h++; } } file.close(); columns_a = size_a/rows_a; columns_h = size_h/rows_h; int op_size = ((rows_a+rows_h-1)*(columns_a+columns_h-1)); size_t size_ax = size_a*sizeof(double); size_t size_hx = size_h*sizeof(double); size_t size_cx = op_size*sizeof(double); // Allocate the host input vector a double *h_a; hipMallocManaged(&h_a, size_ax); // Allocate the host input vector h double *h_h; hipMallocManaged(&h_h, size_hx); // Allocate the host output vector c double *h_c; hipMallocManaged(&h_c, size_cx); for (int i = 0; i < rows_a; i++) { for(int j=0;j< columns_a; j++) { h_a[i*columns_a + j] = a[i][j]; } } for (int i = 0; i < rows_h; i++) { for(int j=0;j< columns_h; j++) { h_h[i*columns_h + j] = h[i][j]; } } // Launch the CUDA Kernel dim3 blocksPerGrid(((rows_a + rows_h - 2) / 32) + 1, ((columns_h + columns_a - 2) / 32) + 1, 1); dim3 threadsPerBlock(32,32,1); convolution2D<<<blocksPerGrid, threadsPerBlock>>>(h_a, h_h, h_c, rows_a, columns_a, rows_h, columns_h); err = hipGetLastError(); if (err != hipSuccess) { fprintf(stderr, "Failed to launch convolution2D kernel (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipDeviceSynchronize(); if (err != hipSuccess) { fprintf(stderr, "Failed to synchronize (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } ///////////////////////////////////////////////////////////////////////////////////////////////////////// for (int i = 0; i < rows_a + rows_h - 1; i++) { for (int j = 0; j < columns_a + columns_h - 1; j++) { cout << fixed << setprecision(3) << h_c[(i*(columns_h+columns_a-1))+j] << " "; } cout << endl; } // Free device global memory err = hipFree(h_a); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector a (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipFree(h_h); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector h (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipFree(h_c); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector c (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } // Reset the device and exit // cudaDeviceReset causes the driver to clean up all state. While // not mandatory in normal operation, it is good practice. It is also // needed to ensure correct operation when the application is being // profiled. Calling cudaDeviceReset causes all profile data to be // flushed before the application exits err = hipDeviceReset(); if (err != hipSuccess) { fprintf(stderr, "Failed to deinitialize the device! error=%s\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } printf("Done\n"); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <stdio.h> #include <iostream> #include <fstream> #include <string> #include <hip/hip_runtime.h> #include <cstdlib> #include <sstream> #include <iomanip> using namespace std; //input, filter and output considering maximum dimension possible double a[1000][1000], h[10][10], c[1000][1000]; /** * CUDA Kernel Device code * * Computes the 2D convolution of input b with filter h and output c */ __global__ void convolution2D(double *a_1, double *h_1, double *c_1, int rows_a, int columns_a, int rows_h, int columns_h) { int idx = blockDim.x * blockIdx.x + threadIdx.x; int idy = blockDim.y * blockIdx.y + threadIdx.y; double sum; if((idx < (columns_h+columns_a - 1)) && (idy < (rows_h + rows_a - 1))) { sum = 0; for(int k=0; k< rows_h; k++) { for(int j=0; j< columns_h; j++) { if ((idy - k) >= 0 && (idx - j) >= 0 && (idy - k) < rows_a && (idx - j) < columns_a) { sum += a_1[((idy - k)*columns_a) + (idx - j)] * h_1[k*columns_h + j]; } } } c_1[idy*(columns_a + columns_h - 1) + idx] = sum; __syncthreads(); } } /** * Host main routine */ int main(int argc, char** argv) { // Error code to check return values for CUDA calls hipError_t err = hipSuccess; int columns_a, columns_h, rows_a, rows_h, size_a, size_h; columns_a = columns_h = 0; size_a = size_h = rows_a = rows_h = 0; char *ip_file; string line; double input; ip_file = argv[1]; int i, j, k; ifstream file(ip_file); if(file.is_open()) { i=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { a[i][j] = input; size_a++; j++; } i++; rows_a++; } k=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { h[k][j] = input; j++; size_h++; } k++; rows_h++; } } file.close(); columns_a = size_a/rows_a; columns_h = size_h/rows_h; int op_size = ((rows_a+rows_h-1)*(columns_a+columns_h-1)); size_t size_ax = size_a*sizeof(double); size_t size_hx = size_h*sizeof(double); size_t size_cx = op_size*sizeof(double); // Allocate the host input vector a double *h_a; hipMallocManaged(&h_a, size_ax); // Allocate the host input vector h double *h_h; hipMallocManaged(&h_h, size_hx); // Allocate the host output vector c double *h_c; hipMallocManaged(&h_c, size_cx); for (int i = 0; i < rows_a; i++) { for(int j=0;j< columns_a; j++) { h_a[i*columns_a + j] = a[i][j]; } } for (int i = 0; i < rows_h; i++) { for(int j=0;j< columns_h; j++) { h_h[i*columns_h + j] = h[i][j]; } } // Launch the CUDA Kernel dim3 blocksPerGrid(((rows_a + rows_h - 2) / 32) + 1, ((columns_h + columns_a - 2) / 32) + 1, 1); dim3 threadsPerBlock(32,32,1); convolution2D<<<blocksPerGrid, threadsPerBlock>>>(h_a, h_h, h_c, rows_a, columns_a, rows_h, columns_h); err = hipGetLastError(); if (err != hipSuccess) { fprintf(stderr, "Failed to launch convolution2D kernel (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipDeviceSynchronize(); if (err != hipSuccess) { fprintf(stderr, "Failed to synchronize (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } ///////////////////////////////////////////////////////////////////////////////////////////////////////// for (int i = 0; i < rows_a + rows_h - 1; i++) { for (int j = 0; j < columns_a + columns_h - 1; j++) { cout << fixed << setprecision(3) << h_c[(i*(columns_h+columns_a-1))+j] << " "; } cout << endl; } // Free device global memory err = hipFree(h_a); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector a (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipFree(h_h); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector h (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipFree(h_c); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector c (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } // Reset the device and exit // cudaDeviceReset causes the driver to clean up all state. While // not mandatory in normal operation, it is good practice. It is also // needed to ensure correct operation when the application is being // profiled. Calling cudaDeviceReset causes all profile data to be // flushed before the application exits err = hipDeviceReset(); if (err != hipSuccess) { fprintf(stderr, "Failed to deinitialize the device! error=%s\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } printf("Done\n"); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z13convolution2DPdS_S_iiii .globl _Z13convolution2DPdS_S_iiii .p2align 8 .type _Z13convolution2DPdS_S_iiii,@function _Z13convolution2DPdS_S_iiii: s_clause 0x2 s_load_b32 s4, s[0:1], 0x34 s_load_b32 s12, s[0:1], 0x1c s_load_b32 s13, s[0:1], 0x24 v_and_b32_e32 v3, 0x3ff, v0 s_add_u32 s2, s0, 40 s_addc_u32 s3, s1, 0 s_waitcnt lgkmcnt(0) s_and_b32 s4, s4, 0xffff s_delay_alu instid0(SALU_CYCLE_1) v_mad_u64_u32 v[1:2], null, s14, s4, v[3:4] s_add_i32 s5, s12, s13 s_mov_b32 s4, exec_lo s_add_i32 s5, s5, -1 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_cmpx_gt_i32_e64 s5, v1 s_cbranch_execz .LBB0_14 s_load_b32 s2, s[2:3], 0xc s_clause 0x1 s_load_b32 s14, s[0:1], 0x18 s_load_b32 s16, s[0:1], 0x20 v_bfe_u32 v0, v0, 10, 10 s_waitcnt lgkmcnt(0) s_lshr_b32 s2, s2, 16 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_mad_u64_u32 v[2:3], null, s15, s2, v[0:1] s_add_i32 s2, s14, s16 s_delay_alu instid0(SALU_CYCLE_1) s_add_i32 s2, s2, -1 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_cmp_gt_i32_e32 vcc_lo, s2, v2 s_and_b32 exec_lo, exec_lo, vcc_lo s_cbranch_execz .LBB0_14 s_cmp_lt_i32 s16, 1 s_cbranch_scc1 .LBB0_12 s_load_b128 s[8:11], s[0:1], 0x0 v_mul_lo_u32 v0, s12, v2 v_mov_b32_e32 v3, 0 v_mov_b32_e32 v4, 0 s_cmp_gt_i32 s13, 0 s_mov_b32 s7, 0 s_cselect_b32 s15, -1, 0 s_mov_b32 s17, 0 s_mov_b32 s18, 0 s_branch .LBB0_5 .LBB0_4: s_set_inst_prefetch_distance 0x2 v_subrev_nc_u32_e32 v0, s12, v0 s_add_i32 s18, s18, 1 s_add_i32 s17, s17, s13 s_cmp_eq_u32 s18, s16 s_cbranch_scc1 .LBB0_13 .LBB0_5: s_and_not1_b32 vcc_lo, exec_lo, s15 s_cbranch_vccnz .LBB0_4 v_subrev_nc_u32_e32 v5, s18, v2 s_mov_b32 s6, s17 s_mov_b32 s19, s13 s_delay_alu instid0(VALU_DEP_1) v_cmp_lt_i32_e32 vcc_lo, -1, v5 v_cmp_gt_i32_e64 s2, s14, v5 v_mov_b32_e32 v5, v1 s_set_inst_prefetch_distance 0x1 s_branch .LBB0_9 .p2align 6 .LBB0_7: s_or_b32 exec_lo, exec_lo, s4 .LBB0_8: s_delay_alu instid0(SALU_CYCLE_1) s_or_b32 exec_lo, exec_lo, s20 v_add_nc_u32_e32 v5, -1, v5 s_add_i32 s19, s19, -1 s_add_i32 s6, s6, 1 s_cmp_eq_u32 s19, 0 s_cbranch_scc1 .LBB0_4 .LBB0_9: s_and_saveexec_b32 s20, vcc_lo s_cbranch_execz .LBB0_8 v_cmp_lt_i32_e64 s3, -1, v5 v_cmp_gt_i32_e64 s4, s12, v5 s_delay_alu instid0(VALU_DEP_2) s_and_b32 s3, s2, s3 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) s_and_b32 s3, s4, s3 s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s4, s3 s_cbranch_execz .LBB0_7 v_add_nc_u32_e32 v6, v0, v5 s_lshl_b64 s[22:23], s[6:7], 3 s_waitcnt lgkmcnt(0) s_add_u32 s22, s10, s22 s_addc_u32 s23, s11, s23 v_ashrrev_i32_e32 v7, 31, v6 s_load_b64 s[22:23], s[22:23], 0x0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[6:7], 3, v[6:7] v_add_co_u32 v6, s3, s8, v6 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v7, s3, s9, v7, s3 global_load_b64 v[6:7], v[6:7], off s_waitcnt vmcnt(0) lgkmcnt(0) v_fma_f64 v[3:4], v[6:7], s[22:23], v[3:4] s_branch .LBB0_7 .LBB0_12: v_mov_b32_e32 v3, 0 v_mov_b32_e32 v4, 0 .LBB0_13: s_load_b64 s[0:1], s[0:1], 0x10 v_mad_u64_u32 v[5:6], null, v2, s5, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v6, 31, v5 v_lshlrev_b64 v[0:1], 3, v[5:6] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v0, vcc_lo, s0, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo global_store_b64 v[0:1], v[3:4], off s_waitcnt_vscnt null, 0x0 s_barrier buffer_gl0_inv .LBB0_14: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z13convolution2DPdS_S_iiii .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 296 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 8 .amdhsa_next_free_sgpr 24 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z13convolution2DPdS_S_iiii, .Lfunc_end0-_Z13convolution2DPdS_S_iiii .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value - .offset: 28 .size: 4 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: by_value - .offset: 36 .size: 4 .value_kind: by_value - .offset: 40 .size: 4 .value_kind: hidden_block_count_x - .offset: 44 .size: 4 .value_kind: hidden_block_count_y - .offset: 48 .size: 4 .value_kind: hidden_block_count_z - .offset: 52 .size: 2 .value_kind: hidden_group_size_x - .offset: 54 .size: 2 .value_kind: hidden_group_size_y - .offset: 56 .size: 2 .value_kind: hidden_group_size_z - .offset: 58 .size: 2 .value_kind: hidden_remainder_x - .offset: 60 .size: 2 .value_kind: hidden_remainder_y - .offset: 62 .size: 2 .value_kind: hidden_remainder_z - .offset: 80 .size: 8 .value_kind: hidden_global_offset_x - .offset: 88 .size: 8 .value_kind: hidden_global_offset_y - .offset: 96 .size: 8 .value_kind: hidden_global_offset_z - .offset: 104 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 296 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z13convolution2DPdS_S_iiii .private_segment_fixed_size: 0 .sgpr_count: 26 .sgpr_spill_count: 0 .symbol: _Z13convolution2DPdS_S_iiii.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 8 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <stdio.h> #include <iostream> #include <fstream> #include <string> #include <hip/hip_runtime.h> #include <cstdlib> #include <sstream> #include <iomanip> using namespace std; //input, filter and output considering maximum dimension possible double a[1000][1000], h[10][10], c[1000][1000]; /** * CUDA Kernel Device code * * Computes the 2D convolution of input b with filter h and output c */ __global__ void convolution2D(double *a_1, double *h_1, double *c_1, int rows_a, int columns_a, int rows_h, int columns_h) { int idx = blockDim.x * blockIdx.x + threadIdx.x; int idy = blockDim.y * blockIdx.y + threadIdx.y; double sum; if((idx < (columns_h+columns_a - 1)) && (idy < (rows_h + rows_a - 1))) { sum = 0; for(int k=0; k< rows_h; k++) { for(int j=0; j< columns_h; j++) { if ((idy - k) >= 0 && (idx - j) >= 0 && (idy - k) < rows_a && (idx - j) < columns_a) { sum += a_1[((idy - k)*columns_a) + (idx - j)] * h_1[k*columns_h + j]; } } } c_1[idy*(columns_a + columns_h - 1) + idx] = sum; __syncthreads(); } } /** * Host main routine */ int main(int argc, char** argv) { // Error code to check return values for CUDA calls hipError_t err = hipSuccess; int columns_a, columns_h, rows_a, rows_h, size_a, size_h; columns_a = columns_h = 0; size_a = size_h = rows_a = rows_h = 0; char *ip_file; string line; double input; ip_file = argv[1]; int i, j, k; ifstream file(ip_file); if(file.is_open()) { i=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { a[i][j] = input; size_a++; j++; } i++; rows_a++; } k=0; while(getline(file, line) && line != "") { j=0; stringstream ss(line); while(ss >> input) { h[k][j] = input; j++; size_h++; } k++; rows_h++; } } file.close(); columns_a = size_a/rows_a; columns_h = size_h/rows_h; int op_size = ((rows_a+rows_h-1)*(columns_a+columns_h-1)); size_t size_ax = size_a*sizeof(double); size_t size_hx = size_h*sizeof(double); size_t size_cx = op_size*sizeof(double); // Allocate the host input vector a double *h_a; hipMallocManaged(&h_a, size_ax); // Allocate the host input vector h double *h_h; hipMallocManaged(&h_h, size_hx); // Allocate the host output vector c double *h_c; hipMallocManaged(&h_c, size_cx); for (int i = 0; i < rows_a; i++) { for(int j=0;j< columns_a; j++) { h_a[i*columns_a + j] = a[i][j]; } } for (int i = 0; i < rows_h; i++) { for(int j=0;j< columns_h; j++) { h_h[i*columns_h + j] = h[i][j]; } } // Launch the CUDA Kernel dim3 blocksPerGrid(((rows_a + rows_h - 2) / 32) + 1, ((columns_h + columns_a - 2) / 32) + 1, 1); dim3 threadsPerBlock(32,32,1); convolution2D<<<blocksPerGrid, threadsPerBlock>>>(h_a, h_h, h_c, rows_a, columns_a, rows_h, columns_h); err = hipGetLastError(); if (err != hipSuccess) { fprintf(stderr, "Failed to launch convolution2D kernel (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipDeviceSynchronize(); if (err != hipSuccess) { fprintf(stderr, "Failed to synchronize (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } ///////////////////////////////////////////////////////////////////////////////////////////////////////// for (int i = 0; i < rows_a + rows_h - 1; i++) { for (int j = 0; j < columns_a + columns_h - 1; j++) { cout << fixed << setprecision(3) << h_c[(i*(columns_h+columns_a-1))+j] << " "; } cout << endl; } // Free device global memory err = hipFree(h_a); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector a (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipFree(h_h); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector h (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } err = hipFree(h_c); if (err != hipSuccess) { fprintf(stderr, "Failed to free device vector c (error code %s)!\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } // Reset the device and exit // cudaDeviceReset causes the driver to clean up all state. While // not mandatory in normal operation, it is good practice. It is also // needed to ensure correct operation when the application is being // profiled. Calling cudaDeviceReset causes all profile data to be // flushed before the application exits err = hipDeviceReset(); if (err != hipSuccess) { fprintf(stderr, "Failed to deinitialize the device! error=%s\n", hipGetErrorString(err)); exit(EXIT_FAILURE); } printf("Done\n"); return 0; }
.text .file "exec2.hip" # Start of file scope inline assembly .globl _ZSt21ios_base_library_initv # End of file scope inline assembly .globl _Z28__device_stub__convolution2DPdS_S_iiii # -- Begin function _Z28__device_stub__convolution2DPdS_S_iiii .p2align 4, 0x90 .type _Z28__device_stub__convolution2DPdS_S_iiii,@function _Z28__device_stub__convolution2DPdS_S_iiii: # @_Z28__device_stub__convolution2DPdS_S_iiii .cfi_startproc # %bb.0: subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 20(%rsp), %rax movq %rax, 120(%rsp) leaq 16(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%rsp) leaq 160(%rsp), %rax movq %rax, 144(%rsp) leaq 56(%rsp), %rdi leaq 40(%rsp), %rsi leaq 32(%rsp), %rdx leaq 24(%rsp), %rcx callq __hipPopCallConfiguration movq 56(%rsp), %rsi movl 64(%rsp), %edx movq 40(%rsp), %rcx movl 48(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z13convolution2DPdS_S_iiii, %edi pushq 24(%rsp) .cfi_adjust_cfa_offset 8 pushq 40(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $168, %rsp .cfi_adjust_cfa_offset -168 retq .Lfunc_end0: .size _Z28__device_stub__convolution2DPdS_S_iiii, .Lfunc_end0-_Z28__device_stub__convolution2DPdS_S_iiii .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .Lfunc_begin0: .cfi_startproc .cfi_personality 3, __gxx_personality_v0 .cfi_lsda 3, .Lexception0 # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 subq $1096, %rsp # imm = 0x448 .cfi_def_cfa_offset 1152 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 leaq 80(%rsp), %rax movq %rax, 64(%rsp) movq $0, 72(%rsp) movb $0, 80(%rsp) movq 8(%rsi), %rsi .Ltmp0: .cfi_escape 0x2e, 0x00 leaq 576(%rsp), %rdi movl $8, %edx callq _ZNSt14basic_ifstreamIcSt11char_traitsIcEEC1EPKcSt13_Ios_Openmode .Ltmp1: # %bb.1: leaq 696(%rsp), %rdi .cfi_escape 0x2e, 0x00 callq _ZNKSt12__basic_fileIcE7is_openEv xorl %ebp, %ebp movl $0, %ebx movl $0, %ecx movq %rcx, (%rsp) # 8-byte Spill movl $0, %ecx movq %rcx, 8(%rsp) # 8-byte Spill testb %al, %al je .LBB1_47 # %bb.2: # %.preheader218 movq 576(%rsp), %rax movq -24(%rax), %rax movq 816(%rsp,%rax), %r15 testq %r15, %r15 je .LBB1_30 # %bb.3: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i.lr.ph xorl %eax, %eax movq %rax, 8(%rsp) # 8-byte Spill movl $a, %r13d leaq 64(%rsp), %r12 leaq 176(%rsp), %rbp leaq 40(%rsp), %r14 xorl %ebx, %ebx jmp .LBB1_4 .p2align 4, 0x90 .LBB1_29: # %_ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEED1Ev.exit # in Loop: Header=BB1_4 Depth=1 incq 8(%rsp) # 8-byte Folded Spill movq $_ZTVSt15basic_streambufIcSt11char_traitsIcEE+16, 200(%rsp) .cfi_escape 0x2e, 0x00 leaq 256(%rsp), %rdi callq _ZNSt6localeD1Ev movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+16(%rip), %rax movq %rax, 176(%rsp) movq -24(%rax), %rax movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+24(%rip), %rcx movq %rcx, 176(%rsp,%rax) movq $0, 184(%rsp) .cfi_escape 0x2e, 0x00 leaq 304(%rsp), %rdi callq _ZNSt8ios_baseD2Ev movq 576(%rsp), %rax movq -24(%rax), %rax movq 816(%rsp,%rax), %r15 addq $8000, %r13 # imm = 0x1F40 testq %r15, %r15 je .LBB1_30 .LBB1_4: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i # =>This Loop Header: Depth=1 # Child Loop BB1_17 Depth 2 cmpb $0, 56(%r15) je .LBB1_6 # %bb.5: # in Loop: Header=BB1_4 Depth=1 movzbl 67(%r15), %eax jmp .LBB1_8 .p2align 4, 0x90 .LBB1_6: # in Loop: Header=BB1_4 Depth=1 .Ltmp3: .cfi_escape 0x2e, 0x00 movq %r15, %rdi callq _ZNKSt5ctypeIcE13_M_widen_initEv .Ltmp4: # %bb.7: # %.noexc149 # in Loop: Header=BB1_4 Depth=1 movq (%r15), %rax .Ltmp5: .cfi_escape 0x2e, 0x00 movq %r15, %rdi movl $10, %esi callq *48(%rax) .Ltmp6: .LBB1_8: # %_ZNKSt9basic_iosIcSt11char_traitsIcEE5widenEc.exit.i # in Loop: Header=BB1_4 Depth=1 .Ltmp7: .cfi_escape 0x2e, 0x00 movsbl %al, %edx leaq 576(%rsp), %rdi movq %r12, %rsi callq _ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EES4_ .Ltmp8: # %bb.9: # %_ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EE.exit # in Loop: Header=BB1_4 Depth=1 movq (%rax), %rcx movq -24(%rcx), %rcx testb $5, 32(%rax,%rcx) jne .LBB1_11 # %bb.10: # %_ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EE.exit # in Loop: Header=BB1_4 Depth=1 cmpq $0, 72(%rsp) je .LBB1_11 # %bb.15: # %.critedge202 # in Loop: Header=BB1_4 Depth=1 .Ltmp10: .cfi_escape 0x2e, 0x00 movq %rbp, %rdi movq %r12, %rsi movl $24, %edx callq _ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEC1ERKNS_12basic_stringIcS2_S3_EESt13_Ios_Openmode .Ltmp11: # %bb.16: # %.preheader217.preheader # in Loop: Header=BB1_4 Depth=1 movq %r13, %r15 .p2align 4, 0x90 .LBB1_17: # %.preheader217 # Parent Loop BB1_4 Depth=1 # => This Inner Loop Header: Depth=2 .Ltmp13: .cfi_escape 0x2e, 0x00 movq %rbp, %rdi movq %r14, %rsi callq _ZNSi10_M_extractIdEERSiRT_ .Ltmp14: # %bb.18: # %_ZNSirsERd.exit # in Loop: Header=BB1_17 Depth=2 movq (%rax), %rcx movq -24(%rcx), %rcx testb $5, 32(%rax,%rcx) jne .LBB1_27 # %bb.19: # in Loop: Header=BB1_17 Depth=2 movsd 40(%rsp), %xmm0 # xmm0 = mem[0],zero movsd %xmm0, (%r15) incl %ebx addq $8, %r15 jmp .LBB1_17 .p2align 4, 0x90 .LBB1_27: # in Loop: Header=BB1_4 Depth=1 movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, 176(%rsp) movq -24(%rax), %rax movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+64(%rip), %rcx movq %rcx, 176(%rsp,%rax) movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+72(%rip), %rax movq %rax, 192(%rsp) movq $_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE+16, 200(%rsp) movq 272(%rsp), %rdi leaq 288(%rsp), %rax cmpq %rax, %rdi je .LBB1_29 # %bb.28: # %.critedge.i.i.i.i.i # in Loop: Header=BB1_4 Depth=1 .cfi_escape 0x2e, 0x00 callq _ZdlPv jmp .LBB1_29 .LBB1_11: # %.critedge.preheader movq 576(%rsp), %rax movq -24(%rax), %rax movq 816(%rsp,%rax), %r12 testq %r12, %r12 je .LBB1_30 # %bb.12: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i155.lr.ph xorl %eax, %eax movq %rax, (%rsp) # 8-byte Spill movl $h, %r13d leaq 176(%rsp), %r14 leaq 40(%rsp), %r15 xorl %ebp, %ebp jmp .LBB1_13 .p2align 4, 0x90 .LBB1_46: # %_ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEED1Ev.exit172 # in Loop: Header=BB1_13 Depth=1 incq (%rsp) # 8-byte Folded Spill movq $_ZTVSt15basic_streambufIcSt11char_traitsIcEE+16, 200(%rsp) .cfi_escape 0x2e, 0x00 leaq 256(%rsp), %rdi callq _ZNSt6localeD1Ev movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+16(%rip), %rax movq %rax, 176(%rsp) movq -24(%rax), %rax movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+24(%rip), %rcx movq %rcx, 176(%rsp,%rax) movq $0, 184(%rsp) .cfi_escape 0x2e, 0x00 leaq 304(%rsp), %rdi callq _ZNSt8ios_baseD2Ev movq 576(%rsp), %rax movq -24(%rax), %rax movq 816(%rsp,%rax), %r12 addq $80, %r13 testq %r12, %r12 je .LBB1_30 .LBB1_13: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i155 # =>This Loop Header: Depth=1 # Child Loop BB1_39 Depth 2 cmpb $0, 56(%r12) je .LBB1_32 # %bb.14: # in Loop: Header=BB1_13 Depth=1 movzbl 67(%r12), %eax jmp .LBB1_34 .p2align 4, 0x90 .LBB1_32: # in Loop: Header=BB1_13 Depth=1 .Ltmp16: .cfi_escape 0x2e, 0x00 movq %r12, %rdi callq _ZNKSt5ctypeIcE13_M_widen_initEv .Ltmp17: # %bb.33: # %.noexc160 # in Loop: Header=BB1_13 Depth=1 movq (%r12), %rax .Ltmp18: .cfi_escape 0x2e, 0x00 movq %r12, %rdi movl $10, %esi callq *48(%rax) .Ltmp19: .LBB1_34: # %_ZNKSt9basic_iosIcSt11char_traitsIcEE5widenEc.exit.i157 # in Loop: Header=BB1_13 Depth=1 .Ltmp20: .cfi_escape 0x2e, 0x00 movsbl %al, %edx leaq 576(%rsp), %rdi leaq 64(%rsp), %rsi callq _ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EES4_ .Ltmp21: # %bb.35: # %_ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EE.exit163 # in Loop: Header=BB1_13 Depth=1 movq (%rax), %rcx movq -24(%rcx), %rcx testb $5, 32(%rax,%rcx) jne .LBB1_47 # %bb.36: # %_ZSt7getlineIcSt11char_traitsIcESaIcEERSt13basic_istreamIT_T0_ES7_RNSt7__cxx1112basic_stringIS4_S5_T1_EE.exit163 # in Loop: Header=BB1_13 Depth=1 cmpq $0, 72(%rsp) je .LBB1_47 # %bb.37: # %.critedge203 # in Loop: Header=BB1_13 Depth=1 .Ltmp23: .cfi_escape 0x2e, 0x00 movq %r14, %rdi leaq 64(%rsp), %rsi movl $24, %edx callq _ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEC1ERKNS_12basic_stringIcS2_S3_EESt13_Ios_Openmode .Ltmp24: # %bb.38: # %.preheader211.preheader # in Loop: Header=BB1_13 Depth=1 movq %r13, %r12 .p2align 4, 0x90 .LBB1_39: # %.preheader211 # Parent Loop BB1_13 Depth=1 # => This Inner Loop Header: Depth=2 .Ltmp26: .cfi_escape 0x2e, 0x00 movq %r14, %rdi movq %r15, %rsi callq _ZNSi10_M_extractIdEERSiRT_ .Ltmp27: # %bb.40: # %_ZNSirsERd.exit168 # in Loop: Header=BB1_39 Depth=2 movq (%rax), %rcx movq -24(%rcx), %rcx testb $5, 32(%rax,%rcx) jne .LBB1_44 # %bb.41: # in Loop: Header=BB1_39 Depth=2 movsd 40(%rsp), %xmm0 # xmm0 = mem[0],zero movsd %xmm0, (%r12) incl %ebp addq $8, %r12 jmp .LBB1_39 .p2align 4, 0x90 .LBB1_44: # in Loop: Header=BB1_13 Depth=1 movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE(%rip), %rax movq %rax, 176(%rsp) movq -24(%rax), %rax movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+64(%rip), %rcx movq %rcx, 176(%rsp,%rax) movq _ZTTNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEEE+72(%rip), %rax movq %rax, 192(%rsp) movq $_ZTVNSt7__cxx1115basic_stringbufIcSt11char_traitsIcESaIcEEE+16, 200(%rsp) movq 272(%rsp), %rdi leaq 288(%rsp), %rax cmpq %rax, %rdi je .LBB1_46 # %bb.45: # %.critedge.i.i.i.i.i170 # in Loop: Header=BB1_13 Depth=1 .cfi_escape 0x2e, 0x00 callq _ZdlPv jmp .LBB1_46 .LBB1_47: # %.critedge2 leaq 592(%rsp), %rdi .Ltmp29: .cfi_escape 0x2e, 0x00 callq _ZNSt13basic_filebufIcSt11char_traitsIcEE5closeEv .Ltmp30: # %bb.48: # %.noexc174 testq %rax, %rax jne .LBB1_50 # %bb.49: movq 576(%rsp), %rax movq -24(%rax), %rax leaq (%rsp,%rax), %rdi addq $576, %rdi # imm = 0x240 movl 608(%rsp,%rax), %esi orl $4, %esi .Ltmp31: .cfi_escape 0x2e, 0x00 callq _ZNSt9basic_iosIcSt11char_traitsIcEE5clearESt12_Ios_Iostate .Ltmp32: .LBB1_50: # %_ZNSt14basic_ifstreamIcSt11char_traitsIcEE5closeEv.exit movl %ebx, %eax cltd idivl 8(%rsp) # 4-byte Folded Reload movl %eax, %r14d movl %ebp, %eax cltd idivl (%rsp) # 4-byte Folded Reload movl %eax, %r15d movslq %ebx, %rsi shlq $3, %rsi .Ltmp33: .cfi_escape 0x2e, 0x00 leaq 32(%rsp), %rdi movl $1, %edx callq hipMallocManaged .Ltmp34: # %bb.51: # %_ZL16hipMallocManagedIdE10hipError_tPPT_mj.exit movslq %ebp, %rsi shlq $3, %rsi .Ltmp36: .cfi_escape 0x2e, 0x00 leaq 24(%rsp), %rdi movl $1, %edx callq hipMallocManaged .Ltmp37: # %bb.52: # %_ZL16hipMallocManagedIdE10hipError_tPPT_mj.exit178 .Ltmp39: movq 8(%rsp), %rbx # 8-byte Reload movq (%rsp), %r12 # 8-byte Reload leal (%rbx,%r12), %r13d decl %r13d leal (%r15,%r14), %ebp decl %ebp movl %ebp, %eax imull %r13d, %eax movslq %eax, %rsi shlq $3, %rsi .cfi_escape 0x2e, 0x00 leaq 16(%rsp), %rdi movl $1, %edx callq hipMallocManaged .Ltmp40: # %bb.53: # %_ZL16hipMallocManagedIdE10hipError_tPPT_mj.exit180.preheader movq %rbx, %rcx addl %r12d, %ebx leal (%r15,%r14), %eax movq %rax, 96(%rsp) # 8-byte Spill movq 32(%rsp), %rax cmpl $1, %ecx # kill: def $ecx killed $ecx killed $rcx def $rcx adcl $0, %ecx movl %r14d, %edx xorl %esi, %esi movl $a, %edi xorl %r8d, %r8d jmp .LBB1_54 .p2align 4, 0x90 .LBB1_57: # %_ZL16hipMallocManagedIdE10hipError_tPPT_mj.exit180 # in Loop: Header=BB1_54 Depth=1 incq %r8 addl %r14d, %esi addq $8000, %rdi # imm = 0x1F40 cmpq %rcx, %r8 je .LBB1_58 .LBB1_54: # %.preheader210 # =>This Loop Header: Depth=1 # Child Loop BB1_56 Depth 2 testl %r14d, %r14d jle .LBB1_57 # %bb.55: # %.lr.ph # in Loop: Header=BB1_54 Depth=1 movl %esi, %r9d leaq (%rax,%r9,8), %r9 xorl %r10d, %r10d .p2align 4, 0x90 .LBB1_56: # Parent Loop BB1_54 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%rdi,%r10,8), %xmm0 # xmm0 = mem[0],zero movsd %xmm0, (%r9,%r10,8) incq %r10 cmpq %r10, %rdx jne .LBB1_56 jmp .LBB1_57 .LBB1_58: # %.preheader209 movq 24(%rsp), %rax movq (%rsp), %rcx # 8-byte Reload cmpl $1, %ecx # kill: def $ecx killed $ecx killed $rcx def $rcx adcl $0, %ecx movl %r15d, %edx xorl %esi, %esi movl $h, %edi xorl %r8d, %r8d jmp .LBB1_59 .p2align 4, 0x90 .LBB1_62: # %._crit_edge257 # in Loop: Header=BB1_59 Depth=1 incq %r8 addl %r15d, %esi addq $80, %rdi cmpq %rcx, %r8 je .LBB1_63 .LBB1_59: # %.preheader208 # =>This Loop Header: Depth=1 # Child Loop BB1_61 Depth 2 testl %r15d, %r15d jle .LBB1_62 # %bb.60: # %.lr.ph256 # in Loop: Header=BB1_59 Depth=1 movl %esi, %r9d leaq (%rax,%r9,8), %r9 xorl %r10d, %r10d .p2align 4, 0x90 .LBB1_61: # Parent Loop BB1_59 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%rdi,%r10,8), %xmm0 # xmm0 = mem[0],zero movsd %xmm0, (%r9,%r10,8) incq %r10 cmpq %r10, %rdx jne .LBB1_61 jmp .LBB1_62 .LBB1_63: leal -2(%rbx), %eax leal 29(%rbx), %ecx testl %eax, %eax cmovnsl %eax, %ecx sarl $5, %ecx incl %ecx movq 96(%rsp), %rdx # 8-byte Reload leal -2(%rdx), %eax leal 29(%rdx), %edi testl %eax, %eax cmovnsl %eax, %edi sarl $5, %edi incl %edi shlq $32, %rdi orq %rcx, %rdi .Ltmp42: .cfi_escape 0x2e, 0x00 movabsq $137438953504, %rdx # imm = 0x2000000020 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration .Ltmp43: # %bb.64: testl %eax, %eax jne .LBB1_67 # %bb.65: movq 32(%rsp), %rax movq 24(%rsp), %rcx movq 16(%rsp), %rdx movq %rax, 168(%rsp) movq %rcx, 160(%rsp) movq %rdx, 152(%rsp) movq 8(%rsp), %rax # 8-byte Reload movl %eax, 60(%rsp) movl %r14d, 56(%rsp) movq (%rsp), %rax # 8-byte Reload movl %eax, 52(%rsp) movl %r15d, 48(%rsp) leaq 168(%rsp), %rax movq %rax, 176(%rsp) leaq 160(%rsp), %rax movq %rax, 184(%rsp) leaq 152(%rsp), %rax movq %rax, 192(%rsp) leaq 60(%rsp), %rax movq %rax, 200(%rsp) leaq 56(%rsp), %rax movq %rax, 208(%rsp) leaq 52(%rsp), %rax movq %rax, 216(%rsp) leaq 48(%rsp), %rax movq %rax, 224(%rsp) .Ltmp44: .cfi_escape 0x2e, 0x00 leaq 136(%rsp), %rdi leaq 120(%rsp), %rsi leaq 112(%rsp), %rdx leaq 104(%rsp), %rcx callq __hipPopCallConfiguration .Ltmp45: # %bb.66: # %.noexc181 movq 136(%rsp), %rsi movl 144(%rsp), %edx movq 120(%rsp), %rcx movl 128(%rsp), %r8d .Ltmp46: .cfi_escape 0x2e, 0x10 leaq 176(%rsp), %r9 movl $_Z13convolution2DPdS_S_iiii, %edi pushq 104(%rsp) .cfi_adjust_cfa_offset 8 pushq 120(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .Ltmp47: .LBB1_67: .Ltmp48: .cfi_escape 0x2e, 0x00 callq hipGetLastError .Ltmp49: # %bb.68: testl %eax, %eax jne .LBB1_69 # %bb.74: .Ltmp52: .cfi_escape 0x2e, 0x00 callq hipDeviceSynchronize .Ltmp53: # %bb.75: testl %eax, %eax jne .LBB1_81 # %bb.76: # %.preheader207 cmpl $2, %ebx jb .LBB1_91 # %bb.77: # %.preheader.lr.ph cmpl $2, %ebp movl $1, %ecx movl $1, %eax cmovgel %ebp, %eax movq %rax, (%rsp) # 8-byte Spill movl %r13d, %eax movslq %ebp, %r13 cmpl $2, %eax cmovgel %eax, %ecx movq %rcx, 8(%rsp) # 8-byte Spill shlq $3, %r13 xorl %r15d, %r15d movl $-261, %r12d # imm = 0xFEFB xorl %r14d, %r14d .p2align 4, 0x90 .LBB1_78: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB1_95 Depth 2 movq (%rsp), %rbx # 8-byte Reload movq %r15, %rbp cmpl $2, 96(%rsp) # 4-byte Folded Reload jl .LBB1_79 .p2align 4, 0x90 .LBB1_95: # %_ZNSolsEPFRSt8ios_baseS0_E.exit # Parent Loop BB1_78 Depth=1 # => This Inner Loop Header: Depth=2 movq _ZSt4cout(%rip), %rax movq -24(%rax), %rcx movl _ZSt4cout+24(%rcx), %edx andl %r12d, %edx orl $4, %edx movl %edx, _ZSt4cout+24(%rcx) movq -24(%rax), %rax movq $3, _ZSt4cout+8(%rax) movq 16(%rsp), %rax movsd (%rax,%rbp), %xmm0 # xmm0 = mem[0],zero .Ltmp56: .cfi_escape 0x2e, 0x00 movl $_ZSt4cout, %edi callq _ZNSo9_M_insertIdEERSoT_ .Ltmp57: # %bb.96: # %_ZNSolsEd.exit # in Loop: Header=BB1_95 Depth=2 .Ltmp58: .cfi_escape 0x2e, 0x00 movl $.L.str.3, %esi movl $1, %edx movq %rax, %rdi callq _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l .Ltmp59: # %bb.97: # %_ZStlsISt11char_traitsIcEERSt13basic_ostreamIcT_ES5_PKc.exit # in Loop: Header=BB1_95 Depth=2 addq $8, %rbp decq %rbx jne .LBB1_95 .LBB1_79: # %._crit_edge262 # in Loop: Header=BB1_78 Depth=1 movq _ZSt4cout(%rip), %rax movq -24(%rax), %rax movq _ZSt4cout+240(%rax), %rbx testq %rbx, %rbx je .LBB1_80 # %bb.84: # %_ZSt13__check_facetISt5ctypeIcEERKT_PS3_.exit.i.i191 # in Loop: Header=BB1_78 Depth=1 cmpb $0, 56(%rbx) je .LBB1_86 # %bb.85: # in Loop: Header=BB1_78 Depth=1 movzbl 67(%rbx), %eax jmp .LBB1_88 .p2align 4, 0x90 .LBB1_86: # in Loop: Header=BB1_78 Depth=1 .Ltmp61: .cfi_escape 0x2e, 0x00 movq %rbx, %rdi callq _ZNKSt5ctypeIcE13_M_widen_initEv .Ltmp62: # %bb.87: # %.noexc196 # in Loop: Header=BB1_78 Depth=1 movq (%rbx), %rax .Ltmp63: .cfi_escape 0x2e, 0x00 movq %rbx, %rdi movl $10, %esi callq *48(%rax) .Ltmp64: .LBB1_88: # %_ZNKSt9basic_iosIcSt11char_traitsIcEE5widenEc.exit.i193 # in Loop: Header=BB1_78 Depth=1 .Ltmp65: .cfi_escape 0x2e, 0x00 movsbl %al, %esi movl $_ZSt4cout, %edi callq _ZNSo3putEc .Ltmp66: # %bb.89: # %.noexc198 # in Loop: Header=BB1_78 Depth=1 .Ltmp67: .cfi_escape 0x2e, 0x00 movq %rax, %rdi callq _ZNSo5flushEv .Ltmp68: # %bb.90: # %_ZNSolsEPFRSoS_E.exit # in Loop: Header=BB1_78 Depth=1 incq %r14 addq %r13, %r15 cmpq 8(%rsp), %r14 # 8-byte Folded Reload jne .LBB1_78 .LBB1_91: # %._crit_edge265 movq 32(%rsp), %rdi .Ltmp70: .cfi_escape 0x2e, 0x00 callq hipFree .Ltmp71: # %bb.92: testl %eax, %eax jne .LBB1_93 # %bb.101: movq 24(%rsp), %rdi .Ltmp74: .cfi_escape 0x2e, 0x00 callq hipFree .Ltmp75: # %bb.102: testl %eax, %eax jne .LBB1_103 # %bb.105: movq 16(%rsp), %rdi .Ltmp78: .cfi_escape 0x2e, 0x00 callq hipFree .Ltmp79: # %bb.106: testl %eax, %eax jne .LBB1_107 # %bb.109: .Ltmp82: .cfi_escape 0x2e, 0x00 callq hipDeviceReset .Ltmp83: # %bb.110: testl %eax, %eax jne .LBB1_111 # %bb.113: .cfi_escape 0x2e, 0x00 movl $.Lstr, %edi callq puts@PLT .cfi_escape 0x2e, 0x00 leaq 576(%rsp), %rdi movl $_ZTTSt14basic_ifstreamIcSt11char_traitsIcEE, %esi callq _ZNSt14basic_ifstreamIcSt11char_traitsIcEED2Ev leaq 832(%rsp), %rdi .cfi_escape 0x2e, 0x00 callq _ZNSt8ios_baseD2Ev movq 64(%rsp), %rdi leaq 80(%rsp), %rax cmpq %rax, %rdi je .LBB1_115 # %bb.114: # %.critedge.i.i .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB1_115: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit xorl %eax, %eax addq $1096, %rsp # imm = 0x448 .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .LBB1_30: # %.critedge._crit_edge.invoke .cfi_def_cfa_offset 1152 .Ltmp90: .cfi_escape 0x2e, 0x00 callq _ZSt16__throw_bad_castv .Ltmp91: # %bb.31: # %.critedge._crit_edge.cont .LBB1_80: .Ltmp87: .cfi_escape 0x2e, 0x00 callq _ZSt16__throw_bad_castv .Ltmp88: # %bb.83: # %.noexc195 .LBB1_69: movq stderr(%rip), %rbx .Ltmp50: .cfi_escape 0x2e, 0x00 movl %eax, %edi callq hipGetErrorString .Ltmp51: # %bb.70: .cfi_escape 0x2e, 0x00 movl $.L.str.1, %esi jmp .LBB1_71 .LBB1_81: movq stderr(%rip), %rbx .Ltmp54: .cfi_escape 0x2e, 0x00 movl %eax, %edi callq hipGetErrorString .Ltmp55: # %bb.82: .cfi_escape 0x2e, 0x00 movl $.L.str.2, %esi jmp .LBB1_71 .LBB1_93: movq stderr(%rip), %rbx .Ltmp72: .cfi_escape 0x2e, 0x00 movl %eax, %edi callq hipGetErrorString .Ltmp73: # %bb.94: .cfi_escape 0x2e, 0x00 movl $.L.str.4, %esi jmp .LBB1_71 .LBB1_103: movq stderr(%rip), %rbx .Ltmp76: .cfi_escape 0x2e, 0x00 movl %eax, %edi callq hipGetErrorString .Ltmp77: # %bb.104: .cfi_escape 0x2e, 0x00 movl $.L.str.5, %esi jmp .LBB1_71 .LBB1_107: movq stderr(%rip), %rbx .Ltmp80: .cfi_escape 0x2e, 0x00 movl %eax, %edi callq hipGetErrorString .Ltmp81: # %bb.108: .cfi_escape 0x2e, 0x00 movl $.L.str.6, %esi jmp .LBB1_71 .LBB1_111: movq stderr(%rip), %rbx .Ltmp84: .cfi_escape 0x2e, 0x00 movl %eax, %edi callq hipGetErrorString .Ltmp85: # %bb.112: .cfi_escape 0x2e, 0x00 movl $.L.str.7, %esi .LBB1_71: movq %rbx, %rdi movq %rax, %rdx xorl %eax, %eax callq fprintf .cfi_escape 0x2e, 0x00 movl $1, %edi callq exit .LBB1_116: .Ltmp41: jmp .LBB1_117 .LBB1_73: .Ltmp38: jmp .LBB1_117 .LBB1_72: .Ltmp35: jmp .LBB1_117 .LBB1_20: .Ltmp2: movq %rax, %rbx jmp .LBB1_119 .LBB1_122: .Ltmp86: jmp .LBB1_117 .LBB1_42: .Ltmp25: jmp .LBB1_117 .LBB1_24: .Ltmp12: jmp .LBB1_117 .LBB1_100: # %.loopexit.split-lp .Ltmp89: jmp .LBB1_117 .LBB1_21: # %.loopexit212 .Ltmp22: jmp .LBB1_117 .LBB1_22: # %.loopexit.split-lp213.loopexit .Ltmp9: jmp .LBB1_117 .LBB1_23: # %.loopexit.split-lp213.loopexit.split-lp .Ltmp92: jmp .LBB1_117 .LBB1_99: # %.loopexit .Ltmp69: jmp .LBB1_117 .LBB1_43: .Ltmp28: jmp .LBB1_26 .LBB1_25: .Ltmp15: .LBB1_26: # %.loopexit.split-lp213 movq %rax, %rbx .cfi_escape 0x2e, 0x00 leaq 176(%rsp), %rdi callq _ZNSt7__cxx1118basic_stringstreamIcSt11char_traitsIcESaIcEED1Ev jmp .LBB1_118 .LBB1_98: .Ltmp60: .LBB1_117: # %.loopexit.split-lp213 movq %rax, %rbx .LBB1_118: # %.loopexit.split-lp213 .cfi_escape 0x2e, 0x00 leaq 576(%rsp), %rdi movl $_ZTTSt14basic_ifstreamIcSt11char_traitsIcEE, %esi callq _ZNSt14basic_ifstreamIcSt11char_traitsIcEED2Ev leaq 832(%rsp), %rdi .cfi_escape 0x2e, 0x00 callq _ZNSt8ios_baseD2Ev .LBB1_119: movq 64(%rsp), %rdi leaq 80(%rsp), %rax cmpq %rax, %rdi je .LBB1_121 # %bb.120: # %.critedge.i.i187 .cfi_escape 0x2e, 0x00 callq _ZdlPv .LBB1_121: # %_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEED2Ev.exit189 .cfi_escape 0x2e, 0x00 movq %rbx, %rdi callq _Unwind_Resume@PLT .Lfunc_end1: .size main, .Lfunc_end1-main .cfi_endproc .section .gcc_except_table,"a",@progbits .p2align 2, 0x0 GCC_except_table1: .Lexception0: .byte 255 # @LPStart Encoding = omit .byte 255 # @TType Encoding = omit .byte 1 # Call site Encoding = uleb128 .uleb128 .Lcst_end0-.Lcst_begin0 .Lcst_begin0: .uleb128 .Ltmp0-.Lfunc_begin0 # >> Call Site 1 << .uleb128 .Ltmp1-.Ltmp0 # Call between .Ltmp0 and .Ltmp1 .uleb128 .Ltmp2-.Lfunc_begin0 # jumps to .Ltmp2 .byte 0 # On action: cleanup .uleb128 .Ltmp3-.Lfunc_begin0 # >> Call Site 2 << .uleb128 .Ltmp8-.Ltmp3 # Call between .Ltmp3 and .Ltmp8 .uleb128 .Ltmp9-.Lfunc_begin0 # jumps to .Ltmp9 .byte 0 # On action: cleanup .uleb128 .Ltmp10-.Lfunc_begin0 # >> Call Site 3 << .uleb128 .Ltmp11-.Ltmp10 # Call between .Ltmp10 and .Ltmp11 .uleb128 .Ltmp12-.Lfunc_begin0 # jumps to .Ltmp12 .byte 0 # On action: cleanup .uleb128 .Ltmp13-.Lfunc_begin0 # >> Call Site 4 << .uleb128 .Ltmp14-.Ltmp13 # Call between .Ltmp13 and .Ltmp14 .uleb128 .Ltmp15-.Lfunc_begin0 # jumps to .Ltmp15 .byte 0 # On action: cleanup .uleb128 .Ltmp16-.Lfunc_begin0 # >> Call Site 5 << .uleb128 .Ltmp21-.Ltmp16 # Call between .Ltmp16 and .Ltmp21 .uleb128 .Ltmp22-.Lfunc_begin0 # jumps to .Ltmp22 .byte 0 # On action: cleanup .uleb128 .Ltmp23-.Lfunc_begin0 # >> Call Site 6 << .uleb128 .Ltmp24-.Ltmp23 # Call between .Ltmp23 and .Ltmp24 .uleb128 .Ltmp25-.Lfunc_begin0 # jumps to .Ltmp25 .byte 0 # On action: cleanup .uleb128 .Ltmp26-.Lfunc_begin0 # >> Call Site 7 << .uleb128 .Ltmp27-.Ltmp26 # Call between .Ltmp26 and .Ltmp27 .uleb128 .Ltmp28-.Lfunc_begin0 # jumps to .Ltmp28 .byte 0 # On action: cleanup .uleb128 .Ltmp29-.Lfunc_begin0 # >> Call Site 8 << .uleb128 .Ltmp32-.Ltmp29 # Call between .Ltmp29 and .Ltmp32 .uleb128 .Ltmp92-.Lfunc_begin0 # jumps to .Ltmp92 .byte 0 # On action: cleanup .uleb128 .Ltmp33-.Lfunc_begin0 # >> Call Site 9 << .uleb128 .Ltmp34-.Ltmp33 # Call between .Ltmp33 and .Ltmp34 .uleb128 .Ltmp35-.Lfunc_begin0 # jumps to .Ltmp35 .byte 0 # On action: cleanup .uleb128 .Ltmp36-.Lfunc_begin0 # >> Call Site 10 << .uleb128 .Ltmp37-.Ltmp36 # Call between .Ltmp36 and .Ltmp37 .uleb128 .Ltmp38-.Lfunc_begin0 # jumps to .Ltmp38 .byte 0 # On action: cleanup .uleb128 .Ltmp39-.Lfunc_begin0 # >> Call Site 11 << .uleb128 .Ltmp40-.Ltmp39 # Call between .Ltmp39 and .Ltmp40 .uleb128 .Ltmp41-.Lfunc_begin0 # jumps to .Ltmp41 .byte 0 # On action: cleanup .uleb128 .Ltmp42-.Lfunc_begin0 # >> Call Site 12 << .uleb128 .Ltmp53-.Ltmp42 # Call between .Ltmp42 and .Ltmp53 .uleb128 .Ltmp86-.Lfunc_begin0 # jumps to .Ltmp86 .byte 0 # On action: cleanup .uleb128 .Ltmp56-.Lfunc_begin0 # >> Call Site 13 << .uleb128 .Ltmp59-.Ltmp56 # Call between .Ltmp56 and .Ltmp59 .uleb128 .Ltmp60-.Lfunc_begin0 # jumps to .Ltmp60 .byte 0 # On action: cleanup .uleb128 .Ltmp61-.Lfunc_begin0 # >> Call Site 14 << .uleb128 .Ltmp68-.Ltmp61 # Call between .Ltmp61 and .Ltmp68 .uleb128 .Ltmp69-.Lfunc_begin0 # jumps to .Ltmp69 .byte 0 # On action: cleanup .uleb128 .Ltmp70-.Lfunc_begin0 # >> Call Site 15 << .uleb128 .Ltmp83-.Ltmp70 # Call between .Ltmp70 and .Ltmp83 .uleb128 .Ltmp86-.Lfunc_begin0 # jumps to .Ltmp86 .byte 0 # On action: cleanup .uleb128 .Ltmp90-.Lfunc_begin0 # >> Call Site 16 << .uleb128 .Ltmp91-.Ltmp90 # Call between .Ltmp90 and .Ltmp91 .uleb128 .Ltmp92-.Lfunc_begin0 # jumps to .Ltmp92 .byte 0 # On action: cleanup .uleb128 .Ltmp87-.Lfunc_begin0 # >> Call Site 17 << .uleb128 .Ltmp88-.Ltmp87 # Call between .Ltmp87 and .Ltmp88 .uleb128 .Ltmp89-.Lfunc_begin0 # jumps to .Ltmp89 .byte 0 # On action: cleanup .uleb128 .Ltmp50-.Lfunc_begin0 # >> Call Site 18 << .uleb128 .Ltmp85-.Ltmp50 # Call between .Ltmp50 and .Ltmp85 .uleb128 .Ltmp86-.Lfunc_begin0 # jumps to .Ltmp86 .byte 0 # On action: cleanup .uleb128 .Ltmp85-.Lfunc_begin0 # >> Call Site 19 << .uleb128 .Lfunc_end1-.Ltmp85 # Call between .Ltmp85 and .Lfunc_end1 .byte 0 # has no landing pad .byte 0 # On action: cleanup .Lcst_end0: .p2align 2, 0x0 # -- End function .text .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB2_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB2_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z13convolution2DPdS_S_iiii, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end2: .size __hip_module_ctor, .Lfunc_end2-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB3_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB3_2: retq .Lfunc_end3: .size __hip_module_dtor, .Lfunc_end3-__hip_module_dtor .cfi_endproc # -- End function .type a,@object # @a .bss .globl a .p2align 4, 0x0 a: .zero 8000000 .size a, 8000000 .type h,@object # @h .globl h .p2align 4, 0x0 h: .zero 800 .size h, 800 .type c,@object # @c .globl c .p2align 4, 0x0 c: .zero 8000000 .size c, 8000000 .type _Z13convolution2DPdS_S_iiii,@object # @_Z13convolution2DPdS_S_iiii .section .rodata,"a",@progbits .globl _Z13convolution2DPdS_S_iiii .p2align 3, 0x0 _Z13convolution2DPdS_S_iiii: .quad _Z28__device_stub__convolution2DPdS_S_iiii .size _Z13convolution2DPdS_S_iiii, 8 .type .L.str.1,@object # @.str.1 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.1: .asciz "Failed to launch convolution2D kernel (error code %s)!\n" .size .L.str.1, 56 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "Failed to synchronize (error code %s)!\n" .size .L.str.2, 40 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz " " .size .L.str.3, 2 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "Failed to free device vector a (error code %s)!\n" .size .L.str.4, 49 .type .L.str.5,@object # @.str.5 .L.str.5: .asciz "Failed to free device vector h (error code %s)!\n" .size .L.str.5, 49 .type .L.str.6,@object # @.str.6 .L.str.6: .asciz "Failed to free device vector c (error code %s)!\n" .size .L.str.6, 49 .type .L.str.7,@object # @.str.7 .L.str.7: .asciz "Failed to deinitialize the device! error=%s\n" .size .L.str.7, 45 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z13convolution2DPdS_S_iiii" .size .L__unnamed_1, 28 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "Done" .size .Lstr, 5 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z28__device_stub__convolution2DPdS_S_iiii .addrsig_sym __gxx_personality_v0 .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Unwind_Resume .addrsig_sym _Z13convolution2DPdS_S_iiii .addrsig_sym _ZSt4cout .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z13convolution2DPdS_S_iiii .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R0, SR_CTAID.X ; /* 0x0000000000007919 */ /* 0x000e220000002500 */ /*0020*/ ULDC.64 UR6, c[0x0][0x178] ; /* 0x00005e0000067ab9 */ /* 0x000fe40000000a00 */ /*0030*/ ULDC.64 UR8, c[0x0][0x180] ; /* 0x0000600000087ab9 */ /* 0x000fe20000000a00 */ /*0040*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e220000002100 */ /*0050*/ UIADD3 UR4, UR7, UR9, URZ ; /* 0x0000000907047290 */ /* 0x000fe4000fffe03f */ /*0060*/ UIADD3 UR5, UR6, UR8, URZ ; /* 0x0000000806057290 */ /* 0x000fe2000fffe03f */ /*0070*/ S2R R21, SR_CTAID.Y ; /* 0x0000000000157919 */ /* 0x000e620000002600 */ /*0080*/ UIADD3 UR4, UR4, -0x1, URZ ; /* 0xffffffff04047890 */ /* 0x000fe4000fffe03f */ /*0090*/ UIADD3 UR5, UR5, -0x1, URZ ; /* 0xffffffff05057890 */ /* 0x000fe2000fffe03f */ /*00a0*/ S2R R2, SR_TID.Y ; /* 0x0000000000027919 */ /* 0x000e620000002200 */ /*00b0*/ IMAD R0, R0, c[0x0][0x0], R3 ; /* 0x0000000000007a24 */ /* 0x001fca00078e0203 */ /*00c0*/ ISETP.GE.AND P0, PT, R0, UR4, PT ; /* 0x0000000400007c0c */ /* 0x000fe2000bf06270 */ /*00d0*/ IMAD R21, R21, c[0x0][0x4], R2 ; /* 0x0000010015157a24 */ /* 0x002fca00078e0202 */ /*00e0*/ ISETP.GE.OR P0, PT, R21, UR5, P0 ; /* 0x0000000515007c0c */ /* 0x000fda0008706670 */ /*00f0*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0100*/ IMAD.MOV.U32 R2, RZ, RZ, 0x1 ; /* 0x00000001ff027424 */ /* 0x000fe200078e00ff */ /*0110*/ ULDC.64 UR6, c[0x0][0x118] ; /* 0x0000460000067ab9 */ /* 0x000fc80000000a00 */ /*0120*/ ISETP.LE.AND P0, PT, R2, c[0x0][0x184], PT ; /* 0x0000610002007a0c */ /* 0x000fc80003f03270 */ /*0130*/ ISETP.GT.OR P0, PT, R2, c[0x0][0x180], !P0 ; /* 0x0000600002007a0c */ /* 0x000fe40004704670 */ /*0140*/ CS2R R2, SRZ ; /* 0x0000000000027805 */ /* 0x000fd6000001ff00 */ /*0150*/ @P0 BRA 0x7a0 ; /* 0x0000064000000947 */ /* 0x000fea0003800000 */ /*0160*/ IMAD.MOV.U32 R2, RZ, RZ, 0x1 ; /* 0x00000001ff027424 */ /* 0x000fe400078e00ff */ /*0170*/ IMAD.MOV.U32 R20, RZ, RZ, c[0x0][0x184] ; /* 0x00006100ff147624 */ /* 0x000fe400078e00ff */ /*0180*/ IMAD.MOV.U32 R22, RZ, RZ, RZ ; /* 0x000000ffff167224 */ /* 0x000fe200078e00ff */ /*0190*/ IADD3 R2, -R2, c[0x0][0x184], RZ ; /* 0x0000610002027a10 */ /* 0x000fe40007ffe1ff */ /*01a0*/ LOP3.LUT R20, R20, 0x3, RZ, 0xc0, !PT ; /* 0x0000000314147812 */ /* 0x000fe400078ec0ff */ /*01b0*/ ISETP.GE.U32.AND P3, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe40003f66070 */ /*01c0*/ CS2R R2, SRZ ; /* 0x0000000000027805 */ /* 0x000fe2000001ff00 */ /*01d0*/ IADD3 R23, -R20, c[0x0][0x184], RZ ; /* 0x0000610014177a10 */ /* 0x000fc40007ffe1ff */ /*01e0*/ IMAD.MOV.U32 R26, RZ, RZ, R22.reuse ; /* 0x000000ffff1a7224 */ /* 0x100fe200078e0016 */ /*01f0*/ ISETP.NE.AND P4, PT, R20, RZ, PT ; /* 0x000000ff1400720c */ /* 0x000fe20003f85270 */ /*0200*/ IMAD.IADD R24, R21, 0x1, -R22 ; /* 0x0000000115187824 */ /* 0x000fe200078e0a16 */ /*0210*/ IADD3 R22, R22, 0x1, RZ ; /* 0x0000000116167810 */ /* 0x000fe20007ffe0ff */ /*0220*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fc600078e00ff */ /*0230*/ ISETP.GE.AND P5, PT, R22, c[0x0][0x180], PT ; /* 0x0000600016007a0c */ /* 0x000fe20003fa6270 */ /*0240*/ @!P3 BRA 0x4f0 ; /* 0x000002a00000b947 */ /* 0x00ffee0003800000 */ /*0250*/ ISETP.GE.AND P6, PT, R24, RZ, PT ; /* 0x000000ff1800720c */ /* 0x000fe20003fc6270 */ /*0260*/ IMAD.MOV.U32 R25, RZ, RZ, RZ ; /* 0x000000ffff197224 */ /* 0x000fe400078e00ff */ /*0270*/ IMAD.MOV.U32 R27, RZ, RZ, R23 ; /* 0x000000ffff1b7224 */ /* 0x000fc600078e0017 */ /*0280*/ ISETP.LT.OR P0, PT, R0.reuse, R25, !P6 ; /* 0x000000190000720c */ /* 0x040fe20007701670 */ /*0290*/ IMAD.IADD R5, R0, 0x1, -R25 ; /* 0x0000000100057824 */ /* 0x000fe400078e0a19 */ /*02a0*/ IMAD.MOV.U32 R17, RZ, RZ, 0x8 ; /* 0x00000008ff117424 */ /* 0x000fe200078e00ff */ /*02b0*/ ISETP.GE.OR P0, PT, R24.reuse, c[0x0][0x178], P0 ; /* 0x00005e0018007a0c */ /* 0x040fe20000706670 */ /*02c0*/ IMAD R12, R24, c[0x0][0x17c], R5 ; /* 0x00005f00180c7a24 */ /* 0x000fe400078e0205 */ /*02d0*/ IMAD R16, R26, c[0x0][0x184], R25 ; /* 0x000061001a107a24 */ /* 0x000fe200078e0219 */ /*02e0*/ ISETP.GE.OR P2, PT, R5, c[0x0][0x17c], P0 ; /* 0x00005f0005007a0c */ /* 0x000fe20000746670 */ /*02f0*/ IMAD.WIDE R12, R12, R17, c[0x0][0x160] ; /* 0x000058000c0c7625 */ /* 0x000fc800078e0211 */ /*0300*/ IMAD.WIDE R16, R16, R17, c[0x0][0x168] ; /* 0x00005a0010107625 */ /* 0x000fd000078e0211 */ /*0310*/ @!P2 LDG.E.64 R6, [R12.64] ; /* 0x000000060c06a981 */ /* 0x000ea8000c1e1b00 */ /*0320*/ @!P2 LDG.E.64 R4, [R16.64] ; /* 0x000000061004a981 */ /* 0x000ea2000c1e1b00 */ /*0330*/ ISETP.LE.OR P0, PT, R0.reuse, R25, !P6 ; /* 0x000000190000720c */ /* 0x040fe40007703670 */ /*0340*/ IADD3 R8, R0, -0x1, -R25 ; /* 0xffffffff00087810 */ /* 0x000fe40007ffe819 */ /*0350*/ IADD3 R9, R25, 0x2, RZ ; /* 0x0000000219097810 */ /* 0x000fe40007ffe0ff */ /*0360*/ ISETP.GE.OR P0, PT, R24, c[0x0][0x178], P0 ; /* 0x00005e0018007a0c */ /* 0x000fc40000706670 */ /*0370*/ ISETP.LT.OR P1, PT, R0.reuse, R9, !P6 ; /* 0x000000090000720c */ /* 0x040fe20007721670 */ /*0380*/ IMAD.IADD R9, R0, 0x1, -R9 ; /* 0x0000000100097824 */ /* 0x000fe200078e0a09 */ /*0390*/ ISETP.GE.OR P0, PT, R8, c[0x0][0x17c], P0 ; /* 0x00005f0008007a0c */ /* 0x000fe40000706670 */ /*03a0*/ IADD3 R11, R25, 0x3, RZ ; /* 0x00000003190b7810 */ /* 0x000fe40007ffe0ff */ /*03b0*/ ISETP.GE.OR P1, PT, R24, c[0x0][0x178], P1 ; /* 0x00005e0018007a0c */ /* 0x000fc80000f26670 */ /*03c0*/ ISETP.GE.OR P1, PT, R9, c[0x0][0x17c], P1 ; /* 0x00005f0009007a0c */ /* 0x000fda0000f26670 */ /*03d0*/ @!P1 LDG.E.64 R8, [R16.64+0x10] ; /* 0x0000100610089981 */ /* 0x000ee2000c1e1b00 */ /*03e0*/ @!P2 DFMA R2, R6, R4, R2 ; /* 0x000000040602a22b */ /* 0x0070620000000002 */ /*03f0*/ ISETP.LT.OR P2, PT, R0.reuse, R11, !P6 ; /* 0x0000000b0000720c */ /* 0x040fe20007741670 */ /*0400*/ IMAD.IADD R11, R0, 0x1, -R11 ; /* 0x00000001000b7824 */ /* 0x000fe200078e0a0b */ /*0410*/ @!P0 LDG.E.64 R4, [R16.64+0x8] ; /* 0x0000080610048981 */ /* 0x001e64000c1e1b00 */ /*0420*/ ISETP.GE.OR P2, PT, R24, c[0x0][0x178], P2 ; /* 0x00005e0018007a0c */ /* 0x000fe40001746670 */ /*0430*/ @!P0 LDG.E.64 R6, [R12.64+-0x8] ; /* 0xfffff8060c068981 */ /* 0x000e64000c1e1b00 */ /*0440*/ ISETP.GE.OR P2, PT, R11, c[0x0][0x17c], P2 ; /* 0x00005f000b007a0c */ /* 0x000fc40001746670 */ /*0450*/ @!P1 LDG.E.64 R10, [R12.64+-0x10] ; /* 0xfffff0060c0a9981 */ /* 0x000ef6000c1e1b00 */ /*0460*/ @!P2 LDG.E.64 R14, [R16.64+0x18] ; /* 0x00001806100ea981 */ /* 0x000ea8000c1e1b00 */ /*0470*/ @!P2 LDG.E.64 R18, [R12.64+-0x18] ; /* 0xffffe8060c12a981 */ /* 0x000ea2000c1e1b00 */ /*0480*/ IADD3 R27, R27, -0x4, RZ ; /* 0xfffffffc1b1b7810 */ /* 0x000fc40007ffe0ff */ /*0490*/ IADD3 R25, R25, 0x4, RZ ; /* 0x0000000419197810 */ /* 0x000fe20007ffe0ff */ /*04a0*/ @!P0 DFMA R2, R4, R6, R2 ; /* 0x000000060402822b */ /* 0x002ee20000000002 */ /*04b0*/ ISETP.NE.AND P0, PT, R27, RZ, PT ; /* 0x000000ff1b00720c */ /* 0x000fca0003f05270 */ /*04c0*/ @!P1 DFMA R2, R8, R10, R2 ; /* 0x0000000a0802922b */ /* 0x008e8c0000000002 */ /*04d0*/ @!P2 DFMA R2, R14, R18, R2 ; /* 0x000000120e02a22b */ /* 0x0040640000000002 */ /*04e0*/ @P0 BRA 0x280 ; /* 0xfffffd9000000947 */ /* 0x000fea000383ffff */ /*04f0*/ @!P4 BRA 0x790 ; /* 0x000002900000c947 */ /* 0x000fea0003800000 */ /*0500*/ ISETP.GE.AND P0, PT, R24, RZ, PT ; /* 0x000000ff1800720c */ /* 0x000fe20003f06270 */ /*0510*/ IMAD.IADD R5, R0.reuse, 0x1, -R25.reuse ; /* 0x0000000100057824 */ /* 0x140fe200078e0a19 */ /*0520*/ BSSY B0, 0x610 ; /* 0x000000e000007945 */ /* 0x000fe20003800000 */ /*0530*/ IMAD.MOV.U32 R4, RZ, RZ, 0x8 ; /* 0x00000008ff047424 */ /* 0x000fe200078e00ff */ /*0540*/ ISETP.LT.OR P1, PT, R0, R25, !P0 ; /* 0x000000190000720c */ /* 0x000fe20004721670 */ /*0550*/ IMAD R6, R26, c[0x0][0x184], R25 ; /* 0x000061001a067a24 */ /* 0x000fe200078e0219 */ /*0560*/ ISETP.NE.AND P2, PT, R20, 0x1, PT ; /* 0x000000011400780c */ /* 0x000fe40003f45270 */ /*0570*/ ISETP.GE.OR P1, PT, R24, c[0x0][0x178], P1 ; /* 0x00005e0018007a0c */ /* 0x000fe20000f26670 */ /*0580*/ IMAD.WIDE R6, R6, R4, c[0x0][0x168] ; /* 0x00005a0006067625 */ /* 0x000fc600078e0204 */ /*0590*/ ISETP.GE.OR P1, PT, R5, c[0x0][0x17c], P1 ; /* 0x00005f0005007a0c */ /* 0x000fe20000f26670 */ /*05a0*/ IMAD R5, R24, c[0x0][0x17c], R5 ; /* 0x00005f0018057a24 */ /* 0x000fc800078e0205 */ /*05b0*/ IMAD.WIDE R4, R5, R4, c[0x0][0x160] ; /* 0x0000580005047625 */ /* 0x000fd000078e0204 */ /*05c0*/ @P1 BRA 0x600 ; /* 0x0000003000001947 */ /* 0x000fea0003800000 */ /*05d0*/ LDG.E.64 R8, [R4.64] ; /* 0x0000000604087981 */ /* 0x000ea8000c1e1b00 */ /*05e0*/ LDG.E.64 R10, [R6.64] ; /* 0x00000006060a7981 */ /* 0x000ea4000c1e1b00 */ /*05f0*/ DFMA R2, R8, R10, R2 ; /* 0x0000000a0802722b */ /* 0x00628c0000000002 */ /*0600*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0610*/ @!P2 BRA 0x790 ; /* 0x000001700000a947 */ /* 0x000fea0003800000 */ /*0620*/ ISETP.LE.OR P1, PT, R0.reuse, R25, !P0 ; /* 0x000000190000720c */ /* 0x040fe20004723670 */ /*0630*/ BSSY B0, 0x6d0 ; /* 0x0000009000007945 */ /* 0x000fe20003800000 */ /*0640*/ IADD3 R8, R0, -0x1, -R25 ; /* 0xffffffff00087810 */ /* 0x002fe40007ffe819 */ /*0650*/ ISETP.GE.OR P1, PT, R24, c[0x0][0x178], P1 ; /* 0x00005e0018007a0c */ /* 0x000fc40000f26670 */ /*0660*/ ISETP.NE.AND P2, PT, R20, 0x2, PT ; /* 0x000000021400780c */ /* 0x000fe40003f45270 */ /*0670*/ ISETP.GE.OR P1, PT, R8, c[0x0][0x17c], P1 ; /* 0x00005f0008007a0c */ /* 0x000fda0000f26670 */ /*0680*/ @P1 BRA 0x6c0 ; /* 0x0000003000001947 */ /* 0x000fea0003800000 */ /*0690*/ LDG.E.64 R8, [R6.64+0x8] ; /* 0x0000080606087981 */ /* 0x000ee8000c1e1b00 */ /*06a0*/ LDG.E.64 R10, [R4.64+-0x8] ; /* 0xfffff806040a7981 */ /* 0x000ee4000c1e1b00 */ /*06b0*/ DFMA R2, R8, R10, R2 ; /* 0x0000000a0802722b */ /* 0x00c28c0000000002 */ /*06c0*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*06d0*/ @!P2 BRA 0x790 ; /* 0x000000b00000a947 */ /* 0x000fea0003800000 */ /*06e0*/ IADD3 R25, R25, 0x2, RZ ; /* 0x0000000219197810 */ /* 0x000fe20007ffe0ff */ /*06f0*/ BSSY B0, 0x790 ; /* 0x0000009000007945 */ /* 0x000fe60003800000 */ /*0700*/ ISETP.LT.OR P0, PT, R0.reuse, R25, !P0 ; /* 0x000000190000720c */ /* 0x040fe20004701670 */ /*0710*/ IMAD.IADD R25, R0, 0x1, -R25 ; /* 0x0000000100197824 */ /* 0x000fc600078e0a19 */ /*0720*/ ISETP.GE.OR P0, PT, R24, c[0x0][0x178], P0 ; /* 0x00005e0018007a0c */ /* 0x000fc80000706670 */ /*0730*/ ISETP.GE.OR P0, PT, R25, c[0x0][0x17c], P0 ; /* 0x00005f0019007a0c */ /* 0x000fda0000706670 */ /*0740*/ @P0 BRA 0x780 ; /* 0x0000003000000947 */ /* 0x000fea0003800000 */ /*0750*/ LDG.E.64 R6, [R6.64+0x10] ; /* 0x0000100606067981 */ /* 0x000ee8000c1e1b00 */ /*0760*/ LDG.E.64 R4, [R4.64+-0x10] ; /* 0xfffff00604047981 */ /* 0x000ee4000c1e1b00 */ /*0770*/ DFMA R2, R6, R4, R2 ; /* 0x000000040602722b */ /* 0x00c4cc0000000002 */ /*0780*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0790*/ @!P5 BRA 0x1e0 ; /* 0xfffffa400000d947 */ /* 0x000fea000383ffff */ /*07a0*/ IMAD.MOV.U32 R5, RZ, RZ, 0x8 ; /* 0x00000008ff057424 */ /* 0x004fe400078e00ff */ /*07b0*/ IMAD R4, R21, UR4, R0 ; /* 0x0000000415047c24 */ /* 0x000fc8000f8e0200 */ /*07c0*/ IMAD.WIDE R4, R4, R5, c[0x0][0x170] ; /* 0x00005c0004047625 */ /* 0x000fca00078e0205 */ /*07d0*/ STG.E.64 [R4.64], R2 ; /* 0x0000000204007986 */ /* 0x00afe8000c101b06 */ /*07e0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*07f0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0800*/ BRA 0x800; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0810*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0820*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0830*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0840*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0850*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0860*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0870*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0880*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0890*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z13convolution2DPdS_S_iiii .globl _Z13convolution2DPdS_S_iiii .p2align 8 .type _Z13convolution2DPdS_S_iiii,@function _Z13convolution2DPdS_S_iiii: s_clause 0x2 s_load_b32 s4, s[0:1], 0x34 s_load_b32 s12, s[0:1], 0x1c s_load_b32 s13, s[0:1], 0x24 v_and_b32_e32 v3, 0x3ff, v0 s_add_u32 s2, s0, 40 s_addc_u32 s3, s1, 0 s_waitcnt lgkmcnt(0) s_and_b32 s4, s4, 0xffff s_delay_alu instid0(SALU_CYCLE_1) v_mad_u64_u32 v[1:2], null, s14, s4, v[3:4] s_add_i32 s5, s12, s13 s_mov_b32 s4, exec_lo s_add_i32 s5, s5, -1 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_cmpx_gt_i32_e64 s5, v1 s_cbranch_execz .LBB0_14 s_load_b32 s2, s[2:3], 0xc s_clause 0x1 s_load_b32 s14, s[0:1], 0x18 s_load_b32 s16, s[0:1], 0x20 v_bfe_u32 v0, v0, 10, 10 s_waitcnt lgkmcnt(0) s_lshr_b32 s2, s2, 16 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_mad_u64_u32 v[2:3], null, s15, s2, v[0:1] s_add_i32 s2, s14, s16 s_delay_alu instid0(SALU_CYCLE_1) s_add_i32 s2, s2, -1 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_cmp_gt_i32_e32 vcc_lo, s2, v2 s_and_b32 exec_lo, exec_lo, vcc_lo s_cbranch_execz .LBB0_14 s_cmp_lt_i32 s16, 1 s_cbranch_scc1 .LBB0_12 s_load_b128 s[8:11], s[0:1], 0x0 v_mul_lo_u32 v0, s12, v2 v_mov_b32_e32 v3, 0 v_mov_b32_e32 v4, 0 s_cmp_gt_i32 s13, 0 s_mov_b32 s7, 0 s_cselect_b32 s15, -1, 0 s_mov_b32 s17, 0 s_mov_b32 s18, 0 s_branch .LBB0_5 .LBB0_4: s_set_inst_prefetch_distance 0x2 v_subrev_nc_u32_e32 v0, s12, v0 s_add_i32 s18, s18, 1 s_add_i32 s17, s17, s13 s_cmp_eq_u32 s18, s16 s_cbranch_scc1 .LBB0_13 .LBB0_5: s_and_not1_b32 vcc_lo, exec_lo, s15 s_cbranch_vccnz .LBB0_4 v_subrev_nc_u32_e32 v5, s18, v2 s_mov_b32 s6, s17 s_mov_b32 s19, s13 s_delay_alu instid0(VALU_DEP_1) v_cmp_lt_i32_e32 vcc_lo, -1, v5 v_cmp_gt_i32_e64 s2, s14, v5 v_mov_b32_e32 v5, v1 s_set_inst_prefetch_distance 0x1 s_branch .LBB0_9 .p2align 6 .LBB0_7: s_or_b32 exec_lo, exec_lo, s4 .LBB0_8: s_delay_alu instid0(SALU_CYCLE_1) s_or_b32 exec_lo, exec_lo, s20 v_add_nc_u32_e32 v5, -1, v5 s_add_i32 s19, s19, -1 s_add_i32 s6, s6, 1 s_cmp_eq_u32 s19, 0 s_cbranch_scc1 .LBB0_4 .LBB0_9: s_and_saveexec_b32 s20, vcc_lo s_cbranch_execz .LBB0_8 v_cmp_lt_i32_e64 s3, -1, v5 v_cmp_gt_i32_e64 s4, s12, v5 s_delay_alu instid0(VALU_DEP_2) s_and_b32 s3, s2, s3 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) s_and_b32 s3, s4, s3 s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s4, s3 s_cbranch_execz .LBB0_7 v_add_nc_u32_e32 v6, v0, v5 s_lshl_b64 s[22:23], s[6:7], 3 s_waitcnt lgkmcnt(0) s_add_u32 s22, s10, s22 s_addc_u32 s23, s11, s23 v_ashrrev_i32_e32 v7, 31, v6 s_load_b64 s[22:23], s[22:23], 0x0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[6:7], 3, v[6:7] v_add_co_u32 v6, s3, s8, v6 s_delay_alu instid0(VALU_DEP_1) v_add_co_ci_u32_e64 v7, s3, s9, v7, s3 global_load_b64 v[6:7], v[6:7], off s_waitcnt vmcnt(0) lgkmcnt(0) v_fma_f64 v[3:4], v[6:7], s[22:23], v[3:4] s_branch .LBB0_7 .LBB0_12: v_mov_b32_e32 v3, 0 v_mov_b32_e32 v4, 0 .LBB0_13: s_load_b64 s[0:1], s[0:1], 0x10 v_mad_u64_u32 v[5:6], null, v2, s5, v[1:2] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v6, 31, v5 v_lshlrev_b64 v[0:1], 3, v[5:6] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v0, vcc_lo, s0, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo global_store_b64 v[0:1], v[3:4], off s_waitcnt_vscnt null, 0x0 s_barrier buffer_gl0_inv .LBB0_14: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z13convolution2DPdS_S_iiii .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 296 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 8 .amdhsa_next_free_sgpr 24 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z13convolution2DPdS_S_iiii, .Lfunc_end0-_Z13convolution2DPdS_S_iiii .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value - .offset: 28 .size: 4 .value_kind: by_value - .offset: 32 .size: 4 .value_kind: by_value - .offset: 36 .size: 4 .value_kind: by_value - .offset: 40 .size: 4 .value_kind: hidden_block_count_x - .offset: 44 .size: 4 .value_kind: hidden_block_count_y - .offset: 48 .size: 4 .value_kind: hidden_block_count_z - .offset: 52 .size: 2 .value_kind: hidden_group_size_x - .offset: 54 .size: 2 .value_kind: hidden_group_size_y - .offset: 56 .size: 2 .value_kind: hidden_group_size_z - .offset: 58 .size: 2 .value_kind: hidden_remainder_x - .offset: 60 .size: 2 .value_kind: hidden_remainder_y - .offset: 62 .size: 2 .value_kind: hidden_remainder_z - .offset: 80 .size: 8 .value_kind: hidden_global_offset_x - .offset: 88 .size: 8 .value_kind: hidden_global_offset_y - .offset: 96 .size: 8 .value_kind: hidden_global_offset_z - .offset: 104 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 296 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z13convolution2DPdS_S_iiii .private_segment_fixed_size: 0 .sgpr_count: 26 .sgpr_spill_count: 0 .symbol: _Z13convolution2DPdS_S_iiii.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 8 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include "includes.h" __device__ void gpu_bottomUpMerge(long* source, long* dest, long start, long middle, long end) { long i = start; long j = middle; for (long k = start; k < end; k++) { if (i < middle && (j >= end || source[i] < source[j])) { dest[k] = source[i]; i++; } else { dest[k] = source[j]; j++; } } } __device__ unsigned int getIdx(dim3* threads, dim3* blocks) { int x; return threadIdx.x + threadIdx.y * (x = threads->x) + threadIdx.z * (x *= threads->y) + blockIdx.x * (x *= threads->z) + blockIdx.y * (x *= blocks->z) + blockIdx.z * (x *= blocks->y); } __global__ void gpu_mergesort(long* source, long* dest, long size, long width, long slices, dim3* threads, dim3* blocks) { unsigned int idx = getIdx(threads, blocks); long start = width*idx*slices, middle, end; for (long slice = 0; slice < slices; slice++) { if (start >= size) break; middle = min(start + (width >> 1), size); end = min(start + width, size); gpu_bottomUpMerge(source, dest, start, middle, end); start += width; } }
.file "tmpxft_000dd33b_00000000-6_gpu_mergesort.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2031: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2031: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z17gpu_bottomUpMergePlS_lll .type _Z17gpu_bottomUpMergePlS_lll, @function _Z17gpu_bottomUpMergePlS_lll: .LFB2027: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2027: .size _Z17gpu_bottomUpMergePlS_lll, .-_Z17gpu_bottomUpMergePlS_lll .globl _Z6getIdxP4dim3S0_ .type _Z6getIdxP4dim3S0_, @function _Z6getIdxP4dim3S0_: .LFB2028: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2028: .size _Z6getIdxP4dim3S0_, .-_Z6getIdxP4dim3S0_ .globl _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_ .type _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_, @function _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_: .LFB2053: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 56(%rsp) movq %rsi, 48(%rsp) movq %rdx, 40(%rsp) movq %rcx, 32(%rsp) movq %r8, 24(%rsp) movq %r9, 16(%rsp) movq 208(%rsp), %rax movq %rax, 8(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 40(%rsp), %rax movq %rax, 144(%rsp) leaq 32(%rsp), %rax movq %rax, 152(%rsp) leaq 24(%rsp), %rax movq %rax, 160(%rsp) leaq 16(%rsp), %rax movq %rax, 168(%rsp) leaq 8(%rsp), %rax movq %rax, 176(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) movl $1, 88(%rsp) movl $1, 92(%rsp) movl $1, 96(%rsp) movl $1, 100(%rsp) leaq 72(%rsp), %rcx leaq 64(%rsp), %rdx leaq 92(%rsp), %rsi leaq 80(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L11 .L7: movq 184(%rsp), %rax subq %fs:40, %rax jne .L12 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L11: .cfi_restore_state pushq 72(%rsp) .cfi_def_cfa_offset 216 pushq 72(%rsp) .cfi_def_cfa_offset 224 leaq 144(%rsp), %r9 movq 108(%rsp), %rcx movl 116(%rsp), %r8d movq 96(%rsp), %rsi movl 104(%rsp), %edx leaq _Z13gpu_mergesortPlS_lllP4dim3S1_(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L7 .L12: call __stack_chk_fail@PLT .cfi_endproc .LFE2053: .size _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_, .-_Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_ .globl _Z13gpu_mergesortPlS_lllP4dim3S1_ .type _Z13gpu_mergesortPlS_lllP4dim3S1_, @function _Z13gpu_mergesortPlS_lllP4dim3S1_: .LFB2054: .cfi_startproc endbr64 subq $16, %rsp .cfi_def_cfa_offset 24 pushq 24(%rsp) .cfi_def_cfa_offset 32 call _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_ addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _Z13gpu_mergesortPlS_lllP4dim3S1_, .-_Z13gpu_mergesortPlS_lllP4dim3S1_ .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z13gpu_mergesortPlS_lllP4dim3S1_" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2056: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z13gpu_mergesortPlS_lllP4dim3S1_(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2056: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include "includes.h" __device__ void gpu_bottomUpMerge(long* source, long* dest, long start, long middle, long end) { long i = start; long j = middle; for (long k = start; k < end; k++) { if (i < middle && (j >= end || source[i] < source[j])) { dest[k] = source[i]; i++; } else { dest[k] = source[j]; j++; } } } __device__ unsigned int getIdx(dim3* threads, dim3* blocks) { int x; return threadIdx.x + threadIdx.y * (x = threads->x) + threadIdx.z * (x *= threads->y) + blockIdx.x * (x *= threads->z) + blockIdx.y * (x *= blocks->z) + blockIdx.z * (x *= blocks->y); } __global__ void gpu_mergesort(long* source, long* dest, long size, long width, long slices, dim3* threads, dim3* blocks) { unsigned int idx = getIdx(threads, blocks); long start = width*idx*slices, middle, end; for (long slice = 0; slice < slices; slice++) { if (start >= size) break; middle = min(start + (width >> 1), size); end = min(start + width, size); gpu_bottomUpMerge(source, dest, start, middle, end); start += width; } }
#include <hip/hip_runtime.h> #include "includes.h" __device__ void gpu_bottomUpMerge(long* source, long* dest, long start, long middle, long end) { long i = start; long j = middle; for (long k = start; k < end; k++) { if (i < middle && (j >= end || source[i] < source[j])) { dest[k] = source[i]; i++; } else { dest[k] = source[j]; j++; } } } __device__ unsigned int getIdx(dim3* threads, dim3* blocks) { int x; return threadIdx.x + threadIdx.y * (x = threads->x) + threadIdx.z * (x *= threads->y) + blockIdx.x * (x *= threads->z) + blockIdx.y * (x *= blocks->z) + blockIdx.z * (x *= blocks->y); } __global__ void gpu_mergesort(long* source, long* dest, long size, long width, long slices, dim3* threads, dim3* blocks) { unsigned int idx = getIdx(threads, blocks); long start = width*idx*slices, middle, end; for (long slice = 0; slice < slices; slice++) { if (start >= size) break; middle = min(start + (width >> 1), size); end = min(start + width, size); gpu_bottomUpMerge(source, dest, start, middle, end); start += width; } }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include "includes.h" __device__ void gpu_bottomUpMerge(long* source, long* dest, long start, long middle, long end) { long i = start; long j = middle; for (long k = start; k < end; k++) { if (i < middle && (j >= end || source[i] < source[j])) { dest[k] = source[i]; i++; } else { dest[k] = source[j]; j++; } } } __device__ unsigned int getIdx(dim3* threads, dim3* blocks) { int x; return threadIdx.x + threadIdx.y * (x = threads->x) + threadIdx.z * (x *= threads->y) + blockIdx.x * (x *= threads->z) + blockIdx.y * (x *= blocks->z) + blockIdx.z * (x *= blocks->y); } __global__ void gpu_mergesort(long* source, long* dest, long size, long width, long slices, dim3* threads, dim3* blocks) { unsigned int idx = getIdx(threads, blocks); long start = width*idx*slices, middle, end; for (long slice = 0; slice < slices; slice++) { if (start >= size) break; middle = min(start + (width >> 1), size); end = min(start + width, size); gpu_bottomUpMerge(source, dest, start, middle, end); start += width; } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z13gpu_mergesortPlS_lllP4dim3S1_ .globl _Z13gpu_mergesortPlS_lllP4dim3S1_ .p2align 8 .type _Z13gpu_mergesortPlS_lllP4dim3S1_,@function _Z13gpu_mergesortPlS_lllP4dim3S1_: s_clause 0x1 s_load_b64 s[2:3], s[0:1], 0x30 s_load_b256 s[4:11], s[0:1], 0x10 v_bfe_u32 v1, v0, 20, 10 s_waitcnt lgkmcnt(0) s_load_b64 s[2:3], s[2:3], 0x4 s_clause 0x1 s_load_b32 s12, s[10:11], 0x8 s_load_b64 s[10:11], s[10:11], 0x0 s_waitcnt lgkmcnt(0) s_mul_i32 s2, s2, s15 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_add_i32 s2, s2, s14 s_mul_i32 s2, s2, s3 s_mul_hi_u32 s3, s8, s6 s_add_i32 s2, s2, s13 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(VALU_DEP_2) v_mad_u64_u32 v[2:3], null, s2, s12, v[1:2] v_bfe_u32 v3, v0, 10, 10 s_mul_i32 s2, s8, s6 v_mul_lo_u32 v1, v2, s11 v_and_b32_e32 v2, 0x3ff, v0 v_cmp_gt_i64_e64 s11, s[8:9], 0 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) v_add_nc_u32_e32 v3, v1, v3 v_mad_u64_u32 v[4:5], null, v3, s10, v[2:3] s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(SALU_CYCLE_1) v_mad_u64_u32 v[0:1], null, s2, v4, 0 s_mul_i32 s2, s8, s7 s_add_i32 s2, s3, s2 s_mul_i32 s3, s9, s6 s_delay_alu instid0(SALU_CYCLE_1) s_add_i32 s2, s2, s3 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_mad_u64_u32 v[5:6], null, s2, v4, v[1:2] s_mov_b64 s[2:3], 0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mov_b32_e32 v1, v5 v_cmp_gt_i64_e32 vcc_lo, s[4:5], v[0:1] s_and_b32 s11, s11, vcc_lo s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s12, s11 s_cbranch_execz .LBB0_15 v_mad_u64_u32 v[4:5], null, s10, v3, v[2:3] s_mul_i32 s10, s8, s6 s_mul_hi_u32 s11, s8, s6 s_mul_i32 s16, s9, s6 s_load_b128 s[12:15], s[0:1], 0x0 v_cmp_gt_i64_e64 s1, s[6:7], 0 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(SALU_CYCLE_1) v_mad_u64_u32 v[2:3], null, s10, v4, 0 s_mul_i32 s10, s8, s7 s_add_i32 s10, s11, s10 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) s_add_i32 s10, s10, s16 s_mov_b32 s16, 0 v_mad_u64_u32 v[5:6], null, s10, v4, v[3:4] s_ashr_i64 s[10:11], s[6:7], 1 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mov_b32_e32 v3, v5 v_lshlrev_b64 v[2:3], 3, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v2, vcc_lo, s14, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s15, v3, vcc_lo s_lshl_b64 s[14:15], s[6:7], 3 s_branch .LBB0_4 .LBB0_2: s_or_b32 exec_lo, exec_lo, s17 .LBB0_3: s_add_u32 s2, s2, 1 s_addc_u32 s3, s3, 0 v_cmp_le_i64_e32 vcc_lo, s[4:5], v[4:5] v_cmp_ge_i64_e64 s0, s[2:3], s[8:9] v_dual_mov_b32 v0, v4 :: v_dual_mov_b32 v1, v5 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_3) | instid1(SALU_CYCLE_1) s_or_b32 s0, s0, vcc_lo v_add_co_u32 v2, vcc_lo, v2, s14 v_add_co_ci_u32_e32 v3, vcc_lo, s15, v3, vcc_lo s_and_b32 s0, exec_lo, s0 s_or_b32 s16, s0, s16 s_delay_alu instid0(SALU_CYCLE_1) s_and_not1_b32 exec_lo, exec_lo, s16 s_cbranch_execz .LBB0_15 .LBB0_4: v_add_co_u32 v4, vcc_lo, v0, s6 v_add_co_ci_u32_e32 v5, vcc_lo, s7, v1, vcc_lo s_and_not1_b32 vcc_lo, exec_lo, s1 s_cbranch_vccnz .LBB0_3 v_add_co_u32 v6, vcc_lo, v0, s10 v_add_co_ci_u32_e32 v7, vcc_lo, s11, v1, vcc_lo v_cmp_gt_i64_e64 s0, s[4:5], v[4:5] v_dual_mov_b32 v13, v1 :: v_dual_mov_b32 v12, v0 s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_2) | instid1(VALU_DEP_4) v_cmp_gt_i64_e32 vcc_lo, s[4:5], v[6:7] s_mov_b32 s17, 0 v_mov_b32_e32 v11, v3 v_cndmask_b32_e64 v9, s5, v5, s0 v_cndmask_b32_e64 v8, s4, v4, s0 v_dual_mov_b32 v10, v2 :: v_dual_cndmask_b32 v7, s5, v7 v_cndmask_b32_e32 v6, s4, v6, vcc_lo s_delay_alu instid0(VALU_DEP_1) v_dual_mov_b32 v15, v7 :: v_dual_mov_b32 v14, v6 s_branch .LBB0_7 .LBB0_6: s_or_b32 exec_lo, exec_lo, s18 v_add_co_u32 v0, vcc_lo, v0, 1 v_add_co_ci_u32_e32 v1, vcc_lo, 0, v1, vcc_lo v_add_co_u32 v10, s0, v10, 8 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_ci_u32_e64 v11, s0, 0, v11, s0 v_cmp_ge_i64_e32 vcc_lo, v[0:1], v[8:9] v_dual_mov_b32 v14, v16 :: v_dual_mov_b32 v15, v17 s_or_b32 s17, vcc_lo, s17 s_delay_alu instid0(SALU_CYCLE_1) s_and_not1_b32 exec_lo, exec_lo, s17 s_cbranch_execz .LBB0_2 .LBB0_7: s_mov_b32 s19, -1 s_mov_b32 s0, 0 s_mov_b32 s18, exec_lo v_cmpx_lt_i64_e64 v[12:13], v[6:7] s_cbranch_execz .LBB0_11 v_cmp_ge_i64_e64 s0, v[14:15], v[8:9] s_mov_b32 s20, 0 s_mov_b32 s19, exec_lo v_cmpx_lt_i64_e64 v[14:15], v[8:9] s_cbranch_execz .LBB0_10 v_lshlrev_b64 v[16:17], 3, v[12:13] v_lshlrev_b64 v[18:19], 3, v[14:15] s_or_b32 s0, s0, exec_lo s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_u32 v16, vcc_lo, s12, v16 v_add_co_ci_u32_e32 v17, vcc_lo, s13, v17, vcc_lo s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_4) v_add_co_u32 v18, vcc_lo, s12, v18 v_add_co_ci_u32_e32 v19, vcc_lo, s13, v19, vcc_lo s_clause 0x1 global_load_b64 v[16:17], v[16:17], off global_load_b64 v[18:19], v[18:19], off s_waitcnt vmcnt(0) v_cmp_ge_i64_e32 vcc_lo, v[16:17], v[18:19] s_and_b32 s20, vcc_lo, exec_lo .LBB0_10: s_or_b32 exec_lo, exec_lo, s19 s_delay_alu instid0(SALU_CYCLE_1) s_and_b32 s0, s0, exec_lo s_or_not1_b32 s19, s20, exec_lo .LBB0_11: s_or_b32 exec_lo, exec_lo, s18 s_and_saveexec_b32 s18, s19 s_cbranch_execz .LBB0_13 v_lshlrev_b64 v[16:17], 3, v[14:15] s_and_not1_b32 s0, s0, exec_lo s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v16, vcc_lo, s12, v16 v_add_co_ci_u32_e32 v17, vcc_lo, s13, v17, vcc_lo global_load_b64 v[18:19], v[16:17], off v_add_co_u32 v16, vcc_lo, v14, 1 v_add_co_ci_u32_e32 v17, vcc_lo, 0, v15, vcc_lo s_waitcnt vmcnt(0) global_store_b64 v[10:11], v[18:19], off .LBB0_13: s_or_b32 exec_lo, exec_lo, s18 s_and_saveexec_b32 s18, s0 s_cbranch_execz .LBB0_6 v_lshlrev_b64 v[16:17], 3, v[12:13] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v16, vcc_lo, s12, v16 v_add_co_ci_u32_e32 v17, vcc_lo, s13, v17, vcc_lo v_add_co_u32 v12, vcc_lo, v12, 1 v_add_co_ci_u32_e32 v13, vcc_lo, 0, v13, vcc_lo global_load_b64 v[18:19], v[16:17], off v_dual_mov_b32 v17, v15 :: v_dual_mov_b32 v16, v14 s_waitcnt vmcnt(0) global_store_b64 v[10:11], v[18:19], off s_branch .LBB0_6 .LBB0_15: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z13gpu_mergesortPlS_lllP4dim3S1_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 56 .amdhsa_user_sgpr_count 13 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 1 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 2 .amdhsa_next_free_vgpr 20 .amdhsa_next_free_sgpr 21 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z13gpu_mergesortPlS_lllP4dim3S1_, .Lfunc_end0-_Z13gpu_mergesortPlS_lllP4dim3S1_ .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .offset: 16 .size: 8 .value_kind: by_value - .offset: 24 .size: 8 .value_kind: by_value - .offset: 32 .size: 8 .value_kind: by_value - .address_space: global .offset: 40 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 48 .size: 8 .value_kind: global_buffer .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 56 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z13gpu_mergesortPlS_lllP4dim3S1_ .private_segment_fixed_size: 0 .sgpr_count: 23 .sgpr_spill_count: 0 .symbol: _Z13gpu_mergesortPlS_lllP4dim3S1_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 20 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include "includes.h" __device__ void gpu_bottomUpMerge(long* source, long* dest, long start, long middle, long end) { long i = start; long j = middle; for (long k = start; k < end; k++) { if (i < middle && (j >= end || source[i] < source[j])) { dest[k] = source[i]; i++; } else { dest[k] = source[j]; j++; } } } __device__ unsigned int getIdx(dim3* threads, dim3* blocks) { int x; return threadIdx.x + threadIdx.y * (x = threads->x) + threadIdx.z * (x *= threads->y) + blockIdx.x * (x *= threads->z) + blockIdx.y * (x *= blocks->z) + blockIdx.z * (x *= blocks->y); } __global__ void gpu_mergesort(long* source, long* dest, long size, long width, long slices, dim3* threads, dim3* blocks) { unsigned int idx = getIdx(threads, blocks); long start = width*idx*slices, middle, end; for (long slice = 0; slice < slices; slice++) { if (start >= size) break; middle = min(start + (width >> 1), size); end = min(start + width, size); gpu_bottomUpMerge(source, dest, start, middle, end); start += width; } }
.text .file "gpu_mergesort.hip" .globl _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ # -- Begin function _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .p2align 4, 0x90 .type _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_,@function _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_: # @_Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .cfi_startproc # %bb.0: subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movq %r9, 48(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 160(%rsp), %rax movq %rax, 144(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z13gpu_mergesortPlS_lllP4dim3S1_, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $168, %rsp .cfi_adjust_cfa_offset -168 retq .Lfunc_end0: .size _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_, .Lfunc_end0-_Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z13gpu_mergesortPlS_lllP4dim3S1_, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z13gpu_mergesortPlS_lllP4dim3S1_,@object # @_Z13gpu_mergesortPlS_lllP4dim3S1_ .section .rodata,"a",@progbits .globl _Z13gpu_mergesortPlS_lllP4dim3S1_ .p2align 3, 0x0 _Z13gpu_mergesortPlS_lllP4dim3S1_: .quad _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .size _Z13gpu_mergesortPlS_lllP4dim3S1_, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z13gpu_mergesortPlS_lllP4dim3S1_" .size .L__unnamed_1, 34 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z13gpu_mergesortPlS_lllP4dim3S1_ .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_000dd33b_00000000-6_gpu_mergesort.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2031: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2031: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z17gpu_bottomUpMergePlS_lll .type _Z17gpu_bottomUpMergePlS_lll, @function _Z17gpu_bottomUpMergePlS_lll: .LFB2027: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2027: .size _Z17gpu_bottomUpMergePlS_lll, .-_Z17gpu_bottomUpMergePlS_lll .globl _Z6getIdxP4dim3S0_ .type _Z6getIdxP4dim3S0_, @function _Z6getIdxP4dim3S0_: .LFB2028: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2028: .size _Z6getIdxP4dim3S0_, .-_Z6getIdxP4dim3S0_ .globl _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_ .type _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_, @function _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_: .LFB2053: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 56(%rsp) movq %rsi, 48(%rsp) movq %rdx, 40(%rsp) movq %rcx, 32(%rsp) movq %r8, 24(%rsp) movq %r9, 16(%rsp) movq 208(%rsp), %rax movq %rax, 8(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 40(%rsp), %rax movq %rax, 144(%rsp) leaq 32(%rsp), %rax movq %rax, 152(%rsp) leaq 24(%rsp), %rax movq %rax, 160(%rsp) leaq 16(%rsp), %rax movq %rax, 168(%rsp) leaq 8(%rsp), %rax movq %rax, 176(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) movl $1, 88(%rsp) movl $1, 92(%rsp) movl $1, 96(%rsp) movl $1, 100(%rsp) leaq 72(%rsp), %rcx leaq 64(%rsp), %rdx leaq 92(%rsp), %rsi leaq 80(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L11 .L7: movq 184(%rsp), %rax subq %fs:40, %rax jne .L12 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L11: .cfi_restore_state pushq 72(%rsp) .cfi_def_cfa_offset 216 pushq 72(%rsp) .cfi_def_cfa_offset 224 leaq 144(%rsp), %r9 movq 108(%rsp), %rcx movl 116(%rsp), %r8d movq 96(%rsp), %rsi movl 104(%rsp), %edx leaq _Z13gpu_mergesortPlS_lllP4dim3S1_(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L7 .L12: call __stack_chk_fail@PLT .cfi_endproc .LFE2053: .size _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_, .-_Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_ .globl _Z13gpu_mergesortPlS_lllP4dim3S1_ .type _Z13gpu_mergesortPlS_lllP4dim3S1_, @function _Z13gpu_mergesortPlS_lllP4dim3S1_: .LFB2054: .cfi_startproc endbr64 subq $16, %rsp .cfi_def_cfa_offset 24 pushq 24(%rsp) .cfi_def_cfa_offset 32 call _Z47__device_stub__Z13gpu_mergesortPlS_lllP4dim3S1_PlS_lllP4dim3S1_ addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _Z13gpu_mergesortPlS_lllP4dim3S1_, .-_Z13gpu_mergesortPlS_lllP4dim3S1_ .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z13gpu_mergesortPlS_lllP4dim3S1_" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2056: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z13gpu_mergesortPlS_lllP4dim3S1_(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2056: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "gpu_mergesort.hip" .globl _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ # -- Begin function _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .p2align 4, 0x90 .type _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_,@function _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_: # @_Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .cfi_startproc # %bb.0: subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movq %r9, 48(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 160(%rsp), %rax movq %rax, 144(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z13gpu_mergesortPlS_lllP4dim3S1_, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $168, %rsp .cfi_adjust_cfa_offset -168 retq .Lfunc_end0: .size _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_, .Lfunc_end0-_Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z13gpu_mergesortPlS_lllP4dim3S1_, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z13gpu_mergesortPlS_lllP4dim3S1_,@object # @_Z13gpu_mergesortPlS_lllP4dim3S1_ .section .rodata,"a",@progbits .globl _Z13gpu_mergesortPlS_lllP4dim3S1_ .p2align 3, 0x0 _Z13gpu_mergesortPlS_lllP4dim3S1_: .quad _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .size _Z13gpu_mergesortPlS_lllP4dim3S1_, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z13gpu_mergesortPlS_lllP4dim3S1_" .size .L__unnamed_1, 34 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z28__device_stub__gpu_mergesortPlS_lllP4dim3S1_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z13gpu_mergesortPlS_lllP4dim3S1_ .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include "includes.h" __global__ void MatrixMulKernel(float* M, float* N, float* P, int Width) { //We need to iterate with tiles - starting point and end needed for tiles int Mstart=Width*BLOCK_SIZE*blockIdx.y;//rows of matrix M int Mend=Mstart+Width-1; int mstep=BLOCK_SIZE; int Nstart=BLOCK_SIZE*blockIdx.x;//cols of matrix N int nstep=BLOCK_SIZE*Width; float temp=0; //loop through tiles for(int m=Mstart,n=Nstart;m<Mend;m+=mstep,n+=nstep){ __shared__ float Ms[BLOCK_SIZE][BLOCK_SIZE]; __shared__ float Ns[BLOCK_SIZE][BLOCK_SIZE]; Ms[threadIdx.y][threadIdx.x]=M[m+Width*threadIdx.y+threadIdx.x]; Ns[threadIdx.y][threadIdx.x]=N[n+Width*threadIdx.y+threadIdx.x]; __syncthreads(); for (int i = 0; i < BLOCK_SIZE; ++i) { temp += Ms[threadIdx.y][i] * Ns[i][threadIdx.x]; } __syncthreads(); } P[Width * BLOCK_SIZE * blockIdx.y + BLOCK_SIZE * blockIdx.x + Width * threadIdx.y + threadIdx.x] = temp; }
code for sm_80 Function : _Z15MatrixMulKernelPfS_S_i .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R5, SR_CTAID.Y ; /* 0x0000000000057919 */ /* 0x000e220000002600 */ /*0020*/ ULDC UR4, c[0x0][0x178] ; /* 0x00005e0000047ab9 */ /* 0x000fe20000000800 */ /*0030*/ HFMA2.MMA R21, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff157435 */ /* 0x000fe200000001ff */ /*0040*/ USHF.L.U32 UR4, UR4, 0x5, URZ ; /* 0x0000000504047899 */ /* 0x000fe2000800063f */ /*0050*/ S2R R7, SR_CTAID.X ; /* 0x0000000000077919 */ /* 0x000e620000002500 */ /*0060*/ ULDC.64 UR6, c[0x0][0x118] ; /* 0x0000460000067ab9 */ /* 0x000fc60000000a00 */ /*0070*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e680000002100 */ /*0080*/ S2R R9, SR_TID.Y ; /* 0x0000000000097919 */ /* 0x000ea20000002200 */ /*0090*/ IMAD R3, R5, UR4, RZ ; /* 0x0000000405037c24 */ /* 0x001fca000f8e02ff */ /*00a0*/ IADD3 R18, R3, c[0x0][0x178], RZ ; /* 0x00005e0003127a10 */ /* 0x000fe40007ffe0ff */ /*00b0*/ LEA R0, R7, R0, 0x5 ; /* 0x0000000007007211 */ /* 0x002fe400078e28ff */ /*00c0*/ IADD3 R18, R18, -0x1, RZ ; /* 0xffffffff12127810 */ /* 0x000fc60007ffe0ff */ /*00d0*/ IMAD R0, R9, c[0x0][0x178], R0 ; /* 0x00005e0009007a24 */ /* 0x004fe200078e0200 */ /*00e0*/ ISETP.GE.AND P0, PT, R3, R18, PT ; /* 0x000000120300720c */ /* 0x000fc80003f06270 */ /*00f0*/ IADD3 R0, R3, R0, RZ ; /* 0x0000000003007210 */ /* 0x000fca0007ffe0ff */ /*0100*/ IMAD.WIDE.U32 R20, R0, R21, c[0x0][0x170] ; /* 0x00005c0000147625 */ /* 0x000fc800078e0015 */ /*0110*/ @P0 MOV R13, RZ ; /* 0x000000ff000d0202 */ /* 0x000fe20000000f00 */ /*0120*/ @P0 BRA 0x720 ; /* 0x000005f000000947 */ /* 0x000fea0003800000 */ /*0130*/ S2R R19, SR_TID.Y ; /* 0x0000000000137919 */ /* 0x000e220000002200 */ /*0140*/ MOV R13, RZ ; /* 0x000000ff000d7202 */ /* 0x000fc60000000f00 */ /*0150*/ S2R R2, SR_TID.X ; /* 0x0000000000027919 */ /* 0x000e620000002100 */ /*0160*/ LEA R5, R5, R19, 0x5 ; /* 0x0000001305057211 */ /* 0x001fe200078e28ff */ /*0170*/ IMAD R16, R19.reuse, c[0x0][0x178], R2 ; /* 0x00005e0013107a24 */ /* 0x042fe200078e0202 */ /*0180*/ SHF.L.U32 R19, R19, 0x7, RZ ; /* 0x0000000713137819 */ /* 0x000fc600000006ff */ /*0190*/ IMAD R0, R5, c[0x0][0x178], R2 ; /* 0x00005e0005007a24 */ /* 0x000fe200078e0202 */ /*01a0*/ LEA R16, R7, R16, 0x5 ; /* 0x0000001007107211 */ /* 0x000fe400078e28ff */ /*01b0*/ LEA R17, R2, R19, 0x2 ; /* 0x0000001302117211 */ /* 0x000fe400078e10ff */ /*01c0*/ HFMA2.MMA R15, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff0f7435 */ /* 0x000fd400000001ff */ /*01d0*/ IMAD.WIDE.U32 R8, R0, R15, c[0x0][0x160] ; /* 0x0000580000087625 */ /* 0x000fc800078e000f */ /*01e0*/ IMAD.WIDE.U32 R14, R16.reuse, R15, c[0x0][0x168] ; /* 0x00005a00100e7625 */ /* 0x040fe200078e000f */ /*01f0*/ LDG.E R12, [R8.64] ; /* 0x00000006080c7981 */ /* 0x000eaa000c1e1900 */ /*0200*/ LDG.E R14, [R14.64] ; /* 0x000000060e0e7981 */ /* 0x000ee2000c1e1900 */ /*0210*/ IADD3 R3, R3, 0x20, RZ ; /* 0x0000002003037810 */ /* 0x000fe40007ffe0ff */ /*0220*/ IADD3 R16, R16, UR4, RZ ; /* 0x0000000410107c10 */ /* 0x000fe4000fffe0ff */ /*0230*/ ISETP.GE.AND P0, PT, R3, R18, PT ; /* 0x000000120300720c */ /* 0x000fc40003f06270 */ /*0240*/ IADD3 R0, R0, 0x20, RZ ; /* 0x0000002000007810 */ /* 0x000fe20007ffe0ff */ /*0250*/ STS [R17], R12 ; /* 0x0000000c11007388 */ /* 0x004fe80000000800 */ /*0260*/ STS [R17+0x1000], R14 ; /* 0x0010000e11007388 */ /* 0x008fe80000000800 */ /*0270*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0280*/ LDS R23, [R2.X4+0x1000] ; /* 0x0010000002177984 */ /* 0x000fe80000004800 */ /*0290*/ LDS.128 R4, [R19] ; /* 0x0000000013047984 */ /* 0x000e280000000c00 */ /*02a0*/ LDS R28, [R2.X4+0x1080] ; /* 0x00108000021c7984 */ /* 0x000e680000004800 */ /*02b0*/ LDS R29, [R2.X4+0x1100] ; /* 0x00110000021d7984 */ /* 0x000ea80000004800 */ /*02c0*/ LDS R26, [R2.X4+0x1180] ; /* 0x00118000021a7984 */ /* 0x000ee80000004800 */ /*02d0*/ LDS R27, [R2.X4+0x1200] ; /* 0x00120000021b7984 */ /* 0x000fe80000004800 */ /*02e0*/ LDS.128 R8, [R19+0x10] ; /* 0x0000100013087984 */ /* 0x000f280000000c00 */ /*02f0*/ LDS R24, [R2.X4+0x1280] ; /* 0x0012800002187984 */ /* 0x000f680000004800 */ /*0300*/ LDS R25, [R2.X4+0x1300] ; /* 0x0013000002197984 */ /* 0x000f680000004800 */ /*0310*/ LDS R22, [R2.X4+0x1380] ; /* 0x0013800002167984 */ /* 0x000f620000004800 */ /*0320*/ FFMA R4, R23, R4, R13 ; /* 0x0000000417047223 */ /* 0x001fc6000000000d */ /*0330*/ LDS R23, [R2.X4+0x1400] ; /* 0x0014000002177984 */ /* 0x000fe20000004800 */ /*0340*/ FFMA R5, R28, R5, R4 ; /* 0x000000051c057223 */ /* 0x002fc60000000004 */ /*0350*/ LDS.128 R12, [R19+0x20] ; /* 0x00002000130c7984 */ /* 0x000e220000000c00 */ /*0360*/ FFMA R5, R29, R6, R5 ; /* 0x000000061d057223 */ /* 0x004fc60000000005 */ /*0370*/ LDS R28, [R2.X4+0x1480] ; /* 0x00148000021c7984 */ /* 0x000e620000004800 */ /*0380*/ FFMA R5, R26, R7, R5 ; /* 0x000000071a057223 */ /* 0x008fc60000000005 */ /*0390*/ LDS R29, [R2.X4+0x1500] ; /* 0x00150000021d7984 */ /* 0x000ea80000004800 */ /*03a0*/ LDS R26, [R2.X4+0x1680] ; /* 0x00168000021a7984 */ /* 0x000fe20000004800 */ /*03b0*/ FFMA R5, R27, R8, R5 ; /* 0x000000081b057223 */ /* 0x010fc60000000005 */ /*03c0*/ LDS R27, [R2.X4+0x1700] ; /* 0x00170000021b7984 */ /* 0x000fe20000004800 */ /*03d0*/ FFMA R5, R24, R9, R5 ; /* 0x0000000918057223 */ /* 0x020fc60000000005 */ /*03e0*/ LDS R24, [R2.X4+0x1580] ; /* 0x0015800002187984 */ /* 0x000ee20000004800 */ /*03f0*/ FFMA R10, R25, R10, R5 ; /* 0x0000000a190a7223 */ /* 0x000fc60000000005 */ /*0400*/ LDS R25, [R2.X4+0x1600] ; /* 0x0016000002197984 */ /* 0x000fe20000004800 */ /*0410*/ FFMA R10, R22, R11, R10 ; /* 0x0000000b160a7223 */ /* 0x000fc6000000000a */ /*0420*/ LDS.128 R4, [R19+0x30] ; /* 0x0000300013047984 */ /* 0x000f280000000c00 */ /*0430*/ LDS R22, [R2.X4+0x1780] ; /* 0x0017800002167984 */ /* 0x000f620000004800 */ /*0440*/ FFMA R10, R23, R12, R10 ; /* 0x0000000c170a7223 */ /* 0x001fc6000000000a */ /*0450*/ LDS R23, [R2.X4+0x1800] ; /* 0x0018000002177984 */ /* 0x000fe20000004800 */ /*0460*/ FFMA R10, R28, R13, R10 ; /* 0x0000000d1c0a7223 */ /* 0x002fc6000000000a */ /*0470*/ LDS R28, [R2.X4+0x1a80] ; /* 0x001a8000021c7984 */ /* 0x000fe20000004800 */ /*0480*/ FFMA R14, R29, R14, R10 ; /* 0x0000000e1d0e7223 */ /* 0x004fc6000000000a */ /*0490*/ LDS.128 R8, [R19+0x40] ; /* 0x0000400013087984 */ /* 0x000e280000000c00 */ /*04a0*/ LDS R29, [R2.X4+0x1900] ; /* 0x00190000021d7984 */ /* 0x000fe20000004800 */ /*04b0*/ FFMA R14, R24, R15, R14 ; /* 0x0000000f180e7223 */ /* 0x008fc6000000000e */ /*04c0*/ LDS R24, [R2.X4+0x1880] ; /* 0x0018800002187984 */ /* 0x000e620000004800 */ /*04d0*/ FFMA R4, R25, R4, R14 ; /* 0x0000000419047223 */ /* 0x010fc6000000000e */ /*04e0*/ LDS R25, [R2.X4+0x1a00] ; /* 0x001a000002197984 */ /* 0x000fe20000004800 */ /*04f0*/ FFMA R4, R26, R5, R4 ; /* 0x000000051a047223 */ /* 0x000fc60000000004 */ /*0500*/ LDS R26, [R2.X4+0x1980] ; /* 0x00198000021a7984 */ /* 0x000ea20000004800 */ /*0510*/ FFMA R4, R27, R6, R4 ; /* 0x000000061b047223 */ /* 0x000fc60000000004 */ /*0520*/ LDS.128 R12, [R19+0x50] ; /* 0x00005000130c7984 */ /* 0x000ee20000000c00 */ /*0530*/ FFMA R4, R22, R7, R4 ; /* 0x0000000716047223 */ /* 0x020fc60000000004 */ /*0540*/ LDS R27, [R2.X4+0x1b00] ; /* 0x001b0000021b7984 */ /* 0x000f280000004800 */ /*0550*/ LDS R22, [R2.X4+0x1b80] ; /* 0x001b800002167984 */ /* 0x000f620000004800 */ /*0560*/ FFMA R4, R23, R8, R4 ; /* 0x0000000817047223 */ /* 0x001fc60000000004 */ /*0570*/ LDS R23, [R2.X4+0x1c00] ; /* 0x001c000002177984 */ /* 0x000fe20000004800 */ /*0580*/ FFMA R4, R24, R9, R4 ; /* 0x0000000918047223 */ /* 0x002fc60000000004 */ /*0590*/ LDS R24, [R2.X4+0x1c80] ; /* 0x001c800002187984 */ /* 0x000fe20000004800 */ /*05a0*/ FFMA R10, R29, R10, R4 ; /* 0x0000000a1d0a7223 */ /* 0x000fc60000000004 */ /*05b0*/ LDS.128 R4, [R19+0x60] ; /* 0x0000600013047984 */ /* 0x000e220000000c00 */ /*05c0*/ FFMA R10, R26, R11, R10 ; /* 0x0000000b1a0a7223 */ /* 0x004fc8000000000a */ /*05d0*/ FFMA R10, R25, R12, R10 ; /* 0x0000000c190a7223 */ /* 0x008fe4000000000a */ /*05e0*/ LDS R25, [R2.X4+0x1d00] ; /* 0x001d000002197984 */ /* 0x000e640000004800 */ /*05f0*/ FFMA R10, R28, R13, R10 ; /* 0x0000000d1c0a7223 */ /* 0x000fe4000000000a */ /*0600*/ LDS R12, [R2.X4+0x1d80] ; /* 0x001d8000020c7984 */ /* 0x000ea40000004800 */ /*0610*/ FFMA R14, R27, R14, R10 ; /* 0x0000000e1b0e7223 */ /* 0x010fe4000000000a */ /*0620*/ LDS R13, [R2.X4+0x1e00] ; /* 0x001e0000020d7984 */ /* 0x000fe40000004800 */ /*0630*/ FFMA R26, R22, R15, R14 ; /* 0x0000000f161a7223 */ /* 0x020fc4000000000e */ /*0640*/ LDS.128 R8, [R19+0x70] ; /* 0x0000700013087984 */ /* 0x000ee80000000c00 */ /*0650*/ LDS R22, [R2.X4+0x1e80] ; /* 0x001e800002167984 */ /* 0x000f280000004800 */ /*0660*/ LDS R15, [R2.X4+0x1f00] ; /* 0x001f0000020f7984 */ /* 0x000f680000004800 */ /*0670*/ LDS R14, [R2.X4+0x1f80] ; /* 0x001f8000020e7984 */ /* 0x000f620000004800 */ /*0680*/ FFMA R4, R23, R4, R26 ; /* 0x0000000417047223 */ /* 0x001fc8000000001a */ /*0690*/ FFMA R4, R24, R5, R4 ; /* 0x0000000518047223 */ /* 0x000fc80000000004 */ /*06a0*/ FFMA R4, R25, R6, R4 ; /* 0x0000000619047223 */ /* 0x002fc80000000004 */ /*06b0*/ FFMA R4, R12, R7, R4 ; /* 0x000000070c047223 */ /* 0x004fc80000000004 */ /*06c0*/ FFMA R4, R13, R8, R4 ; /* 0x000000080d047223 */ /* 0x008fc80000000004 */ /*06d0*/ FFMA R4, R22, R9, R4 ; /* 0x0000000916047223 */ /* 0x010fc80000000004 */ /*06e0*/ FFMA R4, R15, R10, R4 ; /* 0x0000000a0f047223 */ /* 0x020fc80000000004 */ /*06f0*/ FFMA R13, R14, R11, R4 ; /* 0x0000000b0e0d7223 */ /* 0x000fe20000000004 */ /*0700*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0710*/ @!P0 BRA 0x1c0 ; /* 0xfffffaa000008947 */ /* 0x000fea000383ffff */ /*0720*/ STG.E [R20.64], R13 ; /* 0x0000000d14007986 */ /* 0x000fe2000c101906 */ /*0730*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0740*/ BRA 0x740; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0750*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0760*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0770*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0780*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0790*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include "includes.h" __global__ void MatrixMulKernel(float* M, float* N, float* P, int Width) { //We need to iterate with tiles - starting point and end needed for tiles int Mstart=Width*BLOCK_SIZE*blockIdx.y;//rows of matrix M int Mend=Mstart+Width-1; int mstep=BLOCK_SIZE; int Nstart=BLOCK_SIZE*blockIdx.x;//cols of matrix N int nstep=BLOCK_SIZE*Width; float temp=0; //loop through tiles for(int m=Mstart,n=Nstart;m<Mend;m+=mstep,n+=nstep){ __shared__ float Ms[BLOCK_SIZE][BLOCK_SIZE]; __shared__ float Ns[BLOCK_SIZE][BLOCK_SIZE]; Ms[threadIdx.y][threadIdx.x]=M[m+Width*threadIdx.y+threadIdx.x]; Ns[threadIdx.y][threadIdx.x]=N[n+Width*threadIdx.y+threadIdx.x]; __syncthreads(); for (int i = 0; i < BLOCK_SIZE; ++i) { temp += Ms[threadIdx.y][i] * Ns[i][threadIdx.x]; } __syncthreads(); } P[Width * BLOCK_SIZE * blockIdx.y + BLOCK_SIZE * blockIdx.x + Width * threadIdx.y + threadIdx.x] = temp; }
.file "tmpxft_00177e16_00000000-6_MatrixMulKernel.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i .type _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i, @function _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movl %ecx, 4(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) leaq 4(%rsp), %rax movq %rax, 120(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 136(%rsp), %rax subq %fs:40, %rax jne .L8 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 168 pushq 40(%rsp) .cfi_def_cfa_offset 176 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z15MatrixMulKernelPfS_S_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i, .-_Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i .globl _Z15MatrixMulKernelPfS_S_i .type _Z15MatrixMulKernelPfS_S_i, @function _Z15MatrixMulKernelPfS_S_i: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z15MatrixMulKernelPfS_S_i, .-_Z15MatrixMulKernelPfS_S_i .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z15MatrixMulKernelPfS_S_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z15MatrixMulKernelPfS_S_i(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include "includes.h" __global__ void MatrixMulKernel(float* M, float* N, float* P, int Width) { //We need to iterate with tiles - starting point and end needed for tiles int Mstart=Width*BLOCK_SIZE*blockIdx.y;//rows of matrix M int Mend=Mstart+Width-1; int mstep=BLOCK_SIZE; int Nstart=BLOCK_SIZE*blockIdx.x;//cols of matrix N int nstep=BLOCK_SIZE*Width; float temp=0; //loop through tiles for(int m=Mstart,n=Nstart;m<Mend;m+=mstep,n+=nstep){ __shared__ float Ms[BLOCK_SIZE][BLOCK_SIZE]; __shared__ float Ns[BLOCK_SIZE][BLOCK_SIZE]; Ms[threadIdx.y][threadIdx.x]=M[m+Width*threadIdx.y+threadIdx.x]; Ns[threadIdx.y][threadIdx.x]=N[n+Width*threadIdx.y+threadIdx.x]; __syncthreads(); for (int i = 0; i < BLOCK_SIZE; ++i) { temp += Ms[threadIdx.y][i] * Ns[i][threadIdx.x]; } __syncthreads(); } P[Width * BLOCK_SIZE * blockIdx.y + BLOCK_SIZE * blockIdx.x + Width * threadIdx.y + threadIdx.x] = temp; }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void MatrixMulKernel(float* M, float* N, float* P, int Width) { //We need to iterate with tiles - starting point and end needed for tiles int Mstart=Width*BLOCK_SIZE*blockIdx.y;//rows of matrix M int Mend=Mstart+Width-1; int mstep=BLOCK_SIZE; int Nstart=BLOCK_SIZE*blockIdx.x;//cols of matrix N int nstep=BLOCK_SIZE*Width; float temp=0; //loop through tiles for(int m=Mstart,n=Nstart;m<Mend;m+=mstep,n+=nstep){ __shared__ float Ms[BLOCK_SIZE][BLOCK_SIZE]; __shared__ float Ns[BLOCK_SIZE][BLOCK_SIZE]; Ms[threadIdx.y][threadIdx.x]=M[m+Width*threadIdx.y+threadIdx.x]; Ns[threadIdx.y][threadIdx.x]=N[n+Width*threadIdx.y+threadIdx.x]; __syncthreads(); for (int i = 0; i < BLOCK_SIZE; ++i) { temp += Ms[threadIdx.y][i] * Ns[i][threadIdx.x]; } __syncthreads(); } P[Width * BLOCK_SIZE * blockIdx.y + BLOCK_SIZE * blockIdx.x + Width * threadIdx.y + threadIdx.x] = temp; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void MatrixMulKernel(float* M, float* N, float* P, int Width) { //We need to iterate with tiles - starting point and end needed for tiles int Mstart=Width*BLOCK_SIZE*blockIdx.y;//rows of matrix M int Mend=Mstart+Width-1; int mstep=BLOCK_SIZE; int Nstart=BLOCK_SIZE*blockIdx.x;//cols of matrix N int nstep=BLOCK_SIZE*Width; float temp=0; //loop through tiles for(int m=Mstart,n=Nstart;m<Mend;m+=mstep,n+=nstep){ __shared__ float Ms[BLOCK_SIZE][BLOCK_SIZE]; __shared__ float Ns[BLOCK_SIZE][BLOCK_SIZE]; Ms[threadIdx.y][threadIdx.x]=M[m+Width*threadIdx.y+threadIdx.x]; Ns[threadIdx.y][threadIdx.x]=N[n+Width*threadIdx.y+threadIdx.x]; __syncthreads(); for (int i = 0; i < BLOCK_SIZE; ++i) { temp += Ms[threadIdx.y][i] * Ns[i][threadIdx.x]; } __syncthreads(); } P[Width * BLOCK_SIZE * blockIdx.y + BLOCK_SIZE * blockIdx.x + Width * threadIdx.y + threadIdx.x] = temp; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z15MatrixMulKernelPfS_S_i .globl _Z15MatrixMulKernelPfS_S_i .p2align 8 .type _Z15MatrixMulKernelPfS_S_i,@function _Z15MatrixMulKernelPfS_S_i: s_load_b32 s3, s[0:1], 0x18 v_mov_b32_e32 v5, 0 v_bfe_u32 v4, v0, 10, 10 s_lshl_b32 s8, s14, 5 s_waitcnt lgkmcnt(0) s_lshl_b32 s9, s3, 5 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_mul_i32 s2, s9, s15 s_add_i32 s10, s3, s2 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_add_i32 s10, s10, -1 s_cmp_ge_i32 s2, s10 s_cbranch_scc1 .LBB0_5 v_and_b32_e32 v3, 0x3ff, v0 s_load_b128 s[4:7], s[0:1], 0x0 v_lshlrev_b32_e32 v6, 7, v4 s_mov_b32 s11, s8 s_mov_b32 s12, s2 v_dual_mov_b32 v5, 0 :: v_dual_lshlrev_b32 v8, 2, v3 v_mad_u64_u32 v[1:2], null, v4, s3, v[3:4] v_mov_b32_e32 v3, 0 s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_or_b32_e32 v7, 0x1000, v8 v_add_nc_u32_e32 v8, v6, v8 v_add_nc_u32_e32 v9, v7, v6 s_set_inst_prefetch_distance 0x1 .p2align 6 .LBB0_2: v_add_nc_u32_e32 v2, s12, v1 s_mov_b32 s13, 0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[10:11], 2, v[2:3] v_add_nc_u32_e32 v2, s11, v1 v_lshlrev_b64 v[12:13], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_4) v_add_co_u32 v10, vcc_lo, s4, v10 v_add_co_ci_u32_e32 v11, vcc_lo, s5, v11, vcc_lo v_mov_b32_e32 v2, v7 s_delay_alu instid0(VALU_DEP_4) v_add_co_u32 v12, vcc_lo, s6, v12 v_add_co_ci_u32_e32 v13, vcc_lo, s7, v13, vcc_lo global_load_b32 v10, v[10:11], off global_load_b32 v11, v[12:13], off s_waitcnt vmcnt(1) ds_store_b32 v8, v10 s_waitcnt vmcnt(0) ds_store_b32 v9, v11 s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv .LBB0_3: v_add_nc_u32_e32 v10, s13, v6 s_add_i32 s13, s13, 4 ds_load_b32 v11, v2 ds_load_b32 v10, v10 v_add_nc_u32_e32 v2, 0x80, v2 s_cmpk_eq_i32 s13, 0x80 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v5, v10, v11 s_cbranch_scc0 .LBB0_3 s_add_i32 s12, s12, 32 s_add_i32 s11, s11, s9 s_cmp_ge_i32 s12, s10 s_barrier buffer_gl0_inv s_cbranch_scc0 .LBB0_2 .LBB0_5: s_set_inst_prefetch_distance 0x2 v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x3ff, v0 s_load_b64 s[0:1], s[0:1], 0x10 v_mul_lo_u32 v2, v4, s3 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_add_nc_u32_e32 v0, s8, v0 v_add3_u32 v0, v0, v2, s2 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[0:1] s_waitcnt lgkmcnt(0) v_add_co_u32 v0, vcc_lo, s0, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo global_store_b32 v[0:1], v5, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z15MatrixMulKernelPfS_S_i .amdhsa_group_segment_fixed_size 8192 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 28 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z15MatrixMulKernelPfS_S_i, .Lfunc_end0-_Z15MatrixMulKernelPfS_S_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value .group_segment_fixed_size: 8192 .kernarg_segment_align: 8 .kernarg_segment_size: 28 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z15MatrixMulKernelPfS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z15MatrixMulKernelPfS_S_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void MatrixMulKernel(float* M, float* N, float* P, int Width) { //We need to iterate with tiles - starting point and end needed for tiles int Mstart=Width*BLOCK_SIZE*blockIdx.y;//rows of matrix M int Mend=Mstart+Width-1; int mstep=BLOCK_SIZE; int Nstart=BLOCK_SIZE*blockIdx.x;//cols of matrix N int nstep=BLOCK_SIZE*Width; float temp=0; //loop through tiles for(int m=Mstart,n=Nstart;m<Mend;m+=mstep,n+=nstep){ __shared__ float Ms[BLOCK_SIZE][BLOCK_SIZE]; __shared__ float Ns[BLOCK_SIZE][BLOCK_SIZE]; Ms[threadIdx.y][threadIdx.x]=M[m+Width*threadIdx.y+threadIdx.x]; Ns[threadIdx.y][threadIdx.x]=N[n+Width*threadIdx.y+threadIdx.x]; __syncthreads(); for (int i = 0; i < BLOCK_SIZE; ++i) { temp += Ms[threadIdx.y][i] * Ns[i][threadIdx.x]; } __syncthreads(); } P[Width * BLOCK_SIZE * blockIdx.y + BLOCK_SIZE * blockIdx.x + Width * threadIdx.y + threadIdx.x] = temp; }
.text .file "MatrixMulKernel.hip" .globl _Z30__device_stub__MatrixMulKernelPfS_S_i # -- Begin function _Z30__device_stub__MatrixMulKernelPfS_S_i .p2align 4, 0x90 .type _Z30__device_stub__MatrixMulKernelPfS_S_i,@function _Z30__device_stub__MatrixMulKernelPfS_S_i: # @_Z30__device_stub__MatrixMulKernelPfS_S_i .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) movl %ecx, 4(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 4(%rsp), %rax movq %rax, 104(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z15MatrixMulKernelPfS_S_i, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z30__device_stub__MatrixMulKernelPfS_S_i, .Lfunc_end0-_Z30__device_stub__MatrixMulKernelPfS_S_i .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z15MatrixMulKernelPfS_S_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z15MatrixMulKernelPfS_S_i,@object # @_Z15MatrixMulKernelPfS_S_i .section .rodata,"a",@progbits .globl _Z15MatrixMulKernelPfS_S_i .p2align 3, 0x0 _Z15MatrixMulKernelPfS_S_i: .quad _Z30__device_stub__MatrixMulKernelPfS_S_i .size _Z15MatrixMulKernelPfS_S_i, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z15MatrixMulKernelPfS_S_i" .size .L__unnamed_1, 27 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z30__device_stub__MatrixMulKernelPfS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z15MatrixMulKernelPfS_S_i .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z15MatrixMulKernelPfS_S_i .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R5, SR_CTAID.Y ; /* 0x0000000000057919 */ /* 0x000e220000002600 */ /*0020*/ ULDC UR4, c[0x0][0x178] ; /* 0x00005e0000047ab9 */ /* 0x000fe20000000800 */ /*0030*/ HFMA2.MMA R21, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff157435 */ /* 0x000fe200000001ff */ /*0040*/ USHF.L.U32 UR4, UR4, 0x5, URZ ; /* 0x0000000504047899 */ /* 0x000fe2000800063f */ /*0050*/ S2R R7, SR_CTAID.X ; /* 0x0000000000077919 */ /* 0x000e620000002500 */ /*0060*/ ULDC.64 UR6, c[0x0][0x118] ; /* 0x0000460000067ab9 */ /* 0x000fc60000000a00 */ /*0070*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e680000002100 */ /*0080*/ S2R R9, SR_TID.Y ; /* 0x0000000000097919 */ /* 0x000ea20000002200 */ /*0090*/ IMAD R3, R5, UR4, RZ ; /* 0x0000000405037c24 */ /* 0x001fca000f8e02ff */ /*00a0*/ IADD3 R18, R3, c[0x0][0x178], RZ ; /* 0x00005e0003127a10 */ /* 0x000fe40007ffe0ff */ /*00b0*/ LEA R0, R7, R0, 0x5 ; /* 0x0000000007007211 */ /* 0x002fe400078e28ff */ /*00c0*/ IADD3 R18, R18, -0x1, RZ ; /* 0xffffffff12127810 */ /* 0x000fc60007ffe0ff */ /*00d0*/ IMAD R0, R9, c[0x0][0x178], R0 ; /* 0x00005e0009007a24 */ /* 0x004fe200078e0200 */ /*00e0*/ ISETP.GE.AND P0, PT, R3, R18, PT ; /* 0x000000120300720c */ /* 0x000fc80003f06270 */ /*00f0*/ IADD3 R0, R3, R0, RZ ; /* 0x0000000003007210 */ /* 0x000fca0007ffe0ff */ /*0100*/ IMAD.WIDE.U32 R20, R0, R21, c[0x0][0x170] ; /* 0x00005c0000147625 */ /* 0x000fc800078e0015 */ /*0110*/ @P0 MOV R13, RZ ; /* 0x000000ff000d0202 */ /* 0x000fe20000000f00 */ /*0120*/ @P0 BRA 0x720 ; /* 0x000005f000000947 */ /* 0x000fea0003800000 */ /*0130*/ S2R R19, SR_TID.Y ; /* 0x0000000000137919 */ /* 0x000e220000002200 */ /*0140*/ MOV R13, RZ ; /* 0x000000ff000d7202 */ /* 0x000fc60000000f00 */ /*0150*/ S2R R2, SR_TID.X ; /* 0x0000000000027919 */ /* 0x000e620000002100 */ /*0160*/ LEA R5, R5, R19, 0x5 ; /* 0x0000001305057211 */ /* 0x001fe200078e28ff */ /*0170*/ IMAD R16, R19.reuse, c[0x0][0x178], R2 ; /* 0x00005e0013107a24 */ /* 0x042fe200078e0202 */ /*0180*/ SHF.L.U32 R19, R19, 0x7, RZ ; /* 0x0000000713137819 */ /* 0x000fc600000006ff */ /*0190*/ IMAD R0, R5, c[0x0][0x178], R2 ; /* 0x00005e0005007a24 */ /* 0x000fe200078e0202 */ /*01a0*/ LEA R16, R7, R16, 0x5 ; /* 0x0000001007107211 */ /* 0x000fe400078e28ff */ /*01b0*/ LEA R17, R2, R19, 0x2 ; /* 0x0000001302117211 */ /* 0x000fe400078e10ff */ /*01c0*/ HFMA2.MMA R15, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff0f7435 */ /* 0x000fd400000001ff */ /*01d0*/ IMAD.WIDE.U32 R8, R0, R15, c[0x0][0x160] ; /* 0x0000580000087625 */ /* 0x000fc800078e000f */ /*01e0*/ IMAD.WIDE.U32 R14, R16.reuse, R15, c[0x0][0x168] ; /* 0x00005a00100e7625 */ /* 0x040fe200078e000f */ /*01f0*/ LDG.E R12, [R8.64] ; /* 0x00000006080c7981 */ /* 0x000eaa000c1e1900 */ /*0200*/ LDG.E R14, [R14.64] ; /* 0x000000060e0e7981 */ /* 0x000ee2000c1e1900 */ /*0210*/ IADD3 R3, R3, 0x20, RZ ; /* 0x0000002003037810 */ /* 0x000fe40007ffe0ff */ /*0220*/ IADD3 R16, R16, UR4, RZ ; /* 0x0000000410107c10 */ /* 0x000fe4000fffe0ff */ /*0230*/ ISETP.GE.AND P0, PT, R3, R18, PT ; /* 0x000000120300720c */ /* 0x000fc40003f06270 */ /*0240*/ IADD3 R0, R0, 0x20, RZ ; /* 0x0000002000007810 */ /* 0x000fe20007ffe0ff */ /*0250*/ STS [R17], R12 ; /* 0x0000000c11007388 */ /* 0x004fe80000000800 */ /*0260*/ STS [R17+0x1000], R14 ; /* 0x0010000e11007388 */ /* 0x008fe80000000800 */ /*0270*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0280*/ LDS R23, [R2.X4+0x1000] ; /* 0x0010000002177984 */ /* 0x000fe80000004800 */ /*0290*/ LDS.128 R4, [R19] ; /* 0x0000000013047984 */ /* 0x000e280000000c00 */ /*02a0*/ LDS R28, [R2.X4+0x1080] ; /* 0x00108000021c7984 */ /* 0x000e680000004800 */ /*02b0*/ LDS R29, [R2.X4+0x1100] ; /* 0x00110000021d7984 */ /* 0x000ea80000004800 */ /*02c0*/ LDS R26, [R2.X4+0x1180] ; /* 0x00118000021a7984 */ /* 0x000ee80000004800 */ /*02d0*/ LDS R27, [R2.X4+0x1200] ; /* 0x00120000021b7984 */ /* 0x000fe80000004800 */ /*02e0*/ LDS.128 R8, [R19+0x10] ; /* 0x0000100013087984 */ /* 0x000f280000000c00 */ /*02f0*/ LDS R24, [R2.X4+0x1280] ; /* 0x0012800002187984 */ /* 0x000f680000004800 */ /*0300*/ LDS R25, [R2.X4+0x1300] ; /* 0x0013000002197984 */ /* 0x000f680000004800 */ /*0310*/ LDS R22, [R2.X4+0x1380] ; /* 0x0013800002167984 */ /* 0x000f620000004800 */ /*0320*/ FFMA R4, R23, R4, R13 ; /* 0x0000000417047223 */ /* 0x001fc6000000000d */ /*0330*/ LDS R23, [R2.X4+0x1400] ; /* 0x0014000002177984 */ /* 0x000fe20000004800 */ /*0340*/ FFMA R5, R28, R5, R4 ; /* 0x000000051c057223 */ /* 0x002fc60000000004 */ /*0350*/ LDS.128 R12, [R19+0x20] ; /* 0x00002000130c7984 */ /* 0x000e220000000c00 */ /*0360*/ FFMA R5, R29, R6, R5 ; /* 0x000000061d057223 */ /* 0x004fc60000000005 */ /*0370*/ LDS R28, [R2.X4+0x1480] ; /* 0x00148000021c7984 */ /* 0x000e620000004800 */ /*0380*/ FFMA R5, R26, R7, R5 ; /* 0x000000071a057223 */ /* 0x008fc60000000005 */ /*0390*/ LDS R29, [R2.X4+0x1500] ; /* 0x00150000021d7984 */ /* 0x000ea80000004800 */ /*03a0*/ LDS R26, [R2.X4+0x1680] ; /* 0x00168000021a7984 */ /* 0x000fe20000004800 */ /*03b0*/ FFMA R5, R27, R8, R5 ; /* 0x000000081b057223 */ /* 0x010fc60000000005 */ /*03c0*/ LDS R27, [R2.X4+0x1700] ; /* 0x00170000021b7984 */ /* 0x000fe20000004800 */ /*03d0*/ FFMA R5, R24, R9, R5 ; /* 0x0000000918057223 */ /* 0x020fc60000000005 */ /*03e0*/ LDS R24, [R2.X4+0x1580] ; /* 0x0015800002187984 */ /* 0x000ee20000004800 */ /*03f0*/ FFMA R10, R25, R10, R5 ; /* 0x0000000a190a7223 */ /* 0x000fc60000000005 */ /*0400*/ LDS R25, [R2.X4+0x1600] ; /* 0x0016000002197984 */ /* 0x000fe20000004800 */ /*0410*/ FFMA R10, R22, R11, R10 ; /* 0x0000000b160a7223 */ /* 0x000fc6000000000a */ /*0420*/ LDS.128 R4, [R19+0x30] ; /* 0x0000300013047984 */ /* 0x000f280000000c00 */ /*0430*/ LDS R22, [R2.X4+0x1780] ; /* 0x0017800002167984 */ /* 0x000f620000004800 */ /*0440*/ FFMA R10, R23, R12, R10 ; /* 0x0000000c170a7223 */ /* 0x001fc6000000000a */ /*0450*/ LDS R23, [R2.X4+0x1800] ; /* 0x0018000002177984 */ /* 0x000fe20000004800 */ /*0460*/ FFMA R10, R28, R13, R10 ; /* 0x0000000d1c0a7223 */ /* 0x002fc6000000000a */ /*0470*/ LDS R28, [R2.X4+0x1a80] ; /* 0x001a8000021c7984 */ /* 0x000fe20000004800 */ /*0480*/ FFMA R14, R29, R14, R10 ; /* 0x0000000e1d0e7223 */ /* 0x004fc6000000000a */ /*0490*/ LDS.128 R8, [R19+0x40] ; /* 0x0000400013087984 */ /* 0x000e280000000c00 */ /*04a0*/ LDS R29, [R2.X4+0x1900] ; /* 0x00190000021d7984 */ /* 0x000fe20000004800 */ /*04b0*/ FFMA R14, R24, R15, R14 ; /* 0x0000000f180e7223 */ /* 0x008fc6000000000e */ /*04c0*/ LDS R24, [R2.X4+0x1880] ; /* 0x0018800002187984 */ /* 0x000e620000004800 */ /*04d0*/ FFMA R4, R25, R4, R14 ; /* 0x0000000419047223 */ /* 0x010fc6000000000e */ /*04e0*/ LDS R25, [R2.X4+0x1a00] ; /* 0x001a000002197984 */ /* 0x000fe20000004800 */ /*04f0*/ FFMA R4, R26, R5, R4 ; /* 0x000000051a047223 */ /* 0x000fc60000000004 */ /*0500*/ LDS R26, [R2.X4+0x1980] ; /* 0x00198000021a7984 */ /* 0x000ea20000004800 */ /*0510*/ FFMA R4, R27, R6, R4 ; /* 0x000000061b047223 */ /* 0x000fc60000000004 */ /*0520*/ LDS.128 R12, [R19+0x50] ; /* 0x00005000130c7984 */ /* 0x000ee20000000c00 */ /*0530*/ FFMA R4, R22, R7, R4 ; /* 0x0000000716047223 */ /* 0x020fc60000000004 */ /*0540*/ LDS R27, [R2.X4+0x1b00] ; /* 0x001b0000021b7984 */ /* 0x000f280000004800 */ /*0550*/ LDS R22, [R2.X4+0x1b80] ; /* 0x001b800002167984 */ /* 0x000f620000004800 */ /*0560*/ FFMA R4, R23, R8, R4 ; /* 0x0000000817047223 */ /* 0x001fc60000000004 */ /*0570*/ LDS R23, [R2.X4+0x1c00] ; /* 0x001c000002177984 */ /* 0x000fe20000004800 */ /*0580*/ FFMA R4, R24, R9, R4 ; /* 0x0000000918047223 */ /* 0x002fc60000000004 */ /*0590*/ LDS R24, [R2.X4+0x1c80] ; /* 0x001c800002187984 */ /* 0x000fe20000004800 */ /*05a0*/ FFMA R10, R29, R10, R4 ; /* 0x0000000a1d0a7223 */ /* 0x000fc60000000004 */ /*05b0*/ LDS.128 R4, [R19+0x60] ; /* 0x0000600013047984 */ /* 0x000e220000000c00 */ /*05c0*/ FFMA R10, R26, R11, R10 ; /* 0x0000000b1a0a7223 */ /* 0x004fc8000000000a */ /*05d0*/ FFMA R10, R25, R12, R10 ; /* 0x0000000c190a7223 */ /* 0x008fe4000000000a */ /*05e0*/ LDS R25, [R2.X4+0x1d00] ; /* 0x001d000002197984 */ /* 0x000e640000004800 */ /*05f0*/ FFMA R10, R28, R13, R10 ; /* 0x0000000d1c0a7223 */ /* 0x000fe4000000000a */ /*0600*/ LDS R12, [R2.X4+0x1d80] ; /* 0x001d8000020c7984 */ /* 0x000ea40000004800 */ /*0610*/ FFMA R14, R27, R14, R10 ; /* 0x0000000e1b0e7223 */ /* 0x010fe4000000000a */ /*0620*/ LDS R13, [R2.X4+0x1e00] ; /* 0x001e0000020d7984 */ /* 0x000fe40000004800 */ /*0630*/ FFMA R26, R22, R15, R14 ; /* 0x0000000f161a7223 */ /* 0x020fc4000000000e */ /*0640*/ LDS.128 R8, [R19+0x70] ; /* 0x0000700013087984 */ /* 0x000ee80000000c00 */ /*0650*/ LDS R22, [R2.X4+0x1e80] ; /* 0x001e800002167984 */ /* 0x000f280000004800 */ /*0660*/ LDS R15, [R2.X4+0x1f00] ; /* 0x001f0000020f7984 */ /* 0x000f680000004800 */ /*0670*/ LDS R14, [R2.X4+0x1f80] ; /* 0x001f8000020e7984 */ /* 0x000f620000004800 */ /*0680*/ FFMA R4, R23, R4, R26 ; /* 0x0000000417047223 */ /* 0x001fc8000000001a */ /*0690*/ FFMA R4, R24, R5, R4 ; /* 0x0000000518047223 */ /* 0x000fc80000000004 */ /*06a0*/ FFMA R4, R25, R6, R4 ; /* 0x0000000619047223 */ /* 0x002fc80000000004 */ /*06b0*/ FFMA R4, R12, R7, R4 ; /* 0x000000070c047223 */ /* 0x004fc80000000004 */ /*06c0*/ FFMA R4, R13, R8, R4 ; /* 0x000000080d047223 */ /* 0x008fc80000000004 */ /*06d0*/ FFMA R4, R22, R9, R4 ; /* 0x0000000916047223 */ /* 0x010fc80000000004 */ /*06e0*/ FFMA R4, R15, R10, R4 ; /* 0x0000000a0f047223 */ /* 0x020fc80000000004 */ /*06f0*/ FFMA R13, R14, R11, R4 ; /* 0x0000000b0e0d7223 */ /* 0x000fe20000000004 */ /*0700*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0710*/ @!P0 BRA 0x1c0 ; /* 0xfffffaa000008947 */ /* 0x000fea000383ffff */ /*0720*/ STG.E [R20.64], R13 ; /* 0x0000000d14007986 */ /* 0x000fe2000c101906 */ /*0730*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0740*/ BRA 0x740; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0750*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0760*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0770*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0780*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0790*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z15MatrixMulKernelPfS_S_i .globl _Z15MatrixMulKernelPfS_S_i .p2align 8 .type _Z15MatrixMulKernelPfS_S_i,@function _Z15MatrixMulKernelPfS_S_i: s_load_b32 s3, s[0:1], 0x18 v_mov_b32_e32 v5, 0 v_bfe_u32 v4, v0, 10, 10 s_lshl_b32 s8, s14, 5 s_waitcnt lgkmcnt(0) s_lshl_b32 s9, s3, 5 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_mul_i32 s2, s9, s15 s_add_i32 s10, s3, s2 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1) s_add_i32 s10, s10, -1 s_cmp_ge_i32 s2, s10 s_cbranch_scc1 .LBB0_5 v_and_b32_e32 v3, 0x3ff, v0 s_load_b128 s[4:7], s[0:1], 0x0 v_lshlrev_b32_e32 v6, 7, v4 s_mov_b32 s11, s8 s_mov_b32 s12, s2 v_dual_mov_b32 v5, 0 :: v_dual_lshlrev_b32 v8, 2, v3 v_mad_u64_u32 v[1:2], null, v4, s3, v[3:4] v_mov_b32_e32 v3, 0 s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_or_b32_e32 v7, 0x1000, v8 v_add_nc_u32_e32 v8, v6, v8 v_add_nc_u32_e32 v9, v7, v6 s_set_inst_prefetch_distance 0x1 .p2align 6 .LBB0_2: v_add_nc_u32_e32 v2, s12, v1 s_mov_b32 s13, 0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[10:11], 2, v[2:3] v_add_nc_u32_e32 v2, s11, v1 v_lshlrev_b64 v[12:13], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_4) v_add_co_u32 v10, vcc_lo, s4, v10 v_add_co_ci_u32_e32 v11, vcc_lo, s5, v11, vcc_lo v_mov_b32_e32 v2, v7 s_delay_alu instid0(VALU_DEP_4) v_add_co_u32 v12, vcc_lo, s6, v12 v_add_co_ci_u32_e32 v13, vcc_lo, s7, v13, vcc_lo global_load_b32 v10, v[10:11], off global_load_b32 v11, v[12:13], off s_waitcnt vmcnt(1) ds_store_b32 v8, v10 s_waitcnt vmcnt(0) ds_store_b32 v9, v11 s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv .LBB0_3: v_add_nc_u32_e32 v10, s13, v6 s_add_i32 s13, s13, 4 ds_load_b32 v11, v2 ds_load_b32 v10, v10 v_add_nc_u32_e32 v2, 0x80, v2 s_cmpk_eq_i32 s13, 0x80 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v5, v10, v11 s_cbranch_scc0 .LBB0_3 s_add_i32 s12, s12, 32 s_add_i32 s11, s11, s9 s_cmp_ge_i32 s12, s10 s_barrier buffer_gl0_inv s_cbranch_scc0 .LBB0_2 .LBB0_5: s_set_inst_prefetch_distance 0x2 v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x3ff, v0 s_load_b64 s[0:1], s[0:1], 0x10 v_mul_lo_u32 v2, v4, s3 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_add_nc_u32_e32 v0, s8, v0 v_add3_u32 v0, v0, v2, s2 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[0:1] s_waitcnt lgkmcnt(0) v_add_co_u32 v0, vcc_lo, s0, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo global_store_b32 v[0:1], v5, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z15MatrixMulKernelPfS_S_i .amdhsa_group_segment_fixed_size 8192 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 28 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z15MatrixMulKernelPfS_S_i, .Lfunc_end0-_Z15MatrixMulKernelPfS_S_i .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .address_space: global .offset: 0 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: by_value .group_segment_fixed_size: 8192 .kernarg_segment_align: 8 .kernarg_segment_size: 28 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z15MatrixMulKernelPfS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z15MatrixMulKernelPfS_S_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_00177e16_00000000-6_MatrixMulKernel.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i .type _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i, @function _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movl %ecx, 4(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) leaq 4(%rsp), %rax movq %rax, 120(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 136(%rsp), %rax subq %fs:40, %rax jne .L8 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 168 pushq 40(%rsp) .cfi_def_cfa_offset 176 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z15MatrixMulKernelPfS_S_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i, .-_Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i .globl _Z15MatrixMulKernelPfS_S_i .type _Z15MatrixMulKernelPfS_S_i, @function _Z15MatrixMulKernelPfS_S_i: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z40__device_stub__Z15MatrixMulKernelPfS_S_iPfS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z15MatrixMulKernelPfS_S_i, .-_Z15MatrixMulKernelPfS_S_i .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z15MatrixMulKernelPfS_S_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z15MatrixMulKernelPfS_S_i(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "MatrixMulKernel.hip" .globl _Z30__device_stub__MatrixMulKernelPfS_S_i # -- Begin function _Z30__device_stub__MatrixMulKernelPfS_S_i .p2align 4, 0x90 .type _Z30__device_stub__MatrixMulKernelPfS_S_i,@function _Z30__device_stub__MatrixMulKernelPfS_S_i: # @_Z30__device_stub__MatrixMulKernelPfS_S_i .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) movl %ecx, 4(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 4(%rsp), %rax movq %rax, 104(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 80(%rsp), %r9 movl $_Z15MatrixMulKernelPfS_S_i, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z30__device_stub__MatrixMulKernelPfS_S_i, .Lfunc_end0-_Z30__device_stub__MatrixMulKernelPfS_S_i .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z15MatrixMulKernelPfS_S_i, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type _Z15MatrixMulKernelPfS_S_i,@object # @_Z15MatrixMulKernelPfS_S_i .section .rodata,"a",@progbits .globl _Z15MatrixMulKernelPfS_S_i .p2align 3, 0x0 _Z15MatrixMulKernelPfS_S_i: .quad _Z30__device_stub__MatrixMulKernelPfS_S_i .size _Z15MatrixMulKernelPfS_S_i, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z15MatrixMulKernelPfS_S_i" .size .L__unnamed_1, 27 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z30__device_stub__MatrixMulKernelPfS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z15MatrixMulKernelPfS_S_i .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
// 1 / (1 + e^(-x)) extern "C" __global__ void logistic(size_t n, double *result, double *x) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i<n) { result[i] = 1.0 / (1.0 + exp(-x[i])); } }
code for sm_80 Function : logistic .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e280000002500 */ /*0020*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R2, R2, c[0x0][0x0], R3 ; /* 0x0000000002027a24 */ /* 0x001fca00078e0203 */ /*0040*/ ISETP.GE.U32.AND P0, PT, R2, c[0x0][0x160], PT ; /* 0x0000580002007a0c */ /* 0x000fe40003f06070 */ /*0050*/ SHF.R.S32.HI R3, RZ, 0x1f, R2 ; /* 0x0000001fff037819 */ /* 0x000fc80000011402 */ /*0060*/ ISETP.GE.U32.AND.EX P0, PT, R3, c[0x0][0x164], PT, P0 ; /* 0x0000590003007a0c */ /* 0x000fda0003f06100 */ /*0070*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0080*/ IMAD.SHL.U32 R0, R2.reuse, 0x8, RZ ; /* 0x0000000802007824 */ /* 0x040fe200078e00ff */ /*0090*/ SHF.L.U64.HI R2, R2, 0x3, R3 ; /* 0x0000000302027819 */ /* 0x000fe20000010203 */ /*00a0*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*00b0*/ IADD3 R4, P0, R0, c[0x0][0x170], RZ ; /* 0x00005c0000047a10 */ /* 0x000fc80007f1e0ff */ /*00c0*/ IADD3.X R5, R2, c[0x0][0x174], RZ, P0, !PT ; /* 0x00005d0002057a10 */ /* 0x000fcc00007fe4ff */ /*00d0*/ LDG.E.64 R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000ea2000c1e1b00 */ /*00e0*/ IMAD.MOV.U32 R6, RZ, RZ, 0x652b82fe ; /* 0x652b82feff067424 */ /* 0x000fe200078e00ff */ /*00f0*/ BSSY B0, 0x340 ; /* 0x0000024000007945 */ /* 0x000fe20003800000 */ /*0100*/ IMAD.MOV.U32 R7, RZ, RZ, 0x3ff71547 ; /* 0x3ff71547ff077424 */ /* 0x000fe400078e00ff */ /*0110*/ IMAD.MOV.U32 R12, RZ, RZ, 0x69ce2bdf ; /* 0x69ce2bdfff0c7424 */ /* 0x000fe400078e00ff */ /*0120*/ IMAD.MOV.U32 R13, RZ, RZ, 0x3e5ade15 ; /* 0x3e5ade15ff0d7424 */ /* 0x000fe400078e00ff */ /*0130*/ DFMA R6, -R4, R6, 6.75539944105574400000e+15 ; /* 0x433800000406742b */ /* 0x004e0c0000000106 */ /*0140*/ DADD R8, R6, -6.75539944105574400000e+15 ; /* 0xc338000006087429 */ /* 0x001e0c0000000000 */ /*0150*/ DFMA R10, R8, c[0x2][0x0], -R4 ; /* 0x00800000080a7a2b */ /* 0x001e0c0000000804 */ /*0160*/ DFMA R8, R8, c[0x2][0x8], R10 ; /* 0x0080020008087a2b */ /* 0x001e0c000000000a */ /*0170*/ DFMA R10, R8, R12, c[0x2][0x10] ; /* 0x00800400080a762b */ /* 0x001e08000000000c */ /*0180*/ DADD R12, -RZ, -R4 ; /* 0x00000000ff0c7229 */ /* 0x000fc80000000904 */ /*0190*/ DFMA R10, R8, R10, c[0x2][0x18] ; /* 0x00800600080a762b */ /* 0x001e0c000000000a */ /*01a0*/ DFMA R10, R8, R10, c[0x2][0x20] ; /* 0x00800800080a762b */ /* 0x001e22000000000a */ /*01b0*/ FSETP.GEU.AND P0, PT, |R13|, 4.1917929649353027344, PT ; /* 0x4086232b0d00780b */ /* 0x000fca0003f0e200 */ /*01c0*/ DFMA R10, R8, R10, c[0x2][0x28] ; /* 0x00800a00080a762b */ /* 0x001e0c000000000a */ /*01d0*/ DFMA R10, R8, R10, c[0x2][0x30] ; /* 0x00800c00080a762b */ /* 0x001e0c000000000a */ /*01e0*/ DFMA R10, R8, R10, c[0x2][0x38] ; /* 0x00800e00080a762b */ /* 0x001e0c000000000a */ /*01f0*/ DFMA R10, R8, R10, c[0x2][0x40] ; /* 0x00801000080a762b */ /* 0x001e0c000000000a */ /*0200*/ DFMA R10, R8, R10, c[0x2][0x48] ; /* 0x00801200080a762b */ /* 0x001e0c000000000a */ /*0210*/ DFMA R10, R8, R10, c[0x2][0x50] ; /* 0x00801400080a762b */ /* 0x001e0c000000000a */ /*0220*/ DFMA R10, R8, R10, 1 ; /* 0x3ff00000080a742b */ /* 0x001e0c000000000a */ /*0230*/ DFMA R10, R8, R10, 1 ; /* 0x3ff00000080a742b */ /* 0x001e14000000000a */ /*0240*/ IMAD R9, R6, 0x100000, R11 ; /* 0x0010000006097824 */ /* 0x001fe400078e020b */ /*0250*/ IMAD.MOV.U32 R8, RZ, RZ, R10 ; /* 0x000000ffff087224 */ /* 0x000fe200078e000a */ /*0260*/ @!P0 BRA 0x330 ; /* 0x000000c000008947 */ /* 0x000fea0003800000 */ /*0270*/ FSETP.GEU.AND P1, PT, |R13|, 4.2275390625, PT ; /* 0x408748000d00780b */ /* 0x000fe20003f2e200 */ /*0280*/ DADD R8, -R4, +INF ; /* 0x7ff0000004087429 */ /* 0x000fc80000000100 */ /*0290*/ DSETP.GT.AND P0, PT, R4, RZ, PT ; /* 0x000000ff0400722a */ /* 0x000e0c0003f04000 */ /*02a0*/ FSEL R8, R8, RZ, !P0 ; /* 0x000000ff08087208 */ /* 0x001fe40004000000 */ /*02b0*/ @!P1 LEA.HI R3, R6, R6, RZ, 0x1 ; /* 0x0000000606039211 */ /* 0x000fe400078f08ff */ /*02c0*/ FSEL R9, R9, RZ, !P0 ; /* 0x000000ff09097208 */ /* 0x000fe40004000000 */ /*02d0*/ @!P1 SHF.R.S32.HI R3, RZ, 0x1, R3 ; /* 0x00000001ff039819 */ /* 0x000fca0000011403 */ /*02e0*/ @!P1 IMAD.IADD R4, R6, 0x1, -R3 ; /* 0x0000000106049824 */ /* 0x000fe400078e0a03 */ /*02f0*/ @!P1 IMAD R11, R3, 0x100000, R11 ; /* 0x00100000030b9824 */ /* 0x000fc600078e020b */ /*0300*/ @!P1 LEA R5, R4, 0x3ff00000, 0x14 ; /* 0x3ff0000004059811 */ /* 0x000fe200078ea0ff */ /*0310*/ @!P1 IMAD.MOV.U32 R4, RZ, RZ, RZ ; /* 0x000000ffff049224 */ /* 0x000fcc00078e00ff */ /*0320*/ @!P1 DMUL R8, R10, R4 ; /* 0x000000040a089228 */ /* 0x00004c0000000000 */ /*0330*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0340*/ DADD R10, R8, 1 ; /* 0x3ff00000080a7429 */ /* 0x003e220000000000 */ /*0350*/ BSSY B0, 0x440 ; /* 0x000000e000007945 */ /* 0x000fea0003800000 */ /*0360*/ MUFU.RCP64H R5, R11 ; /* 0x0000000b00057308 */ /* 0x001e280000001800 */ /*0370*/ IADD3 R4, R11, 0x300402, RZ ; /* 0x003004020b047810 */ /* 0x000fc80007ffe0ff */ /*0380*/ FSETP.GEU.AND P0, PT, |R4|, 5.8789094863358348022e-39, PT ; /* 0x004004020400780b */ /* 0x000fe40003f0e200 */ /*0390*/ DFMA R6, -R10, R4, 1 ; /* 0x3ff000000a06742b */ /* 0x001e0c0000000104 */ /*03a0*/ DFMA R6, R6, R6, R6 ; /* 0x000000060606722b */ /* 0x001e0c0000000006 */ /*03b0*/ DFMA R6, R4, R6, R4 ; /* 0x000000060406722b */ /* 0x001e0c0000000004 */ /*03c0*/ DFMA R8, -R10, R6, 1 ; /* 0x3ff000000a08742b */ /* 0x001e0c0000000106 */ /*03d0*/ DFMA R6, R6, R8, R6 ; /* 0x000000080606722b */ /* 0x0010620000000006 */ /*03e0*/ @P0 BRA 0x430 ; /* 0x0000004000000947 */ /* 0x000fea0003800000 */ /*03f0*/ LOP3.LUT R3, R11, 0x7fffffff, RZ, 0xc0, !PT ; /* 0x7fffffff0b037812 */ /* 0x000fe400078ec0ff */ /*0400*/ MOV R4, 0x430 ; /* 0x0000043000047802 */ /* 0x000fe40000000f00 */ /*0410*/ IADD3 R3, R3, -0x100000, RZ ; /* 0xfff0000003037810 */ /* 0x000fe40007ffe0ff */ /*0420*/ CALL.REL.NOINC 0x480 ; /* 0x0000005000007944 */ /* 0x003fea0003c00000 */ /*0430*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0440*/ IADD3 R4, P0, R0, c[0x0][0x168], RZ ; /* 0x00005a0000047a10 */ /* 0x000fc80007f1e0ff */ /*0450*/ IADD3.X R5, R2, c[0x0][0x16c], RZ, P0, !PT ; /* 0x00005b0002057a10 */ /* 0x000fca00007fe4ff */ /*0460*/ STG.E.64 [R4.64], R6 ; /* 0x0000000604007986 */ /* 0x002fe2000c101b04 */ /*0470*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0480*/ IMAD.MOV.U32 R6, RZ, RZ, R10 ; /* 0x000000ffff067224 */ /* 0x000fe200078e000a */ /*0490*/ BSSY B1, 0x700 ; /* 0x0000026000017945 */ /* 0x000fe20003800000 */ /*04a0*/ IMAD.MOV.U32 R7, RZ, RZ, R11 ; /* 0x000000ffff077224 */ /* 0x000fcc00078e000b */ /*04b0*/ DSETP.GTU.AND P0, PT, |R6|, +INF , PT ; /* 0x7ff000000600742a */ /* 0x000e1c0003f0c200 */ /*04c0*/ @P0 BRA 0x6d0 ; /* 0x0000020000000947 */ /* 0x001fea0003800000 */ /*04d0*/ LOP3.LUT R10, R11, 0x7fffffff, RZ, 0xc0, !PT ; /* 0x7fffffff0b0a7812 */ /* 0x000fc800078ec0ff */ /*04e0*/ IADD3 R5, R10, -0x1, RZ ; /* 0xffffffff0a057810 */ /* 0x000fc80007ffe0ff */ /*04f0*/ ISETP.GE.U32.AND P0, PT, R5, 0x7fefffff, PT ; /* 0x7fefffff0500780c */ /* 0x000fda0003f06070 */ /*0500*/ @P0 LOP3.LUT R9, R7, 0x7ff00000, RZ, 0x3c, !PT ; /* 0x7ff0000007090812 */ /* 0x000fe200078e3cff */ /*0510*/ @P0 IMAD.MOV.U32 R8, RZ, RZ, RZ ; /* 0x000000ffff080224 */ /* 0x000fe200078e00ff */ /*0520*/ @P0 BRA 0x6f0 ; /* 0x000001c000000947 */ /* 0x000fea0003800000 */ /*0530*/ ISETP.GE.U32.AND P0, PT, R10, 0x1000001, PT ; /* 0x010000010a00780c */ /* 0x000fda0003f06070 */ /*0540*/ @!P0 BRA 0x630 ; /* 0x000000e000008947 */ /* 0x000fea0003800000 */ /*0550*/ IADD3 R9, R7, -0x3fe00000, RZ ; /* 0xc020000007097810 */ /* 0x000fe20007ffe0ff */ /*0560*/ IMAD.MOV.U32 R8, RZ, RZ, R6 ; /* 0x000000ffff087224 */ /* 0x000fe400078e0006 */ /*0570*/ IMAD.MOV.U32 R10, RZ, RZ, R3 ; /* 0x000000ffff0a7224 */ /* 0x000fe200078e0003 */ /*0580*/ MUFU.RCP64H R11, R9 ; /* 0x00000009000b7308 */ /* 0x000e2a0000001800 */ /*0590*/ DFMA R12, -R8, R10, 1 ; /* 0x3ff00000080c742b */ /* 0x001e0c000000010a */ /*05a0*/ DFMA R12, R12, R12, R12 ; /* 0x0000000c0c0c722b */ /* 0x001e0c000000000c */ /*05b0*/ DFMA R12, R10, R12, R10 ; /* 0x0000000c0a0c722b */ /* 0x001e0c000000000a */ /*05c0*/ DFMA R10, -R8, R12, 1 ; /* 0x3ff00000080a742b */ /* 0x001e0c000000010c */ /*05d0*/ DFMA R10, R12, R10, R12 ; /* 0x0000000a0c0a722b */ /* 0x001e0c000000000c */ /*05e0*/ DMUL R10, R10, 2.2250738585072013831e-308 ; /* 0x001000000a0a7828 */ /* 0x001e0c0000000000 */ /*05f0*/ DFMA R6, -R6, R10, 1 ; /* 0x3ff000000606742b */ /* 0x001e0c000000010a */ /*0600*/ DFMA R6, R6, R6, R6 ; /* 0x000000060606722b */ /* 0x001e0c0000000006 */ /*0610*/ DFMA R8, R10, R6, R10 ; /* 0x000000060a08722b */ /* 0x001062000000000a */ /*0620*/ BRA 0x6f0 ; /* 0x000000c000007947 */ /* 0x000fea0003800000 */ /*0630*/ DMUL R6, R6, 8.11296384146066816958e+31 ; /* 0x4690000006067828 */ /* 0x000e220000000000 */ /*0640*/ IMAD.MOV.U32 R8, RZ, RZ, R3 ; /* 0x000000ffff087224 */ /* 0x000fca00078e0003 */ /*0650*/ MUFU.RCP64H R9, R7 ; /* 0x0000000700097308 */ /* 0x001e240000001800 */ /*0660*/ DFMA R10, -R6, R8, 1 ; /* 0x3ff00000060a742b */ /* 0x001e0c0000000108 */ /*0670*/ DFMA R10, R10, R10, R10 ; /* 0x0000000a0a0a722b */ /* 0x001e0c000000000a */ /*0680*/ DFMA R10, R8, R10, R8 ; /* 0x0000000a080a722b */ /* 0x001e0c0000000008 */ /*0690*/ DFMA R8, -R6, R10, 1 ; /* 0x3ff000000608742b */ /* 0x001e0c000000010a */ /*06a0*/ DFMA R8, R10, R8, R10 ; /* 0x000000080a08722b */ /* 0x001e0c000000000a */ /*06b0*/ DMUL R8, R8, 8.11296384146066816958e+31 ; /* 0x4690000008087828 */ /* 0x001e220000000000 */ /*06c0*/ BRA 0x6f0 ; /* 0x0000002000007947 */ /* 0x000fea0003800000 */ /*06d0*/ LOP3.LUT R9, R7, 0x80000, RZ, 0xfc, !PT ; /* 0x0008000007097812 */ /* 0x000fe200078efcff */ /*06e0*/ IMAD.MOV.U32 R8, RZ, RZ, R6 ; /* 0x000000ffff087224 */ /* 0x000fe400078e0006 */ /*06f0*/ BSYNC B1 ; /* 0x0000000000017941 */ /* 0x000fea0003800000 */ /*0700*/ IMAD.MOV.U32 R5, RZ, RZ, 0x0 ; /* 0x00000000ff057424 */ /* 0x000fe400078e00ff */ /*0710*/ IMAD.MOV.U32 R6, RZ, RZ, R8 ; /* 0x000000ffff067224 */ /* 0x003fe400078e0008 */ /*0720*/ IMAD.MOV.U32 R7, RZ, RZ, R9 ; /* 0x000000ffff077224 */ /* 0x000fe200078e0009 */ /*0730*/ RET.REL.NODEC R4 0x0 ; /* 0xfffff8c004007950 */ /* 0x000fec0003c3ffff */ /*0740*/ BRA 0x740; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0750*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0760*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0770*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0780*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0790*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
// 1 / (1 + e^(-x)) extern "C" __global__ void logistic(size_t n, double *result, double *x) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i<n) { result[i] = 1.0 / (1.0 + exp(-x[i])); } }
.file "tmpxft_0000d6ec_00000000-6_LogisticFunctionKernel_double.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z30__device_stub__Z8logisticmPdS_mPdS_ .type _Z30__device_stub__Z8logisticmPdS_mPdS_, @function _Z30__device_stub__Z8logisticmPdS_mPdS_: .LFB2051: .cfi_startproc endbr64 subq $136, %rsp .cfi_def_cfa_offset 144 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movq %fs:40, %rax movq %rax, 120(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 120(%rsp), %rax subq %fs:40, %rax jne .L8 addq $136, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq logistic(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z30__device_stub__Z8logisticmPdS_mPdS_, .-_Z30__device_stub__Z8logisticmPdS_mPdS_ .globl logistic .type logistic, @function logistic: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z30__device_stub__Z8logisticmPdS_mPdS_ addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size logistic, .-logistic .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "logistic" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq logistic(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
// 1 / (1 + e^(-x)) extern "C" __global__ void logistic(size_t n, double *result, double *x) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i<n) { result[i] = 1.0 / (1.0 + exp(-x[i])); } }
#include <hip/hip_runtime.h> // 1 / (1 + e^(-x)) extern "C" __global__ void logistic(size_t n, double *result, double *x) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i<n) { result[i] = 1.0 / (1.0 + exp(-x[i])); } }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> // 1 / (1 + e^(-x)) extern "C" __global__ void logistic(size_t n, double *result, double *x) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i<n) { result[i] = 1.0 / (1.0 + exp(-x[i])); } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected logistic .globl logistic .p2align 8 .type logistic,@function logistic: s_clause 0x1 s_load_b32 s4, s[0:1], 0x24 s_load_b64 s[2:3], s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_and_b32 s4, s4, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s4, v[0:1] v_ashrrev_i32_e32 v2, 31, v1 s_delay_alu instid0(VALU_DEP_1) v_cmp_gt_u64_e32 vcc_lo, s[2:3], v[1:2] s_and_saveexec_b32 s2, vcc_lo s_cbranch_execz .LBB0_2 s_load_b128 s[4:7], s[0:1], 0x8 v_lshlrev_b64 v[0:1], 3, v[1:2] s_mov_b32 s1, 0xbff71547 s_mov_b32 s0, 0x652b82fe s_mov_b32 s3, 0x3e5ade15 s_mov_b32 s2, 0x6a5dcb37 s_waitcnt lgkmcnt(0) v_add_co_u32 v2, vcc_lo, s6, v0 v_add_co_ci_u32_e32 v3, vcc_lo, s7, v1, vcc_lo global_load_b64 v[2:3], v[2:3], off s_waitcnt vmcnt(0) v_mul_f64 v[4:5], v[2:3], s[0:1] s_mov_b32 s1, 0xbfe62e42 s_mov_b32 s0, 0xfefa39ef s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_rndne_f64_e32 v[4:5], v[4:5] v_fma_f64 v[6:7], v[4:5], s[0:1], -v[2:3] s_mov_b32 s1, 0xbc7abc9e s_mov_b32 s0, 0x3b39803f v_cvt_i32_f64_e32 v10, v[4:5] s_delay_alu instid0(VALU_DEP_2) v_fma_f64 v[6:7], v[4:5], s[0:1], v[6:7] s_mov_b32 s1, 0x3e928af3 s_mov_b32 s0, 0xfca7ab0c s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], s[2:3], s[0:1] s_mov_b32 s1, 0x3ec71dee s_mov_b32 s0, 0x623fde64 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3efa0199 s_mov_b32 s0, 0x7c89e6b0 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3f2a01a0 s_mov_b32 s0, 0x14761f6e s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3f56c16c s_mov_b32 s0, 0x1852b7b0 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3f811111 s_mov_b32 s0, 0x11122322 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3fa55555 s_mov_b32 s0, 0x555502a1 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3fc55555 s_mov_b32 s0, 0x55555511 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3fe00000 s_mov_b32 s0, 11 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] v_cmp_nlt_f64_e64 s0, 0x4090cc00, v[2:3] s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[6:7], v[8:9], 1.0 v_fma_f64 v[4:5], v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ldexp_f64 v[4:5], v[4:5], v10 v_add_f64 v[4:5], v[4:5], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_div_scale_f64 v[6:7], null, v[4:5], v[4:5], 1.0 v_div_scale_f64 v[12:13], vcc_lo, 1.0, v[4:5], 1.0 v_rcp_f64_e32 v[8:9], v[6:7] s_waitcnt_depctr 0xfff v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_mul_f64 v[10:11], v[12:13], v[8:9] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[6:7], -v[6:7], v[10:11], v[12:13] v_div_fmas_f64 v[6:7], v[6:7], v[8:9], v[10:11] v_cmp_ngt_f64_e32 vcc_lo, 0xc0900000, v[2:3] s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_div_fixup_f64 v[4:5], v[6:7], v[4:5], 1.0 v_cndmask_b32_e32 v5, 0, v5, vcc_lo s_and_b32 vcc_lo, s0, vcc_lo s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_cndmask_b32_e32 v2, 0, v4, vcc_lo v_add_co_u32 v0, vcc_lo, s4, v0 v_cndmask_b32_e64 v3, 0x3ff00000, v5, s0 v_add_co_ci_u32_e32 v1, vcc_lo, s5, v1, vcc_lo global_store_b64 v[0:1], v[2:3], off .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel logistic .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 280 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size logistic, .Lfunc_end0-logistic .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 8 .value_kind: by_value - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: hidden_block_count_x - .offset: 28 .size: 4 .value_kind: hidden_block_count_y - .offset: 32 .size: 4 .value_kind: hidden_block_count_z - .offset: 36 .size: 2 .value_kind: hidden_group_size_x - .offset: 38 .size: 2 .value_kind: hidden_group_size_y - .offset: 40 .size: 2 .value_kind: hidden_group_size_z - .offset: 42 .size: 2 .value_kind: hidden_remainder_x - .offset: 44 .size: 2 .value_kind: hidden_remainder_y - .offset: 46 .size: 2 .value_kind: hidden_remainder_z - .offset: 64 .size: 8 .value_kind: hidden_global_offset_x - .offset: 72 .size: 8 .value_kind: hidden_global_offset_y - .offset: 80 .size: 8 .value_kind: hidden_global_offset_z - .offset: 88 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 280 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: logistic .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: logistic.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> // 1 / (1 + e^(-x)) extern "C" __global__ void logistic(size_t n, double *result, double *x) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i<n) { result[i] = 1.0 / (1.0 + exp(-x[i])); } }
.text .file "LogisticFunctionKernel_double.hip" .globl __device_stub__logistic # -- Begin function __device_stub__logistic .p2align 4, 0x90 .type __device_stub__logistic,@function __device_stub__logistic: # @__device_stub__logistic .cfi_startproc # %bb.0: subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 80(%rsp), %r9 movl $logistic, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $120, %rsp .cfi_adjust_cfa_offset -120 retq .Lfunc_end0: .size __device_stub__logistic, .Lfunc_end0-__device_stub__logistic .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $logistic, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type logistic,@object # @logistic .section .rodata,"a",@progbits .globl logistic .p2align 3, 0x0 logistic: .quad __device_stub__logistic .size logistic, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "logistic" .size .L__unnamed_1, 9 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __device_stub__logistic .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym logistic .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : logistic .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e280000002500 */ /*0020*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R2, R2, c[0x0][0x0], R3 ; /* 0x0000000002027a24 */ /* 0x001fca00078e0203 */ /*0040*/ ISETP.GE.U32.AND P0, PT, R2, c[0x0][0x160], PT ; /* 0x0000580002007a0c */ /* 0x000fe40003f06070 */ /*0050*/ SHF.R.S32.HI R3, RZ, 0x1f, R2 ; /* 0x0000001fff037819 */ /* 0x000fc80000011402 */ /*0060*/ ISETP.GE.U32.AND.EX P0, PT, R3, c[0x0][0x164], PT, P0 ; /* 0x0000590003007a0c */ /* 0x000fda0003f06100 */ /*0070*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0080*/ IMAD.SHL.U32 R0, R2.reuse, 0x8, RZ ; /* 0x0000000802007824 */ /* 0x040fe200078e00ff */ /*0090*/ SHF.L.U64.HI R2, R2, 0x3, R3 ; /* 0x0000000302027819 */ /* 0x000fe20000010203 */ /*00a0*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*00b0*/ IADD3 R4, P0, R0, c[0x0][0x170], RZ ; /* 0x00005c0000047a10 */ /* 0x000fc80007f1e0ff */ /*00c0*/ IADD3.X R5, R2, c[0x0][0x174], RZ, P0, !PT ; /* 0x00005d0002057a10 */ /* 0x000fcc00007fe4ff */ /*00d0*/ LDG.E.64 R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000ea2000c1e1b00 */ /*00e0*/ IMAD.MOV.U32 R6, RZ, RZ, 0x652b82fe ; /* 0x652b82feff067424 */ /* 0x000fe200078e00ff */ /*00f0*/ BSSY B0, 0x340 ; /* 0x0000024000007945 */ /* 0x000fe20003800000 */ /*0100*/ IMAD.MOV.U32 R7, RZ, RZ, 0x3ff71547 ; /* 0x3ff71547ff077424 */ /* 0x000fe400078e00ff */ /*0110*/ IMAD.MOV.U32 R12, RZ, RZ, 0x69ce2bdf ; /* 0x69ce2bdfff0c7424 */ /* 0x000fe400078e00ff */ /*0120*/ IMAD.MOV.U32 R13, RZ, RZ, 0x3e5ade15 ; /* 0x3e5ade15ff0d7424 */ /* 0x000fe400078e00ff */ /*0130*/ DFMA R6, -R4, R6, 6.75539944105574400000e+15 ; /* 0x433800000406742b */ /* 0x004e0c0000000106 */ /*0140*/ DADD R8, R6, -6.75539944105574400000e+15 ; /* 0xc338000006087429 */ /* 0x001e0c0000000000 */ /*0150*/ DFMA R10, R8, c[0x2][0x0], -R4 ; /* 0x00800000080a7a2b */ /* 0x001e0c0000000804 */ /*0160*/ DFMA R8, R8, c[0x2][0x8], R10 ; /* 0x0080020008087a2b */ /* 0x001e0c000000000a */ /*0170*/ DFMA R10, R8, R12, c[0x2][0x10] ; /* 0x00800400080a762b */ /* 0x001e08000000000c */ /*0180*/ DADD R12, -RZ, -R4 ; /* 0x00000000ff0c7229 */ /* 0x000fc80000000904 */ /*0190*/ DFMA R10, R8, R10, c[0x2][0x18] ; /* 0x00800600080a762b */ /* 0x001e0c000000000a */ /*01a0*/ DFMA R10, R8, R10, c[0x2][0x20] ; /* 0x00800800080a762b */ /* 0x001e22000000000a */ /*01b0*/ FSETP.GEU.AND P0, PT, |R13|, 4.1917929649353027344, PT ; /* 0x4086232b0d00780b */ /* 0x000fca0003f0e200 */ /*01c0*/ DFMA R10, R8, R10, c[0x2][0x28] ; /* 0x00800a00080a762b */ /* 0x001e0c000000000a */ /*01d0*/ DFMA R10, R8, R10, c[0x2][0x30] ; /* 0x00800c00080a762b */ /* 0x001e0c000000000a */ /*01e0*/ DFMA R10, R8, R10, c[0x2][0x38] ; /* 0x00800e00080a762b */ /* 0x001e0c000000000a */ /*01f0*/ DFMA R10, R8, R10, c[0x2][0x40] ; /* 0x00801000080a762b */ /* 0x001e0c000000000a */ /*0200*/ DFMA R10, R8, R10, c[0x2][0x48] ; /* 0x00801200080a762b */ /* 0x001e0c000000000a */ /*0210*/ DFMA R10, R8, R10, c[0x2][0x50] ; /* 0x00801400080a762b */ /* 0x001e0c000000000a */ /*0220*/ DFMA R10, R8, R10, 1 ; /* 0x3ff00000080a742b */ /* 0x001e0c000000000a */ /*0230*/ DFMA R10, R8, R10, 1 ; /* 0x3ff00000080a742b */ /* 0x001e14000000000a */ /*0240*/ IMAD R9, R6, 0x100000, R11 ; /* 0x0010000006097824 */ /* 0x001fe400078e020b */ /*0250*/ IMAD.MOV.U32 R8, RZ, RZ, R10 ; /* 0x000000ffff087224 */ /* 0x000fe200078e000a */ /*0260*/ @!P0 BRA 0x330 ; /* 0x000000c000008947 */ /* 0x000fea0003800000 */ /*0270*/ FSETP.GEU.AND P1, PT, |R13|, 4.2275390625, PT ; /* 0x408748000d00780b */ /* 0x000fe20003f2e200 */ /*0280*/ DADD R8, -R4, +INF ; /* 0x7ff0000004087429 */ /* 0x000fc80000000100 */ /*0290*/ DSETP.GT.AND P0, PT, R4, RZ, PT ; /* 0x000000ff0400722a */ /* 0x000e0c0003f04000 */ /*02a0*/ FSEL R8, R8, RZ, !P0 ; /* 0x000000ff08087208 */ /* 0x001fe40004000000 */ /*02b0*/ @!P1 LEA.HI R3, R6, R6, RZ, 0x1 ; /* 0x0000000606039211 */ /* 0x000fe400078f08ff */ /*02c0*/ FSEL R9, R9, RZ, !P0 ; /* 0x000000ff09097208 */ /* 0x000fe40004000000 */ /*02d0*/ @!P1 SHF.R.S32.HI R3, RZ, 0x1, R3 ; /* 0x00000001ff039819 */ /* 0x000fca0000011403 */ /*02e0*/ @!P1 IMAD.IADD R4, R6, 0x1, -R3 ; /* 0x0000000106049824 */ /* 0x000fe400078e0a03 */ /*02f0*/ @!P1 IMAD R11, R3, 0x100000, R11 ; /* 0x00100000030b9824 */ /* 0x000fc600078e020b */ /*0300*/ @!P1 LEA R5, R4, 0x3ff00000, 0x14 ; /* 0x3ff0000004059811 */ /* 0x000fe200078ea0ff */ /*0310*/ @!P1 IMAD.MOV.U32 R4, RZ, RZ, RZ ; /* 0x000000ffff049224 */ /* 0x000fcc00078e00ff */ /*0320*/ @!P1 DMUL R8, R10, R4 ; /* 0x000000040a089228 */ /* 0x00004c0000000000 */ /*0330*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0340*/ DADD R10, R8, 1 ; /* 0x3ff00000080a7429 */ /* 0x003e220000000000 */ /*0350*/ BSSY B0, 0x440 ; /* 0x000000e000007945 */ /* 0x000fea0003800000 */ /*0360*/ MUFU.RCP64H R5, R11 ; /* 0x0000000b00057308 */ /* 0x001e280000001800 */ /*0370*/ IADD3 R4, R11, 0x300402, RZ ; /* 0x003004020b047810 */ /* 0x000fc80007ffe0ff */ /*0380*/ FSETP.GEU.AND P0, PT, |R4|, 5.8789094863358348022e-39, PT ; /* 0x004004020400780b */ /* 0x000fe40003f0e200 */ /*0390*/ DFMA R6, -R10, R4, 1 ; /* 0x3ff000000a06742b */ /* 0x001e0c0000000104 */ /*03a0*/ DFMA R6, R6, R6, R6 ; /* 0x000000060606722b */ /* 0x001e0c0000000006 */ /*03b0*/ DFMA R6, R4, R6, R4 ; /* 0x000000060406722b */ /* 0x001e0c0000000004 */ /*03c0*/ DFMA R8, -R10, R6, 1 ; /* 0x3ff000000a08742b */ /* 0x001e0c0000000106 */ /*03d0*/ DFMA R6, R6, R8, R6 ; /* 0x000000080606722b */ /* 0x0010620000000006 */ /*03e0*/ @P0 BRA 0x430 ; /* 0x0000004000000947 */ /* 0x000fea0003800000 */ /*03f0*/ LOP3.LUT R3, R11, 0x7fffffff, RZ, 0xc0, !PT ; /* 0x7fffffff0b037812 */ /* 0x000fe400078ec0ff */ /*0400*/ MOV R4, 0x430 ; /* 0x0000043000047802 */ /* 0x000fe40000000f00 */ /*0410*/ IADD3 R3, R3, -0x100000, RZ ; /* 0xfff0000003037810 */ /* 0x000fe40007ffe0ff */ /*0420*/ CALL.REL.NOINC 0x480 ; /* 0x0000005000007944 */ /* 0x003fea0003c00000 */ /*0430*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*0440*/ IADD3 R4, P0, R0, c[0x0][0x168], RZ ; /* 0x00005a0000047a10 */ /* 0x000fc80007f1e0ff */ /*0450*/ IADD3.X R5, R2, c[0x0][0x16c], RZ, P0, !PT ; /* 0x00005b0002057a10 */ /* 0x000fca00007fe4ff */ /*0460*/ STG.E.64 [R4.64], R6 ; /* 0x0000000604007986 */ /* 0x002fe2000c101b04 */ /*0470*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0480*/ IMAD.MOV.U32 R6, RZ, RZ, R10 ; /* 0x000000ffff067224 */ /* 0x000fe200078e000a */ /*0490*/ BSSY B1, 0x700 ; /* 0x0000026000017945 */ /* 0x000fe20003800000 */ /*04a0*/ IMAD.MOV.U32 R7, RZ, RZ, R11 ; /* 0x000000ffff077224 */ /* 0x000fcc00078e000b */ /*04b0*/ DSETP.GTU.AND P0, PT, |R6|, +INF , PT ; /* 0x7ff000000600742a */ /* 0x000e1c0003f0c200 */ /*04c0*/ @P0 BRA 0x6d0 ; /* 0x0000020000000947 */ /* 0x001fea0003800000 */ /*04d0*/ LOP3.LUT R10, R11, 0x7fffffff, RZ, 0xc0, !PT ; /* 0x7fffffff0b0a7812 */ /* 0x000fc800078ec0ff */ /*04e0*/ IADD3 R5, R10, -0x1, RZ ; /* 0xffffffff0a057810 */ /* 0x000fc80007ffe0ff */ /*04f0*/ ISETP.GE.U32.AND P0, PT, R5, 0x7fefffff, PT ; /* 0x7fefffff0500780c */ /* 0x000fda0003f06070 */ /*0500*/ @P0 LOP3.LUT R9, R7, 0x7ff00000, RZ, 0x3c, !PT ; /* 0x7ff0000007090812 */ /* 0x000fe200078e3cff */ /*0510*/ @P0 IMAD.MOV.U32 R8, RZ, RZ, RZ ; /* 0x000000ffff080224 */ /* 0x000fe200078e00ff */ /*0520*/ @P0 BRA 0x6f0 ; /* 0x000001c000000947 */ /* 0x000fea0003800000 */ /*0530*/ ISETP.GE.U32.AND P0, PT, R10, 0x1000001, PT ; /* 0x010000010a00780c */ /* 0x000fda0003f06070 */ /*0540*/ @!P0 BRA 0x630 ; /* 0x000000e000008947 */ /* 0x000fea0003800000 */ /*0550*/ IADD3 R9, R7, -0x3fe00000, RZ ; /* 0xc020000007097810 */ /* 0x000fe20007ffe0ff */ /*0560*/ IMAD.MOV.U32 R8, RZ, RZ, R6 ; /* 0x000000ffff087224 */ /* 0x000fe400078e0006 */ /*0570*/ IMAD.MOV.U32 R10, RZ, RZ, R3 ; /* 0x000000ffff0a7224 */ /* 0x000fe200078e0003 */ /*0580*/ MUFU.RCP64H R11, R9 ; /* 0x00000009000b7308 */ /* 0x000e2a0000001800 */ /*0590*/ DFMA R12, -R8, R10, 1 ; /* 0x3ff00000080c742b */ /* 0x001e0c000000010a */ /*05a0*/ DFMA R12, R12, R12, R12 ; /* 0x0000000c0c0c722b */ /* 0x001e0c000000000c */ /*05b0*/ DFMA R12, R10, R12, R10 ; /* 0x0000000c0a0c722b */ /* 0x001e0c000000000a */ /*05c0*/ DFMA R10, -R8, R12, 1 ; /* 0x3ff00000080a742b */ /* 0x001e0c000000010c */ /*05d0*/ DFMA R10, R12, R10, R12 ; /* 0x0000000a0c0a722b */ /* 0x001e0c000000000c */ /*05e0*/ DMUL R10, R10, 2.2250738585072013831e-308 ; /* 0x001000000a0a7828 */ /* 0x001e0c0000000000 */ /*05f0*/ DFMA R6, -R6, R10, 1 ; /* 0x3ff000000606742b */ /* 0x001e0c000000010a */ /*0600*/ DFMA R6, R6, R6, R6 ; /* 0x000000060606722b */ /* 0x001e0c0000000006 */ /*0610*/ DFMA R8, R10, R6, R10 ; /* 0x000000060a08722b */ /* 0x001062000000000a */ /*0620*/ BRA 0x6f0 ; /* 0x000000c000007947 */ /* 0x000fea0003800000 */ /*0630*/ DMUL R6, R6, 8.11296384146066816958e+31 ; /* 0x4690000006067828 */ /* 0x000e220000000000 */ /*0640*/ IMAD.MOV.U32 R8, RZ, RZ, R3 ; /* 0x000000ffff087224 */ /* 0x000fca00078e0003 */ /*0650*/ MUFU.RCP64H R9, R7 ; /* 0x0000000700097308 */ /* 0x001e240000001800 */ /*0660*/ DFMA R10, -R6, R8, 1 ; /* 0x3ff00000060a742b */ /* 0x001e0c0000000108 */ /*0670*/ DFMA R10, R10, R10, R10 ; /* 0x0000000a0a0a722b */ /* 0x001e0c000000000a */ /*0680*/ DFMA R10, R8, R10, R8 ; /* 0x0000000a080a722b */ /* 0x001e0c0000000008 */ /*0690*/ DFMA R8, -R6, R10, 1 ; /* 0x3ff000000608742b */ /* 0x001e0c000000010a */ /*06a0*/ DFMA R8, R10, R8, R10 ; /* 0x000000080a08722b */ /* 0x001e0c000000000a */ /*06b0*/ DMUL R8, R8, 8.11296384146066816958e+31 ; /* 0x4690000008087828 */ /* 0x001e220000000000 */ /*06c0*/ BRA 0x6f0 ; /* 0x0000002000007947 */ /* 0x000fea0003800000 */ /*06d0*/ LOP3.LUT R9, R7, 0x80000, RZ, 0xfc, !PT ; /* 0x0008000007097812 */ /* 0x000fe200078efcff */ /*06e0*/ IMAD.MOV.U32 R8, RZ, RZ, R6 ; /* 0x000000ffff087224 */ /* 0x000fe400078e0006 */ /*06f0*/ BSYNC B1 ; /* 0x0000000000017941 */ /* 0x000fea0003800000 */ /*0700*/ IMAD.MOV.U32 R5, RZ, RZ, 0x0 ; /* 0x00000000ff057424 */ /* 0x000fe400078e00ff */ /*0710*/ IMAD.MOV.U32 R6, RZ, RZ, R8 ; /* 0x000000ffff067224 */ /* 0x003fe400078e0008 */ /*0720*/ IMAD.MOV.U32 R7, RZ, RZ, R9 ; /* 0x000000ffff077224 */ /* 0x000fe200078e0009 */ /*0730*/ RET.REL.NODEC R4 0x0 ; /* 0xfffff8c004007950 */ /* 0x000fec0003c3ffff */ /*0740*/ BRA 0x740; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0750*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0760*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0770*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0780*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0790*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected logistic .globl logistic .p2align 8 .type logistic,@function logistic: s_clause 0x1 s_load_b32 s4, s[0:1], 0x24 s_load_b64 s[2:3], s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_and_b32 s4, s4, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s4, v[0:1] v_ashrrev_i32_e32 v2, 31, v1 s_delay_alu instid0(VALU_DEP_1) v_cmp_gt_u64_e32 vcc_lo, s[2:3], v[1:2] s_and_saveexec_b32 s2, vcc_lo s_cbranch_execz .LBB0_2 s_load_b128 s[4:7], s[0:1], 0x8 v_lshlrev_b64 v[0:1], 3, v[1:2] s_mov_b32 s1, 0xbff71547 s_mov_b32 s0, 0x652b82fe s_mov_b32 s3, 0x3e5ade15 s_mov_b32 s2, 0x6a5dcb37 s_waitcnt lgkmcnt(0) v_add_co_u32 v2, vcc_lo, s6, v0 v_add_co_ci_u32_e32 v3, vcc_lo, s7, v1, vcc_lo global_load_b64 v[2:3], v[2:3], off s_waitcnt vmcnt(0) v_mul_f64 v[4:5], v[2:3], s[0:1] s_mov_b32 s1, 0xbfe62e42 s_mov_b32 s0, 0xfefa39ef s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_rndne_f64_e32 v[4:5], v[4:5] v_fma_f64 v[6:7], v[4:5], s[0:1], -v[2:3] s_mov_b32 s1, 0xbc7abc9e s_mov_b32 s0, 0x3b39803f v_cvt_i32_f64_e32 v10, v[4:5] s_delay_alu instid0(VALU_DEP_2) v_fma_f64 v[6:7], v[4:5], s[0:1], v[6:7] s_mov_b32 s1, 0x3e928af3 s_mov_b32 s0, 0xfca7ab0c s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], s[2:3], s[0:1] s_mov_b32 s1, 0x3ec71dee s_mov_b32 s0, 0x623fde64 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3efa0199 s_mov_b32 s0, 0x7c89e6b0 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3f2a01a0 s_mov_b32 s0, 0x14761f6e s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3f56c16c s_mov_b32 s0, 0x1852b7b0 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3f811111 s_mov_b32 s0, 0x11122322 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3fa55555 s_mov_b32 s0, 0x555502a1 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3fc55555 s_mov_b32 s0, 0x55555511 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] s_mov_b32 s1, 0x3fe00000 s_mov_b32 s0, 11 s_delay_alu instid0(VALU_DEP_1) | instid1(SALU_CYCLE_1) v_fma_f64 v[8:9], v[6:7], v[8:9], s[0:1] v_cmp_nlt_f64_e64 s0, 0x4090cc00, v[2:3] s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[6:7], v[8:9], 1.0 v_fma_f64 v[4:5], v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_ldexp_f64 v[4:5], v[4:5], v10 v_add_f64 v[4:5], v[4:5], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_div_scale_f64 v[6:7], null, v[4:5], v[4:5], 1.0 v_div_scale_f64 v[12:13], vcc_lo, 1.0, v[4:5], 1.0 v_rcp_f64_e32 v[8:9], v[6:7] s_waitcnt_depctr 0xfff v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_fma_f64 v[10:11], -v[6:7], v[8:9], 1.0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[8:9], v[8:9], v[10:11], v[8:9] v_mul_f64 v[10:11], v[12:13], v[8:9] s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fma_f64 v[6:7], -v[6:7], v[10:11], v[12:13] v_div_fmas_f64 v[6:7], v[6:7], v[8:9], v[10:11] v_cmp_ngt_f64_e32 vcc_lo, 0xc0900000, v[2:3] s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_div_fixup_f64 v[4:5], v[6:7], v[4:5], 1.0 v_cndmask_b32_e32 v5, 0, v5, vcc_lo s_and_b32 vcc_lo, s0, vcc_lo s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_cndmask_b32_e32 v2, 0, v4, vcc_lo v_add_co_u32 v0, vcc_lo, s4, v0 v_cndmask_b32_e64 v3, 0x3ff00000, v5, s0 v_add_co_ci_u32_e32 v1, vcc_lo, s5, v1, vcc_lo global_store_b64 v[0:1], v[2:3], off .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel logistic .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 280 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 0 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 0 .amdhsa_next_free_vgpr 14 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size logistic, .Lfunc_end0-logistic .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 8 .value_kind: by_value - .address_space: global .offset: 8 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .offset: 24 .size: 4 .value_kind: hidden_block_count_x - .offset: 28 .size: 4 .value_kind: hidden_block_count_y - .offset: 32 .size: 4 .value_kind: hidden_block_count_z - .offset: 36 .size: 2 .value_kind: hidden_group_size_x - .offset: 38 .size: 2 .value_kind: hidden_group_size_y - .offset: 40 .size: 2 .value_kind: hidden_group_size_z - .offset: 42 .size: 2 .value_kind: hidden_remainder_x - .offset: 44 .size: 2 .value_kind: hidden_remainder_y - .offset: 46 .size: 2 .value_kind: hidden_remainder_z - .offset: 64 .size: 8 .value_kind: hidden_global_offset_x - .offset: 72 .size: 8 .value_kind: hidden_global_offset_y - .offset: 80 .size: 8 .value_kind: hidden_global_offset_z - .offset: 88 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 280 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: logistic .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: logistic.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 14 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_0000d6ec_00000000-6_LogisticFunctionKernel_double.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z30__device_stub__Z8logisticmPdS_mPdS_ .type _Z30__device_stub__Z8logisticmPdS_mPdS_, @function _Z30__device_stub__Z8logisticmPdS_mPdS_: .LFB2051: .cfi_startproc endbr64 subq $136, %rsp .cfi_def_cfa_offset 144 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movq %fs:40, %rax movq %rax, 120(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 120(%rsp), %rax subq %fs:40, %rax jne .L8 addq $136, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq logistic(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z30__device_stub__Z8logisticmPdS_mPdS_, .-_Z30__device_stub__Z8logisticmPdS_mPdS_ .globl logistic .type logistic, @function logistic: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z30__device_stub__Z8logisticmPdS_mPdS_ addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size logistic, .-logistic .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "logistic" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq logistic(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "LogisticFunctionKernel_double.hip" .globl __device_stub__logistic # -- Begin function __device_stub__logistic .p2align 4, 0x90 .type __device_stub__logistic,@function __device_stub__logistic: # @__device_stub__logistic .cfi_startproc # %bb.0: subq $104, %rsp .cfi_def_cfa_offset 112 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movq %rdx, 56(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 56(%rsp), %rax movq %rax, 96(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 80(%rsp), %r9 movl $logistic, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $120, %rsp .cfi_adjust_cfa_offset -120 retq .Lfunc_end0: .size __device_stub__logistic, .Lfunc_end0-__device_stub__logistic .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB1_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB1_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $logistic, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end1: .size __hip_module_ctor, .Lfunc_end1-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB2_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB2_2: retq .Lfunc_end2: .size __hip_module_dtor, .Lfunc_end2-__hip_module_dtor .cfi_endproc # -- End function .type logistic,@object # @logistic .section .rodata,"a",@progbits .globl logistic .p2align 3, 0x0 logistic: .quad __device_stub__logistic .size logistic, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "logistic" .size .L__unnamed_1, 9 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __device_stub__logistic .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym logistic .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
// Kernel code for Gaussian Mixture Model Expectation Maximization. // // Andrew Harp (andrew.harp@gmail.com) // http://andrewharp.com/gmmcuda // This effects the maximum dimensionality of your data. // Has to be hardcoded because it affects memory allocation. // Due to the way I clear out the lower triangle of the cholesky (and // possibly other) places MAX_DIM * MAX_DIM needs to be less than // BLOCK_SIZE. #define MAX_DIM 16 // You can change this, but only to a power of 2. #define BLOCK_SIZE 256 // Estep-normalize is broken down into an arbitrary number of chunks. #define NUM_CHUNKS 32 #include "cuda_runtime.h" /////////////////////////////////////////////////////////////////////////// __device__ void cholesky(float* el, const unsigned int ndim) { const unsigned int tid = threadIdx.x; // Dunno how to parallelize this part... if (tid == 0) { float sum = 0; int i, j, k; //if (el.ncols() != n) throw("need square matrix"); for (i=0; i<ndim; i++) { for (j=i; j<ndim; j++) { sum = el[__umul24(i, ndim)+j]; for (k=i-1; k >= 0; k--) { sum -= el[__umul24(i, ndim)+k] * el[__umul24(j, ndim)+k]; } if (i == j) { //if (sum <= 0.0) // throw("Cholesky failed"); el[__umul24(i, ndim)+i] = sqrt(sum); } else { el[__umul24(j, ndim)+i] = sum/el[__umul24(i, ndim)+i]; } } } } __syncthreads(); // Clear lower triangular part. if ((tid/ndim) < (tid%ndim)) { el[__umul24((tid/ndim), ndim) + (tid%ndim)] = 0.0f; } } /////////////////////////////////////////////////////////////////////////// __device__ float logdet(float* el, const unsigned int ndim) { float sum = 0.0f; for (unsigned int i=0; i<ndim; ++i) { sum += __logf(el[(i*ndim)+i]); } return 2.*sum; } /////////////////////////////////////////////////////////////////////////// __device__ float parallelSum(float* data, const unsigned int ndata) { const unsigned int tid = threadIdx.x; float t; __syncthreads(); // Butterfly sum. ndata MUST be a power of 2. for(unsigned int bit = ndata >> 1; bit > 0; bit >>= 1) { t = data[tid] + data[tid^bit]; __syncthreads(); data[tid] = t; __syncthreads(); } return data[tid]; } /////////////////////////////////////////////////////////////////////////// __device__ void copyArray(float* fromArr, float* toArr, unsigned int ndata=BLOCK_SIZE) { unsigned int n; unsigned int base_off; for (base_off=0; base_off < ndata; base_off += blockDim.x) { n = base_off + threadIdx.x; if (n < ndata) { toArr[n] = fromArr[n]; } } } /////////////////////////////////////////////////////////////////////////// // Parallel reduction, for when all you want is the sum of a certain // quantity computed for every 1 to N. CODE should be something in terms // of n. The resulting sum will be placed in RESULT. // tmp_buff, base_off, RESULT, and n must be previously defined, however // they will be overwritten during the execution of the macro. #define REDUCE(N, CODE, RESULT) \ base_off = 0; \ RESULT = 0.0f; \ while (base_off + BLOCK_SIZE < N) { \ n = base_off + tid; \ tmp_buff[tid] = CODE; \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); \ base_off += BLOCK_SIZE; \ } \ n = base_off + tid; \ if (n < N) {tmp_buff[tid] = CODE;} \ else {tmp_buff[tid] = 0.0f;} \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); /////////////////////////////////////////////////////////////////////////// // This function computes for a single cluster k. __global__ void estep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int tid = threadIdx.x; const unsigned int k = blockIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int lndb = __umul24(try_num, num_clusts); unsigned int n, base_off; float tmp, sum; float v[MAX_DIM]; __shared__ float lndet_k; __shared__ float frac_k; __shared__ float chol[MAX_DIM * MAX_DIM]; __syncthreads(); copyArray(_sig_ + sb + __umul24(k, sigsize), chol, sigsize); cholesky(chol, num_dims); __syncthreads(); if (tid == 0) { frac_k = _frac_[fb + k]; lndet_k = logdet(chol, num_dims); _lndets_[lndb + k] = lndet_k; } __syncthreads(); // Loop through data. for (base_off=0; base_off < num_data; base_off += BLOCK_SIZE) { n = base_off + tid; sum=0.0f; //if (b.size() != n || y.size() != n) throw("bad lengths"); if (n < num_data) { for (unsigned int i=0; i<num_dims; ++i) { tmp = _data_[__umul24(i, num_data) + n] - _means_[mb + __umul24(k, num_dims)+i]; for (unsigned int j=0; j<i; j++) { tmp -= chol[__umul24(i, num_dims)+j] * v[j]; } v[i] = tmp/chol[__umul24(i, num_dims)+i]; sum += v[i] * v[i]; } // Assign likelihood of this data being in this cluster. _resp_[rb + n] = -0.5f*(sum + lndet_k) + __logf(frac_k); } } // (n < num_data) } /////////////////////////////////////////////////////////////////////////// // Loop through data again, normalizing probabilities. // We are looping across clusters here as well as data, since every data // point needs to know its potential parents. __global__ void estep_normalize_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, float* _loglike_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int num_chunks = gridDim.x; const unsigned int chunk_num = blockIdx.x; const unsigned int tid = threadIdx.x; // We're only handling so many data points per block in this kernel, since // data is independant of other data here. const unsigned int n_per_block = ceil((float)num_data / (float)num_chunks); const unsigned int start_off = __umul24(n_per_block, chunk_num); const unsigned int end_off = min(start_off + n_per_block, num_data); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data; const unsigned int lb = __umul24(try_num, num_chunks); unsigned int n, base_off, k; float sum, max, tmp; __shared__ float loglike[BLOCK_SIZE]; loglike[tid] = 0.0f; __syncthreads(); // Loop through data. for (base_off = start_off; base_off < end_off; base_off += BLOCK_SIZE) { n = base_off + tid; if (n < end_off) { max = -99.9e30f; // Find cluster with maximum likelihood for this data point. for (k=0; k<num_clusts; ++k) { tmp = _resp_[rb + (k*num_data) + n]; if (tmp > max) { max = tmp; } } // Sum marginal probabilities. sum = 0.0f; for (k=0; k<num_clusts; ++k) { sum += __expf(_resp_[rb + (k*num_data) + n] - max); } // Assign probabilities of point belonging to each cluster. tmp = max + __logf(sum); for (k = 0; k < num_clusts; ++k) { _resp_[rb + (k*num_data) + n] = __expf(_resp_[rb + (k*num_data) + n] - tmp); } loglike[tid] += tmp; } } tmp = parallelSum(loglike, BLOCK_SIZE); if (tid == 0) { _loglike_[lb + chunk_num] = tmp; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int tid = threadIdx.x; // Every block is mapped to cluster and dimension. const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y; // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int fb = __umul24(try_num, num_clusts); unsigned int n, base_off; float wgt_k, sum_k_m; __shared__ float tmp_buff[BLOCK_SIZE]; // Sum all weight assigned to cluster k. REDUCE(num_data, _resp_[rb + n], wgt_k); __syncthreads(); // Update fractional prior. if (tid == 0 && m == 0) { _frac_[fb + k] = wgt_k / (float)num_data; } __syncthreads(); // Only concerned about dimension m in this block. // Sum will become the sum of movement in that direction for this cluster. REDUCE(num_data, _resp_[rb + n] * _data_[(m*num_data)+n], sum_k_m); __syncthreads(); if (tid == 0) { _means_[mb + __umul24(k, num_dims) + m] = sum_k_m / wgt_k; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_sigma_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { // Every block is mapped to cluster and dimension pair. const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y / num_dims; const unsigned int j = blockIdx.y % num_dims; const unsigned int tid = threadIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int db_m = (m*num_data); const unsigned int db_j = (j*num_data); unsigned int n, base_off; __shared__ float tmp_buff[BLOCK_SIZE]; __shared__ float wgt_k; __shared__ float mean_k_m; if (tid == 0) { wgt_k = _frac_[fb + k] * num_data; mean_k_m = _means_[mb + __umul24(k, num_dims) + m]; } __syncthreads(); float sum; REDUCE(num_data, _resp_[rb + n] * (_data_[db_m + n] - mean_k_m) * (_data_[db_j + n]), sum); // Set this block's Sigma val. if (tid == 0) { _sig_[sb + (__umul24(k, sigsize)) + (__umul24(m, num_dims)) + j] = sum / wgt_k; } }
.file "tmpxft_00021a4a_00000000-6_gpugaumixmod_kernel.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2033: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2033: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z8choleskyPfj .type _Z8choleskyPfj, @function _Z8choleskyPfj: .LFB2027: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2027: .size _Z8choleskyPfj, .-_Z8choleskyPfj .globl _Z6logdetPfj .type _Z6logdetPfj, @function _Z6logdetPfj: .LFB2028: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2028: .size _Z6logdetPfj, .-_Z6logdetPfj .globl _Z11parallelSumPfj .type _Z11parallelSumPfj, @function _Z11parallelSumPfj: .LFB2029: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2029: .size _Z11parallelSumPfj, .-_Z11parallelSumPfj .globl _Z9copyArrayPfS_j .type _Z9copyArrayPfS_j, @function _Z9copyArrayPfS_j: .LFB2030: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2030: .size _Z9copyArrayPfS_j, .-_Z9copyArrayPfS_j .globl _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj .type _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj, @function _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj: .LFB2055: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movq %rcx, 16(%rsp) movq %r8, 8(%rsp) movq %r9, (%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 16(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) movq %rsp, %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) leaq 224(%rsp), %rax movq %rax, 176(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L15 .L11: movq 184(%rsp), %rax subq %fs:40, %rax jne .L16 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L15: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z12estep_kernelPfS_S_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L11 .L16: call __stack_chk_fail@PLT .cfi_endproc .LFE2055: .size _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj, .-_Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj .globl _Z12estep_kernelPfS_S_S_S_S_jjj .type _Z12estep_kernelPfS_S_S_S_S_jjj, @function _Z12estep_kernelPfS_S_S_S_S_jjj: .LFB2056: .cfi_startproc endbr64 subq $16, %rsp .cfi_def_cfa_offset 24 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 40 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 48 call _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj addq $40, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2056: .size _Z12estep_kernelPfS_S_S_S_S_jjj, .-_Z12estep_kernelPfS_S_S_S_S_jjj .globl _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj .type _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj, @function _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj: .LFB2057: .cfi_startproc endbr64 subq $232, %rsp .cfi_def_cfa_offset 240 movq %rdi, 56(%rsp) movq %rsi, 48(%rsp) movq %rdx, 40(%rsp) movq %rcx, 32(%rsp) movq %r8, 24(%rsp) movq %r9, 16(%rsp) movq 240(%rsp), %rax movq %rax, 8(%rsp) movq %fs:40, %rax movq %rax, 216(%rsp) xorl %eax, %eax leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 40(%rsp), %rax movq %rax, 144(%rsp) leaq 32(%rsp), %rax movq %rax, 152(%rsp) leaq 24(%rsp), %rax movq %rax, 160(%rsp) leaq 16(%rsp), %rax movq %rax, 168(%rsp) leaq 8(%rsp), %rax movq %rax, 176(%rsp) leaq 248(%rsp), %rax movq %rax, 184(%rsp) leaq 256(%rsp), %rax movq %rax, 192(%rsp) leaq 264(%rsp), %rax movq %rax, 200(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) movl $1, 88(%rsp) movl $1, 92(%rsp) movl $1, 96(%rsp) movl $1, 100(%rsp) leaq 72(%rsp), %rcx leaq 64(%rsp), %rdx leaq 92(%rsp), %rsi leaq 80(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L23 .L19: movq 216(%rsp), %rax subq %fs:40, %rax jne .L24 addq $232, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L23: .cfi_restore_state pushq 72(%rsp) .cfi_def_cfa_offset 248 pushq 72(%rsp) .cfi_def_cfa_offset 256 leaq 144(%rsp), %r9 movq 108(%rsp), %rcx movl 116(%rsp), %r8d movq 96(%rsp), %rsi movl 104(%rsp), %edx leaq _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 240 jmp .L19 .L24: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj, .-_Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj .globl _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .type _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, @function _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj: .LFB2058: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 24 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 40 pushq 40(%rsp) .cfi_def_cfa_offset 48 call _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj addq $40, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2058: .size _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, .-_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .globl _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .type _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, @function _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj: .LFB2059: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movq %rcx, 16(%rsp) movq %r8, 8(%rsp) movl %r9d, 4(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 16(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 4(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L31 .L27: movq 184(%rsp), %rax subq %fs:40, %rax jne .L32 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L31: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z12mstep_kernelPfS_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L27 .L32: call __stack_chk_fail@PLT .cfi_endproc .LFE2059: .size _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, .-_Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .globl _Z12mstep_kernelPfS_S_S_S_jjj .type _Z12mstep_kernelPfS_S_S_S_jjj, @function _Z12mstep_kernelPfS_S_S_S_jjj: .LFB2060: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 24 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 call _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2060: .size _Z12mstep_kernelPfS_S_S_S_jjj, .-_Z12mstep_kernelPfS_S_S_S_jjj .globl _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .type _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, @function _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj: .LFB2061: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movq %rcx, 16(%rsp) movq %r8, 8(%rsp) movl %r9d, 4(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 16(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 4(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L39 .L35: movq 184(%rsp), %rax subq %fs:40, %rax jne .L40 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L39: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z18mstep_sigma_kernelPfS_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L35 .L40: call __stack_chk_fail@PLT .cfi_endproc .LFE2061: .size _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, .-_Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .globl _Z18mstep_sigma_kernelPfS_S_S_S_jjj .type _Z18mstep_sigma_kernelPfS_S_S_S_jjj, @function _Z18mstep_sigma_kernelPfS_S_S_S_jjj: .LFB2062: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 24 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 call _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2062: .size _Z18mstep_sigma_kernelPfS_S_S_S_jjj, .-_Z18mstep_sigma_kernelPfS_S_S_S_jjj .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z18mstep_sigma_kernelPfS_S_S_S_jjj" .section .rodata.str1.1,"aMS",@progbits,1 .LC1: .string "_Z12mstep_kernelPfS_S_S_S_jjj" .section .rodata.str1.8 .align 8 .LC2: .string "_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj" .align 8 .LC3: .string "_Z12estep_kernelPfS_S_S_S_S_jjj" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2064: .cfi_startproc endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rbx movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z18mstep_sigma_kernelPfS_S_S_S_jjj(%rip), %rsi movq %rax, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC1(%rip), %rdx movq %rdx, %rcx leaq _Z12mstep_kernelPfS_S_S_S_jjj(%rip), %rsi movq %rbx, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC2(%rip), %rdx movq %rdx, %rcx leaq _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj(%rip), %rsi movq %rbx, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC3(%rip), %rdx movq %rdx, %rcx leaq _Z12estep_kernelPfS_S_S_S_S_jjj(%rip), %rsi movq %rbx, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT popq %rbx .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2064: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
// Kernel code for Gaussian Mixture Model Expectation Maximization. // // Andrew Harp (andrew.harp@gmail.com) // http://andrewharp.com/gmmcuda // This effects the maximum dimensionality of your data. // Has to be hardcoded because it affects memory allocation. // Due to the way I clear out the lower triangle of the cholesky (and // possibly other) places MAX_DIM * MAX_DIM needs to be less than // BLOCK_SIZE. #define MAX_DIM 16 // You can change this, but only to a power of 2. #define BLOCK_SIZE 256 // Estep-normalize is broken down into an arbitrary number of chunks. #define NUM_CHUNKS 32 #include "cuda_runtime.h" /////////////////////////////////////////////////////////////////////////// __device__ void cholesky(float* el, const unsigned int ndim) { const unsigned int tid = threadIdx.x; // Dunno how to parallelize this part... if (tid == 0) { float sum = 0; int i, j, k; //if (el.ncols() != n) throw("need square matrix"); for (i=0; i<ndim; i++) { for (j=i; j<ndim; j++) { sum = el[__umul24(i, ndim)+j]; for (k=i-1; k >= 0; k--) { sum -= el[__umul24(i, ndim)+k] * el[__umul24(j, ndim)+k]; } if (i == j) { //if (sum <= 0.0) // throw("Cholesky failed"); el[__umul24(i, ndim)+i] = sqrt(sum); } else { el[__umul24(j, ndim)+i] = sum/el[__umul24(i, ndim)+i]; } } } } __syncthreads(); // Clear lower triangular part. if ((tid/ndim) < (tid%ndim)) { el[__umul24((tid/ndim), ndim) + (tid%ndim)] = 0.0f; } } /////////////////////////////////////////////////////////////////////////// __device__ float logdet(float* el, const unsigned int ndim) { float sum = 0.0f; for (unsigned int i=0; i<ndim; ++i) { sum += __logf(el[(i*ndim)+i]); } return 2.*sum; } /////////////////////////////////////////////////////////////////////////// __device__ float parallelSum(float* data, const unsigned int ndata) { const unsigned int tid = threadIdx.x; float t; __syncthreads(); // Butterfly sum. ndata MUST be a power of 2. for(unsigned int bit = ndata >> 1; bit > 0; bit >>= 1) { t = data[tid] + data[tid^bit]; __syncthreads(); data[tid] = t; __syncthreads(); } return data[tid]; } /////////////////////////////////////////////////////////////////////////// __device__ void copyArray(float* fromArr, float* toArr, unsigned int ndata=BLOCK_SIZE) { unsigned int n; unsigned int base_off; for (base_off=0; base_off < ndata; base_off += blockDim.x) { n = base_off + threadIdx.x; if (n < ndata) { toArr[n] = fromArr[n]; } } } /////////////////////////////////////////////////////////////////////////// // Parallel reduction, for when all you want is the sum of a certain // quantity computed for every 1 to N. CODE should be something in terms // of n. The resulting sum will be placed in RESULT. // tmp_buff, base_off, RESULT, and n must be previously defined, however // they will be overwritten during the execution of the macro. #define REDUCE(N, CODE, RESULT) \ base_off = 0; \ RESULT = 0.0f; \ while (base_off + BLOCK_SIZE < N) { \ n = base_off + tid; \ tmp_buff[tid] = CODE; \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); \ base_off += BLOCK_SIZE; \ } \ n = base_off + tid; \ if (n < N) {tmp_buff[tid] = CODE;} \ else {tmp_buff[tid] = 0.0f;} \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); /////////////////////////////////////////////////////////////////////////// // This function computes for a single cluster k. __global__ void estep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int tid = threadIdx.x; const unsigned int k = blockIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int lndb = __umul24(try_num, num_clusts); unsigned int n, base_off; float tmp, sum; float v[MAX_DIM]; __shared__ float lndet_k; __shared__ float frac_k; __shared__ float chol[MAX_DIM * MAX_DIM]; __syncthreads(); copyArray(_sig_ + sb + __umul24(k, sigsize), chol, sigsize); cholesky(chol, num_dims); __syncthreads(); if (tid == 0) { frac_k = _frac_[fb + k]; lndet_k = logdet(chol, num_dims); _lndets_[lndb + k] = lndet_k; } __syncthreads(); // Loop through data. for (base_off=0; base_off < num_data; base_off += BLOCK_SIZE) { n = base_off + tid; sum=0.0f; //if (b.size() != n || y.size() != n) throw("bad lengths"); if (n < num_data) { for (unsigned int i=0; i<num_dims; ++i) { tmp = _data_[__umul24(i, num_data) + n] - _means_[mb + __umul24(k, num_dims)+i]; for (unsigned int j=0; j<i; j++) { tmp -= chol[__umul24(i, num_dims)+j] * v[j]; } v[i] = tmp/chol[__umul24(i, num_dims)+i]; sum += v[i] * v[i]; } // Assign likelihood of this data being in this cluster. _resp_[rb + n] = -0.5f*(sum + lndet_k) + __logf(frac_k); } } // (n < num_data) } /////////////////////////////////////////////////////////////////////////// // Loop through data again, normalizing probabilities. // We are looping across clusters here as well as data, since every data // point needs to know its potential parents. __global__ void estep_normalize_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, float* _loglike_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int num_chunks = gridDim.x; const unsigned int chunk_num = blockIdx.x; const unsigned int tid = threadIdx.x; // We're only handling so many data points per block in this kernel, since // data is independant of other data here. const unsigned int n_per_block = ceil((float)num_data / (float)num_chunks); const unsigned int start_off = __umul24(n_per_block, chunk_num); const unsigned int end_off = min(start_off + n_per_block, num_data); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data; const unsigned int lb = __umul24(try_num, num_chunks); unsigned int n, base_off, k; float sum, max, tmp; __shared__ float loglike[BLOCK_SIZE]; loglike[tid] = 0.0f; __syncthreads(); // Loop through data. for (base_off = start_off; base_off < end_off; base_off += BLOCK_SIZE) { n = base_off + tid; if (n < end_off) { max = -99.9e30f; // Find cluster with maximum likelihood for this data point. for (k=0; k<num_clusts; ++k) { tmp = _resp_[rb + (k*num_data) + n]; if (tmp > max) { max = tmp; } } // Sum marginal probabilities. sum = 0.0f; for (k=0; k<num_clusts; ++k) { sum += __expf(_resp_[rb + (k*num_data) + n] - max); } // Assign probabilities of point belonging to each cluster. tmp = max + __logf(sum); for (k = 0; k < num_clusts; ++k) { _resp_[rb + (k*num_data) + n] = __expf(_resp_[rb + (k*num_data) + n] - tmp); } loglike[tid] += tmp; } } tmp = parallelSum(loglike, BLOCK_SIZE); if (tid == 0) { _loglike_[lb + chunk_num] = tmp; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int tid = threadIdx.x; // Every block is mapped to cluster and dimension. const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y; // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int fb = __umul24(try_num, num_clusts); unsigned int n, base_off; float wgt_k, sum_k_m; __shared__ float tmp_buff[BLOCK_SIZE]; // Sum all weight assigned to cluster k. REDUCE(num_data, _resp_[rb + n], wgt_k); __syncthreads(); // Update fractional prior. if (tid == 0 && m == 0) { _frac_[fb + k] = wgt_k / (float)num_data; } __syncthreads(); // Only concerned about dimension m in this block. // Sum will become the sum of movement in that direction for this cluster. REDUCE(num_data, _resp_[rb + n] * _data_[(m*num_data)+n], sum_k_m); __syncthreads(); if (tid == 0) { _means_[mb + __umul24(k, num_dims) + m] = sum_k_m / wgt_k; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_sigma_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { // Every block is mapped to cluster and dimension pair. const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y / num_dims; const unsigned int j = blockIdx.y % num_dims; const unsigned int tid = threadIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int db_m = (m*num_data); const unsigned int db_j = (j*num_data); unsigned int n, base_off; __shared__ float tmp_buff[BLOCK_SIZE]; __shared__ float wgt_k; __shared__ float mean_k_m; if (tid == 0) { wgt_k = _frac_[fb + k] * num_data; mean_k_m = _means_[mb + __umul24(k, num_dims) + m]; } __syncthreads(); float sum; REDUCE(num_data, _resp_[rb + n] * (_data_[db_m + n] - mean_k_m) * (_data_[db_j + n]), sum); // Set this block's Sigma val. if (tid == 0) { _sig_[sb + (__umul24(k, sigsize)) + (__umul24(m, num_dims)) + j] = sum / wgt_k; } }
// Kernel code for Gaussian Mixture Model Expectation Maximization. // // Andrew Harp (andrew.harp@gmail.com) // http://andrewharp.com/gmmcuda // This effects the maximum dimensionality of your data. // Has to be hardcoded because it affects memory allocation. // Due to the way I clear out the lower triangle of the cholesky (and // possibly other) places MAX_DIM * MAX_DIM needs to be less than // BLOCK_SIZE. #define MAX_DIM 16 // You can change this, but only to a power of 2. #define BLOCK_SIZE 256 // Estep-normalize is broken down into an arbitrary number of chunks. #define NUM_CHUNKS 32 #include "hip/hip_runtime.h" /////////////////////////////////////////////////////////////////////////// __device__ void cholesky(float* el, const unsigned int ndim) { const unsigned int tid = threadIdx.x; // Dunno how to parallelize this part... if (tid == 0) { float sum = 0; int i, j, k; //if (el.ncols() != n) throw("need square matrix"); for (i=0; i<ndim; i++) { for (j=i; j<ndim; j++) { sum = el[__umul24(i, ndim)+j]; for (k=i-1; k >= 0; k--) { sum -= el[__umul24(i, ndim)+k] * el[__umul24(j, ndim)+k]; } if (i == j) { //if (sum <= 0.0) // throw("Cholesky failed"); el[__umul24(i, ndim)+i] = sqrt(sum); } else { el[__umul24(j, ndim)+i] = sum/el[__umul24(i, ndim)+i]; } } } } __syncthreads(); // Clear lower triangular part. if ((tid/ndim) < (tid%ndim)) { el[__umul24((tid/ndim), ndim) + (tid%ndim)] = 0.0f; } } /////////////////////////////////////////////////////////////////////////// __device__ float logdet(float* el, const unsigned int ndim) { float sum = 0.0f; for (unsigned int i=0; i<ndim; ++i) { sum += __logf(el[(i*ndim)+i]); } return 2.*sum; } /////////////////////////////////////////////////////////////////////////// __device__ float parallelSum(float* data, const unsigned int ndata) { const unsigned int tid = threadIdx.x; float t; __syncthreads(); // Butterfly sum. ndata MUST be a power of 2. for(unsigned int bit = ndata >> 1; bit > 0; bit >>= 1) { t = data[tid] + data[tid^bit]; __syncthreads(); data[tid] = t; __syncthreads(); } return data[tid]; } /////////////////////////////////////////////////////////////////////////// __device__ void copyArray(float* fromArr, float* toArr, unsigned int ndata=BLOCK_SIZE) { unsigned int n; unsigned int base_off; for (base_off=0; base_off < ndata; base_off += blockDim.x) { n = base_off + threadIdx.x; if (n < ndata) { toArr[n] = fromArr[n]; } } } /////////////////////////////////////////////////////////////////////////// // Parallel reduction, for when all you want is the sum of a certain // quantity computed for every 1 to N. CODE should be something in terms // of n. The resulting sum will be placed in RESULT. // tmp_buff, base_off, RESULT, and n must be previously defined, however // they will be overwritten during the execution of the macro. #define REDUCE(N, CODE, RESULT) \ base_off = 0; \ RESULT = 0.0f; \ while (base_off + BLOCK_SIZE < N) { \ n = base_off + tid; \ tmp_buff[tid] = CODE; \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); \ base_off += BLOCK_SIZE; \ } \ n = base_off + tid; \ if (n < N) {tmp_buff[tid] = CODE;} \ else {tmp_buff[tid] = 0.0f;} \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); /////////////////////////////////////////////////////////////////////////// // This function computes for a single cluster k. __global__ void estep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int tid = threadIdx.x; const unsigned int k = blockIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int lndb = __umul24(try_num, num_clusts); unsigned int n, base_off; float tmp, sum; float v[MAX_DIM]; __shared__ float lndet_k; __shared__ float frac_k; __shared__ float chol[MAX_DIM * MAX_DIM]; __syncthreads(); copyArray(_sig_ + sb + __umul24(k, sigsize), chol, sigsize); cholesky(chol, num_dims); __syncthreads(); if (tid == 0) { frac_k = _frac_[fb + k]; lndet_k = logdet(chol, num_dims); _lndets_[lndb + k] = lndet_k; } __syncthreads(); // Loop through data. for (base_off=0; base_off < num_data; base_off += BLOCK_SIZE) { n = base_off + tid; sum=0.0f; //if (b.size() != n || y.size() != n) throw("bad lengths"); if (n < num_data) { for (unsigned int i=0; i<num_dims; ++i) { tmp = _data_[__umul24(i, num_data) + n] - _means_[mb + __umul24(k, num_dims)+i]; for (unsigned int j=0; j<i; j++) { tmp -= chol[__umul24(i, num_dims)+j] * v[j]; } v[i] = tmp/chol[__umul24(i, num_dims)+i]; sum += v[i] * v[i]; } // Assign likelihood of this data being in this cluster. _resp_[rb + n] = -0.5f*(sum + lndet_k) + __logf(frac_k); } } // (n < num_data) } /////////////////////////////////////////////////////////////////////////// // Loop through data again, normalizing probabilities. // We are looping across clusters here as well as data, since every data // point needs to know its potential parents. __global__ void estep_normalize_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, float* _loglike_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int num_chunks = gridDim.x; const unsigned int chunk_num = blockIdx.x; const unsigned int tid = threadIdx.x; // We're only handling so many data points per block in this kernel, since // data is independant of other data here. const unsigned int n_per_block = ceil((float)num_data / (float)num_chunks); const unsigned int start_off = __umul24(n_per_block, chunk_num); const unsigned int end_off = min(start_off + n_per_block, num_data); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data; const unsigned int lb = __umul24(try_num, num_chunks); unsigned int n, base_off, k; float sum, max, tmp; __shared__ float loglike[BLOCK_SIZE]; loglike[tid] = 0.0f; __syncthreads(); // Loop through data. for (base_off = start_off; base_off < end_off; base_off += BLOCK_SIZE) { n = base_off + tid; if (n < end_off) { max = -99.9e30f; // Find cluster with maximum likelihood for this data point. for (k=0; k<num_clusts; ++k) { tmp = _resp_[rb + (k*num_data) + n]; if (tmp > max) { max = tmp; } } // Sum marginal probabilities. sum = 0.0f; for (k=0; k<num_clusts; ++k) { sum += __expf(_resp_[rb + (k*num_data) + n] - max); } // Assign probabilities of point belonging to each cluster. tmp = max + __logf(sum); for (k = 0; k < num_clusts; ++k) { _resp_[rb + (k*num_data) + n] = __expf(_resp_[rb + (k*num_data) + n] - tmp); } loglike[tid] += tmp; } } tmp = parallelSum(loglike, BLOCK_SIZE); if (tid == 0) { _loglike_[lb + chunk_num] = tmp; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int tid = threadIdx.x; // Every block is mapped to cluster and dimension. const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y; // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int fb = __umul24(try_num, num_clusts); unsigned int n, base_off; float wgt_k, sum_k_m; __shared__ float tmp_buff[BLOCK_SIZE]; // Sum all weight assigned to cluster k. REDUCE(num_data, _resp_[rb + n], wgt_k); __syncthreads(); // Update fractional prior. if (tid == 0 && m == 0) { _frac_[fb + k] = wgt_k / (float)num_data; } __syncthreads(); // Only concerned about dimension m in this block. // Sum will become the sum of movement in that direction for this cluster. REDUCE(num_data, _resp_[rb + n] * _data_[(m*num_data)+n], sum_k_m); __syncthreads(); if (tid == 0) { _means_[mb + __umul24(k, num_dims) + m] = sum_k_m / wgt_k; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_sigma_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { // Every block is mapped to cluster and dimension pair. const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y / num_dims; const unsigned int j = blockIdx.y % num_dims; const unsigned int tid = threadIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int db_m = (m*num_data); const unsigned int db_j = (j*num_data); unsigned int n, base_off; __shared__ float tmp_buff[BLOCK_SIZE]; __shared__ float wgt_k; __shared__ float mean_k_m; if (tid == 0) { wgt_k = _frac_[fb + k] * num_data; mean_k_m = _means_[mb + __umul24(k, num_dims) + m]; } __syncthreads(); float sum; REDUCE(num_data, _resp_[rb + n] * (_data_[db_m + n] - mean_k_m) * (_data_[db_j + n]), sum); // Set this block's Sigma val. if (tid == 0) { _sig_[sb + (__umul24(k, sigsize)) + (__umul24(m, num_dims)) + j] = sum / wgt_k; } }
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
// Kernel code for Gaussian Mixture Model Expectation Maximization. // // Andrew Harp (andrew.harp@gmail.com) // http://andrewharp.com/gmmcuda // This effects the maximum dimensionality of your data. // Has to be hardcoded because it affects memory allocation. // Due to the way I clear out the lower triangle of the cholesky (and // possibly other) places MAX_DIM * MAX_DIM needs to be less than // BLOCK_SIZE. #define MAX_DIM 16 // You can change this, but only to a power of 2. #define BLOCK_SIZE 256 // Estep-normalize is broken down into an arbitrary number of chunks. #define NUM_CHUNKS 32 #include "hip/hip_runtime.h" /////////////////////////////////////////////////////////////////////////// __device__ void cholesky(float* el, const unsigned int ndim) { const unsigned int tid = threadIdx.x; // Dunno how to parallelize this part... if (tid == 0) { float sum = 0; int i, j, k; //if (el.ncols() != n) throw("need square matrix"); for (i=0; i<ndim; i++) { for (j=i; j<ndim; j++) { sum = el[__umul24(i, ndim)+j]; for (k=i-1; k >= 0; k--) { sum -= el[__umul24(i, ndim)+k] * el[__umul24(j, ndim)+k]; } if (i == j) { //if (sum <= 0.0) // throw("Cholesky failed"); el[__umul24(i, ndim)+i] = sqrt(sum); } else { el[__umul24(j, ndim)+i] = sum/el[__umul24(i, ndim)+i]; } } } } __syncthreads(); // Clear lower triangular part. if ((tid/ndim) < (tid%ndim)) { el[__umul24((tid/ndim), ndim) + (tid%ndim)] = 0.0f; } } /////////////////////////////////////////////////////////////////////////// __device__ float logdet(float* el, const unsigned int ndim) { float sum = 0.0f; for (unsigned int i=0; i<ndim; ++i) { sum += __logf(el[(i*ndim)+i]); } return 2.*sum; } /////////////////////////////////////////////////////////////////////////// __device__ float parallelSum(float* data, const unsigned int ndata) { const unsigned int tid = threadIdx.x; float t; __syncthreads(); // Butterfly sum. ndata MUST be a power of 2. for(unsigned int bit = ndata >> 1; bit > 0; bit >>= 1) { t = data[tid] + data[tid^bit]; __syncthreads(); data[tid] = t; __syncthreads(); } return data[tid]; } /////////////////////////////////////////////////////////////////////////// __device__ void copyArray(float* fromArr, float* toArr, unsigned int ndata=BLOCK_SIZE) { unsigned int n; unsigned int base_off; for (base_off=0; base_off < ndata; base_off += blockDim.x) { n = base_off + threadIdx.x; if (n < ndata) { toArr[n] = fromArr[n]; } } } /////////////////////////////////////////////////////////////////////////// // Parallel reduction, for when all you want is the sum of a certain // quantity computed for every 1 to N. CODE should be something in terms // of n. The resulting sum will be placed in RESULT. // tmp_buff, base_off, RESULT, and n must be previously defined, however // they will be overwritten during the execution of the macro. #define REDUCE(N, CODE, RESULT) \ base_off = 0; \ RESULT = 0.0f; \ while (base_off + BLOCK_SIZE < N) { \ n = base_off + tid; \ tmp_buff[tid] = CODE; \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); \ base_off += BLOCK_SIZE; \ } \ n = base_off + tid; \ if (n < N) {tmp_buff[tid] = CODE;} \ else {tmp_buff[tid] = 0.0f;} \ RESULT += parallelSum(tmp_buff, BLOCK_SIZE); /////////////////////////////////////////////////////////////////////////// // This function computes for a single cluster k. __global__ void estep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int tid = threadIdx.x; const unsigned int k = blockIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int lndb = __umul24(try_num, num_clusts); unsigned int n, base_off; float tmp, sum; float v[MAX_DIM]; __shared__ float lndet_k; __shared__ float frac_k; __shared__ float chol[MAX_DIM * MAX_DIM]; __syncthreads(); copyArray(_sig_ + sb + __umul24(k, sigsize), chol, sigsize); cholesky(chol, num_dims); __syncthreads(); if (tid == 0) { frac_k = _frac_[fb + k]; lndet_k = logdet(chol, num_dims); _lndets_[lndb + k] = lndet_k; } __syncthreads(); // Loop through data. for (base_off=0; base_off < num_data; base_off += BLOCK_SIZE) { n = base_off + tid; sum=0.0f; //if (b.size() != n || y.size() != n) throw("bad lengths"); if (n < num_data) { for (unsigned int i=0; i<num_dims; ++i) { tmp = _data_[__umul24(i, num_data) + n] - _means_[mb + __umul24(k, num_dims)+i]; for (unsigned int j=0; j<i; j++) { tmp -= chol[__umul24(i, num_dims)+j] * v[j]; } v[i] = tmp/chol[__umul24(i, num_dims)+i]; sum += v[i] * v[i]; } // Assign likelihood of this data being in this cluster. _resp_[rb + n] = -0.5f*(sum + lndet_k) + __logf(frac_k); } } // (n < num_data) } /////////////////////////////////////////////////////////////////////////// // Loop through data again, normalizing probabilities. // We are looping across clusters here as well as data, since every data // point needs to know its potential parents. __global__ void estep_normalize_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, float* _lndets_, float* _loglike_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.y; const unsigned int num_chunks = gridDim.x; const unsigned int chunk_num = blockIdx.x; const unsigned int tid = threadIdx.x; // We're only handling so many data points per block in this kernel, since // data is independant of other data here. const unsigned int n_per_block = ceil((float)num_data / (float)num_chunks); const unsigned int start_off = __umul24(n_per_block, chunk_num); const unsigned int end_off = min(start_off + n_per_block, num_data); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data; const unsigned int lb = __umul24(try_num, num_chunks); unsigned int n, base_off, k; float sum, max, tmp; __shared__ float loglike[BLOCK_SIZE]; loglike[tid] = 0.0f; __syncthreads(); // Loop through data. for (base_off = start_off; base_off < end_off; base_off += BLOCK_SIZE) { n = base_off + tid; if (n < end_off) { max = -99.9e30f; // Find cluster with maximum likelihood for this data point. for (k=0; k<num_clusts; ++k) { tmp = _resp_[rb + (k*num_data) + n]; if (tmp > max) { max = tmp; } } // Sum marginal probabilities. sum = 0.0f; for (k=0; k<num_clusts; ++k) { sum += __expf(_resp_[rb + (k*num_data) + n] - max); } // Assign probabilities of point belonging to each cluster. tmp = max + __logf(sum); for (k = 0; k < num_clusts; ++k) { _resp_[rb + (k*num_data) + n] = __expf(_resp_[rb + (k*num_data) + n] - tmp); } loglike[tid] += tmp; } } tmp = parallelSum(loglike, BLOCK_SIZE); if (tid == 0) { _loglike_[lb + chunk_num] = tmp; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int tid = threadIdx.x; // Every block is mapped to cluster and dimension. const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y; // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int fb = __umul24(try_num, num_clusts); unsigned int n, base_off; float wgt_k, sum_k_m; __shared__ float tmp_buff[BLOCK_SIZE]; // Sum all weight assigned to cluster k. REDUCE(num_data, _resp_[rb + n], wgt_k); __syncthreads(); // Update fractional prior. if (tid == 0 && m == 0) { _frac_[fb + k] = wgt_k / (float)num_data; } __syncthreads(); // Only concerned about dimension m in this block. // Sum will become the sum of movement in that direction for this cluster. REDUCE(num_data, _resp_[rb + n] * _data_[(m*num_data)+n], sum_k_m); __syncthreads(); if (tid == 0) { _means_[mb + __umul24(k, num_dims) + m] = sum_k_m / wgt_k; } } /////////////////////////////////////////////////////////////////////////// __global__ void mstep_sigma_kernel(float* _resp_, float* _frac_, float* _data_, float* _means_, float* _sig_, const unsigned int num_clusts, const unsigned int num_dims, const unsigned int num_data) { // Every block is mapped to cluster and dimension pair. const unsigned int try_num = blockIdx.x / num_clusts; const unsigned int k = blockIdx.x % num_clusts; const unsigned int m = blockIdx.y / num_dims; const unsigned int j = blockIdx.y % num_dims; const unsigned int tid = threadIdx.x; const unsigned int sigsize = __umul24(num_dims, num_dims); // Base offsets. const unsigned int rb = __umul24(try_num, num_clusts) * num_data + (k*num_data); const unsigned int mb = __umul24(try_num, num_clusts) * num_dims; const unsigned int sb = __umul24(try_num, num_clusts) * sigsize; const unsigned int fb = __umul24(try_num, num_clusts); const unsigned int db_m = (m*num_data); const unsigned int db_j = (j*num_data); unsigned int n, base_off; __shared__ float tmp_buff[BLOCK_SIZE]; __shared__ float wgt_k; __shared__ float mean_k_m; if (tid == 0) { wgt_k = _frac_[fb + k] * num_data; mean_k_m = _means_[mb + __umul24(k, num_dims) + m]; } __syncthreads(); float sum; REDUCE(num_data, _resp_[rb + n] * (_data_[db_m + n] - mean_k_m) * (_data_[db_j + n]), sum); // Set this block's Sigma val. if (tid == 0) { _sig_[sb + (__umul24(k, sigsize)) + (__umul24(m, num_dims)) + j] = sum / wgt_k; } }
.text .file "gpugaumixmod_kernel.hip" .globl _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj # -- Begin function _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .p2align 4, 0x90 .type _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj,@function _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj: # @_Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movq %r9, 48(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 192(%rsp), %rax movq %rax, 160(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z12estep_kernelPfS_S_S_S_S_jjj, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end0: .size _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj, .Lfunc_end0-_Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .cfi_endproc # -- End function .globl _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj # -- Begin function _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .p2align 4, 0x90 .type _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj,@function _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj: # @_Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .cfi_startproc # %bb.0: subq $184, %rsp .cfi_def_cfa_offset 192 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movq %r9, 48(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 192(%rsp), %rax movq %rax, 144(%rsp) leaq 200(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $200, %rsp .cfi_adjust_cfa_offset -200 retq .Lfunc_end1: .size _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj, .Lfunc_end1-_Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .cfi_endproc # -- End function .globl _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj # -- Begin function _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .p2align 4, 0x90 .type _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj,@function _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj: # @_Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movl %r9d, 4(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 4(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z12mstep_kernelPfS_S_S_S_jjj, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end2: .size _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj, .Lfunc_end2-_Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .cfi_endproc # -- End function .globl _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj # -- Begin function _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .p2align 4, 0x90 .type _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj,@function _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj: # @_Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movl %r9d, 4(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 4(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z18mstep_sigma_kernelPfS_S_S_S_jjj, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end3: .size _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj, .Lfunc_end3-_Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: pushq %rbx .cfi_def_cfa_offset 16 subq $32, %rsp .cfi_def_cfa_offset 48 .cfi_offset %rbx, -16 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB4_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB4_2: movq __hip_gpubin_handle(%rip), %rbx xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12estep_kernelPfS_S_S_S_S_jjj, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, %esi movl $.L__unnamed_2, %edx movl $.L__unnamed_2, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12mstep_kernelPfS_S_S_S_jjj, %esi movl $.L__unnamed_3, %edx movl $.L__unnamed_3, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z18mstep_sigma_kernelPfS_S_S_S_jjj, %esi movl $.L__unnamed_4, %edx movl $.L__unnamed_4, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $32, %rsp .cfi_def_cfa_offset 16 popq %rbx .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end4: .size __hip_module_ctor, .Lfunc_end4-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB5_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB5_2: retq .Lfunc_end5: .size __hip_module_dtor, .Lfunc_end5-__hip_module_dtor .cfi_endproc # -- End function .type _Z12estep_kernelPfS_S_S_S_S_jjj,@object # @_Z12estep_kernelPfS_S_S_S_S_jjj .section .rodata,"a",@progbits .globl _Z12estep_kernelPfS_S_S_S_S_jjj .p2align 3, 0x0 _Z12estep_kernelPfS_S_S_S_S_jjj: .quad _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .size _Z12estep_kernelPfS_S_S_S_S_jjj, 8 .type _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj,@object # @_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .globl _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .p2align 3, 0x0 _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj: .quad _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .size _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, 8 .type _Z12mstep_kernelPfS_S_S_S_jjj,@object # @_Z12mstep_kernelPfS_S_S_S_jjj .globl _Z12mstep_kernelPfS_S_S_S_jjj .p2align 3, 0x0 _Z12mstep_kernelPfS_S_S_S_jjj: .quad _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .size _Z12mstep_kernelPfS_S_S_S_jjj, 8 .type _Z18mstep_sigma_kernelPfS_S_S_S_jjj,@object # @_Z18mstep_sigma_kernelPfS_S_S_S_jjj .globl _Z18mstep_sigma_kernelPfS_S_S_S_jjj .p2align 3, 0x0 _Z18mstep_sigma_kernelPfS_S_S_S_jjj: .quad _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .size _Z18mstep_sigma_kernelPfS_S_S_S_jjj, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z12estep_kernelPfS_S_S_S_S_jjj" .size .L__unnamed_1, 32 .type .L__unnamed_2,@object # @1 .L__unnamed_2: .asciz "_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj" .size .L__unnamed_2, 44 .type .L__unnamed_3,@object # @2 .L__unnamed_3: .asciz "_Z12mstep_kernelPfS_S_S_S_jjj" .size .L__unnamed_3, 30 .type .L__unnamed_4,@object # @3 .L__unnamed_4: .asciz "_Z18mstep_sigma_kernelPfS_S_S_S_jjj" .size .L__unnamed_4, 36 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .addrsig_sym _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .addrsig_sym _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .addrsig_sym _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z12estep_kernelPfS_S_S_S_S_jjj .addrsig_sym _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .addrsig_sym _Z12mstep_kernelPfS_S_S_S_jjj .addrsig_sym _Z18mstep_sigma_kernelPfS_S_S_S_jjj .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_00021a4a_00000000-6_gpugaumixmod_kernel.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2033: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2033: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z8choleskyPfj .type _Z8choleskyPfj, @function _Z8choleskyPfj: .LFB2027: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2027: .size _Z8choleskyPfj, .-_Z8choleskyPfj .globl _Z6logdetPfj .type _Z6logdetPfj, @function _Z6logdetPfj: .LFB2028: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2028: .size _Z6logdetPfj, .-_Z6logdetPfj .globl _Z11parallelSumPfj .type _Z11parallelSumPfj, @function _Z11parallelSumPfj: .LFB2029: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2029: .size _Z11parallelSumPfj, .-_Z11parallelSumPfj .globl _Z9copyArrayPfS_j .type _Z9copyArrayPfS_j, @function _Z9copyArrayPfS_j: .LFB2030: .cfi_startproc endbr64 pushq %rax .cfi_def_cfa_offset 16 popq %rax .cfi_def_cfa_offset 8 subq $24, %rsp .cfi_def_cfa_offset 32 movl $1, 12(%rsp) movl 12(%rsp), %edi call exit@PLT .cfi_endproc .LFE2030: .size _Z9copyArrayPfS_j, .-_Z9copyArrayPfS_j .globl _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj .type _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj, @function _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj: .LFB2055: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movq %rcx, 16(%rsp) movq %r8, 8(%rsp) movq %r9, (%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 16(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) movq %rsp, %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) leaq 224(%rsp), %rax movq %rax, 176(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L15 .L11: movq 184(%rsp), %rax subq %fs:40, %rax jne .L16 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L15: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z12estep_kernelPfS_S_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L11 .L16: call __stack_chk_fail@PLT .cfi_endproc .LFE2055: .size _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj, .-_Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj .globl _Z12estep_kernelPfS_S_S_S_S_jjj .type _Z12estep_kernelPfS_S_S_S_S_jjj, @function _Z12estep_kernelPfS_S_S_S_S_jjj: .LFB2056: .cfi_startproc endbr64 subq $16, %rsp .cfi_def_cfa_offset 24 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 40 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 48 call _Z45__device_stub__Z12estep_kernelPfS_S_S_S_S_jjjPfS_S_S_S_S_jjj addq $40, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2056: .size _Z12estep_kernelPfS_S_S_S_S_jjj, .-_Z12estep_kernelPfS_S_S_S_S_jjj .globl _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj .type _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj, @function _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj: .LFB2057: .cfi_startproc endbr64 subq $232, %rsp .cfi_def_cfa_offset 240 movq %rdi, 56(%rsp) movq %rsi, 48(%rsp) movq %rdx, 40(%rsp) movq %rcx, 32(%rsp) movq %r8, 24(%rsp) movq %r9, 16(%rsp) movq 240(%rsp), %rax movq %rax, 8(%rsp) movq %fs:40, %rax movq %rax, 216(%rsp) xorl %eax, %eax leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 40(%rsp), %rax movq %rax, 144(%rsp) leaq 32(%rsp), %rax movq %rax, 152(%rsp) leaq 24(%rsp), %rax movq %rax, 160(%rsp) leaq 16(%rsp), %rax movq %rax, 168(%rsp) leaq 8(%rsp), %rax movq %rax, 176(%rsp) leaq 248(%rsp), %rax movq %rax, 184(%rsp) leaq 256(%rsp), %rax movq %rax, 192(%rsp) leaq 264(%rsp), %rax movq %rax, 200(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) movl $1, 88(%rsp) movl $1, 92(%rsp) movl $1, 96(%rsp) movl $1, 100(%rsp) leaq 72(%rsp), %rcx leaq 64(%rsp), %rdx leaq 92(%rsp), %rsi leaq 80(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L23 .L19: movq 216(%rsp), %rax subq %fs:40, %rax jne .L24 addq $232, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L23: .cfi_restore_state pushq 72(%rsp) .cfi_def_cfa_offset 248 pushq 72(%rsp) .cfi_def_cfa_offset 256 leaq 144(%rsp), %r9 movq 108(%rsp), %rcx movl 116(%rsp), %r8d movq 96(%rsp), %rsi movl 104(%rsp), %edx leaq _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 240 jmp .L19 .L24: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj, .-_Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj .globl _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .type _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, @function _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj: .LFB2058: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 24 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 movl 40(%rsp), %eax pushq %rax .cfi_def_cfa_offset 40 pushq 40(%rsp) .cfi_def_cfa_offset 48 call _Z57__device_stub__Z22estep_normalize_kernelPfS_S_S_S_S_S_jjjPfS_S_S_S_S_S_jjj addq $40, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2058: .size _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, .-_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .globl _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .type _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, @function _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj: .LFB2059: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movq %rcx, 16(%rsp) movq %r8, 8(%rsp) movl %r9d, 4(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 16(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 4(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L31 .L27: movq 184(%rsp), %rax subq %fs:40, %rax jne .L32 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L31: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z12mstep_kernelPfS_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L27 .L32: call __stack_chk_fail@PLT .cfi_endproc .LFE2059: .size _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, .-_Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .globl _Z12mstep_kernelPfS_S_S_S_jjj .type _Z12mstep_kernelPfS_S_S_S_jjj, @function _Z12mstep_kernelPfS_S_S_S_jjj: .LFB2060: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 24 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 call _Z43__device_stub__Z12mstep_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2060: .size _Z12mstep_kernelPfS_S_S_S_jjj, .-_Z12mstep_kernelPfS_S_S_S_jjj .globl _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .type _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, @function _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj: .LFB2061: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movq %rdx, 24(%rsp) movq %rcx, 16(%rsp) movq %r8, 8(%rsp) movl %r9d, 4(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 40(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) leaq 24(%rsp), %rax movq %rax, 128(%rsp) leaq 16(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 4(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L39 .L35: movq 184(%rsp), %rax subq %fs:40, %rax jne .L40 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L39: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _Z18mstep_sigma_kernelPfS_S_S_S_jjj(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L35 .L40: call __stack_chk_fail@PLT .cfi_endproc .LFE2061: .size _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj, .-_Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj .globl _Z18mstep_sigma_kernelPfS_S_S_S_jjj .type _Z18mstep_sigma_kernelPfS_S_S_S_jjj, @function _Z18mstep_sigma_kernelPfS_S_S_S_jjj: .LFB2062: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 24 movl 24(%rsp), %eax pushq %rax .cfi_def_cfa_offset 32 call _Z49__device_stub__Z18mstep_sigma_kernelPfS_S_S_S_jjjPfS_S_S_S_jjj addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2062: .size _Z18mstep_sigma_kernelPfS_S_S_S_jjj, .-_Z18mstep_sigma_kernelPfS_S_S_S_jjj .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z18mstep_sigma_kernelPfS_S_S_S_jjj" .section .rodata.str1.1,"aMS",@progbits,1 .LC1: .string "_Z12mstep_kernelPfS_S_S_S_jjj" .section .rodata.str1.8 .align 8 .LC2: .string "_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj" .align 8 .LC3: .string "_Z12estep_kernelPfS_S_S_S_S_jjj" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2064: .cfi_startproc endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rbx movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z18mstep_sigma_kernelPfS_S_S_S_jjj(%rip), %rsi movq %rax, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC1(%rip), %rdx movq %rdx, %rcx leaq _Z12mstep_kernelPfS_S_S_S_jjj(%rip), %rsi movq %rbx, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC2(%rip), %rdx movq %rdx, %rcx leaq _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj(%rip), %rsi movq %rbx, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC3(%rip), %rdx movq %rdx, %rcx leaq _Z12estep_kernelPfS_S_S_S_S_jjj(%rip), %rsi movq %rbx, %rdi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT popq %rbx .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2064: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "gpugaumixmod_kernel.hip" .globl _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj # -- Begin function _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .p2align 4, 0x90 .type _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj,@function _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj: # @_Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movq %r9, 48(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 192(%rsp), %rax movq %rax, 160(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z12estep_kernelPfS_S_S_S_S_jjj, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end0: .size _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj, .Lfunc_end0-_Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .cfi_endproc # -- End function .globl _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj # -- Begin function _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .p2align 4, 0x90 .type _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj,@function _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj: # @_Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .cfi_startproc # %bb.0: subq $184, %rsp .cfi_def_cfa_offset 192 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movq %r9, 48(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 48(%rsp), %rax movq %rax, 136(%rsp) leaq 192(%rsp), %rax movq %rax, 144(%rsp) leaq 200(%rsp), %rax movq %rax, 152(%rsp) leaq 208(%rsp), %rax movq %rax, 160(%rsp) leaq 216(%rsp), %rax movq %rax, 168(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $200, %rsp .cfi_adjust_cfa_offset -200 retq .Lfunc_end1: .size _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj, .Lfunc_end1-_Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .cfi_endproc # -- End function .globl _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj # -- Begin function _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .p2align 4, 0x90 .type _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj,@function _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj: # @_Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movl %r9d, 4(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 4(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z12mstep_kernelPfS_S_S_S_jjj, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end2: .size _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj, .Lfunc_end2-_Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .cfi_endproc # -- End function .globl _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj # -- Begin function _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .p2align 4, 0x90 .type _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj,@function _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj: # @_Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movq %rcx, 64(%rsp) movq %r8, 56(%rsp) movl %r9d, 4(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 64(%rsp), %rax movq %rax, 120(%rsp) leaq 56(%rsp), %rax movq %rax, 128(%rsp) leaq 4(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 40(%rsp), %rdi leaq 24(%rsp), %rsi leaq 16(%rsp), %rdx leaq 8(%rsp), %rcx callq __hipPopCallConfiguration movq 40(%rsp), %rsi movl 48(%rsp), %edx movq 24(%rsp), %rcx movl 32(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z18mstep_sigma_kernelPfS_S_S_S_jjj, %edi pushq 8(%rsp) .cfi_adjust_cfa_offset 8 pushq 24(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end3: .size _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj, .Lfunc_end3-_Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: pushq %rbx .cfi_def_cfa_offset 16 subq $32, %rsp .cfi_def_cfa_offset 48 .cfi_offset %rbx, -16 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB4_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB4_2: movq __hip_gpubin_handle(%rip), %rbx xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12estep_kernelPfS_S_S_S_S_jjj, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, %esi movl $.L__unnamed_2, %edx movl $.L__unnamed_2, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12mstep_kernelPfS_S_S_S_jjj, %esi movl $.L__unnamed_3, %edx movl $.L__unnamed_3, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z18mstep_sigma_kernelPfS_S_S_S_jjj, %esi movl $.L__unnamed_4, %edx movl $.L__unnamed_4, %ecx movq %rbx, %rdi movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $32, %rsp .cfi_def_cfa_offset 16 popq %rbx .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end4: .size __hip_module_ctor, .Lfunc_end4-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB5_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB5_2: retq .Lfunc_end5: .size __hip_module_dtor, .Lfunc_end5-__hip_module_dtor .cfi_endproc # -- End function .type _Z12estep_kernelPfS_S_S_S_S_jjj,@object # @_Z12estep_kernelPfS_S_S_S_S_jjj .section .rodata,"a",@progbits .globl _Z12estep_kernelPfS_S_S_S_S_jjj .p2align 3, 0x0 _Z12estep_kernelPfS_S_S_S_S_jjj: .quad _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .size _Z12estep_kernelPfS_S_S_S_S_jjj, 8 .type _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj,@object # @_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .globl _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .p2align 3, 0x0 _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj: .quad _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .size _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj, 8 .type _Z12mstep_kernelPfS_S_S_S_jjj,@object # @_Z12mstep_kernelPfS_S_S_S_jjj .globl _Z12mstep_kernelPfS_S_S_S_jjj .p2align 3, 0x0 _Z12mstep_kernelPfS_S_S_S_jjj: .quad _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .size _Z12mstep_kernelPfS_S_S_S_jjj, 8 .type _Z18mstep_sigma_kernelPfS_S_S_S_jjj,@object # @_Z18mstep_sigma_kernelPfS_S_S_S_jjj .globl _Z18mstep_sigma_kernelPfS_S_S_S_jjj .p2align 3, 0x0 _Z18mstep_sigma_kernelPfS_S_S_S_jjj: .quad _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .size _Z18mstep_sigma_kernelPfS_S_S_S_jjj, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z12estep_kernelPfS_S_S_S_S_jjj" .size .L__unnamed_1, 32 .type .L__unnamed_2,@object # @1 .L__unnamed_2: .asciz "_Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj" .size .L__unnamed_2, 44 .type .L__unnamed_3,@object # @2 .L__unnamed_3: .asciz "_Z12mstep_kernelPfS_S_S_S_jjj" .size .L__unnamed_3, 30 .type .L__unnamed_4,@object # @3 .L__unnamed_4: .asciz "_Z18mstep_sigma_kernelPfS_S_S_S_jjj" .size .L__unnamed_4, 36 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z27__device_stub__estep_kernelPfS_S_S_S_S_jjj .addrsig_sym _Z37__device_stub__estep_normalize_kernelPfS_S_S_S_S_S_jjj .addrsig_sym _Z27__device_stub__mstep_kernelPfS_S_S_S_jjj .addrsig_sym _Z33__device_stub__mstep_sigma_kernelPfS_S_S_S_jjj .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z12estep_kernelPfS_S_S_S_S_jjj .addrsig_sym _Z22estep_normalize_kernelPfS_S_S_S_S_S_jjj .addrsig_sym _Z12mstep_kernelPfS_S_S_S_jjj .addrsig_sym _Z18mstep_sigma_kernelPfS_S_S_S_jjj .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
// #include <ATen/ATen.h> #include <cuda.h> #include <cuda_runtime.h> #include <cstdio> #include <cmath> #include <iostream> namespace { template <typename scalar_t> __device__ __forceinline__ void single_mul( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { scalar_t uavc = x_re * (y_re + y_im); *out_re += uavc - (x_re + x_im) * y_im; *out_im += (x_im - x_re) * y_re + uavc; } template <typename scalar_t> __device__ __forceinline__ void single_add( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { *out_re += x_re + y_re; *out_im += x_im + y_im; } /** Complex multiplication of tensors using shared memory and barrier synchronization. Compute the element wise complex multiplication for each thread in the block and write the result to the shared memory. Then synchronize the threads and in the log based fashion sum up the results for each output pixel through its channels, if they are present in the cache. The stride is the number of threads per block times the I (the two float representation of the complex numbers). */ template <typename scalar_t> __global__ void complex_mul_cuda_kernel( const scalar_t* __restrict__ x, const scalar_t* __restrict__ y, scalar_t* __restrict__ out, const int N, const int F, const int C, const int H, const int W) { // The size of the shared memory cache should be twice the number of threads // per block as we store the real and imaginary part of the result. extern __shared__ float cache[]; // cache for the result of the complex multiplication const int I = 2; // the last dimension for the complex number const int plane_size = H * W; const int channel_size = plane_size * I; const int image_size = C * channel_size; // size of the image from the batch // number of complex values in the input that we iterate through const int nr_values = C * H * W; const int n = blockIdx.x; // current index of an image/input map in the batch const int f = blockIdx.y; // current index of a filter from the filter bank const int block_size = blockDim.x; const int thread_nr = threadIdx.x; // stride for the H*W map is equal to the number of threads declared in a block const int stride = block_size * I; // we need H*W threads per plane, each deals with I numbers const int n_idx = n * image_size; // start index in the batch for this input map const int f_idx = f * image_size; // start index in the bank for this filter // find index for the output const int no_idx = n * (F * channel_size); // output index for the batch data point const int fo_idx = f * channel_size; // output index for the filter/channel // Each H*W plane contains H*W*I elements in depth. // We linearize it and start from 0, move by #threads*I steps in outer loop. const int start_idx = threadIdx.x*I; // index in the input map int N_idx = n_idx + start_idx; // index across the first channel plane (in the input map n). const int last_N_idx = n_idx + image_size; // last index for the starting position to compute the sum through each channel for this pixel // To prevent us from a deadlock, we have to always execute __syncthreads(); // for all the threads in the block. Each thread has to do the same number of // iterations for any loop. To ensure that, we keep all threads running, // even though, some of them are really idle. We keep the loop running to // the multiple of the block size that is greater than the number of values // in the input map in total: C*H*W - this is a number of complex cells in the // input map. const int num_blocks = (nr_values + block_size - 1) / block_size; const int last_block_idx = n_idx + num_blocks * block_size * I; // index in the filter int F_idx = f_idx + start_idx; // index in the output, we compute cells on a flat plane (no channels) int base_O_idx = no_idx + fo_idx; int run_O_idx = (start_idx % channel_size); int thread_cidx = thread_nr * I; printf("N_idx:%d, last_block_idx:%d, last_N_idx:%d\n", N_idx, last_block_idx, last_N_idx); while (N_idx < last_block_idx) { // Zero out caches. cache[thread_cidx] = 0; cache[thread_cidx + 1] = 0; if (N_idx < last_N_idx - 1) { scalar_t out_re = 0; scalar_t out_im = 0; scalar_t x_re = x[N_idx]; scalar_t x_im = x[N_idx + 1]; scalar_t y_re = y[F_idx]; scalar_t y_im = y[F_idx + 1]; single_mul(x_re, x_im, y_re, y_im, &out_re, &out_im); cache[thread_cidx] = out_re; cache[thread_cidx + 1] = out_im; } __syncthreads(); // Make the results visible to all threads. // It is not O(logN) but O(N) as of now. For each element in the output // map we have a dedicated thread. The thread goes through all the // channels present in the cache. if (thread_nr < plane_size) { for (int cache_index = thread_nr + plane_size; cache_index < block_size; cache_index += plane_size) { cache[thread_cidx] += cache[cache_index*I]; cache[thread_cidx + 1] += cache[cache_index*I + 1]; } // Move the summed values (across the channels) for each pixel to // the output. const int O_idx = base_O_idx + run_O_idx; out[O_idx] += cache[thread_cidx]; out[O_idx + 1] = cache[thread_cidx + 1]; } N_idx += stride; F_idx += stride; run_O_idx = (run_O_idx + stride) % channel_size; // Make sure that all cache cells are zeroed out before moving on. // We need this as in the second part we access cache cells that do not // belong only to this thread. __syncthreads(); } } } // namespace //void complex_mul_stride_no_permute_cuda( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // AT_DISPATCH_FLOATING_TYPES(x.type(), "complex_mul_cuda", // ([&] { // complex_mul_cuda_kernel<scalar_t><<<blocks, threads>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); // })); //} //template <typename scalar_t> //void complex_mul_stride_no_permute_cuda_pure( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // // Run kernel on the GPU // complex_mul_cuda_kernel<scalar_t><<<blocks, 1024>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); //} /** Uncomment the pytorch related stuff. Compile: ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvcc complex_mul_kernel_stride_no_permute.cu -o complex_mul_profile.out ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvprof ./complex_mul_profile.out nvidia /usr/local/cuda/bin/nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m -c complex_mul_kernel.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m complex_mul_kernel_stride_no_permute.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 Segmentation fault */ int main(void) { int N = 1; int F = 1; int C = 4; int H = 3; int W = 2; int I = 2; int size_input = N * C * H * W * I; int size_filter = F * C * H * W * I; int size_output = N * F * H * W * I; int cuda_block_threads = 16; // auto dims = {128, 32, 16, 8, 2}; // at::Tensor x = at::randn({128, 32, 16, 8, 2}); // at::Tensor y = at::randn({128, 32, 16, 8, 2}); // at::Tensor out = at::zeros({128, 32, 16, 8, 2}); float *x, *y, * out; // Allocate unified memory - accessible from cpu or gpu cudaMallocManaged(&x, size_input*sizeof(float)); cudaMallocManaged(&y, size_filter*sizeof(float)); cudaMallocManaged(&out, size_output*sizeof(float)); for (int j=0; j<H; ++j) { for (int i=0; i<W; ++i) { x[(j*W+i)*2] = 3; x[(j*W+i)*2 + 1] = 1; y[(j*W+i)*2] = 4; y[(j*W+i)*2 + 1] = 2; } } for (int i=0; i<H*W*2; i+=2) { printf("%p %d: %f, %f, %f, %f\n", x, i, x[i], x[i+1], y[i], y[i+1]); } // float *dz; // device z // cudaMalloc(&dz, 9*sizeof(float)); // cudaMemcpy(dz, hz, 9*sizeof(float), cudaMemcpyHostToDevice); const dim3 blocks(N, F); complex_mul_cuda_kernel<float><<<blocks, cuda_block_threads, cuda_block_threads*2>>>(x, y, out, N, F, C, H, W); for (int i=0; i<H*W*C; i+=2) { printf("%d: %f, %f\n", i, out[i], out[i+1]); } cudaFree(x); cudaFree(y); cudaFree(out); // cudaFree(dz); printf("finished computation\n"); return 0; }
code for sm_80 Function : _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ IMAD.MOV.U32 R1, RZ, RZ, c[0x0][0x28] ; /* 0x00000a00ff017624 */ /* 0x000fe400078e00ff */ /*0010*/ IABS R6, c[0x0][0x0] ; /* 0x0000000000067a13 */ /* 0x000fe20000000000 */ /*0020*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x188] ; /* 0x00006200ff027624 */ /* 0x000fe200078e00ff */ /*0030*/ MOV R7, c[0x0][0x0] ; /* 0x0000000000077a02 */ /* 0x000fe20000000f00 */ /*0040*/ S2UR UR36, SR_CTAID.X ; /* 0x00000000002479c3 */ /* 0x000e220000002500 */ /*0050*/ I2F.RP R3, R6 ; /* 0x0000000600037306 */ /* 0x000e620000209400 */ /*0060*/ IMAD R2, R2, c[0x0][0x184], RZ ; /* 0x0000610002027a24 */ /* 0x000fe200078e02ff */ /*0070*/ S2R R19, SR_TID.X ; /* 0x0000000000137919 */ /* 0x000ea20000002100 */ /*0080*/ IADD3 R1, R1, -0x10, RZ ; /* 0xfffffff001017810 */ /* 0x000fe40007ffe0ff */ /*0090*/ IMAD R0, R2.reuse, c[0x0][0x180], R7 ; /* 0x0000600002007a24 */ /* 0x040fe400078e0207 */ /*00a0*/ IMAD.SHL.U32 R18, R2, 0x2, RZ ; /* 0x0000000202127824 */ /* 0x000fc600078e00ff */ /*00b0*/ IADD3 R0, R0, -0x1, RZ ; /* 0xffffffff00007810 */ /* 0x000fe20007ffe0ff */ /*00c0*/ IMAD R22, R18, c[0x0][0x180], RZ ; /* 0x0000600012167a24 */ /* 0x000fc600078e02ff */ /*00d0*/ ISETP.GE.AND P2, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fe20003f46270 */ /*00e0*/ MUFU.RCP R3, R3 ; /* 0x0000000300037308 */ /* 0x002e620000001000 */ /*00f0*/ IMAD R16, R22, UR36, RZ ; /* 0x0000002416107c24 */ /* 0x001fe4000f8e02ff */ /*0100*/ IMAD.SHL.U32 R25, R19, 0x2, RZ ; /* 0x0000000213197824 */ /* 0x004fc600078e00ff */ /*0110*/ IADD3 R24, R22, R16, RZ ; /* 0x0000001016187210 */ /* 0x000fe40007ffe0ff */ /*0120*/ IADD3 R4, R3, 0xffffffe, RZ ; /* 0x0ffffffe03047810 */ /* 0x002fc60007ffe0ff */ /*0130*/ STL [R1+0x8], R24 ; /* 0x0000081801007387 */ /* 0x0001e20000100800 */ /*0140*/ IABS R3, R0 ; /* 0x0000000000037213 */ /* 0x000fe20000000000 */ /*0150*/ F2I.FTZ.U32.TRUNC.NTZ R5, R4 ; /* 0x0000000400057305 */ /* 0x0002a4000021f000 */ /*0160*/ HFMA2.MMA R4, -RZ, RZ, 0, 0 ; /* 0x00000000ff047435 */ /* 0x002fe200000001ff */ /*0170*/ IMAD.MOV R8, RZ, RZ, -R5 ; /* 0x000000ffff087224 */ /* 0x004fc800078e0a05 */ /*0180*/ IMAD.MOV.U32 R7, RZ, RZ, R8 ; /* 0x000000ffff077224 */ /* 0x000fe200078e0008 */ /*0190*/ MOV R8, 0x0 ; /* 0x0000000000087802 */ /* 0x000fc60000000f00 */ /*01a0*/ IMAD R7, R7, R6, RZ ; /* 0x0000000607077224 */ /* 0x000fc600078e02ff */ /*01b0*/ LDC.64 R8, c[0x4][R8] ; /* 0x0100000008087b82 */ /* 0x000e620000000a00 */ /*01c0*/ IMAD.HI.U32 R4, R5, R7, R4 ; /* 0x0000000705047227 */ /* 0x000fe200078e0004 */ /*01d0*/ MOV R5, c[0x4][0xc] ; /* 0x0100030000057a02 */ /* 0x000fca0000000f00 */ /*01e0*/ IMAD.HI.U32 R4, R4, R3, RZ ; /* 0x0000000304047227 */ /* 0x000fc800078e00ff */ /*01f0*/ IMAD.MOV R4, RZ, RZ, -R4 ; /* 0x000000ffff047224 */ /* 0x000fc800078e0a04 */ /*0200*/ IMAD R3, R6, R4, R3 ; /* 0x0000000406037224 */ /* 0x000fe400078e0203 */ /*0210*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x4][0x8] ; /* 0x01000200ff047624 */ /* 0x000fc600078e00ff */ /*0220*/ ISETP.GT.U32.AND P0, PT, R6, R3, PT ; /* 0x000000030600720c */ /* 0x000fda0003f04070 */ /*0230*/ @!P0 IMAD.IADD R3, R3, 0x1, -R6 ; /* 0x0000000103038824 */ /* 0x000fe200078e0a06 */ /*0240*/ ISETP.NE.AND P0, PT, RZ, c[0x0][0x0], PT ; /* 0x00000000ff007a0c */ /* 0x000fc80003f05270 */ /*0250*/ ISETP.GT.U32.AND P1, PT, R6, R3, PT ; /* 0x000000030600720c */ /* 0x000fda0003f24070 */ /*0260*/ @!P1 IADD3 R3, R3, -R6, RZ ; /* 0x8000000603039210 */ /* 0x000fca0007ffe0ff */ /*0270*/ @!P2 IMAD.MOV R3, RZ, RZ, -R3 ; /* 0x000000ffff03a224 */ /* 0x000fe200078e0a03 */ /*0280*/ @!P0 LOP3.LUT R3, RZ, c[0x0][0x0], RZ, 0x33, !PT ; /* 0x00000000ff038a12 */ /* 0x000fe400078e33ff */ /*0290*/ IADD3 R6, P0, R1, c[0x0][0x20], RZ ; /* 0x0000080001067a10 */ /* 0x000fe40007f1e0ff */ /*02a0*/ IADD3 R17, R0, -R3, RZ ; /* 0x8000000300117210 */ /* 0x000fc60007ffe0ff */ /*02b0*/ IMAD.X R7, RZ, RZ, c[0x0][0x24], P0 ; /* 0x00000900ff077624 */ /* 0x000fe400000e06ff */ /*02c0*/ IMAD R17, R17, 0x2, R16 ; /* 0x0000000211117824 */ /* 0x000fe200078e0210 */ /*02d0*/ IADD3 R16, R16, R25, RZ ; /* 0x0000001910107210 */ /* 0x000fca0007ffe0ff */ /*02e0*/ STL.64 [R1], R16 ; /* 0x0000001001007387 */ /* 0x0001e40000100a00 */ /*02f0*/ LEPC R10 ; /* 0x00000000000a734e */ /* 0x002fe20000000000 */ /*0300*/ MOV R3, 0x370 ; /* 0x0000037000037802 */ /* 0x000fe40000000f00 */ /*0310*/ MOV R20, 0x2f0 ; /* 0x000002f000147802 */ /* 0x000fe40000000f00 */ /*0320*/ MOV R21, 0x0 ; /* 0x0000000000157802 */ /* 0x000fe40000000f00 */ /*0330*/ MOV R0, 0x0 ; /* 0x0000000000007802 */ /* 0x000fe40000000f00 */ /*0340*/ IADD3 R20, P0, P1, -R20, R3, R10 ; /* 0x0000000314147210 */ /* 0x000fc8000791e10a */ /*0350*/ IADD3.X R21, ~R0, R21, R11, P0, P1 ; /* 0x0000001500157210 */ /* 0x000fc800007e250b */ /*0360*/ CALL.ABS.NOINC R8 ; /* 0x0000000008007343 */ /* 0x001fea0003c00000 */ /*0370*/ ISETP.GE.AND P0, PT, R16, R17, PT ; /* 0x000000111000720c */ /* 0x000fda0003f06270 */ /*0380*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0390*/ I2F.U32.RP R6, R2 ; /* 0x0000000200067306 */ /* 0x000e220000209000 */ /*03a0*/ IADD3 R7, RZ, -R2, RZ ; /* 0x80000002ff077210 */ /* 0x000fe20007ffe0ff */ /*03b0*/ IMAD.IADD R3, R2.reuse, 0x1, R19 ; /* 0x0000000102037824 */ /* 0x040fe200078e0213 */ /*03c0*/ S2UR UR5, SR_CTAID.Y ; /* 0x00000000000579c3 */ /* 0x000e620000002600 */ /*03d0*/ ISETP.NE.U32.AND P2, PT, R2, RZ, PT ; /* 0x000000ff0200720c */ /* 0x000fe20003f45070 */ /*03e0*/ ULDC UR4, c[0x0][0x17c] ; /* 0x00005f0000047ab9 */ /* 0x000fe20000000800 */ /*03f0*/ IADD3 R23, R24, -0x1, RZ ; /* 0xffffffff18177810 */ /* 0x000fe40007ffe0ff */ /*0400*/ LOP3.LUT R0, RZ, R3, RZ, 0x33, !PT ; /* 0x00000003ff007212 */ /* 0x000fe200078e33ff */ /*0410*/ MUFU.RCP R6, R6 ; /* 0x0000000600067308 */ /* 0x001e220000001000 */ /*0420*/ UIMAD UR4, UR36, UR4, UR5 ; /* 0x00000004240472a4 */ /* 0x002fe2000f8e0205 */ /*0430*/ IADD3 R4, R6, 0xffffffe, RZ ; /* 0x0ffffffe06047810 */ /* 0x001fcc0007ffe0ff */ /*0440*/ F2I.FTZ.U32.TRUNC.NTZ R5, R4 ; /* 0x0000000400057305 */ /* 0x000064000021f000 */ /*0450*/ HFMA2.MMA R4, -RZ, RZ, 0, 0 ; /* 0x00000000ff047435 */ /* 0x001fe200000001ff */ /*0460*/ IMAD R7, R7, R5, RZ ; /* 0x0000000507077224 */ /* 0x002fd200078e02ff */ /*0470*/ IMAD.HI.U32 R8, R5, R7, R4 ; /* 0x0000000705087227 */ /* 0x000fe200078e0004 */ /*0480*/ IADD3 R5, R0, c[0x0][0x0], RZ ; /* 0x0000000000057a10 */ /* 0x000fe40007ffe0ff */ /*0490*/ SHF.L.U32 R7, R25, 0x2, RZ ; /* 0x0000000219077819 */ /* 0x000fe200000006ff */ /*04a0*/ IMAD R4, R22, UR5, R25 ; /* 0x0000000516047c24 */ /* 0x000fe4000f8e0219 */ /*04b0*/ IMAD.HI.U32 R0, R8, R5, RZ ; /* 0x0000000508007227 */ /* 0x000fc800078e00ff */ /*04c0*/ IMAD.MOV R6, RZ, RZ, -R0 ; /* 0x000000ffff067224 */ /* 0x000fe400078e0a00 */ /*04d0*/ IMAD R22, R18, UR4, RZ ; /* 0x0000000412167c24 */ /* 0x000fe4000f8e02ff */ /*04e0*/ IMAD R5, R2, R6, R5 ; /* 0x0000000602057224 */ /* 0x000fca00078e0205 */ /*04f0*/ ISETP.GE.U32.AND P0, PT, R5, R2, PT ; /* 0x000000020500720c */ /* 0x000fda0003f06070 */ /*0500*/ @P0 IADD3 R5, -R2, R5, RZ ; /* 0x0000000502050210 */ /* 0x000fe40007ffe1ff */ /*0510*/ @P0 IADD3 R0, R0, 0x1, RZ ; /* 0x0000000100000810 */ /* 0x000fe40007ffe0ff */ /*0520*/ ISETP.GE.U32.AND P1, PT, R5, R2, PT ; /* 0x000000020500720c */ /* 0x000fe20003f26070 */ /*0530*/ IMAD.IADD R5, R2, 0x1, R3 ; /* 0x0000000102057824 */ /* 0x000fc800078e0203 */ /*0540*/ IMAD.IADD R9, R2, 0x1, R5 ; /* 0x0000000102097824 */ /* 0x000fca00078e0205 */ /*0550*/ IADD3 R6, R2, R9, RZ ; /* 0x0000000902067210 */ /* 0x000fc60007ffe0ff */ /*0560*/ @P1 IADD3 R0, R0, 0x1, RZ ; /* 0x0000000100001810 */ /* 0x000fe40007ffe0ff */ /*0570*/ @!P2 LOP3.LUT R0, RZ, R2, RZ, 0x33, !PT ; /* 0x00000002ff00a212 */ /* 0x000fc800078e33ff */ /*0580*/ IADD3 R8, R0, 0x1, RZ ; /* 0x0000000100087810 */ /* 0x000fc80007ffe0ff */ /*0590*/ LOP3.LUT R8, R8, 0x3, RZ, 0xc0, !PT ; /* 0x0000000308087812 */ /* 0x000fe400078ec0ff */ /*05a0*/ ISETP.GE.AND P1, PT, R16, R23, PT ; /* 0x000000171000720c */ /* 0x000fe20003f26270 */ /*05b0*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd80000000a00 */ /*05c0*/ @!P1 IMAD.MOV.U32 R13, RZ, RZ, 0x4 ; /* 0x00000004ff0d9424 */ /* 0x000fc800078e00ff */ /*05d0*/ @!P1 IMAD.WIDE R10, R4, R13, c[0x0][0x168] ; /* 0x00005a00040a9625 */ /* 0x000fc800078e020d */ /*05e0*/ @!P1 IMAD.WIDE R12, R16, R13, c[0x0][0x160] ; /* 0x00005800100c9625 */ /* 0x000fe200078e020d */ /*05f0*/ @!P1 LDG.E.CONSTANT R20, [R10.64+0x4] ; /* 0x000004040a149981 */ /* 0x0000a8000c1e9900 */ /*0600*/ @!P1 LDG.E.CONSTANT R21, [R10.64] ; /* 0x000000040a159981 */ /* 0x0000a8000c1e9900 */ /*0610*/ @!P1 LDG.E.CONSTANT R26, [R12.64] ; /* 0x000000040c1a9981 */ /* 0x0002e8000c1e9900 */ /*0620*/ @!P1 LDG.E.CONSTANT R27, [R12.64+0x4] ; /* 0x000004040c1b9981 */ /* 0x000322000c1e9900 */ /*0630*/ IABS R24, R18 ; /* 0x0000001200187213 */ /* 0x000fe20000000000 */ /*0640*/ WARPSYNC 0xffffffff ; /* 0xffffffff00007948 */ /* 0x000fe20003800000 */ /*0650*/ IABS R10, R25 ; /* 0x00000019000a7213 */ /* 0x001fe20000000000 */ /*0660*/ BSSY B0, 0xce0 ; /* 0x0000067000007945 */ /* 0x000fe20003800000 */ /*0670*/ I2F.RP R28, R24 ; /* 0x00000018001c7306 */ /* 0x000e220000209400 */ /*0680*/ ISETP.GE.AND P2, PT, R25, RZ, PT ; /* 0x000000ff1900720c */ /* 0x000fce0003f46270 */ /*0690*/ MUFU.RCP R28, R28 ; /* 0x0000001c001c7308 */ /* 0x001e240000001000 */ /*06a0*/ IADD3 R14, R28, 0xffffffe, RZ ; /* 0x0ffffffe1c0e7810 */ /* 0x001fcc0007ffe0ff */ /*06b0*/ F2I.FTZ.U32.TRUNC.NTZ R15, R14 ; /* 0x0000000e000f7305 */ /* 0x000164000021f000 */ /*06c0*/ IMAD.MOV.U32 R14, RZ, RZ, RZ ; /* 0x000000ffff0e7224 */ /* 0x001fe200078e00ff */ /*06d0*/ IADD3 R29, RZ, -R15, RZ ; /* 0x8000000fff1d7210 */ /* 0x020fca0007ffe0ff */ /*06e0*/ IMAD R11, R29, R24, RZ ; /* 0x000000181d0b7224 */ /* 0x000fc800078e02ff */ /*06f0*/ IMAD.HI.U32 R15, R15, R11, R14 ; /* 0x0000000b0f0f7227 */ /* 0x000fe200078e000e */ /*0700*/ IABS R11, R18 ; /* 0x00000012000b7213 */ /* 0x000fc80000000000 */ /*0710*/ IADD3 R13, RZ, -R11, RZ ; /* 0x8000000bff0d7210 */ /* 0x002fe20007ffe0ff */ /*0720*/ IMAD.HI.U32 R15, R15, R10, RZ ; /* 0x0000000a0f0f7227 */ /* 0x000fc800078e00ff */ /*0730*/ IMAD R13, R15, R13, R10 ; /* 0x0000000d0f0d7224 */ /* 0x000fca00078e020a */ /*0740*/ ISETP.GT.U32.AND P0, PT, R24, R13, PT ; /* 0x0000000d1800720c */ /* 0x000fda0003f04070 */ /*0750*/ @!P0 IMAD.IADD R13, R13, 0x1, -R24 ; /* 0x000000010d0d8824 */ /* 0x000fca00078e0a18 */ /*0760*/ ISETP.GT.U32.AND P0, PT, R24, R13, PT ; /* 0x0000000d1800720c */ /* 0x000fda0003f04070 */ /*0770*/ @!P0 IADD3 R13, R13, -R24, RZ ; /* 0x800000180d0d8210 */ /* 0x000fe40007ffe0ff */ /*0780*/ ISETP.GE.AND P0, PT, R19, R2, PT ; /* 0x000000021300720c */ /* 0x000fc60003f06270 */ /*0790*/ IMAD.MOV.U32 R25, RZ, RZ, R13 ; /* 0x000000ffff197224 */ /* 0x000fca00078e000d */ /*07a0*/ @!P2 IADD3 R25, -R25, RZ, RZ ; /* 0x000000ff1919a210 */ /* 0x000fe20007ffe1ff */ /*07b0*/ @!P1 FADD R11, R20, R21 ; /* 0x00000015140b9221 */ /* 0x004fc80000000000 */ /*07c0*/ @!P1 FMUL R12, R11, R26 ; /* 0x0000001a0b0c9220 */ /* 0x008fe40000400000 */ /*07d0*/ @!P1 FADD R10, -R26.reuse, R27.reuse ; /* 0x0000001b1a0a9221 */ /* 0x150fe40000000100 */ /*07e0*/ @!P1 FADD R27, R26, R27 ; /* 0x0000001b1a1b9221 */ /* 0x000fe40000000000 */ /*07f0*/ @!P1 FFMA R21, R21, R10, R12.reuse ; /* 0x0000000a15159223 */ /* 0x100fe4000000000c */ /*0800*/ @!P1 FFMA R27, R20, -R27, R12 ; /* 0x8000001b141b9223 */ /* 0x000fe2000000000c */ /*0810*/ CS2R R10, SRZ ; /* 0x00000000000a7805 */ /* 0x000fe4000001ff00 */ /*0820*/ @!P1 FADD R11, RZ, R21 ; /* 0x00000015ff0b9221 */ /* 0x000fc40000000000 */ /*0830*/ @!P1 FADD R10, RZ, R27 ; /* 0x0000001bff0a9221 */ /* 0x000fe20000000000 */ /*0840*/ ISETP.NE.AND P1, PT, R18, RZ, PT ; /* 0x000000ff1200720c */ /* 0x000fc80003f25270 */ /*0850*/ STS.64 [R7], R10 ; /* 0x0000000a07007388 */ /* 0x0001f20000000a00 */ /*0860*/ @!P1 LOP3.LUT R25, RZ, R18, RZ, 0x33, !PT ; /* 0x00000012ff199212 */ /* 0x000fe200078e33ff */ /*0870*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0880*/ @P0 BRA 0xcd0 ; /* 0x0000044000000947 */ /* 0x000fea0003800000 */ /*0890*/ ISETP.GE.AND P0, PT, R3, c[0x0][0x0], PT ; /* 0x0000000003007a0c */ /* 0x001fe20003f06270 */ /*08a0*/ BSSY B1, 0xc40 ; /* 0x0000039000017945 */ /* 0x000fd80003800000 */ /*08b0*/ @P0 BRA 0xc30 ; /* 0x0000037000000947 */ /* 0x000fea0003800000 */ /*08c0*/ ISETP.NE.AND P1, PT, R8, RZ, PT ; /* 0x000000ff0800720c */ /* 0x000fe20003f25270 */ /*08d0*/ BSSY B2, 0xa80 ; /* 0x000001a000027945 */ /* 0x000fe20003800000 */ /*08e0*/ ISETP.GE.U32.AND P0, PT, R0, 0x3, PT ; /* 0x000000030000780c */ /* 0x000fe20003f06070 */ /*08f0*/ IMAD.MOV.U32 R27, RZ, RZ, R3 ; /* 0x000000ffff1b7224 */ /* 0x000fd400078e0003 */ /*0900*/ @!P1 BRA 0xa70 ; /* 0x0000016000009947 */ /* 0x000fea0003800000 */ /*0910*/ LEA R15, R2, R7, 0x3 ; /* 0x00000007020f7211 */ /* 0x000fe200078e18ff */ /*0920*/ LDS.64 R10, [R7] ; /* 0x00000000070a7984 */ /* 0x000fe20000000a00 */ /*0930*/ ISETP.NE.AND P1, PT, R8, 0x1, PT ; /* 0x000000010800780c */ /* 0x000fe20003f25270 */ /*0940*/ IMAD.MOV.U32 R27, RZ, RZ, R5 ; /* 0x000000ffff1b7224 */ /* 0x000fe400078e0005 */ /*0950*/ LDS.64 R12, [R15] ; /* 0x000000000f0c7984 */ /* 0x000e240000000a00 */ /*0960*/ FADD R10, R12, R10 ; /* 0x0000000a0c0a7221 */ /* 0x001fe40000000000 */ /*0970*/ FADD R11, R13, R11 ; /* 0x0000000b0d0b7221 */ /* 0x000fca0000000000 */ /*0980*/ STS.64 [R7], R10 ; /* 0x0000000a07007388 */ /* 0x0001e20000000a00 */ /*0990*/ @!P1 BRA 0xa70 ; /* 0x000000d000009947 */ /* 0x000fea0003800000 */ /*09a0*/ LEA R15, R2, R15, 0x3 ; /* 0x0000000f020f7211 */ /* 0x000fe400078e18ff */ /*09b0*/ ISETP.NE.AND P1, PT, R8, 0x2, PT ; /* 0x000000020800780c */ /* 0x000fe40003f25270 */ /*09c0*/ MOV R27, R9 ; /* 0x00000009001b7202 */ /* 0x000fe20000000f00 */ /*09d0*/ LDS.64 R12, [R15] ; /* 0x000000000f0c7984 */ /* 0x000e740000000a00 */ /*09e0*/ @P1 IMAD R14, R2, 0x8, R15 ; /* 0x00000008020e1824 */ /* 0x000fc400078e020f */ /*09f0*/ @P1 IMAD.MOV.U32 R27, RZ, RZ, R6 ; /* 0x000000ffff1b1224 */ /* 0x000fe400078e0006 */ /*0a00*/ FADD R12, R10, R12 ; /* 0x0000000c0a0c7221 */ /* 0x002fe40000000000 */ /*0a10*/ FADD R13, R11, R13 ; /* 0x0000000d0b0d7221 */ /* 0x000fca0000000000 */ /*0a20*/ STS.64 [R7], R12 ; /* 0x0000000c07007388 */ /* 0x000fe80000000a00 */ /*0a30*/ @P1 LDS.64 R10, [R14] ; /* 0x000000000e0a1984 */ /* 0x001e240000000a00 */ /*0a40*/ @P1 FADD R11, R13, R11 ; /* 0x0000000b0d0b1221 */ /* 0x001fe40000000000 */ /*0a50*/ @P1 FADD R10, R12, R10 ; /* 0x0000000a0c0a1221 */ /* 0x000fca0000000000 */ /*0a60*/ @P1 STS.64 [R7], R10 ; /* 0x0000000a07001388 */ /* 0x0001e40000000a00 */ /*0a70*/ BSYNC B2 ; /* 0x0000000000027941 */ /* 0x000fea0003800000 */ /*0a80*/ @!P0 BRA 0xc30 ; /* 0x000001a000008947 */ /* 0x000fea0003800000 */ /*0a90*/ LDS.64 R10, [R7] ; /* 0x00000000070a7984 */ /* 0x0010620000000a00 */ /*0aa0*/ SHF.L.U32 R29, R27, 0x3, RZ ; /* 0x000000031b1d7819 */ /* 0x000fca00000006ff */ /*0ab0*/ LDS.64 R12, [R29] ; /* 0x000000001d0c7984 */ /* 0x0004e20000000a00 */ /*0ac0*/ IMAD R24, R2.reuse, 0x8, R29 ; /* 0x0000000802187824 */ /* 0x040fe200078e021d */ /*0ad0*/ IADD3 R27, R2, R27, R2 ; /* 0x0000001b021b7210 */ /* 0x000fc80007ffe002 */ /*0ae0*/ LEA R26, R2.reuse, R24, 0x3 ; /* 0x00000018021a7211 */ /* 0x040fe400078e18ff */ /*0af0*/ IADD3 R27, R2, R27, R2 ; /* 0x0000001b021b7210 */ /* 0x000fc60007ffe002 */ /*0b00*/ IMAD R28, R2, 0x8, R26 ; /* 0x00000008021c7824 */ /* 0x000fe200078e021a */ /*0b10*/ ISETP.GE.AND P0, PT, R27, c[0x0][0x0], PT ; /* 0x000000001b007a0c */ /* 0x000fc80003f06270 */ /*0b20*/ LEA R29, R2, R28, 0x3 ; /* 0x0000001c021d7211 */ /* 0x004fe200078e18ff */ /*0b30*/ FADD R13, R13, R11 ; /* 0x0000000b0d0d7221 */ /* 0x00afe40000000000 */ /*0b40*/ FADD R12, R12, R10 ; /* 0x0000000a0c0c7221 */ /* 0x000fca0000000000 */ /*0b50*/ STS.64 [R7], R12 ; /* 0x0000000c07007388 */ /* 0x000fe80000000a00 */ /*0b60*/ LDS.64 R10, [R24] ; /* 0x00000000180a7984 */ /* 0x000e640000000a00 */ /*0b70*/ FADD R15, R13, R11 ; /* 0x0000000b0d0f7221 */ /* 0x002fe40000000000 */ /*0b80*/ FADD R14, R12, R10 ; /* 0x0000000a0c0e7221 */ /* 0x000fca0000000000 */ /*0b90*/ STS.64 [R7], R14 ; /* 0x0000000e07007388 */ /* 0x000fe80000000a00 */ /*0ba0*/ LDS.64 R10, [R26] ; /* 0x000000001a0a7984 */ /* 0x000e640000000a00 */ /*0bb0*/ FADD R21, R15, R11 ; /* 0x0000000b0f157221 */ /* 0x002fe40000000000 */ /*0bc0*/ FADD R20, R14, R10 ; /* 0x0000000a0e147221 */ /* 0x000fca0000000000 */ /*0bd0*/ STS.64 [R7], R20 ; /* 0x0000001407007388 */ /* 0x000fe80000000a00 */ /*0be0*/ LDS.64 R10, [R28] ; /* 0x000000001c0a7984 */ /* 0x000e640000000a00 */ /*0bf0*/ FADD R10, R20, R10 ; /* 0x0000000a140a7221 */ /* 0x002fe40000000000 */ /*0c00*/ FADD R11, R21, R11 ; /* 0x0000000b150b7221 */ /* 0x000fca0000000000 */ /*0c10*/ STS.64 [R7], R10 ; /* 0x0000000a07007388 */ /* 0x0003e20000000a00 */ /*0c20*/ @!P0 BRA 0xab0 ; /* 0xfffffe8000008947 */ /* 0x000fea000383ffff */ /*0c30*/ BSYNC B1 ; /* 0x0000000000017941 */ /* 0x000fea0003800000 */ /*0c40*/ HFMA2.MMA R13, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff0d7435 */ /* 0x000fe200000001ff */ /*0c50*/ IMAD.IADD R12, R22, 0x1, R25 ; /* 0x00000001160c7824 */ /* 0x000fe200078e0219 */ /*0c60*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0c70*/ LDS.64 R10, [R7] ; /* 0x00000000070a7984 */ /* 0x003e2e0000000a00 */ /*0c80*/ IMAD.WIDE R12, R12, R13, c[0x0][0x170] ; /* 0x00005c000c0c7625 */ /* 0x000fca00078e020d */ /*0c90*/ LDG.E R15, [R12.64] ; /* 0x000000040c0f7981 */ /* 0x000ea8000c1e1900 */ /*0ca0*/ STG.E [R12.64+0x4], R11 ; /* 0x0000040b0c007986 */ /* 0x0011e2000c101904 */ /*0cb0*/ FADD R15, R15, R10 ; /* 0x0000000a0f0f7221 */ /* 0x004fca0000000000 */ /*0cc0*/ STG.E [R12.64], R15 ; /* 0x0000000f0c007986 */ /* 0x0001e4000c101904 */ /*0cd0*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x001fea0003800000 */ /*0ce0*/ IMAD.MOV.U32 R11, RZ, RZ, c[0x0][0x0] ; /* 0x00000000ff0b7624 */ /* 0x000fe200078e00ff */ /*0cf0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fe60000010000 */ /*0d00*/ IMAD R4, R11.reuse, 0x2, R4 ; /* 0x000000020b047824 */ /* 0x040fe200078e0204 */ /*0d10*/ LEA R16, R11.reuse, R16, 0x1 ; /* 0x000000100b107211 */ /* 0x040fe400078e08ff */ /*0d20*/ LEA R25, R11, R25, 0x1 ; /* 0x000000190b197211 */ /* 0x000fe400078e08ff */ /*0d30*/ ISETP.GE.AND P0, PT, R16, R17, PT ; /* 0x000000111000720c */ /* 0x000fda0003f06270 */ /*0d40*/ @!P0 BRA 0x5a0 ; /* 0xfffff85000008947 */ /* 0x000fea000383ffff */ /*0d50*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0d60*/ BRA 0xd60; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0d70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0d90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0da0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0db0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0dc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0dd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0de0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0df0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
// #include <ATen/ATen.h> #include <cuda.h> #include <cuda_runtime.h> #include <cstdio> #include <cmath> #include <iostream> namespace { template <typename scalar_t> __device__ __forceinline__ void single_mul( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { scalar_t uavc = x_re * (y_re + y_im); *out_re += uavc - (x_re + x_im) * y_im; *out_im += (x_im - x_re) * y_re + uavc; } template <typename scalar_t> __device__ __forceinline__ void single_add( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { *out_re += x_re + y_re; *out_im += x_im + y_im; } /** Complex multiplication of tensors using shared memory and barrier synchronization. Compute the element wise complex multiplication for each thread in the block and write the result to the shared memory. Then synchronize the threads and in the log based fashion sum up the results for each output pixel through its channels, if they are present in the cache. The stride is the number of threads per block times the I (the two float representation of the complex numbers). */ template <typename scalar_t> __global__ void complex_mul_cuda_kernel( const scalar_t* __restrict__ x, const scalar_t* __restrict__ y, scalar_t* __restrict__ out, const int N, const int F, const int C, const int H, const int W) { // The size of the shared memory cache should be twice the number of threads // per block as we store the real and imaginary part of the result. extern __shared__ float cache[]; // cache for the result of the complex multiplication const int I = 2; // the last dimension for the complex number const int plane_size = H * W; const int channel_size = plane_size * I; const int image_size = C * channel_size; // size of the image from the batch // number of complex values in the input that we iterate through const int nr_values = C * H * W; const int n = blockIdx.x; // current index of an image/input map in the batch const int f = blockIdx.y; // current index of a filter from the filter bank const int block_size = blockDim.x; const int thread_nr = threadIdx.x; // stride for the H*W map is equal to the number of threads declared in a block const int stride = block_size * I; // we need H*W threads per plane, each deals with I numbers const int n_idx = n * image_size; // start index in the batch for this input map const int f_idx = f * image_size; // start index in the bank for this filter // find index for the output const int no_idx = n * (F * channel_size); // output index for the batch data point const int fo_idx = f * channel_size; // output index for the filter/channel // Each H*W plane contains H*W*I elements in depth. // We linearize it and start from 0, move by #threads*I steps in outer loop. const int start_idx = threadIdx.x*I; // index in the input map int N_idx = n_idx + start_idx; // index across the first channel plane (in the input map n). const int last_N_idx = n_idx + image_size; // last index for the starting position to compute the sum through each channel for this pixel // To prevent us from a deadlock, we have to always execute __syncthreads(); // for all the threads in the block. Each thread has to do the same number of // iterations for any loop. To ensure that, we keep all threads running, // even though, some of them are really idle. We keep the loop running to // the multiple of the block size that is greater than the number of values // in the input map in total: C*H*W - this is a number of complex cells in the // input map. const int num_blocks = (nr_values + block_size - 1) / block_size; const int last_block_idx = n_idx + num_blocks * block_size * I; // index in the filter int F_idx = f_idx + start_idx; // index in the output, we compute cells on a flat plane (no channels) int base_O_idx = no_idx + fo_idx; int run_O_idx = (start_idx % channel_size); int thread_cidx = thread_nr * I; printf("N_idx:%d, last_block_idx:%d, last_N_idx:%d\n", N_idx, last_block_idx, last_N_idx); while (N_idx < last_block_idx) { // Zero out caches. cache[thread_cidx] = 0; cache[thread_cidx + 1] = 0; if (N_idx < last_N_idx - 1) { scalar_t out_re = 0; scalar_t out_im = 0; scalar_t x_re = x[N_idx]; scalar_t x_im = x[N_idx + 1]; scalar_t y_re = y[F_idx]; scalar_t y_im = y[F_idx + 1]; single_mul(x_re, x_im, y_re, y_im, &out_re, &out_im); cache[thread_cidx] = out_re; cache[thread_cidx + 1] = out_im; } __syncthreads(); // Make the results visible to all threads. // It is not O(logN) but O(N) as of now. For each element in the output // map we have a dedicated thread. The thread goes through all the // channels present in the cache. if (thread_nr < plane_size) { for (int cache_index = thread_nr + plane_size; cache_index < block_size; cache_index += plane_size) { cache[thread_cidx] += cache[cache_index*I]; cache[thread_cidx + 1] += cache[cache_index*I + 1]; } // Move the summed values (across the channels) for each pixel to // the output. const int O_idx = base_O_idx + run_O_idx; out[O_idx] += cache[thread_cidx]; out[O_idx + 1] = cache[thread_cidx + 1]; } N_idx += stride; F_idx += stride; run_O_idx = (run_O_idx + stride) % channel_size; // Make sure that all cache cells are zeroed out before moving on. // We need this as in the second part we access cache cells that do not // belong only to this thread. __syncthreads(); } } } // namespace //void complex_mul_stride_no_permute_cuda( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // AT_DISPATCH_FLOATING_TYPES(x.type(), "complex_mul_cuda", // ([&] { // complex_mul_cuda_kernel<scalar_t><<<blocks, threads>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); // })); //} //template <typename scalar_t> //void complex_mul_stride_no_permute_cuda_pure( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // // Run kernel on the GPU // complex_mul_cuda_kernel<scalar_t><<<blocks, 1024>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); //} /** Uncomment the pytorch related stuff. Compile: ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvcc complex_mul_kernel_stride_no_permute.cu -o complex_mul_profile.out ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvprof ./complex_mul_profile.out nvidia /usr/local/cuda/bin/nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m -c complex_mul_kernel.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m complex_mul_kernel_stride_no_permute.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 Segmentation fault */ int main(void) { int N = 1; int F = 1; int C = 4; int H = 3; int W = 2; int I = 2; int size_input = N * C * H * W * I; int size_filter = F * C * H * W * I; int size_output = N * F * H * W * I; int cuda_block_threads = 16; // auto dims = {128, 32, 16, 8, 2}; // at::Tensor x = at::randn({128, 32, 16, 8, 2}); // at::Tensor y = at::randn({128, 32, 16, 8, 2}); // at::Tensor out = at::zeros({128, 32, 16, 8, 2}); float *x, *y, * out; // Allocate unified memory - accessible from cpu or gpu cudaMallocManaged(&x, size_input*sizeof(float)); cudaMallocManaged(&y, size_filter*sizeof(float)); cudaMallocManaged(&out, size_output*sizeof(float)); for (int j=0; j<H; ++j) { for (int i=0; i<W; ++i) { x[(j*W+i)*2] = 3; x[(j*W+i)*2 + 1] = 1; y[(j*W+i)*2] = 4; y[(j*W+i)*2 + 1] = 2; } } for (int i=0; i<H*W*2; i+=2) { printf("%p %d: %f, %f, %f, %f\n", x, i, x[i], x[i+1], y[i], y[i+1]); } // float *dz; // device z // cudaMalloc(&dz, 9*sizeof(float)); // cudaMemcpy(dz, hz, 9*sizeof(float), cudaMemcpyHostToDevice); const dim3 blocks(N, F); complex_mul_cuda_kernel<float><<<blocks, cuda_block_threads, cuda_block_threads*2>>>(x, y, out, N, F, C, H, W); for (int i=0; i<H*W*C; i+=2) { printf("%d: %f, %f\n", i, out[i], out[i+1]); } cudaFree(x); cudaFree(y); cudaFree(out); // cudaFree(dz); printf("finished computation\n"); return 0; }
.file "tmpxft_00187060_00000000-6_complex_mul_shared_mem.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .section .text._ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,"axG",@progbits,_ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,comdat .weak _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .hidden _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .type _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, @function _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii: .LFB4002: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax movl %ecx, 4(%rsp) movl %r8d, 8(%rsp) movl %r9d, 12(%rsp) movl 208(%rsp), %eax movl %eax, 16(%rsp) movl 216(%rsp), %eax movl %eax, 20(%rsp) movq %rdi, 24(%rsp) leaq 24(%rsp), %rax movq %rax, 112(%rsp) movq %rsi, 32(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) movq %rdx, 40(%rsp) leaq 40(%rsp), %rax movq %rax, 128(%rsp) leaq 4(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%rsp) leaq 16(%rsp), %rax movq %rax, 160(%rsp) leaq 20(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L5 .L1: movq 184(%rsp), %rax subq %fs:40, %rax jne .L6 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L5: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L1 .L6: call __stack_chk_fail@PLT .cfi_endproc .LFE4002: .size _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, .-_ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .text .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB3676: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3676: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC4: .string "%p %d: %f, %f, %f, %f\n" .LC5: .string "%d: %f, %f\n" .LC6: .string "finished computation\n" .text .globl main .type main, @function main: .LFB3673: .cfi_startproc endbr64 pushq %r12 .cfi_def_cfa_offset 16 .cfi_offset 12, -16 pushq %rbp .cfi_def_cfa_offset 24 .cfi_offset 6, -24 pushq %rbx .cfi_def_cfa_offset 32 .cfi_offset 3, -32 subq $64, %rsp .cfi_def_cfa_offset 96 movq %fs:40, %rax movq %rax, 56(%rsp) xorl %eax, %eax leaq 8(%rsp), %rdi movl $1, %edx movl $192, %esi call cudaMallocManaged@PLT leaq 16(%rsp), %rdi movl $1, %edx movl $192, %esi call cudaMallocManaged@PLT leaq 24(%rsp), %rdi movl $1, %edx movl $48, %esi call cudaMallocManaged@PLT movl $16, %edx movss .LC0(%rip), %xmm3 movss .LC1(%rip), %xmm2 movss .LC2(%rip), %xmm1 movss .LC3(%rip), %xmm0 jmp .L10 .L21: addq $16, %rdx cmpq $64, %rdx je .L16 .L10: leaq -16(%rdx), %rax .L11: movq 8(%rsp), %rcx movss %xmm3, (%rcx,%rax) movq 8(%rsp), %rcx movss %xmm2, 4(%rcx,%rax) movq 16(%rsp), %rcx movss %xmm1, (%rcx,%rax) movq 16(%rsp), %rcx movss %xmm0, 4(%rcx,%rax) addq $8, %rax cmpq %rdx, %rax jne .L11 jmp .L21 .L16: movl $4, %ebx movl $0, %ebp leaq .LC4(%rip), %r12 .L12: movq 16(%rsp), %rax movq 8(%rsp), %rdx pxor %xmm0, %xmm0 cvtss2sd -4(%rdx,%rbx), %xmm0 pxor %xmm3, %xmm3 cvtss2sd (%rax,%rbx), %xmm3 pxor %xmm2, %xmm2 cvtss2sd -4(%rax,%rbx), %xmm2 pxor %xmm1, %xmm1 cvtss2sd (%rdx,%rbx), %xmm1 movl %ebp, %ecx movq %r12, %rsi movl $2, %edi movl $4, %eax call __printf_chk@PLT addl $2, %ebp addq $8, %rbx cmpl $12, %ebp jne .L12 movl $1, 32(%rsp) movl $1, 36(%rsp) movl $1, 40(%rsp) movl $16, 44(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $0, %r9d movl $32, %r8d movq 44(%rsp), %rdx movl $1, %ecx movq 32(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L22 .L13: movl $4, %ebp movl $0, %ebx leaq .LC5(%rip), %r12 .L14: movq 24(%rsp), %rax pxor %xmm0, %xmm0 cvtss2sd -4(%rax,%rbp), %xmm0 pxor %xmm1, %xmm1 cvtss2sd (%rax,%rbp), %xmm1 movl %ebx, %edx movq %r12, %rsi movl $2, %edi movl $2, %eax call __printf_chk@PLT addl $2, %ebx addq $8, %rbp cmpl $24, %ebx jne .L14 movq 8(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call cudaFree@PLT movq 24(%rsp), %rdi call cudaFree@PLT leaq .LC6(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 56(%rsp), %rax subq %fs:40, %rax jne .L23 movl $0, %eax addq $64, %rsp .cfi_remember_state .cfi_def_cfa_offset 32 popq %rbx .cfi_def_cfa_offset 24 popq %rbp .cfi_def_cfa_offset 16 popq %r12 .cfi_def_cfa_offset 8 ret .L22: .cfi_restore_state pushq $2 .cfi_def_cfa_offset 104 pushq $3 .cfi_def_cfa_offset 112 movl $4, %r9d movl $1, %r8d movl $1, %ecx movq 40(%rsp), %rdx movq 32(%rsp), %rsi movq 24(%rsp), %rdi call _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii addq $16, %rsp .cfi_def_cfa_offset 96 jmp .L13 .L23: call __stack_chk_fail@PLT .cfi_endproc .LFE3673: .size main, .-main .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC7: .string "_ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB3701: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC7(%rip), %rdx movq %rdx, %rcx leaq _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3701: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .section .rodata.cst4,"aM",@progbits,4 .align 4 .LC0: .long 1077936128 .align 4 .LC1: .long 1065353216 .align 4 .LC2: .long 1082130432 .align 4 .LC3: .long 1073741824 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
// #include <ATen/ATen.h> #include <cuda.h> #include <cuda_runtime.h> #include <cstdio> #include <cmath> #include <iostream> namespace { template <typename scalar_t> __device__ __forceinline__ void single_mul( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { scalar_t uavc = x_re * (y_re + y_im); *out_re += uavc - (x_re + x_im) * y_im; *out_im += (x_im - x_re) * y_re + uavc; } template <typename scalar_t> __device__ __forceinline__ void single_add( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { *out_re += x_re + y_re; *out_im += x_im + y_im; } /** Complex multiplication of tensors using shared memory and barrier synchronization. Compute the element wise complex multiplication for each thread in the block and write the result to the shared memory. Then synchronize the threads and in the log based fashion sum up the results for each output pixel through its channels, if they are present in the cache. The stride is the number of threads per block times the I (the two float representation of the complex numbers). */ template <typename scalar_t> __global__ void complex_mul_cuda_kernel( const scalar_t* __restrict__ x, const scalar_t* __restrict__ y, scalar_t* __restrict__ out, const int N, const int F, const int C, const int H, const int W) { // The size of the shared memory cache should be twice the number of threads // per block as we store the real and imaginary part of the result. extern __shared__ float cache[]; // cache for the result of the complex multiplication const int I = 2; // the last dimension for the complex number const int plane_size = H * W; const int channel_size = plane_size * I; const int image_size = C * channel_size; // size of the image from the batch // number of complex values in the input that we iterate through const int nr_values = C * H * W; const int n = blockIdx.x; // current index of an image/input map in the batch const int f = blockIdx.y; // current index of a filter from the filter bank const int block_size = blockDim.x; const int thread_nr = threadIdx.x; // stride for the H*W map is equal to the number of threads declared in a block const int stride = block_size * I; // we need H*W threads per plane, each deals with I numbers const int n_idx = n * image_size; // start index in the batch for this input map const int f_idx = f * image_size; // start index in the bank for this filter // find index for the output const int no_idx = n * (F * channel_size); // output index for the batch data point const int fo_idx = f * channel_size; // output index for the filter/channel // Each H*W plane contains H*W*I elements in depth. // We linearize it and start from 0, move by #threads*I steps in outer loop. const int start_idx = threadIdx.x*I; // index in the input map int N_idx = n_idx + start_idx; // index across the first channel plane (in the input map n). const int last_N_idx = n_idx + image_size; // last index for the starting position to compute the sum through each channel for this pixel // To prevent us from a deadlock, we have to always execute __syncthreads(); // for all the threads in the block. Each thread has to do the same number of // iterations for any loop. To ensure that, we keep all threads running, // even though, some of them are really idle. We keep the loop running to // the multiple of the block size that is greater than the number of values // in the input map in total: C*H*W - this is a number of complex cells in the // input map. const int num_blocks = (nr_values + block_size - 1) / block_size; const int last_block_idx = n_idx + num_blocks * block_size * I; // index in the filter int F_idx = f_idx + start_idx; // index in the output, we compute cells on a flat plane (no channels) int base_O_idx = no_idx + fo_idx; int run_O_idx = (start_idx % channel_size); int thread_cidx = thread_nr * I; printf("N_idx:%d, last_block_idx:%d, last_N_idx:%d\n", N_idx, last_block_idx, last_N_idx); while (N_idx < last_block_idx) { // Zero out caches. cache[thread_cidx] = 0; cache[thread_cidx + 1] = 0; if (N_idx < last_N_idx - 1) { scalar_t out_re = 0; scalar_t out_im = 0; scalar_t x_re = x[N_idx]; scalar_t x_im = x[N_idx + 1]; scalar_t y_re = y[F_idx]; scalar_t y_im = y[F_idx + 1]; single_mul(x_re, x_im, y_re, y_im, &out_re, &out_im); cache[thread_cidx] = out_re; cache[thread_cidx + 1] = out_im; } __syncthreads(); // Make the results visible to all threads. // It is not O(logN) but O(N) as of now. For each element in the output // map we have a dedicated thread. The thread goes through all the // channels present in the cache. if (thread_nr < plane_size) { for (int cache_index = thread_nr + plane_size; cache_index < block_size; cache_index += plane_size) { cache[thread_cidx] += cache[cache_index*I]; cache[thread_cidx + 1] += cache[cache_index*I + 1]; } // Move the summed values (across the channels) for each pixel to // the output. const int O_idx = base_O_idx + run_O_idx; out[O_idx] += cache[thread_cidx]; out[O_idx + 1] = cache[thread_cidx + 1]; } N_idx += stride; F_idx += stride; run_O_idx = (run_O_idx + stride) % channel_size; // Make sure that all cache cells are zeroed out before moving on. // We need this as in the second part we access cache cells that do not // belong only to this thread. __syncthreads(); } } } // namespace //void complex_mul_stride_no_permute_cuda( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // AT_DISPATCH_FLOATING_TYPES(x.type(), "complex_mul_cuda", // ([&] { // complex_mul_cuda_kernel<scalar_t><<<blocks, threads>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); // })); //} //template <typename scalar_t> //void complex_mul_stride_no_permute_cuda_pure( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // // Run kernel on the GPU // complex_mul_cuda_kernel<scalar_t><<<blocks, 1024>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); //} /** Uncomment the pytorch related stuff. Compile: ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvcc complex_mul_kernel_stride_no_permute.cu -o complex_mul_profile.out ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvprof ./complex_mul_profile.out nvidia /usr/local/cuda/bin/nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m -c complex_mul_kernel.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m complex_mul_kernel_stride_no_permute.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 Segmentation fault */ int main(void) { int N = 1; int F = 1; int C = 4; int H = 3; int W = 2; int I = 2; int size_input = N * C * H * W * I; int size_filter = F * C * H * W * I; int size_output = N * F * H * W * I; int cuda_block_threads = 16; // auto dims = {128, 32, 16, 8, 2}; // at::Tensor x = at::randn({128, 32, 16, 8, 2}); // at::Tensor y = at::randn({128, 32, 16, 8, 2}); // at::Tensor out = at::zeros({128, 32, 16, 8, 2}); float *x, *y, * out; // Allocate unified memory - accessible from cpu or gpu cudaMallocManaged(&x, size_input*sizeof(float)); cudaMallocManaged(&y, size_filter*sizeof(float)); cudaMallocManaged(&out, size_output*sizeof(float)); for (int j=0; j<H; ++j) { for (int i=0; i<W; ++i) { x[(j*W+i)*2] = 3; x[(j*W+i)*2 + 1] = 1; y[(j*W+i)*2] = 4; y[(j*W+i)*2 + 1] = 2; } } for (int i=0; i<H*W*2; i+=2) { printf("%p %d: %f, %f, %f, %f\n", x, i, x[i], x[i+1], y[i], y[i+1]); } // float *dz; // device z // cudaMalloc(&dz, 9*sizeof(float)); // cudaMemcpy(dz, hz, 9*sizeof(float), cudaMemcpyHostToDevice); const dim3 blocks(N, F); complex_mul_cuda_kernel<float><<<blocks, cuda_block_threads, cuda_block_threads*2>>>(x, y, out, N, F, C, H, W); for (int i=0; i<H*W*C; i+=2) { printf("%d: %f, %f\n", i, out[i], out[i+1]); } cudaFree(x); cudaFree(y); cudaFree(out); // cudaFree(dz); printf("finished computation\n"); return 0; }
// #include <ATen/ATen.h> #include <hip/hip_runtime.h> #include <cstdio> #include <cmath> #include <iostream> namespace { template <typename scalar_t> __device__ __forceinline__ void single_mul( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { scalar_t uavc = x_re * (y_re + y_im); *out_re += uavc - (x_re + x_im) * y_im; *out_im += (x_im - x_re) * y_re + uavc; } template <typename scalar_t> __device__ __forceinline__ void single_add( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { *out_re += x_re + y_re; *out_im += x_im + y_im; } /** Complex multiplication of tensors using shared memory and barrier synchronization. Compute the element wise complex multiplication for each thread in the block and write the result to the shared memory. Then synchronize the threads and in the log based fashion sum up the results for each output pixel through its channels, if they are present in the cache. The stride is the number of threads per block times the I (the two float representation of the complex numbers). */ template <typename scalar_t> __global__ void complex_mul_cuda_kernel( const scalar_t* __restrict__ x, const scalar_t* __restrict__ y, scalar_t* __restrict__ out, const int N, const int F, const int C, const int H, const int W) { // The size of the shared memory cache should be twice the number of threads // per block as we store the real and imaginary part of the result. extern __shared__ float cache[]; // cache for the result of the complex multiplication const int I = 2; // the last dimension for the complex number const int plane_size = H * W; const int channel_size = plane_size * I; const int image_size = C * channel_size; // size of the image from the batch // number of complex values in the input that we iterate through const int nr_values = C * H * W; const int n = blockIdx.x; // current index of an image/input map in the batch const int f = blockIdx.y; // current index of a filter from the filter bank const int block_size = blockDim.x; const int thread_nr = threadIdx.x; // stride for the H*W map is equal to the number of threads declared in a block const int stride = block_size * I; // we need H*W threads per plane, each deals with I numbers const int n_idx = n * image_size; // start index in the batch for this input map const int f_idx = f * image_size; // start index in the bank for this filter // find index for the output const int no_idx = n * (F * channel_size); // output index for the batch data point const int fo_idx = f * channel_size; // output index for the filter/channel // Each H*W plane contains H*W*I elements in depth. // We linearize it and start from 0, move by #threads*I steps in outer loop. const int start_idx = threadIdx.x*I; // index in the input map int N_idx = n_idx + start_idx; // index across the first channel plane (in the input map n). const int last_N_idx = n_idx + image_size; // last index for the starting position to compute the sum through each channel for this pixel // To prevent us from a deadlock, we have to always execute __syncthreads(); // for all the threads in the block. Each thread has to do the same number of // iterations for any loop. To ensure that, we keep all threads running, // even though, some of them are really idle. We keep the loop running to // the multiple of the block size that is greater than the number of values // in the input map in total: C*H*W - this is a number of complex cells in the // input map. const int num_blocks = (nr_values + block_size - 1) / block_size; const int last_block_idx = n_idx + num_blocks * block_size * I; // index in the filter int F_idx = f_idx + start_idx; // index in the output, we compute cells on a flat plane (no channels) int base_O_idx = no_idx + fo_idx; int run_O_idx = (start_idx % channel_size); int thread_cidx = thread_nr * I; printf("N_idx:%d, last_block_idx:%d, last_N_idx:%d\n", N_idx, last_block_idx, last_N_idx); while (N_idx < last_block_idx) { // Zero out caches. cache[thread_cidx] = 0; cache[thread_cidx + 1] = 0; if (N_idx < last_N_idx - 1) { scalar_t out_re = 0; scalar_t out_im = 0; scalar_t x_re = x[N_idx]; scalar_t x_im = x[N_idx + 1]; scalar_t y_re = y[F_idx]; scalar_t y_im = y[F_idx + 1]; single_mul(x_re, x_im, y_re, y_im, &out_re, &out_im); cache[thread_cidx] = out_re; cache[thread_cidx + 1] = out_im; } __syncthreads(); // Make the results visible to all threads. // It is not O(logN) but O(N) as of now. For each element in the output // map we have a dedicated thread. The thread goes through all the // channels present in the cache. if (thread_nr < plane_size) { for (int cache_index = thread_nr + plane_size; cache_index < block_size; cache_index += plane_size) { cache[thread_cidx] += cache[cache_index*I]; cache[thread_cidx + 1] += cache[cache_index*I + 1]; } // Move the summed values (across the channels) for each pixel to // the output. const int O_idx = base_O_idx + run_O_idx; out[O_idx] += cache[thread_cidx]; out[O_idx + 1] = cache[thread_cidx + 1]; } N_idx += stride; F_idx += stride; run_O_idx = (run_O_idx + stride) % channel_size; // Make sure that all cache cells are zeroed out before moving on. // We need this as in the second part we access cache cells that do not // belong only to this thread. __syncthreads(); } } } // namespace //void complex_mul_stride_no_permute_cuda( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // AT_DISPATCH_FLOATING_TYPES(x.type(), "complex_mul_cuda", // ([&] { // complex_mul_cuda_kernel<scalar_t><<<blocks, threads>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); // })); //} //template <typename scalar_t> //void complex_mul_stride_no_permute_cuda_pure( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // // Run kernel on the GPU // complex_mul_cuda_kernel<scalar_t><<<blocks, 1024>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); //} /** Uncomment the pytorch related stuff. Compile: ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvcc complex_mul_kernel_stride_no_permute.cu -o complex_mul_profile.out ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvprof ./complex_mul_profile.out nvidia /usr/local/cuda/bin/nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m -c complex_mul_kernel.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m complex_mul_kernel_stride_no_permute.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 Segmentation fault */ int main(void) { int N = 1; int F = 1; int C = 4; int H = 3; int W = 2; int I = 2; int size_input = N * C * H * W * I; int size_filter = F * C * H * W * I; int size_output = N * F * H * W * I; int cuda_block_threads = 16; // auto dims = {128, 32, 16, 8, 2}; // at::Tensor x = at::randn({128, 32, 16, 8, 2}); // at::Tensor y = at::randn({128, 32, 16, 8, 2}); // at::Tensor out = at::zeros({128, 32, 16, 8, 2}); float *x, *y, * out; // Allocate unified memory - accessible from cpu or gpu hipMallocManaged(&x, size_input*sizeof(float)); hipMallocManaged(&y, size_filter*sizeof(float)); hipMallocManaged(&out, size_output*sizeof(float)); for (int j=0; j<H; ++j) { for (int i=0; i<W; ++i) { x[(j*W+i)*2] = 3; x[(j*W+i)*2 + 1] = 1; y[(j*W+i)*2] = 4; y[(j*W+i)*2 + 1] = 2; } } for (int i=0; i<H*W*2; i+=2) { printf("%p %d: %f, %f, %f, %f\n", x, i, x[i], x[i+1], y[i], y[i+1]); } // float *dz; // device z // cudaMalloc(&dz, 9*sizeof(float)); // cudaMemcpy(dz, hz, 9*sizeof(float), cudaMemcpyHostToDevice); const dim3 blocks(N, F); complex_mul_cuda_kernel<float><<<blocks, cuda_block_threads, cuda_block_threads*2>>>(x, y, out, N, F, C, H, W); for (int i=0; i<H*W*C; i+=2) { printf("%d: %f, %f\n", i, out[i], out[i+1]); } hipFree(x); hipFree(y); hipFree(out); // cudaFree(dz); printf("finished computation\n"); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
// #include <ATen/ATen.h> #include <hip/hip_runtime.h> #include <cstdio> #include <cmath> #include <iostream> namespace { template <typename scalar_t> __device__ __forceinline__ void single_mul( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { scalar_t uavc = x_re * (y_re + y_im); *out_re += uavc - (x_re + x_im) * y_im; *out_im += (x_im - x_re) * y_re + uavc; } template <typename scalar_t> __device__ __forceinline__ void single_add( scalar_t x_re, scalar_t x_im, scalar_t y_re, scalar_t y_im, scalar_t* out_re, scalar_t* out_im) { *out_re += x_re + y_re; *out_im += x_im + y_im; } /** Complex multiplication of tensors using shared memory and barrier synchronization. Compute the element wise complex multiplication for each thread in the block and write the result to the shared memory. Then synchronize the threads and in the log based fashion sum up the results for each output pixel through its channels, if they are present in the cache. The stride is the number of threads per block times the I (the two float representation of the complex numbers). */ template <typename scalar_t> __global__ void complex_mul_cuda_kernel( const scalar_t* __restrict__ x, const scalar_t* __restrict__ y, scalar_t* __restrict__ out, const int N, const int F, const int C, const int H, const int W) { // The size of the shared memory cache should be twice the number of threads // per block as we store the real and imaginary part of the result. extern __shared__ float cache[]; // cache for the result of the complex multiplication const int I = 2; // the last dimension for the complex number const int plane_size = H * W; const int channel_size = plane_size * I; const int image_size = C * channel_size; // size of the image from the batch // number of complex values in the input that we iterate through const int nr_values = C * H * W; const int n = blockIdx.x; // current index of an image/input map in the batch const int f = blockIdx.y; // current index of a filter from the filter bank const int block_size = blockDim.x; const int thread_nr = threadIdx.x; // stride for the H*W map is equal to the number of threads declared in a block const int stride = block_size * I; // we need H*W threads per plane, each deals with I numbers const int n_idx = n * image_size; // start index in the batch for this input map const int f_idx = f * image_size; // start index in the bank for this filter // find index for the output const int no_idx = n * (F * channel_size); // output index for the batch data point const int fo_idx = f * channel_size; // output index for the filter/channel // Each H*W plane contains H*W*I elements in depth. // We linearize it and start from 0, move by #threads*I steps in outer loop. const int start_idx = threadIdx.x*I; // index in the input map int N_idx = n_idx + start_idx; // index across the first channel plane (in the input map n). const int last_N_idx = n_idx + image_size; // last index for the starting position to compute the sum through each channel for this pixel // To prevent us from a deadlock, we have to always execute __syncthreads(); // for all the threads in the block. Each thread has to do the same number of // iterations for any loop. To ensure that, we keep all threads running, // even though, some of them are really idle. We keep the loop running to // the multiple of the block size that is greater than the number of values // in the input map in total: C*H*W - this is a number of complex cells in the // input map. const int num_blocks = (nr_values + block_size - 1) / block_size; const int last_block_idx = n_idx + num_blocks * block_size * I; // index in the filter int F_idx = f_idx + start_idx; // index in the output, we compute cells on a flat plane (no channels) int base_O_idx = no_idx + fo_idx; int run_O_idx = (start_idx % channel_size); int thread_cidx = thread_nr * I; printf("N_idx:%d, last_block_idx:%d, last_N_idx:%d\n", N_idx, last_block_idx, last_N_idx); while (N_idx < last_block_idx) { // Zero out caches. cache[thread_cidx] = 0; cache[thread_cidx + 1] = 0; if (N_idx < last_N_idx - 1) { scalar_t out_re = 0; scalar_t out_im = 0; scalar_t x_re = x[N_idx]; scalar_t x_im = x[N_idx + 1]; scalar_t y_re = y[F_idx]; scalar_t y_im = y[F_idx + 1]; single_mul(x_re, x_im, y_re, y_im, &out_re, &out_im); cache[thread_cidx] = out_re; cache[thread_cidx + 1] = out_im; } __syncthreads(); // Make the results visible to all threads. // It is not O(logN) but O(N) as of now. For each element in the output // map we have a dedicated thread. The thread goes through all the // channels present in the cache. if (thread_nr < plane_size) { for (int cache_index = thread_nr + plane_size; cache_index < block_size; cache_index += plane_size) { cache[thread_cidx] += cache[cache_index*I]; cache[thread_cidx + 1] += cache[cache_index*I + 1]; } // Move the summed values (across the channels) for each pixel to // the output. const int O_idx = base_O_idx + run_O_idx; out[O_idx] += cache[thread_cidx]; out[O_idx + 1] = cache[thread_cidx + 1]; } N_idx += stride; F_idx += stride; run_O_idx = (run_O_idx + stride) % channel_size; // Make sure that all cache cells are zeroed out before moving on. // We need this as in the second part we access cache cells that do not // belong only to this thread. __syncthreads(); } } } // namespace //void complex_mul_stride_no_permute_cuda( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // AT_DISPATCH_FLOATING_TYPES(x.type(), "complex_mul_cuda", // ([&] { // complex_mul_cuda_kernel<scalar_t><<<blocks, threads>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); // })); //} //template <typename scalar_t> //void complex_mul_stride_no_permute_cuda_pure( // at::Tensor x, // at::Tensor y, // at::Tensor out, // int threads = 1024) { // // const auto N = x.size(0); // batch_size // const auto F = y.size(0); // filter_bank_size // const auto C = x.size(1); // number of channels // const auto H = x.size(2); // height of the matrix // const auto W = x.size(3); // width of the matrix // // const auto x_blocks = N; // const auto y_blocks = F; // const dim3 blocks(x_blocks, y_blocks); // // // Run kernel on the GPU // complex_mul_cuda_kernel<scalar_t><<<blocks, 1024>>>( // x.data<scalar_t>(), y.data<scalar_t>(), out.data<scalar_t>(), // N, F, C, H, W); //} /** Uncomment the pytorch related stuff. Compile: ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvcc complex_mul_kernel_stride_no_permute.cu -o complex_mul_profile.out ady@skr-compute1:/tmp/pycharm_project_154/cnns/nnlib/pytorch_cuda/complex_mul_cuda$ nvprof ./complex_mul_profile.out nvidia /usr/local/cuda/bin/nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m -c complex_mul_kernel.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 nvcc -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/torch/csrc/api/include -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/TH -I/local/ady/anaconda3/lib/python3.6/site-packages/torch/lib/include/THC -I/usr/local/cuda/include -I/local/ady/anaconda3/include/python3.6m complex_mul_kernel_stride_no_permute.cu -o complex_mul_kernel_stride_no_permute.out -std=c++11 Segmentation fault */ int main(void) { int N = 1; int F = 1; int C = 4; int H = 3; int W = 2; int I = 2; int size_input = N * C * H * W * I; int size_filter = F * C * H * W * I; int size_output = N * F * H * W * I; int cuda_block_threads = 16; // auto dims = {128, 32, 16, 8, 2}; // at::Tensor x = at::randn({128, 32, 16, 8, 2}); // at::Tensor y = at::randn({128, 32, 16, 8, 2}); // at::Tensor out = at::zeros({128, 32, 16, 8, 2}); float *x, *y, * out; // Allocate unified memory - accessible from cpu or gpu hipMallocManaged(&x, size_input*sizeof(float)); hipMallocManaged(&y, size_filter*sizeof(float)); hipMallocManaged(&out, size_output*sizeof(float)); for (int j=0; j<H; ++j) { for (int i=0; i<W; ++i) { x[(j*W+i)*2] = 3; x[(j*W+i)*2 + 1] = 1; y[(j*W+i)*2] = 4; y[(j*W+i)*2 + 1] = 2; } } for (int i=0; i<H*W*2; i+=2) { printf("%p %d: %f, %f, %f, %f\n", x, i, x[i], x[i+1], y[i], y[i+1]); } // float *dz; // device z // cudaMalloc(&dz, 9*sizeof(float)); // cudaMemcpy(dz, hz, 9*sizeof(float), cudaMemcpyHostToDevice); const dim3 blocks(N, F); complex_mul_cuda_kernel<float><<<blocks, cuda_block_threads, cuda_block_threads*2>>>(x, y, out, N, F, C, H, W); for (int i=0; i<H*W*C; i+=2) { printf("%d: %f, %f\n", i, out[i], out[i+1]); } hipFree(x); hipFree(y); hipFree(out); // cudaFree(dz); printf("finished computation\n"); return 0; }
.text .file "complex_mul_shared_mem.hip" # Start of file scope inline assembly .globl _ZSt21ios_base_library_initv # End of file scope inline assembly .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbx .cfi_def_cfa_offset 16 subq $192, %rsp .cfi_def_cfa_offset 208 .cfi_offset %rbx, -16 leaq 16(%rsp), %rdi movl $192, %esi movl $1, %edx callq hipMallocManaged leaq 8(%rsp), %rdi movl $192, %esi movl $1, %edx callq hipMallocManaged leaq 24(%rsp), %rdi movl $48, %esi movl $1, %edx callq hipMallocManaged movq 16(%rsp), %rax movq 8(%rsp), %rcx addq $4, %rcx addq $4, %rax xorl %edx, %edx movabsq $4575657222486360064, %rsi # imm = 0x3F80000040400000 movabsq $4611686019509518336, %rdi # imm = 0x4000000040800000 .p2align 4, 0x90 .LBB0_1: # %.preheader71 # =>This Loop Header: Depth=1 # Child Loop BB0_2 Depth 2 xorl %r8d, %r8d .p2align 4, 0x90 .LBB0_2: # Parent Loop BB0_1 Depth=1 # => This Inner Loop Header: Depth=2 movq %rsi, -4(%rax,%r8,8) movq %rdi, -4(%rcx,%r8,8) incq %r8 cmpq $1, %r8 je .LBB0_2 # %bb.3: # in Loop: Header=BB0_1 Depth=1 incq %rdx addq $16, %rcx addq $16, %rax cmpq $3, %rdx jne .LBB0_1 # %bb.4: # %.preheader.preheader xorl %ebx, %ebx .p2align 4, 0x90 .LBB0_5: # %.preheader # =>This Inner Loop Header: Depth=1 movq 16(%rsp), %rsi movss (%rsi,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss 4(%rsi,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 movq 8(%rsp), %rax movss (%rax,%rbx,4), %xmm2 # xmm2 = mem[0],zero,zero,zero movss 4(%rax,%rbx,4), %xmm3 # xmm3 = mem[0],zero,zero,zero cvtss2sd %xmm2, %xmm2 cvtss2sd %xmm3, %xmm3 movl $.L.str, %edi movl %ebx, %edx movb $4, %al callq printf leaq 2(%rbx), %rax cmpq $10, %rbx movq %rax, %rbx jb .LBB0_5 # %bb.6: movabsq $4294967297, %rdi # imm = 0x100000001 leaq 15(%rdi), %rdx movl $32, %r8d movl $1, %esi movl $1, %ecx xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB0_7 # %bb.10: movq 16(%rsp), %rax movq 8(%rsp), %rcx movq 24(%rsp), %rdx movq %rax, 120(%rsp) movq %rcx, 112(%rsp) movq %rdx, 104(%rsp) movl $1, 52(%rsp) movl $1, 48(%rsp) movl $4, 44(%rsp) movl $3, 40(%rsp) movl $2, 36(%rsp) leaq 120(%rsp), %rax movq %rax, 128(%rsp) leaq 112(%rsp), %rax movq %rax, 136(%rsp) leaq 104(%rsp), %rax movq %rax, 144(%rsp) leaq 52(%rsp), %rax movq %rax, 152(%rsp) leaq 48(%rsp), %rax movq %rax, 160(%rsp) leaq 44(%rsp), %rax movq %rax, 168(%rsp) leaq 40(%rsp), %rax movq %rax, 176(%rsp) leaq 36(%rsp), %rax movq %rax, 184(%rsp) leaq 88(%rsp), %rdi leaq 72(%rsp), %rsi leaq 64(%rsp), %rdx leaq 56(%rsp), %rcx callq __hipPopCallConfiguration movq 88(%rsp), %rsi movl 96(%rsp), %edx movq 72(%rsp), %rcx movl 80(%rsp), %r8d leaq 128(%rsp), %r9 movl $_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, %edi pushq 56(%rsp) .cfi_adjust_cfa_offset 8 pushq 72(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB0_7: # %.preheader86 xorl %ebx, %ebx .p2align 4, 0x90 .LBB0_8: # =>This Inner Loop Header: Depth=1 movq 24(%rsp), %rax movss (%rax,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss 4(%rax,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 movl $.L.str.1, %edi movl %ebx, %esi movb $2, %al callq printf leaq 2(%rbx), %rax cmpq $22, %rbx movq %rax, %rbx jb .LBB0_8 # %bb.9: movq 16(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree movq 24(%rsp), %rdi callq hipFree movl $.Lstr, %edi callq puts@PLT xorl %eax, %eax addq $192, %rsp .cfi_def_cfa_offset 16 popq %rbx .cfi_def_cfa_offset 8 retq .Lfunc_end0: .size main, .Lfunc_end0-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .type _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,@function _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii: # @_ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 20(%rsp), %rax movq %rax, 120(%rsp) leaq 16(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 56(%rsp), %rdi leaq 40(%rsp), %rsi leaq 32(%rsp), %rdx leaq 24(%rsp), %rcx callq __hipPopCallConfiguration movq 56(%rsp), %rsi movl 64(%rsp), %edx movq 40(%rsp), %rcx movl 48(%rsp), %r8d leaq 96(%rsp), %r9 movl $_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, %edi pushq 24(%rsp) .cfi_adjust_cfa_offset 8 pushq 40(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end1: .size _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, .Lfunc_end1-_ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB2_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB2_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end2: .size __hip_module_ctor, .Lfunc_end2-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB3_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB3_2: retq .Lfunc_end3: .size __hip_module_dtor, .Lfunc_end3-__hip_module_dtor .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%p %d: %f, %f, %f, %f\n" .size .L.str, 23 .type _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,@object # @_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .section .rodata,"a",@progbits .p2align 3, 0x0 _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii: .quad _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .size _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, 8 .type .L.str.1,@object # @.str.1 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.1: .asciz "%d: %f, %f\n" .size .L.str.1, 12 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii" .size .L__unnamed_1, 64 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "finished computation" .size .Lstr, 21 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_00187060_00000000-6_complex_mul_shared_mem.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .section .text._ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,"axG",@progbits,_ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,comdat .weak _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .hidden _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .type _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, @function _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii: .LFB4002: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax movl %ecx, 4(%rsp) movl %r8d, 8(%rsp) movl %r9d, 12(%rsp) movl 208(%rsp), %eax movl %eax, 16(%rsp) movl 216(%rsp), %eax movl %eax, 20(%rsp) movq %rdi, 24(%rsp) leaq 24(%rsp), %rax movq %rax, 112(%rsp) movq %rsi, 32(%rsp) leaq 32(%rsp), %rax movq %rax, 120(%rsp) movq %rdx, 40(%rsp) leaq 40(%rsp), %rax movq %rax, 128(%rsp) leaq 4(%rsp), %rax movq %rax, 136(%rsp) leaq 8(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%rsp) leaq 16(%rsp), %rax movq %rax, 160(%rsp) leaq 20(%rsp), %rax movq %rax, 168(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) movl $1, 72(%rsp) movl $1, 76(%rsp) movl $1, 80(%rsp) movl $1, 84(%rsp) leaq 56(%rsp), %rcx leaq 48(%rsp), %rdx leaq 76(%rsp), %rsi leaq 64(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L5 .L1: movq 184(%rsp), %rax subq %fs:40, %rax jne .L6 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L5: .cfi_restore_state pushq 56(%rsp) .cfi_def_cfa_offset 216 pushq 56(%rsp) .cfi_def_cfa_offset 224 leaq 128(%rsp), %r9 movq 92(%rsp), %rcx movl 100(%rsp), %r8d movq 80(%rsp), %rsi movl 88(%rsp), %edx leaq _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L1 .L6: call __stack_chk_fail@PLT .cfi_endproc .LFE4002: .size _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, .-_ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .text .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB3676: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3676: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC4: .string "%p %d: %f, %f, %f, %f\n" .LC5: .string "%d: %f, %f\n" .LC6: .string "finished computation\n" .text .globl main .type main, @function main: .LFB3673: .cfi_startproc endbr64 pushq %r12 .cfi_def_cfa_offset 16 .cfi_offset 12, -16 pushq %rbp .cfi_def_cfa_offset 24 .cfi_offset 6, -24 pushq %rbx .cfi_def_cfa_offset 32 .cfi_offset 3, -32 subq $64, %rsp .cfi_def_cfa_offset 96 movq %fs:40, %rax movq %rax, 56(%rsp) xorl %eax, %eax leaq 8(%rsp), %rdi movl $1, %edx movl $192, %esi call cudaMallocManaged@PLT leaq 16(%rsp), %rdi movl $1, %edx movl $192, %esi call cudaMallocManaged@PLT leaq 24(%rsp), %rdi movl $1, %edx movl $48, %esi call cudaMallocManaged@PLT movl $16, %edx movss .LC0(%rip), %xmm3 movss .LC1(%rip), %xmm2 movss .LC2(%rip), %xmm1 movss .LC3(%rip), %xmm0 jmp .L10 .L21: addq $16, %rdx cmpq $64, %rdx je .L16 .L10: leaq -16(%rdx), %rax .L11: movq 8(%rsp), %rcx movss %xmm3, (%rcx,%rax) movq 8(%rsp), %rcx movss %xmm2, 4(%rcx,%rax) movq 16(%rsp), %rcx movss %xmm1, (%rcx,%rax) movq 16(%rsp), %rcx movss %xmm0, 4(%rcx,%rax) addq $8, %rax cmpq %rdx, %rax jne .L11 jmp .L21 .L16: movl $4, %ebx movl $0, %ebp leaq .LC4(%rip), %r12 .L12: movq 16(%rsp), %rax movq 8(%rsp), %rdx pxor %xmm0, %xmm0 cvtss2sd -4(%rdx,%rbx), %xmm0 pxor %xmm3, %xmm3 cvtss2sd (%rax,%rbx), %xmm3 pxor %xmm2, %xmm2 cvtss2sd -4(%rax,%rbx), %xmm2 pxor %xmm1, %xmm1 cvtss2sd (%rdx,%rbx), %xmm1 movl %ebp, %ecx movq %r12, %rsi movl $2, %edi movl $4, %eax call __printf_chk@PLT addl $2, %ebp addq $8, %rbx cmpl $12, %ebp jne .L12 movl $1, 32(%rsp) movl $1, 36(%rsp) movl $1, 40(%rsp) movl $16, 44(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $0, %r9d movl $32, %r8d movq 44(%rsp), %rdx movl $1, %ecx movq 32(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L22 .L13: movl $4, %ebp movl $0, %ebx leaq .LC5(%rip), %r12 .L14: movq 24(%rsp), %rax pxor %xmm0, %xmm0 cvtss2sd -4(%rax,%rbp), %xmm0 pxor %xmm1, %xmm1 cvtss2sd (%rax,%rbp), %xmm1 movl %ebx, %edx movq %r12, %rsi movl $2, %edi movl $2, %eax call __printf_chk@PLT addl $2, %ebx addq $8, %rbp cmpl $24, %ebx jne .L14 movq 8(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call cudaFree@PLT movq 24(%rsp), %rdi call cudaFree@PLT leaq .LC6(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 56(%rsp), %rax subq %fs:40, %rax jne .L23 movl $0, %eax addq $64, %rsp .cfi_remember_state .cfi_def_cfa_offset 32 popq %rbx .cfi_def_cfa_offset 24 popq %rbp .cfi_def_cfa_offset 16 popq %r12 .cfi_def_cfa_offset 8 ret .L22: .cfi_restore_state pushq $2 .cfi_def_cfa_offset 104 pushq $3 .cfi_def_cfa_offset 112 movl $4, %r9d movl $1, %r8d movl $1, %ecx movq 40(%rsp), %rdx movq 32(%rsp), %rsi movq 24(%rsp), %rdi call _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii addq $16, %rsp .cfi_def_cfa_offset 96 jmp .L13 .L23: call __stack_chk_fail@PLT .cfi_endproc .LFE3673: .size main, .-main .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC7: .string "_ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB3701: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC7(%rip), %rdx movq %rdx, %rcx leaq _ZN54_GLOBAL__N__d33534c5_25_complex_mul_shared_mem_cu_main23complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3701: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .section .rodata.cst4,"aM",@progbits,4 .align 4 .LC0: .long 1077936128 .align 4 .LC1: .long 1065353216 .align 4 .LC2: .long 1082130432 .align 4 .LC3: .long 1073741824 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "complex_mul_shared_mem.hip" # Start of file scope inline assembly .globl _ZSt21ios_base_library_initv # End of file scope inline assembly .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbx .cfi_def_cfa_offset 16 subq $192, %rsp .cfi_def_cfa_offset 208 .cfi_offset %rbx, -16 leaq 16(%rsp), %rdi movl $192, %esi movl $1, %edx callq hipMallocManaged leaq 8(%rsp), %rdi movl $192, %esi movl $1, %edx callq hipMallocManaged leaq 24(%rsp), %rdi movl $48, %esi movl $1, %edx callq hipMallocManaged movq 16(%rsp), %rax movq 8(%rsp), %rcx addq $4, %rcx addq $4, %rax xorl %edx, %edx movabsq $4575657222486360064, %rsi # imm = 0x3F80000040400000 movabsq $4611686019509518336, %rdi # imm = 0x4000000040800000 .p2align 4, 0x90 .LBB0_1: # %.preheader71 # =>This Loop Header: Depth=1 # Child Loop BB0_2 Depth 2 xorl %r8d, %r8d .p2align 4, 0x90 .LBB0_2: # Parent Loop BB0_1 Depth=1 # => This Inner Loop Header: Depth=2 movq %rsi, -4(%rax,%r8,8) movq %rdi, -4(%rcx,%r8,8) incq %r8 cmpq $1, %r8 je .LBB0_2 # %bb.3: # in Loop: Header=BB0_1 Depth=1 incq %rdx addq $16, %rcx addq $16, %rax cmpq $3, %rdx jne .LBB0_1 # %bb.4: # %.preheader.preheader xorl %ebx, %ebx .p2align 4, 0x90 .LBB0_5: # %.preheader # =>This Inner Loop Header: Depth=1 movq 16(%rsp), %rsi movss (%rsi,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss 4(%rsi,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 movq 8(%rsp), %rax movss (%rax,%rbx,4), %xmm2 # xmm2 = mem[0],zero,zero,zero movss 4(%rax,%rbx,4), %xmm3 # xmm3 = mem[0],zero,zero,zero cvtss2sd %xmm2, %xmm2 cvtss2sd %xmm3, %xmm3 movl $.L.str, %edi movl %ebx, %edx movb $4, %al callq printf leaq 2(%rbx), %rax cmpq $10, %rbx movq %rax, %rbx jb .LBB0_5 # %bb.6: movabsq $4294967297, %rdi # imm = 0x100000001 leaq 15(%rdi), %rdx movl $32, %r8d movl $1, %esi movl $1, %ecx xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB0_7 # %bb.10: movq 16(%rsp), %rax movq 8(%rsp), %rcx movq 24(%rsp), %rdx movq %rax, 120(%rsp) movq %rcx, 112(%rsp) movq %rdx, 104(%rsp) movl $1, 52(%rsp) movl $1, 48(%rsp) movl $4, 44(%rsp) movl $3, 40(%rsp) movl $2, 36(%rsp) leaq 120(%rsp), %rax movq %rax, 128(%rsp) leaq 112(%rsp), %rax movq %rax, 136(%rsp) leaq 104(%rsp), %rax movq %rax, 144(%rsp) leaq 52(%rsp), %rax movq %rax, 152(%rsp) leaq 48(%rsp), %rax movq %rax, 160(%rsp) leaq 44(%rsp), %rax movq %rax, 168(%rsp) leaq 40(%rsp), %rax movq %rax, 176(%rsp) leaq 36(%rsp), %rax movq %rax, 184(%rsp) leaq 88(%rsp), %rdi leaq 72(%rsp), %rsi leaq 64(%rsp), %rdx leaq 56(%rsp), %rcx callq __hipPopCallConfiguration movq 88(%rsp), %rsi movl 96(%rsp), %edx movq 72(%rsp), %rcx movl 80(%rsp), %r8d leaq 128(%rsp), %r9 movl $_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, %edi pushq 56(%rsp) .cfi_adjust_cfa_offset 8 pushq 72(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB0_7: # %.preheader86 xorl %ebx, %ebx .p2align 4, 0x90 .LBB0_8: # =>This Inner Loop Header: Depth=1 movq 24(%rsp), %rax movss (%rax,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero movss 4(%rax,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 cvtss2sd %xmm1, %xmm1 movl $.L.str.1, %edi movl %ebx, %esi movb $2, %al callq printf leaq 2(%rbx), %rax cmpq $22, %rbx movq %rax, %rbx jb .LBB0_8 # %bb.9: movq 16(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree movq 24(%rsp), %rdi callq hipFree movl $.Lstr, %edi callq puts@PLT xorl %eax, %eax addq $192, %rsp .cfi_def_cfa_offset 16 popq %rbx .cfi_def_cfa_offset 8 retq .Lfunc_end0: .size main, .Lfunc_end0-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .type _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,@function _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii: # @_ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .cfi_startproc # %bb.0: subq $168, %rsp .cfi_def_cfa_offset 176 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movq %rdx, 72(%rsp) movl %ecx, 20(%rsp) movl %r8d, 16(%rsp) movl %r9d, 12(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 72(%rsp), %rax movq %rax, 112(%rsp) leaq 20(%rsp), %rax movq %rax, 120(%rsp) leaq 16(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%rsp) leaq 176(%rsp), %rax movq %rax, 144(%rsp) leaq 184(%rsp), %rax movq %rax, 152(%rsp) leaq 56(%rsp), %rdi leaq 40(%rsp), %rsi leaq 32(%rsp), %rdx leaq 24(%rsp), %rcx callq __hipPopCallConfiguration movq 56(%rsp), %rsi movl 64(%rsp), %edx movq 40(%rsp), %rcx movl 48(%rsp), %r8d leaq 96(%rsp), %r9 movl $_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, %edi pushq 24(%rsp) .cfi_adjust_cfa_offset 8 pushq 40(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $184, %rsp .cfi_adjust_cfa_offset -184 retq .Lfunc_end1: .size _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, .Lfunc_end1-_ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB2_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB2_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end2: .size __hip_module_ctor, .Lfunc_end2-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB3_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB3_2: retq .Lfunc_end3: .size __hip_module_dtor, .Lfunc_end3-__hip_module_dtor .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%p %d: %f, %f, %f, %f\n" .size .L.str, 23 .type _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii,@object # @_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .section .rodata,"a",@progbits .p2align 3, 0x0 _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii: .quad _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .size _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii, 8 .type .L.str.1,@object # @.str.1 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.1: .asciz "%d: %f, %f\n" .size .L.str.1, 12 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii" .size .L__unnamed_1, 64 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "finished computation" .size .Lstr, 21 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _ZN12_GLOBAL__N_138__device_stub__complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _ZN12_GLOBAL__N_123complex_mul_cuda_kernelIfEEvPKT_S3_PS1_iiiii .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include <stdio.h> #include <stdlib.h> // Matrices are stored in row-major order: // M(row, col) = *(M.elements + row * M.width + col) typedef struct { int width; int height; float *elements; } Matrix; // Thread block size #define BLOCK_SIZE 16 // Forward declaration of the matrix multiplication kernel __global__ void MatMulKernel(const Matrix, const Matrix, Matrix); // Matrix multiplication - Host code // Matrix dimensions are assumed to be multiples of BLOCK_SIZE void MatMul(const Matrix A, const Matrix B, Matrix C) { cudaSetDevice(0); cudaDeviceSynchronize(); size_t available, total; cudaMemGetInfo(&available, &total); // printf("Mem total: %ld Bytes\nMem available: %ld Bytes\n", available, total); // Load A and B to device memory Matrix d_A; d_A.width = A.width; d_A.height = A.height; size_t size = A.width * A.height * sizeof(float); // printf("size of A: %ld\n", size); cudaMalloc(&d_A.elements, size); cudaError_t error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation A %s\n", cudaGetErrorString(error)); exit(-1); } cudaMemcpy(d_A.elements, A.elements, size, cudaMemcpyHostToDevice); Matrix d_B; d_B.width = B.width; d_B.height = B.height; size = B.width * B.height * sizeof(float); cudaMalloc(&d_B.elements, size); error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation B %s\n", cudaGetErrorString(error)); exit(-1); } cudaMemcpy(d_B.elements, B.elements, size, cudaMemcpyHostToDevice); // Allocate C in device memory Matrix d_C; d_C.width = C.width; d_C.height = C.height; size = C.width * C.height * sizeof(float); error = cudaGetLastError(); cudaMalloc(&d_C.elements, size); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation C %s\n", cudaGetErrorString(error)); exit(-1); } // Invoke kernel dim3 dimBlock(BLOCK_SIZE, BLOCK_SIZE); dim3 dimGrid(B.width / dimBlock.x, A.height / dimBlock.y); MatMulKernel<<<dimGrid, dimBlock>>>(d_A, d_B, d_C); cudaDeviceSynchronize(); error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: calculation error %s\n", cudaGetErrorString(error)); exit(-1); } // Read C from device memory cudaMemcpy(C.elements, d_C.elements, size, cudaMemcpyDeviceToHost); if (error != cudaSuccess) { fprintf(stderr, "ERROR: copying C %s\n", cudaGetErrorString(error)); exit(-1); } // Free device memory cudaFree(d_A.elements); cudaFree(d_B.elements); cudaFree(d_C.elements); } // Matrix multiplication kernel called by MatMul() __global__ void MatMulKernel(Matrix A, Matrix B, Matrix C) { // Each thread computes one element of C // by accumulating results into Cvalue float Cvalue = 0; int row = blockIdx.y * blockDim.y + threadIdx.y; int col = blockIdx.x * blockDim.x + threadIdx.x; for (int e = 0; e < A.width; ++e) Cvalue += A.elements[row * A.width + e] * B.elements[e * B.width + col]; C.elements[row * C.width + col] = Cvalue; } int myrand() { return rand() / (RAND_MAX / 10); } int main() { // A x B srand(0); Matrix A, B, C; A.height = 1 * BLOCK_SIZE; A.width = 1 * BLOCK_SIZE; // hB = wA B.height = A.width; B.width = 1 * BLOCK_SIZE; C.height = A.height; // hC = hA C.width = B.width; // wC = wB A.elements = (float *)malloc(A.height * A.width * sizeof(float)); B.elements = (float *)malloc(B.height * B.width * sizeof(float)); C.elements = (float *)malloc(C.height * C.width * sizeof(float)); printf("Content of A: \n"); for (int i = 0; i < A.height; i++) { for (int j = 0; j < A.width; j++) { A.elements[i * A.height + j] = myrand(); printf("%2d", (int)A.elements[i * A.height + j]); } printf("\n"); } printf("\n\nContent of B: \n"); for (int i = 0; i < B.height; i++) { for (int j = 0; j < B.width; j++) { B.elements[i * B.height + j] = myrand(); printf("%2d", (int)B.elements[i * B.height + j]); } printf("\n"); } MatMul(A, B, C); printf("\n\nContent of C: \n"); for (int i = 0; i < C.height; i++) { for (int j = 0; j < C.width; j++) { printf("%4d", (int)C.elements[i * C.height + j]); } printf("\n"); } return 0; }
code for sm_80 Function : _Z12MatMulKernel6MatrixS_S_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R0, SR_CTAID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002600 */ /*0020*/ MOV R4, c[0x0][0x160] ; /* 0x0000580000047a02 */ /* 0x000fe20000000f00 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ HFMA2.MMA R24, -RZ, RZ, 0, 0 ; /* 0x00000000ff187435 */ /* 0x000fe200000001ff */ /*0050*/ S2R R3, SR_TID.Y ; /* 0x0000000000037919 */ /* 0x000e220000002200 */ /*0060*/ ISETP.GE.AND P0, PT, R4, 0x1, PT ; /* 0x000000010400780c */ /* 0x000fc60003f06270 */ /*0070*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e680000002500 */ /*0080*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0090*/ IMAD R0, R0, c[0x0][0x4], R3 ; /* 0x0000010000007a24 */ /* 0x001fe400078e0203 */ /*00a0*/ IMAD R3, R2, c[0x0][0x0], R5 ; /* 0x0000000002037a24 */ /* 0x002fc600078e0205 */ /*00b0*/ @!P0 BRA 0xc10 ; /* 0x00000b5000008947 */ /* 0x000fea0003800000 */ /*00c0*/ IADD3 R2, R4.reuse, -0x1, RZ ; /* 0xffffffff04027810 */ /* 0x040fe40007ffe0ff */ /*00d0*/ LOP3.LUT R4, R4, 0x3, RZ, 0xc0, !PT ; /* 0x0000000304047812 */ /* 0x000fe400078ec0ff */ /*00e0*/ ISETP.GE.U32.AND P0, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe40003f06070 */ /*00f0*/ MOV R2, RZ ; /* 0x000000ff00027202 */ /* 0x000fe40000000f00 */ /*0100*/ MOV R24, RZ ; /* 0x000000ff00187202 */ /* 0x000fd20000000f00 */ /*0110*/ @!P0 BRA 0xb00 ; /* 0x000009e000008947 */ /* 0x000fea0003800000 */ /*0120*/ IADD3 R5, -R4, c[0x0][0x160], RZ ; /* 0x0000580004057a10 */ /* 0x000fe20007ffe1ff */ /*0130*/ HFMA2.MMA R8, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff087435 */ /* 0x000fe200000001ff */ /*0140*/ ULDC.64 UR6, c[0x0][0x168] ; /* 0x00005a0000067ab9 */ /* 0x000fe20000000a00 */ /*0150*/ HFMA2.MMA R24, -RZ, RZ, 0, 0 ; /* 0x00000000ff187435 */ /* 0x000fe200000001ff */ /*0160*/ ISETP.GT.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fe20003f04270 */ /*0170*/ IMAD R6, R0, c[0x0][0x160], RZ ; /* 0x0000580000067a24 */ /* 0x000fe200078e02ff */ /*0180*/ MOV R2, RZ ; /* 0x000000ff00027202 */ /* 0x000fca0000000f00 */ /*0190*/ IMAD.WIDE R8, R3, R8, c[0x0][0x178] ; /* 0x00005e0003087625 */ /* 0x000fcc00078e0208 */ /*01a0*/ @!P0 BRA 0x960 ; /* 0x000007b000008947 */ /* 0x000fea0003800000 */ /*01b0*/ ISETP.GT.AND P1, PT, R5, 0xc, PT ; /* 0x0000000c0500780c */ /* 0x000fe40003f24270 */ /*01c0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fd60003f0f070 */ /*01d0*/ @!P1 BRA 0x690 ; /* 0x000004b000009947 */ /* 0x000fea0003800000 */ /*01e0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*01f0*/ MOV R12, UR6 ; /* 0x00000006000c7c02 */ /* 0x000fe20008000f00 */ /*0200*/ LDG.E R21, [R8.64] ; /* 0x0000000408157981 */ /* 0x0000a2000c1e1900 */ /*0210*/ MOV R13, UR7 ; /* 0x00000007000d7c02 */ /* 0x000fca0008000f00 */ /*0220*/ IMAD.WIDE R12, R6, 0x4, R12 ; /* 0x00000004060c7825 */ /* 0x000fca00078e020c */ /*0230*/ LDG.E R20, [R12.64] ; /* 0x000000040c147981 */ /* 0x000ea2000c1e1900 */ /*0240*/ MOV R7, c[0x0][0x170] ; /* 0x00005c0000077a02 */ /* 0x000fc60000000f00 */ /*0250*/ LDG.E R14, [R12.64+0x4] ; /* 0x000004040c0e7981 */ /* 0x000ee4000c1e1900 */ /*0260*/ IMAD.WIDE R10, R7.reuse, 0x4, R8 ; /* 0x00000004070a7825 */ /* 0x040fe400078e0208 */ /*0270*/ LDG.E R27, [R12.64+0x8] ; /* 0x000008040c1b7981 */ /* 0x000f28000c1e1900 */ /*0280*/ LDG.E R15, [R10.64] ; /* 0x000000040a0f7981 */ /* 0x0002e2000c1e1900 */ /*0290*/ IMAD.WIDE R22, R7, 0x4, R10 ; /* 0x0000000407167825 */ /* 0x000fc600078e020a */ /*02a0*/ LDG.E R18, [R12.64+0xc] ; /* 0x00000c040c127981 */ /* 0x000f66000c1e1900 */ /*02b0*/ IMAD.WIDE R28, R7.reuse, 0x4, R22 ; /* 0x00000004071c7825 */ /* 0x040fe200078e0216 */ /*02c0*/ LDG.E R26, [R22.64] ; /* 0x00000004161a7981 */ /* 0x000328000c1e1900 */ /*02d0*/ LDG.E R19, [R28.64] ; /* 0x000000041c137981 */ /* 0x000362000c1e1900 */ /*02e0*/ IMAD.WIDE R16, R7, 0x4, R28 ; /* 0x0000000407107825 */ /* 0x000fc600078e021c */ /*02f0*/ LDG.E R8, [R12.64+0x10] ; /* 0x000010040c087981 */ /* 0x001f68000c1e1900 */ /*0300*/ LDG.E R9, [R16.64] ; /* 0x0000000410097981 */ /* 0x000168000c1e1900 */ /*0310*/ LDG.E R10, [R12.64+0x14] ; /* 0x000014040c0a7981 */ /* 0x002f68000c1e1900 */ /*0320*/ LDG.E R28, [R12.64+0x1c] ; /* 0x00001c040c1c7981 */ /* 0x000f62000c1e1900 */ /*0330*/ IMAD.WIDE R16, R7, 0x4, R16 ; /* 0x0000000407107825 */ /* 0x001fca00078e0210 */ /*0340*/ LDG.E R11, [R16.64] ; /* 0x00000004100b7981 */ /* 0x000562000c1e1900 */ /*0350*/ IMAD.WIDE R22, R7, 0x4, R16 ; /* 0x0000000407167825 */ /* 0x000fc800078e0210 */ /*0360*/ FFMA R16, R21, R20, R24 ; /* 0x0000001415107223 */ /* 0x004fe40000000018 */ /*0370*/ LDG.E R20, [R12.64+0x18] ; /* 0x000018040c147981 */ /* 0x000ea2000c1e1900 */ /*0380*/ IMAD.WIDE R24, R7, 0x4, R22 ; /* 0x0000000407187825 */ /* 0x000fc600078e0216 */ /*0390*/ LDG.E R21, [R22.64] ; /* 0x0000000416157981 */ /* 0x0000a8000c1e1900 */ /*03a0*/ LDG.E R29, [R24.64] ; /* 0x00000004181d7981 */ /* 0x0002a2000c1e1900 */ /*03b0*/ FFMA R16, R15, R14, R16 ; /* 0x0000000e0f107223 */ /* 0x008fe40000000010 */ /*03c0*/ IMAD.WIDE R14, R7.reuse, 0x4, R24 ; /* 0x00000004070e7825 */ /* 0x040fe200078e0218 */ /*03d0*/ LDG.E R23, [R12.64+0x20] ; /* 0x000020040c177981 */ /* 0x001ee6000c1e1900 */ /*03e0*/ FFMA R26, R26, R27, R16 ; /* 0x0000001b1a1a7223 */ /* 0x010fe20000000010 */ /*03f0*/ LDG.E R25, [R12.64+0x24] ; /* 0x000024040c197981 */ /* 0x002f22000c1e1900 */ /*0400*/ IMAD.WIDE R16, R7, 0x4, R14 ; /* 0x0000000407107825 */ /* 0x000fc600078e020e */ /*0410*/ LDG.E R14, [R14.64] ; /* 0x000000040e0e7981 */ /* 0x0000e2000c1e1900 */ /*0420*/ FFMA R26, R19, R18, R26 ; /* 0x00000012131a7223 */ /* 0x020fe4000000001a */ /*0430*/ IMAD.WIDE R18, R7, 0x4, R16 ; /* 0x0000000407127825 */ /* 0x000fe200078e0210 */ /*0440*/ LDG.E R22, [R12.64+0x28] ; /* 0x000028040c167981 */ /* 0x000f66000c1e1900 */ /*0450*/ FFMA R26, R9, R8, R26 ; /* 0x00000008091a7223 */ /* 0x000fe2000000001a */ /*0460*/ LDG.E R16, [R16.64] ; /* 0x0000000410107981 */ /* 0x000322000c1e1900 */ /*0470*/ IMAD.WIDE R8, R7, 0x4, R18 ; /* 0x0000000407087825 */ /* 0x000fc600078e0212 */ /*0480*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x000368000c1e1900 */ /*0490*/ LDG.E R24, [R8.64] ; /* 0x0000000408187981 */ /* 0x000568000c1e1900 */ /*04a0*/ LDG.E R15, [R12.64+0x2c] ; /* 0x00002c040c0f7981 */ /* 0x001f62000c1e1900 */ /*04b0*/ FFMA R26, R11, R10, R26 ; /* 0x0000000a0b1a7223 */ /* 0x000fe4000000001a */ /*04c0*/ IMAD.WIDE R10, R7, 0x4, R8 ; /* 0x00000004070a7825 */ /* 0x000fe200078e0208 */ /*04d0*/ LDG.E R17, [R12.64+0x30] ; /* 0x000030040c117981 */ /* 0x002f66000c1e1900 */ /*04e0*/ FFMA R26, R21, R20, R26 ; /* 0x00000014151a7223 */ /* 0x004fc4000000001a */ /*04f0*/ IMAD.WIDE R20, R7, 0x4, R10 ; /* 0x0000000407147825 */ /* 0x000fe400078e020a */ /*0500*/ LDG.E R10, [R10.64] ; /* 0x000000040a0a7981 */ /* 0x0000a4000c1e1900 */ /*0510*/ FFMA R28, R29, R28, R26 ; /* 0x0000001c1d1c7223 */ /* 0x000fe4000000001a */ /*0520*/ IMAD.WIDE R26, R7.reuse, 0x4, R20 ; /* 0x00000004071a7825 */ /* 0x040fe200078e0214 */ /*0530*/ LDG.E R29, [R12.64+0x34] ; /* 0x000034040c1d7981 */ /* 0x000ea8000c1e1900 */ /*0540*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x0002a2000c1e1900 */ /*0550*/ IMAD.WIDE R8, R7, 0x4, R26 ; /* 0x0000000407087825 */ /* 0x000fc600078e021a */ /*0560*/ LDG.E R19, [R26.64] ; /* 0x000000041a137981 */ /* 0x0006a8000c1e1900 */ /*0570*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x0010a8000c1e1900 */ /*0580*/ LDG.E R21, [R12.64+0x38] ; /* 0x000038040c157981 */ /* 0x002ea8000c1e1900 */ /*0590*/ LDG.E R26, [R12.64+0x3c] ; /* 0x00003c040c1a7981 */ /* 0x008ee2000c1e1900 */ /*05a0*/ FFMA R14, R14, R23, R28 ; /* 0x000000170e0e7223 */ /* 0x000fc8000000001c */ /*05b0*/ FFMA R25, R16, R25, R14 ; /* 0x0000001910197223 */ /* 0x010fe2000000000e */ /*05c0*/ IADD3 R5, R5, -0x10, RZ ; /* 0xfffffff005057810 */ /* 0x000fc60007ffe0ff */ /*05d0*/ FFMA R18, R18, R22, R25 ; /* 0x0000001612127223 */ /* 0x020fe20000000019 */ /*05e0*/ ISETP.GT.AND P1, PT, R5, 0xc, PT ; /* 0x0000000c0500780c */ /* 0x000fc60003f24270 */ /*05f0*/ FFMA R15, R24, R15, R18 ; /* 0x0000000f180f7223 */ /* 0x000fe20000000012 */ /*0600*/ UIADD3 UR6, UP0, UR6, 0x40, URZ ; /* 0x0000004006067890 */ /* 0x000fe2000ff1e03f */ /*0610*/ IMAD.WIDE R8, R7, 0x4, R8 ; /* 0x0000000407087825 */ /* 0x001fc600078e0208 */ /*0620*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0630*/ IADD3 R2, R2, 0x10, RZ ; /* 0x0000001002027810 */ /* 0x000fe20007ffe0ff */ /*0640*/ FFMA R10, R10, R17, R15 ; /* 0x000000110a0a7223 */ /* 0x004fc8000000000f */ /*0650*/ FFMA R10, R20, R29, R10 ; /* 0x0000001d140a7223 */ /* 0x000fc8000000000a */ /*0660*/ FFMA R10, R19, R21, R10 ; /* 0x00000015130a7223 */ /* 0x000fc8000000000a */ /*0670*/ FFMA R24, R11, R26, R10 ; /* 0x0000001a0b187223 */ /* 0x008fe2000000000a */ /*0680*/ @P1 BRA 0x1f0 ; /* 0xfffffb6000001947 */ /* 0x000fea000383ffff */ /*0690*/ ISETP.GT.AND P1, PT, R5, 0x4, PT ; /* 0x000000040500780c */ /* 0x000fda0003f24270 */ /*06a0*/ @!P1 BRA 0x940 ; /* 0x0000029000009947 */ /* 0x000fea0003800000 */ /*06b0*/ MOV R7, c[0x0][0x170] ; /* 0x00005c0000077a02 */ /* 0x000fe20000000f00 */ /*06c0*/ LDG.E R23, [R8.64] ; /* 0x0000000408177981 */ /* 0x0000a2000c1e1900 */ /*06d0*/ MOV R10, UR6 ; /* 0x00000006000a7c02 */ /* 0x000fe40008000f00 */ /*06e0*/ MOV R11, UR7 ; /* 0x00000007000b7c02 */ /* 0x000fe20008000f00 */ /*06f0*/ IMAD.WIDE R16, R7, 0x4, R8 ; /* 0x0000000407107825 */ /* 0x000fc800078e0208 */ /*0700*/ IMAD.WIDE R10, R6, 0x4, R10 ; /* 0x00000004060a7825 */ /* 0x000fc800078e020a */ /*0710*/ IMAD.WIDE R12, R7.reuse, 0x4, R16 ; /* 0x00000004070c7825 */ /* 0x040fe200078e0210 */ /*0720*/ LDG.E R22, [R10.64] ; /* 0x000000040a167981 */ /* 0x000ea8000c1e1900 */ /*0730*/ LDG.E R16, [R16.64] ; /* 0x0000000410107981 */ /* 0x0002e2000c1e1900 */ /*0740*/ IMAD.WIDE R14, R7, 0x4, R12 ; /* 0x00000004070e7825 */ /* 0x000fc600078e020c */ /*0750*/ LDG.E R25, [R10.64+0x4] ; /* 0x000004040a197981 */ /* 0x000ee6000c1e1900 */ /*0760*/ IMAD.WIDE R18, R7.reuse, 0x4, R14 ; /* 0x0000000407127825 */ /* 0x040fe200078e020e */ /*0770*/ LDG.E R26, [R12.64] ; /* 0x000000040c1a7981 */ /* 0x000968000c1e1900 */ /*0780*/ LDG.E R27, [R10.64+0x8] ; /* 0x000008040a1b7981 */ /* 0x000f62000c1e1900 */ /*0790*/ IMAD.WIDE R20, R7, 0x4, R18 ; /* 0x0000000407147825 */ /* 0x000fc600078e0212 */ /*07a0*/ LDG.E R14, [R14.64] ; /* 0x000000040e0e7981 */ /* 0x000368000c1e1900 */ /*07b0*/ LDG.E R29, [R10.64+0xc] ; /* 0x00000c040a1d7981 */ /* 0x000f62000c1e1900 */ /*07c0*/ IMAD.WIDE R8, R7, 0x4, R20 ; /* 0x0000000407087825 */ /* 0x001fc600078e0214 */ /*07d0*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x000168000c1e1900 */ /*07e0*/ LDG.E R28, [R10.64+0x10] ; /* 0x000010040a1c7981 */ /* 0x000f62000c1e1900 */ /*07f0*/ IMAD.WIDE R12, R7, 0x4, R8 ; /* 0x00000004070c7825 */ /* 0x010fc600078e0208 */ /*0800*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x000968000c1e1900 */ /*0810*/ LDG.E R15, [R10.64+0x14] ; /* 0x000014040a0f7981 */ /* 0x002f68000c1e1900 */ /*0820*/ LDG.E R17, [R8.64] ; /* 0x0000000408117981 */ /* 0x000368000c1e1900 */ /*0830*/ LDG.E R21, [R10.64+0x1c] ; /* 0x00001c040a157981 */ /* 0x010f28000c1e1900 */ /*0840*/ LDG.E R19, [R12.64] ; /* 0x000000040c137981 */ /* 0x001f28000c1e1900 */ /*0850*/ LDG.E R8, [R10.64+0x18] ; /* 0x000018040a087981 */ /* 0x002f22000c1e1900 */ /*0860*/ UIADD3 UR6, UP0, UR6, 0x20, URZ ; /* 0x0000002006067890 */ /* 0x000fe2000ff1e03f */ /*0870*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fc40003f0e170 */ /*0880*/ IADD3 R2, R2, 0x8, RZ ; /* 0x0000000802027810 */ /* 0x000fe40007ffe0ff */ /*0890*/ IADD3 R5, R5, -0x8, RZ ; /* 0xfffffff805057810 */ /* 0x000fe20007ffe0ff */ /*08a0*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*08b0*/ FFMA R22, R23, R22, R24 ; /* 0x0000001617167223 */ /* 0x004fc80000000018 */ /*08c0*/ FFMA R16, R16, R25, R22 ; /* 0x0000001910107223 */ /* 0x008fc80000000016 */ /*08d0*/ FFMA R16, R26, R27, R16 ; /* 0x0000001b1a107223 */ /* 0x020fc80000000010 */ /*08e0*/ FFMA R29, R14, R29, R16 ; /* 0x0000001d0e1d7223 */ /* 0x000fc80000000010 */ /*08f0*/ FFMA R18, R18, R28, R29 ; /* 0x0000001c12127223 */ /* 0x000fc8000000001d */ /*0900*/ FFMA R15, R20, R15, R18 ; /* 0x0000000f140f7223 */ /* 0x000fc80000000012 */ /*0910*/ FFMA R24, R17, R8, R15 ; /* 0x0000000811187223 */ /* 0x010fe4000000000f */ /*0920*/ IMAD.WIDE R8, R7, 0x4, R12 ; /* 0x0000000407087825 */ /* 0x000fc800078e020c */ /*0930*/ FFMA R24, R19, R21, R24 ; /* 0x0000001513187223 */ /* 0x000fe40000000018 */ /*0940*/ ISETP.NE.OR P0, PT, R5, RZ, P0 ; /* 0x000000ff0500720c */ /* 0x000fda0000705670 */ /*0950*/ @!P0 BRA 0xb00 ; /* 0x000001a000008947 */ /* 0x000fea0003800000 */ /*0960*/ MOV R10, UR6 ; /* 0x00000006000a7c02 */ /* 0x000fe40008000f00 */ /*0970*/ MOV R11, UR7 ; /* 0x00000007000b7c02 */ /* 0x000fe40008000f00 */ /*0980*/ MOV R7, c[0x0][0x170] ; /* 0x00005c0000077a02 */ /* 0x000fc60000000f00 */ /*0990*/ IMAD.WIDE R10, R6, 0x4, R10 ; /* 0x00000004060a7825 */ /* 0x000fc800078e020a */ /*09a0*/ IMAD.WIDE R16, R7.reuse, 0x4, R8 ; /* 0x0000000407107825 */ /* 0x040fe200078e0208 */ /*09b0*/ LDG.E R18, [R10.64] ; /* 0x000000040a127981 */ /* 0x000ea8000c1e1900 */ /*09c0*/ LDG.E R9, [R8.64] ; /* 0x0000000408097981 */ /* 0x000ea2000c1e1900 */ /*09d0*/ IMAD.WIDE R12, R7, 0x4, R16 ; /* 0x00000004070c7825 */ /* 0x000fc600078e0210 */ /*09e0*/ LDG.E R17, [R16.64] ; /* 0x0000000410117981 */ /* 0x000ee8000c1e1900 */ /*09f0*/ LDG.E R19, [R10.64+0x4] ; /* 0x000004040a137981 */ /* 0x000ee2000c1e1900 */ /*0a00*/ IMAD.WIDE R14, R7, 0x4, R12 ; /* 0x00000004070e7825 */ /* 0x000fc600078e020c */ /*0a10*/ LDG.E R21, [R12.64] ; /* 0x000000040c157981 */ /* 0x000f28000c1e1900 */ /*0a20*/ LDG.E R20, [R10.64+0x8] ; /* 0x000008040a147981 */ /* 0x000f28000c1e1900 */ /*0a30*/ LDG.E R22, [R10.64+0xc] ; /* 0x00000c040a167981 */ /* 0x000f68000c1e1900 */ /*0a40*/ LDG.E R23, [R14.64] ; /* 0x000000040e177981 */ /* 0x000f62000c1e1900 */ /*0a50*/ IADD3 R5, R5, -0x4, RZ ; /* 0xfffffffc05057810 */ /* 0x000fc80007ffe0ff */ /*0a60*/ ISETP.NE.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fe20003f05270 */ /*0a70*/ UIADD3 UR6, UP0, UR6, 0x10, URZ ; /* 0x0000001006067890 */ /* 0x000fe2000ff1e03f */ /*0a80*/ IADD3 R2, R2, 0x4, RZ ; /* 0x0000000402027810 */ /* 0x000fc60007ffe0ff */ /*0a90*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0aa0*/ FFMA R18, R9, R18, R24 ; /* 0x0000001209127223 */ /* 0x004fc80000000018 */ /*0ab0*/ FFMA R18, R17, R19, R18 ; /* 0x0000001311127223 */ /* 0x008fe40000000012 */ /*0ac0*/ IMAD.WIDE R8, R7, 0x4, R14 ; /* 0x0000000407087825 */ /* 0x000fc800078e020e */ /*0ad0*/ FFMA R18, R21, R20, R18 ; /* 0x0000001415127223 */ /* 0x010fc80000000012 */ /*0ae0*/ FFMA R24, R23, R22, R18 ; /* 0x0000001617187223 */ /* 0x020fe20000000012 */ /*0af0*/ @P0 BRA 0x960 ; /* 0xfffffe6000000947 */ /* 0x000fea000383ffff */ /*0b00*/ ISETP.NE.AND P0, PT, R4, RZ, PT ; /* 0x000000ff0400720c */ /* 0x000fda0003f05270 */ /*0b10*/ @!P0 BRA 0xc10 ; /* 0x000000f000008947 */ /* 0x000fea0003800000 */ /*0b20*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0b30*/ IMAD R6, R0, c[0x0][0x160], R2 ; /* 0x0000580000067a24 */ /* 0x000fe400078e0202 */ /*0b40*/ IMAD R2, R2, c[0x0][0x170], R3 ; /* 0x00005c0002027a24 */ /* 0x000fce00078e0203 */ /*0b50*/ IMAD.WIDE R6, R6, R9, c[0x0][0x168] ; /* 0x00005a0006067625 */ /* 0x000fc800078e0209 */ /*0b60*/ IMAD.WIDE R8, R2, R9, c[0x0][0x178] ; /* 0x00005e0002087625 */ /* 0x000fca00078e0209 */ /*0b70*/ LDG.E R5, [R8.64] ; /* 0x0000000408057981 */ /* 0x0000a8000c1e1900 */ /*0b80*/ LDG.E R2, [R6.64] ; /* 0x0000000406027981 */ /* 0x0002a2000c1e1900 */ /*0b90*/ IADD3 R4, R4, -0x1, RZ ; /* 0xffffffff04047810 */ /* 0x000fe40007ffe0ff */ /*0ba0*/ MOV R11, c[0x0][0x170] ; /* 0x00005c00000b7a02 */ /* 0x000fe40000000f00 */ /*0bb0*/ ISETP.NE.AND P0, PT, R4, RZ, PT ; /* 0x000000ff0400720c */ /* 0x000fc60003f05270 */ /*0bc0*/ IMAD.WIDE R8, R11, 0x4, R8 ; /* 0x000000040b087825 */ /* 0x001fe200078e0208 */ /*0bd0*/ IADD3 R6, P1, R6, 0x4, RZ ; /* 0x0000000406067810 */ /* 0x002fc80007f3e0ff */ /*0be0*/ IADD3.X R7, RZ, R7, RZ, P1, !PT ; /* 0x00000007ff077210 */ /* 0x000fe20000ffe4ff */ /*0bf0*/ FFMA R24, R5, R2, R24 ; /* 0x0000000205187223 */ /* 0x004fc80000000018 */ /*0c00*/ @P0 BRA 0xb70 ; /* 0xffffff6000000947 */ /* 0x000fea000383ffff */ /*0c10*/ MOV R2, 0x4 ; /* 0x0000000400027802 */ /* 0x000fe20000000f00 */ /*0c20*/ IMAD R3, R0, c[0x0][0x180], R3 ; /* 0x0000600000037a24 */ /* 0x000fc800078e0203 */ /*0c30*/ IMAD.WIDE R2, R3, R2, c[0x0][0x188] ; /* 0x0000620003027625 */ /* 0x000fca00078e0202 */ /*0c40*/ STG.E [R2.64], R24 ; /* 0x0000001802007986 */ /* 0x000fe2000c101904 */ /*0c50*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0c60*/ BRA 0xc60; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0c70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ca0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ce0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <stdio.h> #include <stdlib.h> // Matrices are stored in row-major order: // M(row, col) = *(M.elements + row * M.width + col) typedef struct { int width; int height; float *elements; } Matrix; // Thread block size #define BLOCK_SIZE 16 // Forward declaration of the matrix multiplication kernel __global__ void MatMulKernel(const Matrix, const Matrix, Matrix); // Matrix multiplication - Host code // Matrix dimensions are assumed to be multiples of BLOCK_SIZE void MatMul(const Matrix A, const Matrix B, Matrix C) { cudaSetDevice(0); cudaDeviceSynchronize(); size_t available, total; cudaMemGetInfo(&available, &total); // printf("Mem total: %ld Bytes\nMem available: %ld Bytes\n", available, total); // Load A and B to device memory Matrix d_A; d_A.width = A.width; d_A.height = A.height; size_t size = A.width * A.height * sizeof(float); // printf("size of A: %ld\n", size); cudaMalloc(&d_A.elements, size); cudaError_t error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation A %s\n", cudaGetErrorString(error)); exit(-1); } cudaMemcpy(d_A.elements, A.elements, size, cudaMemcpyHostToDevice); Matrix d_B; d_B.width = B.width; d_B.height = B.height; size = B.width * B.height * sizeof(float); cudaMalloc(&d_B.elements, size); error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation B %s\n", cudaGetErrorString(error)); exit(-1); } cudaMemcpy(d_B.elements, B.elements, size, cudaMemcpyHostToDevice); // Allocate C in device memory Matrix d_C; d_C.width = C.width; d_C.height = C.height; size = C.width * C.height * sizeof(float); error = cudaGetLastError(); cudaMalloc(&d_C.elements, size); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation C %s\n", cudaGetErrorString(error)); exit(-1); } // Invoke kernel dim3 dimBlock(BLOCK_SIZE, BLOCK_SIZE); dim3 dimGrid(B.width / dimBlock.x, A.height / dimBlock.y); MatMulKernel<<<dimGrid, dimBlock>>>(d_A, d_B, d_C); cudaDeviceSynchronize(); error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: calculation error %s\n", cudaGetErrorString(error)); exit(-1); } // Read C from device memory cudaMemcpy(C.elements, d_C.elements, size, cudaMemcpyDeviceToHost); if (error != cudaSuccess) { fprintf(stderr, "ERROR: copying C %s\n", cudaGetErrorString(error)); exit(-1); } // Free device memory cudaFree(d_A.elements); cudaFree(d_B.elements); cudaFree(d_C.elements); } // Matrix multiplication kernel called by MatMul() __global__ void MatMulKernel(Matrix A, Matrix B, Matrix C) { // Each thread computes one element of C // by accumulating results into Cvalue float Cvalue = 0; int row = blockIdx.y * blockDim.y + threadIdx.y; int col = blockIdx.x * blockDim.x + threadIdx.x; for (int e = 0; e < A.width; ++e) Cvalue += A.elements[row * A.width + e] * B.elements[e * B.width + col]; C.elements[row * C.width + col] = Cvalue; } int myrand() { return rand() / (RAND_MAX / 10); } int main() { // A x B srand(0); Matrix A, B, C; A.height = 1 * BLOCK_SIZE; A.width = 1 * BLOCK_SIZE; // hB = wA B.height = A.width; B.width = 1 * BLOCK_SIZE; C.height = A.height; // hC = hA C.width = B.width; // wC = wB A.elements = (float *)malloc(A.height * A.width * sizeof(float)); B.elements = (float *)malloc(B.height * B.width * sizeof(float)); C.elements = (float *)malloc(C.height * C.width * sizeof(float)); printf("Content of A: \n"); for (int i = 0; i < A.height; i++) { for (int j = 0; j < A.width; j++) { A.elements[i * A.height + j] = myrand(); printf("%2d", (int)A.elements[i * A.height + j]); } printf("\n"); } printf("\n\nContent of B: \n"); for (int i = 0; i < B.height; i++) { for (int j = 0; j < B.width; j++) { B.elements[i * B.height + j] = myrand(); printf("%2d", (int)B.elements[i * B.height + j]); } printf("\n"); } MatMul(A, B, C); printf("\n\nContent of C: \n"); for (int i = 0; i < C.height; i++) { for (int j = 0; j < C.width; j++) { printf("%4d", (int)C.elements[i * C.height + j]); } printf("\n"); } return 0; }
.file "tmpxft_001bbc4f_00000000-6_mm.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2062: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2062: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z6myrandv .type _Z6myrandv, @function _Z6myrandv: .LFB2058: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call rand@PLT movl %eax, %edx cltq imulq $-1610612725, %rax, %rax shrq $32, %rax addl %edx, %eax sarl $27, %eax sarl $31, %edx subl %edx, %eax addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2058: .size _Z6myrandv, .-_Z6myrandv .globl _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ .type _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_, @function _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_: .LFB2084: .cfi_startproc endbr64 subq $104, %rsp .cfi_def_cfa_offset 112 movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax movq %rdi, 64(%rsp) movq %rsi, 72(%rsp) movq %rdx, 80(%rsp) movl $1, 16(%rsp) movl $1, 20(%rsp) movl $1, 24(%rsp) movl $1, 28(%rsp) movl $1, 32(%rsp) movl $1, 36(%rsp) leaq 8(%rsp), %rcx movq %rsp, %rdx leaq 28(%rsp), %rsi leaq 16(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L9 .L5: movq 88(%rsp), %rax subq %fs:40, %rax jne .L10 addq $104, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L9: .cfi_restore_state pushq 8(%rsp) .cfi_def_cfa_offset 120 pushq 8(%rsp) .cfi_def_cfa_offset 128 leaq 80(%rsp), %r9 movq 44(%rsp), %rcx movl 52(%rsp), %r8d movq 32(%rsp), %rsi movl 40(%rsp), %edx leaq _Z12MatMulKernel6MatrixS_S_(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 112 jmp .L5 .L10: call __stack_chk_fail@PLT .cfi_endproc .LFE2084: .size _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_, .-_Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ .globl _Z12MatMulKernel6MatrixS_S_ .type _Z12MatMulKernel6MatrixS_S_, @function _Z12MatMulKernel6MatrixS_S_: .LFB2085: .cfi_startproc endbr64 subq $56, %rsp .cfi_def_cfa_offset 64 movq %rdi, 32(%rsp) movq %rsi, 40(%rsp) movq %rdx, 16(%rsp) movq %rcx, 24(%rsp) movq %r8, (%rsp) movq %r9, 8(%rsp) movq %rsp, %rdx leaq 16(%rsp), %rsi leaq 32(%rsp), %rdi call _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ addq $56, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _Z12MatMulKernel6MatrixS_S_, .-_Z12MatMulKernel6MatrixS_S_ .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "ERROR: allocation A %s\n" .LC1: .string "ERROR: allocation B %s\n" .LC2: .string "ERROR: allocation C %s\n" .LC3: .string "ERROR: calculation error %s\n" .text .globl _Z6MatMul6MatrixS_S_ .type _Z6MatMul6MatrixS_S_, @function _Z6MatMul6MatrixS_S_: .LFB2057: .cfi_startproc endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $200, %rsp .cfi_def_cfa_offset 256 movq %rdi, %rbx movq %rsi, 8(%rsp) movq %rdx, %r13 movq %rcx, 16(%rsp) movq %r8, %rbp movq %r9, 24(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax movq %rdi, %r14 sarq $32, %r14 movq %rdx, %r12 sarq $32, %r12 movq %r8, %r15 sarq $32, %r15 movl $0, %edi call cudaSetDevice@PLT call cudaDeviceSynchronize@PLT leaq 48(%rsp), %rsi leaq 40(%rsp), %rdi call cudaMemGetInfo@PLT movl %ebx, 80(%rsp) movl %r14d, 84(%rsp) imull %r14d, %ebx movslq %ebx, %rbx salq $2, %rbx leaq 88(%rsp), %rdi movq %rbx, %rsi call cudaMalloc@PLT call cudaGetLastError@PLT testl %eax, %eax jne .L21 movl $1, %ecx movq %rbx, %rdx movq 8(%rsp), %rsi movq 88(%rsp), %rdi call cudaMemcpy@PLT movl %r13d, 96(%rsp) movl %r12d, 100(%rsp) imull %r13d, %r12d movslq %r12d, %r12 salq $2, %r12 leaq 104(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT call cudaGetLastError@PLT testl %eax, %eax jne .L22 movl $1, %ecx movq %r12, %rdx movq 16(%rsp), %rsi movq 104(%rsp), %rdi call cudaMemcpy@PLT movl %ebp, 112(%rsp) movl %r15d, 116(%rsp) imull %r15d, %ebp movslq %ebp, %rbp salq $2, %rbp call cudaGetLastError@PLT movl %eax, %ebx leaq 120(%rsp), %rdi movq %rbp, %rsi call cudaMalloc@PLT testl %ebx, %ebx jne .L23 shrl $4, %r13d movl %r13d, 68(%rsp) shrl $4, %r14d movl %r14d, 72(%rsp) movl $16, 56(%rsp) movl $16, 60(%rsp) movl $0, %r9d movl $0, %r8d movq 56(%rsp), %rdx movl $1, %ecx movq 68(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L24 .L17: call cudaDeviceSynchronize@PLT call cudaGetLastError@PLT testl %eax, %eax jne .L25 movl $2, %ecx movq %rbp, %rdx movq 120(%rsp), %rsi movq 24(%rsp), %rdi call cudaMemcpy@PLT movq 88(%rsp), %rdi call cudaFree@PLT movq 104(%rsp), %rdi call cudaFree@PLT movq 120(%rsp), %rdi call cudaFree@PLT movq 184(%rsp), %rax subq %fs:40, %rax jne .L26 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .L21: .cfi_restore_state movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC0(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L22: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC1(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L23: movl %ebx, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC2(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L24: movdqa 80(%rsp), %xmm0 movaps %xmm0, 128(%rsp) movdqa 96(%rsp), %xmm1 movaps %xmm1, 144(%rsp) movdqa 112(%rsp), %xmm2 movaps %xmm2, 160(%rsp) leaq 160(%rsp), %rdx leaq 144(%rsp), %rsi leaq 128(%rsp), %rdi call _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ jmp .L17 .L25: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC3(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L26: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size _Z6MatMul6MatrixS_S_, .-_Z6MatMul6MatrixS_S_ .section .rodata.str1.1 .LC4: .string "Content of A: \n" .LC5: .string "%2d" .LC6: .string "\n" .LC7: .string "\n\nContent of B: \n" .LC8: .string "\n\nContent of C: \n" .LC9: .string "%4d" .text .globl main .type main, @function main: .LFB2059: .cfi_startproc endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $24, %rsp .cfi_def_cfa_offset 80 movl $0, %edi call srand@PLT movl $1024, %edi call malloc@PLT movq %rax, %rbx movq %rax, (%rsp) movl $1024, %edi call malloc@PLT movq %rax, %r13 movl $1024, %edi call malloc@PLT movq %rax, 8(%rsp) leaq .LC4(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 64(%rbx), %rbp leaq 1088(%rbx), %r15 leaq .LC5(%rip), %r12 leaq .LC6(%rip), %r14 .L28: leaq -64(%rbp), %rbx .L29: call _Z6myrandv pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, (%rbx) cvttss2sil %xmm0, %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L29 movq %r14, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $64, %rbp cmpq %r15, %rbp jne .L28 leaq .LC7(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 64(%r13), %rbp leaq 1088(%r13), %r15 leaq .LC5(%rip), %r12 leaq .LC6(%rip), %r14 .L31: leaq -64(%rbp), %rbx .L32: call _Z6myrandv pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, (%rbx) cvttss2sil %xmm0, %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L32 movq %r14, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $64, %rbp cmpq %r15, %rbp jne .L31 movabsq $68719476736, %r8 movq %r8, %rdi orq $16, %rdi movq %rdi, %rdx movq %rdi, %r8 movq 8(%rsp), %rbx movq %rbx, %r9 movq %r13, %rcx movq (%rsp), %rsi call _Z6MatMul6MatrixS_S_ leaq .LC8(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 64(%rbx), %rbp leaq 1088(%rbx), %r13 leaq .LC9(%rip), %r12 leaq .LC6(%rip), %r14 .L34: leaq -64(%rbp), %rbx .L35: cvttss2sil (%rbx), %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L35 movq %r14, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $64, %rbp cmpq %r13, %rbp jne .L34 movl $0, %eax addq $24, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2059: .size main, .-main .section .rodata.str1.1 .LC10: .string "_Z12MatMulKernel6MatrixS_S_" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2087: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC10(%rip), %rdx movq %rdx, %rcx leaq _Z12MatMulKernel6MatrixS_S_(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2087: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
You are a helpful coding assistant specialized in converting CUDA source code to HIP/AMD source code.
#include <stdio.h> #include <stdlib.h> // Matrices are stored in row-major order: // M(row, col) = *(M.elements + row * M.width + col) typedef struct { int width; int height; float *elements; } Matrix; // Thread block size #define BLOCK_SIZE 16 // Forward declaration of the matrix multiplication kernel __global__ void MatMulKernel(const Matrix, const Matrix, Matrix); // Matrix multiplication - Host code // Matrix dimensions are assumed to be multiples of BLOCK_SIZE void MatMul(const Matrix A, const Matrix B, Matrix C) { cudaSetDevice(0); cudaDeviceSynchronize(); size_t available, total; cudaMemGetInfo(&available, &total); // printf("Mem total: %ld Bytes\nMem available: %ld Bytes\n", available, total); // Load A and B to device memory Matrix d_A; d_A.width = A.width; d_A.height = A.height; size_t size = A.width * A.height * sizeof(float); // printf("size of A: %ld\n", size); cudaMalloc(&d_A.elements, size); cudaError_t error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation A %s\n", cudaGetErrorString(error)); exit(-1); } cudaMemcpy(d_A.elements, A.elements, size, cudaMemcpyHostToDevice); Matrix d_B; d_B.width = B.width; d_B.height = B.height; size = B.width * B.height * sizeof(float); cudaMalloc(&d_B.elements, size); error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation B %s\n", cudaGetErrorString(error)); exit(-1); } cudaMemcpy(d_B.elements, B.elements, size, cudaMemcpyHostToDevice); // Allocate C in device memory Matrix d_C; d_C.width = C.width; d_C.height = C.height; size = C.width * C.height * sizeof(float); error = cudaGetLastError(); cudaMalloc(&d_C.elements, size); if (error != cudaSuccess) { fprintf(stderr, "ERROR: allocation C %s\n", cudaGetErrorString(error)); exit(-1); } // Invoke kernel dim3 dimBlock(BLOCK_SIZE, BLOCK_SIZE); dim3 dimGrid(B.width / dimBlock.x, A.height / dimBlock.y); MatMulKernel<<<dimGrid, dimBlock>>>(d_A, d_B, d_C); cudaDeviceSynchronize(); error = cudaGetLastError(); if (error != cudaSuccess) { fprintf(stderr, "ERROR: calculation error %s\n", cudaGetErrorString(error)); exit(-1); } // Read C from device memory cudaMemcpy(C.elements, d_C.elements, size, cudaMemcpyDeviceToHost); if (error != cudaSuccess) { fprintf(stderr, "ERROR: copying C %s\n", cudaGetErrorString(error)); exit(-1); } // Free device memory cudaFree(d_A.elements); cudaFree(d_B.elements); cudaFree(d_C.elements); } // Matrix multiplication kernel called by MatMul() __global__ void MatMulKernel(Matrix A, Matrix B, Matrix C) { // Each thread computes one element of C // by accumulating results into Cvalue float Cvalue = 0; int row = blockIdx.y * blockDim.y + threadIdx.y; int col = blockIdx.x * blockDim.x + threadIdx.x; for (int e = 0; e < A.width; ++e) Cvalue += A.elements[row * A.width + e] * B.elements[e * B.width + col]; C.elements[row * C.width + col] = Cvalue; } int myrand() { return rand() / (RAND_MAX / 10); } int main() { // A x B srand(0); Matrix A, B, C; A.height = 1 * BLOCK_SIZE; A.width = 1 * BLOCK_SIZE; // hB = wA B.height = A.width; B.width = 1 * BLOCK_SIZE; C.height = A.height; // hC = hA C.width = B.width; // wC = wB A.elements = (float *)malloc(A.height * A.width * sizeof(float)); B.elements = (float *)malloc(B.height * B.width * sizeof(float)); C.elements = (float *)malloc(C.height * C.width * sizeof(float)); printf("Content of A: \n"); for (int i = 0; i < A.height; i++) { for (int j = 0; j < A.width; j++) { A.elements[i * A.height + j] = myrand(); printf("%2d", (int)A.elements[i * A.height + j]); } printf("\n"); } printf("\n\nContent of B: \n"); for (int i = 0; i < B.height; i++) { for (int j = 0; j < B.width; j++) { B.elements[i * B.height + j] = myrand(); printf("%2d", (int)B.elements[i * B.height + j]); } printf("\n"); } MatMul(A, B, C); printf("\n\nContent of C: \n"); for (int i = 0; i < C.height; i++) { for (int j = 0; j < C.width; j++) { printf("%4d", (int)C.elements[i * C.height + j]); } printf("\n"); } return 0; }
#include <hip/hip_runtime.h> #include <stdio.h> #include <stdlib.h> // Matrices are stored in row-major order: // M(row, col) = *(M.elements + row * M.width + col) typedef struct { int width; int height; float *elements; } Matrix; // Thread block size #define BLOCK_SIZE 16 // Forward declaration of the matrix multiplication kernel __global__ void MatMulKernel(const Matrix, const Matrix, Matrix); // Matrix multiplication - Host code // Matrix dimensions are assumed to be multiples of BLOCK_SIZE void MatMul(const Matrix A, const Matrix B, Matrix C) { hipSetDevice(0); hipDeviceSynchronize(); size_t available, total; hipMemGetInfo(&available, &total); // printf("Mem total: %ld Bytes\nMem available: %ld Bytes\n", available, total); // Load A and B to device memory Matrix d_A; d_A.width = A.width; d_A.height = A.height; size_t size = A.width * A.height * sizeof(float); // printf("size of A: %ld\n", size); hipMalloc(&d_A.elements, size); hipError_t error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation A %s\n", hipGetErrorString(error)); exit(-1); } hipMemcpy(d_A.elements, A.elements, size, hipMemcpyHostToDevice); Matrix d_B; d_B.width = B.width; d_B.height = B.height; size = B.width * B.height * sizeof(float); hipMalloc(&d_B.elements, size); error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation B %s\n", hipGetErrorString(error)); exit(-1); } hipMemcpy(d_B.elements, B.elements, size, hipMemcpyHostToDevice); // Allocate C in device memory Matrix d_C; d_C.width = C.width; d_C.height = C.height; size = C.width * C.height * sizeof(float); error = hipGetLastError(); hipMalloc(&d_C.elements, size); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation C %s\n", hipGetErrorString(error)); exit(-1); } // Invoke kernel dim3 dimBlock(BLOCK_SIZE, BLOCK_SIZE); dim3 dimGrid(B.width / dimBlock.x, A.height / dimBlock.y); MatMulKernel<<<dimGrid, dimBlock>>>(d_A, d_B, d_C); hipDeviceSynchronize(); error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: calculation error %s\n", hipGetErrorString(error)); exit(-1); } // Read C from device memory hipMemcpy(C.elements, d_C.elements, size, hipMemcpyDeviceToHost); if (error != hipSuccess) { fprintf(stderr, "ERROR: copying C %s\n", hipGetErrorString(error)); exit(-1); } // Free device memory hipFree(d_A.elements); hipFree(d_B.elements); hipFree(d_C.elements); } // Matrix multiplication kernel called by MatMul() __global__ void MatMulKernel(Matrix A, Matrix B, Matrix C) { // Each thread computes one element of C // by accumulating results into Cvalue float Cvalue = 0; int row = blockIdx.y * blockDim.y + threadIdx.y; int col = blockIdx.x * blockDim.x + threadIdx.x; for (int e = 0; e < A.width; ++e) Cvalue += A.elements[row * A.width + e] * B.elements[e * B.width + col]; C.elements[row * C.width + col] = Cvalue; } int myrand() { return rand() / (RAND_MAX / 10); } int main() { // A x B srand(0); Matrix A, B, C; A.height = 1 * BLOCK_SIZE; A.width = 1 * BLOCK_SIZE; // hB = wA B.height = A.width; B.width = 1 * BLOCK_SIZE; C.height = A.height; // hC = hA C.width = B.width; // wC = wB A.elements = (float *)malloc(A.height * A.width * sizeof(float)); B.elements = (float *)malloc(B.height * B.width * sizeof(float)); C.elements = (float *)malloc(C.height * C.width * sizeof(float)); printf("Content of A: \n"); for (int i = 0; i < A.height; i++) { for (int j = 0; j < A.width; j++) { A.elements[i * A.height + j] = myrand(); printf("%2d", (int)A.elements[i * A.height + j]); } printf("\n"); } printf("\n\nContent of B: \n"); for (int i = 0; i < B.height; i++) { for (int j = 0; j < B.width; j++) { B.elements[i * B.height + j] = myrand(); printf("%2d", (int)B.elements[i * B.height + j]); } printf("\n"); } MatMul(A, B, C); printf("\n\nContent of C: \n"); for (int i = 0; i < C.height; i++) { for (int j = 0; j < C.width; j++) { printf("%4d", (int)C.elements[i * C.height + j]); } printf("\n"); } return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include <stdio.h> #include <stdlib.h> // Matrices are stored in row-major order: // M(row, col) = *(M.elements + row * M.width + col) typedef struct { int width; int height; float *elements; } Matrix; // Thread block size #define BLOCK_SIZE 16 // Forward declaration of the matrix multiplication kernel __global__ void MatMulKernel(const Matrix, const Matrix, Matrix); // Matrix multiplication - Host code // Matrix dimensions are assumed to be multiples of BLOCK_SIZE void MatMul(const Matrix A, const Matrix B, Matrix C) { hipSetDevice(0); hipDeviceSynchronize(); size_t available, total; hipMemGetInfo(&available, &total); // printf("Mem total: %ld Bytes\nMem available: %ld Bytes\n", available, total); // Load A and B to device memory Matrix d_A; d_A.width = A.width; d_A.height = A.height; size_t size = A.width * A.height * sizeof(float); // printf("size of A: %ld\n", size); hipMalloc(&d_A.elements, size); hipError_t error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation A %s\n", hipGetErrorString(error)); exit(-1); } hipMemcpy(d_A.elements, A.elements, size, hipMemcpyHostToDevice); Matrix d_B; d_B.width = B.width; d_B.height = B.height; size = B.width * B.height * sizeof(float); hipMalloc(&d_B.elements, size); error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation B %s\n", hipGetErrorString(error)); exit(-1); } hipMemcpy(d_B.elements, B.elements, size, hipMemcpyHostToDevice); // Allocate C in device memory Matrix d_C; d_C.width = C.width; d_C.height = C.height; size = C.width * C.height * sizeof(float); error = hipGetLastError(); hipMalloc(&d_C.elements, size); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation C %s\n", hipGetErrorString(error)); exit(-1); } // Invoke kernel dim3 dimBlock(BLOCK_SIZE, BLOCK_SIZE); dim3 dimGrid(B.width / dimBlock.x, A.height / dimBlock.y); MatMulKernel<<<dimGrid, dimBlock>>>(d_A, d_B, d_C); hipDeviceSynchronize(); error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: calculation error %s\n", hipGetErrorString(error)); exit(-1); } // Read C from device memory hipMemcpy(C.elements, d_C.elements, size, hipMemcpyDeviceToHost); if (error != hipSuccess) { fprintf(stderr, "ERROR: copying C %s\n", hipGetErrorString(error)); exit(-1); } // Free device memory hipFree(d_A.elements); hipFree(d_B.elements); hipFree(d_C.elements); } // Matrix multiplication kernel called by MatMul() __global__ void MatMulKernel(Matrix A, Matrix B, Matrix C) { // Each thread computes one element of C // by accumulating results into Cvalue float Cvalue = 0; int row = blockIdx.y * blockDim.y + threadIdx.y; int col = blockIdx.x * blockDim.x + threadIdx.x; for (int e = 0; e < A.width; ++e) Cvalue += A.elements[row * A.width + e] * B.elements[e * B.width + col]; C.elements[row * C.width + col] = Cvalue; } int myrand() { return rand() / (RAND_MAX / 10); } int main() { // A x B srand(0); Matrix A, B, C; A.height = 1 * BLOCK_SIZE; A.width = 1 * BLOCK_SIZE; // hB = wA B.height = A.width; B.width = 1 * BLOCK_SIZE; C.height = A.height; // hC = hA C.width = B.width; // wC = wB A.elements = (float *)malloc(A.height * A.width * sizeof(float)); B.elements = (float *)malloc(B.height * B.width * sizeof(float)); C.elements = (float *)malloc(C.height * C.width * sizeof(float)); printf("Content of A: \n"); for (int i = 0; i < A.height; i++) { for (int j = 0; j < A.width; j++) { A.elements[i * A.height + j] = myrand(); printf("%2d", (int)A.elements[i * A.height + j]); } printf("\n"); } printf("\n\nContent of B: \n"); for (int i = 0; i < B.height; i++) { for (int j = 0; j < B.width; j++) { B.elements[i * B.height + j] = myrand(); printf("%2d", (int)B.elements[i * B.height + j]); } printf("\n"); } MatMul(A, B, C); printf("\n\nContent of C: \n"); for (int i = 0; i < C.height; i++) { for (int j = 0; j < C.width; j++) { printf("%4d", (int)C.elements[i * C.height + j]); } printf("\n"); } return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z12MatMulKernel6MatrixS_S_ .globl _Z12MatMulKernel6MatrixS_S_ .p2align 8 .type _Z12MatMulKernel6MatrixS_S_,@function _Z12MatMulKernel6MatrixS_S_: s_clause 0x2 s_load_b32 s4, s[0:1], 0x3c s_load_b32 s6, s[0:1], 0x0 s_load_b64 s[2:3], s[0:1], 0x28 v_bfe_u32 v2, v0, 10, 10 v_and_b32_e32 v3, 0x3ff, v0 s_waitcnt lgkmcnt(0) s_lshr_b32 s5, s4, 16 s_and_b32 s4, s4, 0xffff s_delay_alu instid0(VALU_DEP_1) v_mad_u64_u32 v[0:1], null, s15, s5, v[2:3] v_mad_u64_u32 v[1:2], null, s14, s4, v[3:4] s_cmp_lt_i32 s6, 1 s_cbranch_scc1 .LBB0_3 s_load_b64 s[8:9], s[0:1], 0x8 s_delay_alu instid0(VALU_DEP_2) v_mul_lo_u32 v2, v0, s6 s_clause 0x1 s_load_b32 s7, s[0:1], 0x10 s_load_b64 s[4:5], s[0:1], 0x18 v_mov_b32_e32 v6, 0 v_mov_b32_e32 v4, v1 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[2:3], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v2, vcc_lo, s8, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s9, v3, vcc_lo .p2align 6 .LBB0_2: v_ashrrev_i32_e32 v5, 31, v4 s_add_i32 s6, s6, -1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) s_cmp_eq_u32 s6, 0 v_lshlrev_b64 v[7:8], 2, v[4:5] v_add_nc_u32_e32 v4, s7, v4 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_u32 v7, vcc_lo, s4, v7 v_add_co_ci_u32_e32 v8, vcc_lo, s5, v8, vcc_lo global_load_b32 v5, v[2:3], off global_load_b32 v7, v[7:8], off v_add_co_u32 v2, vcc_lo, v2, 4 v_add_co_ci_u32_e32 v3, vcc_lo, 0, v3, vcc_lo s_waitcnt vmcnt(0) v_fmac_f32_e32 v6, v5, v7 s_cbranch_scc0 .LBB0_2 s_branch .LBB0_4 .LBB0_3: v_mov_b32_e32 v6, 0 .LBB0_4: s_load_b32 s0, s[0:1], 0x20 s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[2:3], null, v0, s0, v[1:2] v_ashrrev_i32_e32 v3, 31, v2 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[2:3] v_add_co_u32 v0, vcc_lo, s2, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v1, vcc_lo, s3, v1, vcc_lo global_store_b32 v[0:1], v6, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z12MatMulKernel6MatrixS_S_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 304 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 9 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z12MatMulKernel6MatrixS_S_, .Lfunc_end0-_Z12MatMulKernel6MatrixS_S_ .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 16 .value_kind: by_value - .offset: 16 .size: 16 .value_kind: by_value - .offset: 32 .size: 16 .value_kind: by_value - .offset: 48 .size: 4 .value_kind: hidden_block_count_x - .offset: 52 .size: 4 .value_kind: hidden_block_count_y - .offset: 56 .size: 4 .value_kind: hidden_block_count_z - .offset: 60 .size: 2 .value_kind: hidden_group_size_x - .offset: 62 .size: 2 .value_kind: hidden_group_size_y - .offset: 64 .size: 2 .value_kind: hidden_group_size_z - .offset: 66 .size: 2 .value_kind: hidden_remainder_x - .offset: 68 .size: 2 .value_kind: hidden_remainder_y - .offset: 70 .size: 2 .value_kind: hidden_remainder_z - .offset: 88 .size: 8 .value_kind: hidden_global_offset_x - .offset: 96 .size: 8 .value_kind: hidden_global_offset_y - .offset: 104 .size: 8 .value_kind: hidden_global_offset_z - .offset: 112 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 304 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z12MatMulKernel6MatrixS_S_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z12MatMulKernel6MatrixS_S_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 9 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include <stdio.h> #include <stdlib.h> // Matrices are stored in row-major order: // M(row, col) = *(M.elements + row * M.width + col) typedef struct { int width; int height; float *elements; } Matrix; // Thread block size #define BLOCK_SIZE 16 // Forward declaration of the matrix multiplication kernel __global__ void MatMulKernel(const Matrix, const Matrix, Matrix); // Matrix multiplication - Host code // Matrix dimensions are assumed to be multiples of BLOCK_SIZE void MatMul(const Matrix A, const Matrix B, Matrix C) { hipSetDevice(0); hipDeviceSynchronize(); size_t available, total; hipMemGetInfo(&available, &total); // printf("Mem total: %ld Bytes\nMem available: %ld Bytes\n", available, total); // Load A and B to device memory Matrix d_A; d_A.width = A.width; d_A.height = A.height; size_t size = A.width * A.height * sizeof(float); // printf("size of A: %ld\n", size); hipMalloc(&d_A.elements, size); hipError_t error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation A %s\n", hipGetErrorString(error)); exit(-1); } hipMemcpy(d_A.elements, A.elements, size, hipMemcpyHostToDevice); Matrix d_B; d_B.width = B.width; d_B.height = B.height; size = B.width * B.height * sizeof(float); hipMalloc(&d_B.elements, size); error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation B %s\n", hipGetErrorString(error)); exit(-1); } hipMemcpy(d_B.elements, B.elements, size, hipMemcpyHostToDevice); // Allocate C in device memory Matrix d_C; d_C.width = C.width; d_C.height = C.height; size = C.width * C.height * sizeof(float); error = hipGetLastError(); hipMalloc(&d_C.elements, size); if (error != hipSuccess) { fprintf(stderr, "ERROR: allocation C %s\n", hipGetErrorString(error)); exit(-1); } // Invoke kernel dim3 dimBlock(BLOCK_SIZE, BLOCK_SIZE); dim3 dimGrid(B.width / dimBlock.x, A.height / dimBlock.y); MatMulKernel<<<dimGrid, dimBlock>>>(d_A, d_B, d_C); hipDeviceSynchronize(); error = hipGetLastError(); if (error != hipSuccess) { fprintf(stderr, "ERROR: calculation error %s\n", hipGetErrorString(error)); exit(-1); } // Read C from device memory hipMemcpy(C.elements, d_C.elements, size, hipMemcpyDeviceToHost); if (error != hipSuccess) { fprintf(stderr, "ERROR: copying C %s\n", hipGetErrorString(error)); exit(-1); } // Free device memory hipFree(d_A.elements); hipFree(d_B.elements); hipFree(d_C.elements); } // Matrix multiplication kernel called by MatMul() __global__ void MatMulKernel(Matrix A, Matrix B, Matrix C) { // Each thread computes one element of C // by accumulating results into Cvalue float Cvalue = 0; int row = blockIdx.y * blockDim.y + threadIdx.y; int col = blockIdx.x * blockDim.x + threadIdx.x; for (int e = 0; e < A.width; ++e) Cvalue += A.elements[row * A.width + e] * B.elements[e * B.width + col]; C.elements[row * C.width + col] = Cvalue; } int myrand() { return rand() / (RAND_MAX / 10); } int main() { // A x B srand(0); Matrix A, B, C; A.height = 1 * BLOCK_SIZE; A.width = 1 * BLOCK_SIZE; // hB = wA B.height = A.width; B.width = 1 * BLOCK_SIZE; C.height = A.height; // hC = hA C.width = B.width; // wC = wB A.elements = (float *)malloc(A.height * A.width * sizeof(float)); B.elements = (float *)malloc(B.height * B.width * sizeof(float)); C.elements = (float *)malloc(C.height * C.width * sizeof(float)); printf("Content of A: \n"); for (int i = 0; i < A.height; i++) { for (int j = 0; j < A.width; j++) { A.elements[i * A.height + j] = myrand(); printf("%2d", (int)A.elements[i * A.height + j]); } printf("\n"); } printf("\n\nContent of B: \n"); for (int i = 0; i < B.height; i++) { for (int j = 0; j < B.width; j++) { B.elements[i * B.height + j] = myrand(); printf("%2d", (int)B.elements[i * B.height + j]); } printf("\n"); } MatMul(A, B, C); printf("\n\nContent of C: \n"); for (int i = 0; i < C.height; i++) { for (int j = 0; j < C.width; j++) { printf("%4d", (int)C.elements[i * C.height + j]); } printf("\n"); } return 0; }
.text .file "mm.hip" .globl _Z6MatMul6MatrixS_S_ # -- Begin function _Z6MatMul6MatrixS_S_ .p2align 4, 0x90 .type _Z6MatMul6MatrixS_S_,@function _Z6MatMul6MatrixS_S_: # @_Z6MatMul6MatrixS_S_ .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 subq $200, %rsp .cfi_def_cfa_offset 256 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movq %r9, 48(%rsp) # 8-byte Spill movq %r8, %r12 movq %rcx, 56(%rsp) # 8-byte Spill movq %rdx, %r13 movq %rsi, %rbp movq %rdi, %r15 movq %rdi, %rbx shrq $32, %rbx xorl %edi, %edi callq hipSetDevice callq hipDeviceSynchronize leaq 144(%rsp), %rdi leaq 136(%rsp), %rsi callq hipMemGetInfo movq %r15, 32(%rsp) imull %r15d, %ebx movslq %ebx, %rbx shlq $2, %rbx leaq 40(%rsp), %rdi movq %rbx, %rsi callq hipMalloc callq hipGetLastError testl %eax, %eax jne .LBB0_1 # %bb.4: movq %r13, %r14 shrq $32, %r14 movq 40(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx movl $1, %ecx callq hipMemcpy movq %r13, 16(%rsp) imull %r13d, %r14d movslq %r14d, %rbx shlq $2, %rbx leaq 24(%rsp), %rdi movq %rbx, %rsi callq hipMalloc callq hipGetLastError testl %eax, %eax jne .LBB0_5 # %bb.6: movq %r12, %r14 shrq $32, %r14 movq 24(%rsp), %rdi movq 56(%rsp), %rsi # 8-byte Reload movq %rbx, %rdx movl $1, %ecx callq hipMemcpy movl %r12d, (%rsp) movl %r14d, 4(%rsp) imull %r12d, %r14d movslq %r14d, %r12 shlq $2, %r12 callq hipGetLastError movl %eax, %ebx leaq 8(%rsp), %rdi movq %r12, %rsi callq hipMalloc testl %ebx, %ebx jne .LBB0_7 # %bb.8: shrl $4, %r13d shrq $4, %r15 movabsq $1152921500311879680, %rdi # imm = 0xFFFFFFF00000000 andq %r15, %rdi orq %r13, %rdi movabsq $68719476752, %rdx # imm = 0x1000000010 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB0_10 # %bb.9: movups 32(%rsp), %xmm0 movups 16(%rsp), %xmm1 movups (%rsp), %xmm2 movups %xmm0, 184(%rsp) movups %xmm1, 168(%rsp) movups %xmm2, 152(%rsp) leaq 184(%rsp), %rax movq %rax, 112(%rsp) leaq 168(%rsp), %rax movq %rax, 120(%rsp) leaq 152(%rsp), %rax movq %rax, 128(%rsp) leaq 96(%rsp), %rdi leaq 80(%rsp), %rsi leaq 72(%rsp), %rdx leaq 64(%rsp), %rcx callq __hipPopCallConfiguration movq 96(%rsp), %rsi movl 104(%rsp), %edx movq 80(%rsp), %rcx movl 88(%rsp), %r8d leaq 112(%rsp), %r9 movl $_Z12MatMulKernel6MatrixS_S_, %edi pushq 64(%rsp) .cfi_adjust_cfa_offset 8 pushq 80(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB0_10: callq hipDeviceSynchronize callq hipGetLastError testl %eax, %eax jne .LBB0_11 # %bb.12: movq 8(%rsp), %rsi movq 48(%rsp), %rdi # 8-byte Reload movq %r12, %rdx movl $2, %ecx callq hipMemcpy movq 40(%rsp), %rdi callq hipFree movq 24(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree addq $200, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .LBB0_1: .cfi_def_cfa_offset 256 movq stderr(%rip), %rbx movl %eax, %edi callq hipGetErrorString movl $.L.str, %esi jmp .LBB0_2 .LBB0_5: movq stderr(%rip), %rbx movl %eax, %edi callq hipGetErrorString movl $.L.str.1, %esi jmp .LBB0_2 .LBB0_7: movq stderr(%rip), %r14 movl %ebx, %edi callq hipGetErrorString movl $.L.str.2, %esi movq %r14, %rdi jmp .LBB0_3 .LBB0_11: movq stderr(%rip), %rbx movl %eax, %edi callq hipGetErrorString movl $.L.str.3, %esi .LBB0_2: movq %rbx, %rdi .LBB0_3: movq %rax, %rdx xorl %eax, %eax callq fprintf movl $-1, %edi callq exit .Lfunc_end0: .size _Z6MatMul6MatrixS_S_, .Lfunc_end0-_Z6MatMul6MatrixS_S_ .cfi_endproc # -- End function .globl _Z27__device_stub__MatMulKernel6MatrixS_S_ # -- Begin function _Z27__device_stub__MatMulKernel6MatrixS_S_ .p2align 4, 0x90 .type _Z27__device_stub__MatMulKernel6MatrixS_S_,@function _Z27__device_stub__MatMulKernel6MatrixS_S_: # @_Z27__device_stub__MatMulKernel6MatrixS_S_ .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 80(%rsp) movq %rsi, 88(%rsp) movq %rdx, 64(%rsp) movq %rcx, 72(%rsp) movq %r8, 48(%rsp) movq %r9, 56(%rsp) leaq 80(%rsp), %rax movq %rax, 96(%rsp) leaq 64(%rsp), %rax movq %rax, 104(%rsp) leaq 48(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z12MatMulKernel6MatrixS_S_, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end1: .size _Z27__device_stub__MatMulKernel6MatrixS_S_, .Lfunc_end1-_Z27__device_stub__MatMulKernel6MatrixS_S_ .cfi_endproc # -- End function .globl _Z6myrandv # -- Begin function _Z6myrandv .p2align 4, 0x90 .type _Z6myrandv,@function _Z6myrandv: # @_Z6myrandv .cfi_startproc # %bb.0: pushq %rax .cfi_def_cfa_offset 16 callq rand cltq imulq $-1610612725, %rax, %rcx # imm = 0xA000000B shrq $32, %rcx addl %ecx, %eax movl %eax, %ecx shrl $31, %ecx sarl $27, %eax addl %ecx, %eax # kill: def $eax killed $eax killed $rax popq %rcx .cfi_def_cfa_offset 8 retq .Lfunc_end2: .size _Z6myrandv, .Lfunc_end2-_Z6myrandv .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 pushq %rax .cfi_def_cfa_offset 64 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 xorl %edi, %edi callq srand movl $1024, %edi # imm = 0x400 callq malloc movq %rax, %r14 movl $1024, %edi # imm = 0x400 callq malloc movq %rax, %r15 movl $1024, %edi # imm = 0x400 callq malloc movq %rax, %rbx movl $.Lstr, %edi callq puts@PLT movq %r14, %r12 xorl %r13d, %r13d .p2align 4, 0x90 .LBB3_1: # %.preheader82 # =>This Loop Header: Depth=1 # Child Loop BB3_2 Depth 2 xorl %ebp, %ebp .p2align 4, 0x90 .LBB3_2: # Parent Loop BB3_1 Depth=1 # => This Inner Loop Header: Depth=2 callq rand cltq imulq $-1610612725, %rax, %rcx # imm = 0xA000000B shrq $32, %rcx addl %ecx, %eax movl %eax, %ecx shrl $31, %ecx sarl $27, %eax addl %ecx, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r12,%rbp,4) cvttss2si %xmm0, %esi movl $.L.str.6, %edi xorl %eax, %eax callq printf incq %rbp cmpq $16, %rbp jne .LBB3_2 # %bb.3: # in Loop: Header=BB3_1 Depth=1 movl $10, %edi callq putchar@PLT incq %r13 addq $64, %r12 cmpq $16, %r13 jne .LBB3_1 # %bb.4: movl $.Lstr.1, %edi callq puts@PLT movq %r15, %r12 xorl %r13d, %r13d .p2align 4, 0x90 .LBB3_5: # %.preheader81 # =>This Loop Header: Depth=1 # Child Loop BB3_6 Depth 2 xorl %ebp, %ebp .p2align 4, 0x90 .LBB3_6: # Parent Loop BB3_5 Depth=1 # => This Inner Loop Header: Depth=2 callq rand cltq imulq $-1610612725, %rax, %rcx # imm = 0xA000000B shrq $32, %rcx addl %ecx, %eax movl %eax, %ecx shrl $31, %ecx sarl $27, %eax addl %ecx, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r12,%rbp,4) cvttss2si %xmm0, %esi movl $.L.str.6, %edi xorl %eax, %eax callq printf incq %rbp cmpq $16, %rbp jne .LBB3_6 # %bb.7: # in Loop: Header=BB3_5 Depth=1 movl $10, %edi callq putchar@PLT incq %r13 addq $64, %r12 cmpq $16, %r13 jne .LBB3_5 # %bb.8: movabsq $68719476752, %rdi # imm = 0x1000000010 movq %r14, %rsi movq %rdi, %rdx movq %r15, %rcx movq %rdi, %r8 movq %rbx, %r9 callq _Z6MatMul6MatrixS_S_ movl $.Lstr.2, %edi callq puts@PLT xorl %r14d, %r14d .p2align 4, 0x90 .LBB3_9: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB3_10 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB3_10: # Parent Loop BB3_9 Depth=1 # => This Inner Loop Header: Depth=2 cvttss2si (%rbx,%r15,4), %esi movl $.L.str.10, %edi xorl %eax, %eax callq printf incq %r15 cmpq $16, %r15 jne .LBB3_10 # %bb.11: # in Loop: Header=BB3_9 Depth=1 movl $10, %edi callq putchar@PLT incq %r14 addq $64, %rbx cmpq $16, %r14 jne .LBB3_9 # %bb.12: xorl %eax, %eax addq $8, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end3: .size main, .Lfunc_end3-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB4_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB4_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12MatMulKernel6MatrixS_S_, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end4: .size __hip_module_ctor, .Lfunc_end4-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB5_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB5_2: retq .Lfunc_end5: .size __hip_module_dtor, .Lfunc_end5-__hip_module_dtor .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "ERROR: allocation A %s\n" .size .L.str, 24 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "ERROR: allocation B %s\n" .size .L.str.1, 24 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "ERROR: allocation C %s\n" .size .L.str.2, 24 .type _Z12MatMulKernel6MatrixS_S_,@object # @_Z12MatMulKernel6MatrixS_S_ .section .rodata,"a",@progbits .globl _Z12MatMulKernel6MatrixS_S_ .p2align 3, 0x0 _Z12MatMulKernel6MatrixS_S_: .quad _Z27__device_stub__MatMulKernel6MatrixS_S_ .size _Z12MatMulKernel6MatrixS_S_, 8 .type .L.str.3,@object # @.str.3 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.3: .asciz "ERROR: calculation error %s\n" .size .L.str.3, 29 .type .L.str.6,@object # @.str.6 .L.str.6: .asciz "%2d" .size .L.str.6, 4 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "%4d" .size .L.str.10, 4 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z12MatMulKernel6MatrixS_S_" .size .L__unnamed_1, 28 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "Content of A: " .size .Lstr, 15 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "\n\nContent of B: " .size .Lstr.1, 17 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "\n\nContent of C: " .size .Lstr.2, 17 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z27__device_stub__MatMulKernel6MatrixS_S_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z12MatMulKernel6MatrixS_S_ .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80 Function : _Z12MatMulKernel6MatrixS_S_ .headerflags @"EF_CUDA_TEXMODE_UNIFIED EF_CUDA_64BIT_ADDRESS EF_CUDA_SM80 EF_CUDA_VIRTUAL_SM(EF_CUDA_SM80)" /*0000*/ MOV R1, c[0x0][0x28] ; /* 0x00000a0000017a02 */ /* 0x000fe40000000f00 */ /*0010*/ S2R R0, SR_CTAID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002600 */ /*0020*/ MOV R4, c[0x0][0x160] ; /* 0x0000580000047a02 */ /* 0x000fe20000000f00 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ HFMA2.MMA R24, -RZ, RZ, 0, 0 ; /* 0x00000000ff187435 */ /* 0x000fe200000001ff */ /*0050*/ S2R R3, SR_TID.Y ; /* 0x0000000000037919 */ /* 0x000e220000002200 */ /*0060*/ ISETP.GE.AND P0, PT, R4, 0x1, PT ; /* 0x000000010400780c */ /* 0x000fc60003f06270 */ /*0070*/ S2R R2, SR_CTAID.X ; /* 0x0000000000027919 */ /* 0x000e680000002500 */ /*0080*/ S2R R5, SR_TID.X ; /* 0x0000000000057919 */ /* 0x000e620000002100 */ /*0090*/ IMAD R0, R0, c[0x0][0x4], R3 ; /* 0x0000010000007a24 */ /* 0x001fe400078e0203 */ /*00a0*/ IMAD R3, R2, c[0x0][0x0], R5 ; /* 0x0000000002037a24 */ /* 0x002fc600078e0205 */ /*00b0*/ @!P0 BRA 0xc10 ; /* 0x00000b5000008947 */ /* 0x000fea0003800000 */ /*00c0*/ IADD3 R2, R4.reuse, -0x1, RZ ; /* 0xffffffff04027810 */ /* 0x040fe40007ffe0ff */ /*00d0*/ LOP3.LUT R4, R4, 0x3, RZ, 0xc0, !PT ; /* 0x0000000304047812 */ /* 0x000fe400078ec0ff */ /*00e0*/ ISETP.GE.U32.AND P0, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe40003f06070 */ /*00f0*/ MOV R2, RZ ; /* 0x000000ff00027202 */ /* 0x000fe40000000f00 */ /*0100*/ MOV R24, RZ ; /* 0x000000ff00187202 */ /* 0x000fd20000000f00 */ /*0110*/ @!P0 BRA 0xb00 ; /* 0x000009e000008947 */ /* 0x000fea0003800000 */ /*0120*/ IADD3 R5, -R4, c[0x0][0x160], RZ ; /* 0x0000580004057a10 */ /* 0x000fe20007ffe1ff */ /*0130*/ HFMA2.MMA R8, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff087435 */ /* 0x000fe200000001ff */ /*0140*/ ULDC.64 UR6, c[0x0][0x168] ; /* 0x00005a0000067ab9 */ /* 0x000fe20000000a00 */ /*0150*/ HFMA2.MMA R24, -RZ, RZ, 0, 0 ; /* 0x00000000ff187435 */ /* 0x000fe200000001ff */ /*0160*/ ISETP.GT.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fe20003f04270 */ /*0170*/ IMAD R6, R0, c[0x0][0x160], RZ ; /* 0x0000580000067a24 */ /* 0x000fe200078e02ff */ /*0180*/ MOV R2, RZ ; /* 0x000000ff00027202 */ /* 0x000fca0000000f00 */ /*0190*/ IMAD.WIDE R8, R3, R8, c[0x0][0x178] ; /* 0x00005e0003087625 */ /* 0x000fcc00078e0208 */ /*01a0*/ @!P0 BRA 0x960 ; /* 0x000007b000008947 */ /* 0x000fea0003800000 */ /*01b0*/ ISETP.GT.AND P1, PT, R5, 0xc, PT ; /* 0x0000000c0500780c */ /* 0x000fe40003f24270 */ /*01c0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fd60003f0f070 */ /*01d0*/ @!P1 BRA 0x690 ; /* 0x000004b000009947 */ /* 0x000fea0003800000 */ /*01e0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*01f0*/ MOV R12, UR6 ; /* 0x00000006000c7c02 */ /* 0x000fe20008000f00 */ /*0200*/ LDG.E R21, [R8.64] ; /* 0x0000000408157981 */ /* 0x0000a2000c1e1900 */ /*0210*/ MOV R13, UR7 ; /* 0x00000007000d7c02 */ /* 0x000fca0008000f00 */ /*0220*/ IMAD.WIDE R12, R6, 0x4, R12 ; /* 0x00000004060c7825 */ /* 0x000fca00078e020c */ /*0230*/ LDG.E R20, [R12.64] ; /* 0x000000040c147981 */ /* 0x000ea2000c1e1900 */ /*0240*/ MOV R7, c[0x0][0x170] ; /* 0x00005c0000077a02 */ /* 0x000fc60000000f00 */ /*0250*/ LDG.E R14, [R12.64+0x4] ; /* 0x000004040c0e7981 */ /* 0x000ee4000c1e1900 */ /*0260*/ IMAD.WIDE R10, R7.reuse, 0x4, R8 ; /* 0x00000004070a7825 */ /* 0x040fe400078e0208 */ /*0270*/ LDG.E R27, [R12.64+0x8] ; /* 0x000008040c1b7981 */ /* 0x000f28000c1e1900 */ /*0280*/ LDG.E R15, [R10.64] ; /* 0x000000040a0f7981 */ /* 0x0002e2000c1e1900 */ /*0290*/ IMAD.WIDE R22, R7, 0x4, R10 ; /* 0x0000000407167825 */ /* 0x000fc600078e020a */ /*02a0*/ LDG.E R18, [R12.64+0xc] ; /* 0x00000c040c127981 */ /* 0x000f66000c1e1900 */ /*02b0*/ IMAD.WIDE R28, R7.reuse, 0x4, R22 ; /* 0x00000004071c7825 */ /* 0x040fe200078e0216 */ /*02c0*/ LDG.E R26, [R22.64] ; /* 0x00000004161a7981 */ /* 0x000328000c1e1900 */ /*02d0*/ LDG.E R19, [R28.64] ; /* 0x000000041c137981 */ /* 0x000362000c1e1900 */ /*02e0*/ IMAD.WIDE R16, R7, 0x4, R28 ; /* 0x0000000407107825 */ /* 0x000fc600078e021c */ /*02f0*/ LDG.E R8, [R12.64+0x10] ; /* 0x000010040c087981 */ /* 0x001f68000c1e1900 */ /*0300*/ LDG.E R9, [R16.64] ; /* 0x0000000410097981 */ /* 0x000168000c1e1900 */ /*0310*/ LDG.E R10, [R12.64+0x14] ; /* 0x000014040c0a7981 */ /* 0x002f68000c1e1900 */ /*0320*/ LDG.E R28, [R12.64+0x1c] ; /* 0x00001c040c1c7981 */ /* 0x000f62000c1e1900 */ /*0330*/ IMAD.WIDE R16, R7, 0x4, R16 ; /* 0x0000000407107825 */ /* 0x001fca00078e0210 */ /*0340*/ LDG.E R11, [R16.64] ; /* 0x00000004100b7981 */ /* 0x000562000c1e1900 */ /*0350*/ IMAD.WIDE R22, R7, 0x4, R16 ; /* 0x0000000407167825 */ /* 0x000fc800078e0210 */ /*0360*/ FFMA R16, R21, R20, R24 ; /* 0x0000001415107223 */ /* 0x004fe40000000018 */ /*0370*/ LDG.E R20, [R12.64+0x18] ; /* 0x000018040c147981 */ /* 0x000ea2000c1e1900 */ /*0380*/ IMAD.WIDE R24, R7, 0x4, R22 ; /* 0x0000000407187825 */ /* 0x000fc600078e0216 */ /*0390*/ LDG.E R21, [R22.64] ; /* 0x0000000416157981 */ /* 0x0000a8000c1e1900 */ /*03a0*/ LDG.E R29, [R24.64] ; /* 0x00000004181d7981 */ /* 0x0002a2000c1e1900 */ /*03b0*/ FFMA R16, R15, R14, R16 ; /* 0x0000000e0f107223 */ /* 0x008fe40000000010 */ /*03c0*/ IMAD.WIDE R14, R7.reuse, 0x4, R24 ; /* 0x00000004070e7825 */ /* 0x040fe200078e0218 */ /*03d0*/ LDG.E R23, [R12.64+0x20] ; /* 0x000020040c177981 */ /* 0x001ee6000c1e1900 */ /*03e0*/ FFMA R26, R26, R27, R16 ; /* 0x0000001b1a1a7223 */ /* 0x010fe20000000010 */ /*03f0*/ LDG.E R25, [R12.64+0x24] ; /* 0x000024040c197981 */ /* 0x002f22000c1e1900 */ /*0400*/ IMAD.WIDE R16, R7, 0x4, R14 ; /* 0x0000000407107825 */ /* 0x000fc600078e020e */ /*0410*/ LDG.E R14, [R14.64] ; /* 0x000000040e0e7981 */ /* 0x0000e2000c1e1900 */ /*0420*/ FFMA R26, R19, R18, R26 ; /* 0x00000012131a7223 */ /* 0x020fe4000000001a */ /*0430*/ IMAD.WIDE R18, R7, 0x4, R16 ; /* 0x0000000407127825 */ /* 0x000fe200078e0210 */ /*0440*/ LDG.E R22, [R12.64+0x28] ; /* 0x000028040c167981 */ /* 0x000f66000c1e1900 */ /*0450*/ FFMA R26, R9, R8, R26 ; /* 0x00000008091a7223 */ /* 0x000fe2000000001a */ /*0460*/ LDG.E R16, [R16.64] ; /* 0x0000000410107981 */ /* 0x000322000c1e1900 */ /*0470*/ IMAD.WIDE R8, R7, 0x4, R18 ; /* 0x0000000407087825 */ /* 0x000fc600078e0212 */ /*0480*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x000368000c1e1900 */ /*0490*/ LDG.E R24, [R8.64] ; /* 0x0000000408187981 */ /* 0x000568000c1e1900 */ /*04a0*/ LDG.E R15, [R12.64+0x2c] ; /* 0x00002c040c0f7981 */ /* 0x001f62000c1e1900 */ /*04b0*/ FFMA R26, R11, R10, R26 ; /* 0x0000000a0b1a7223 */ /* 0x000fe4000000001a */ /*04c0*/ IMAD.WIDE R10, R7, 0x4, R8 ; /* 0x00000004070a7825 */ /* 0x000fe200078e0208 */ /*04d0*/ LDG.E R17, [R12.64+0x30] ; /* 0x000030040c117981 */ /* 0x002f66000c1e1900 */ /*04e0*/ FFMA R26, R21, R20, R26 ; /* 0x00000014151a7223 */ /* 0x004fc4000000001a */ /*04f0*/ IMAD.WIDE R20, R7, 0x4, R10 ; /* 0x0000000407147825 */ /* 0x000fe400078e020a */ /*0500*/ LDG.E R10, [R10.64] ; /* 0x000000040a0a7981 */ /* 0x0000a4000c1e1900 */ /*0510*/ FFMA R28, R29, R28, R26 ; /* 0x0000001c1d1c7223 */ /* 0x000fe4000000001a */ /*0520*/ IMAD.WIDE R26, R7.reuse, 0x4, R20 ; /* 0x00000004071a7825 */ /* 0x040fe200078e0214 */ /*0530*/ LDG.E R29, [R12.64+0x34] ; /* 0x000034040c1d7981 */ /* 0x000ea8000c1e1900 */ /*0540*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x0002a2000c1e1900 */ /*0550*/ IMAD.WIDE R8, R7, 0x4, R26 ; /* 0x0000000407087825 */ /* 0x000fc600078e021a */ /*0560*/ LDG.E R19, [R26.64] ; /* 0x000000041a137981 */ /* 0x0006a8000c1e1900 */ /*0570*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x0010a8000c1e1900 */ /*0580*/ LDG.E R21, [R12.64+0x38] ; /* 0x000038040c157981 */ /* 0x002ea8000c1e1900 */ /*0590*/ LDG.E R26, [R12.64+0x3c] ; /* 0x00003c040c1a7981 */ /* 0x008ee2000c1e1900 */ /*05a0*/ FFMA R14, R14, R23, R28 ; /* 0x000000170e0e7223 */ /* 0x000fc8000000001c */ /*05b0*/ FFMA R25, R16, R25, R14 ; /* 0x0000001910197223 */ /* 0x010fe2000000000e */ /*05c0*/ IADD3 R5, R5, -0x10, RZ ; /* 0xfffffff005057810 */ /* 0x000fc60007ffe0ff */ /*05d0*/ FFMA R18, R18, R22, R25 ; /* 0x0000001612127223 */ /* 0x020fe20000000019 */ /*05e0*/ ISETP.GT.AND P1, PT, R5, 0xc, PT ; /* 0x0000000c0500780c */ /* 0x000fc60003f24270 */ /*05f0*/ FFMA R15, R24, R15, R18 ; /* 0x0000000f180f7223 */ /* 0x000fe20000000012 */ /*0600*/ UIADD3 UR6, UP0, UR6, 0x40, URZ ; /* 0x0000004006067890 */ /* 0x000fe2000ff1e03f */ /*0610*/ IMAD.WIDE R8, R7, 0x4, R8 ; /* 0x0000000407087825 */ /* 0x001fc600078e0208 */ /*0620*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0630*/ IADD3 R2, R2, 0x10, RZ ; /* 0x0000001002027810 */ /* 0x000fe20007ffe0ff */ /*0640*/ FFMA R10, R10, R17, R15 ; /* 0x000000110a0a7223 */ /* 0x004fc8000000000f */ /*0650*/ FFMA R10, R20, R29, R10 ; /* 0x0000001d140a7223 */ /* 0x000fc8000000000a */ /*0660*/ FFMA R10, R19, R21, R10 ; /* 0x00000015130a7223 */ /* 0x000fc8000000000a */ /*0670*/ FFMA R24, R11, R26, R10 ; /* 0x0000001a0b187223 */ /* 0x008fe2000000000a */ /*0680*/ @P1 BRA 0x1f0 ; /* 0xfffffb6000001947 */ /* 0x000fea000383ffff */ /*0690*/ ISETP.GT.AND P1, PT, R5, 0x4, PT ; /* 0x000000040500780c */ /* 0x000fda0003f24270 */ /*06a0*/ @!P1 BRA 0x940 ; /* 0x0000029000009947 */ /* 0x000fea0003800000 */ /*06b0*/ MOV R7, c[0x0][0x170] ; /* 0x00005c0000077a02 */ /* 0x000fe20000000f00 */ /*06c0*/ LDG.E R23, [R8.64] ; /* 0x0000000408177981 */ /* 0x0000a2000c1e1900 */ /*06d0*/ MOV R10, UR6 ; /* 0x00000006000a7c02 */ /* 0x000fe40008000f00 */ /*06e0*/ MOV R11, UR7 ; /* 0x00000007000b7c02 */ /* 0x000fe20008000f00 */ /*06f0*/ IMAD.WIDE R16, R7, 0x4, R8 ; /* 0x0000000407107825 */ /* 0x000fc800078e0208 */ /*0700*/ IMAD.WIDE R10, R6, 0x4, R10 ; /* 0x00000004060a7825 */ /* 0x000fc800078e020a */ /*0710*/ IMAD.WIDE R12, R7.reuse, 0x4, R16 ; /* 0x00000004070c7825 */ /* 0x040fe200078e0210 */ /*0720*/ LDG.E R22, [R10.64] ; /* 0x000000040a167981 */ /* 0x000ea8000c1e1900 */ /*0730*/ LDG.E R16, [R16.64] ; /* 0x0000000410107981 */ /* 0x0002e2000c1e1900 */ /*0740*/ IMAD.WIDE R14, R7, 0x4, R12 ; /* 0x00000004070e7825 */ /* 0x000fc600078e020c */ /*0750*/ LDG.E R25, [R10.64+0x4] ; /* 0x000004040a197981 */ /* 0x000ee6000c1e1900 */ /*0760*/ IMAD.WIDE R18, R7.reuse, 0x4, R14 ; /* 0x0000000407127825 */ /* 0x040fe200078e020e */ /*0770*/ LDG.E R26, [R12.64] ; /* 0x000000040c1a7981 */ /* 0x000968000c1e1900 */ /*0780*/ LDG.E R27, [R10.64+0x8] ; /* 0x000008040a1b7981 */ /* 0x000f62000c1e1900 */ /*0790*/ IMAD.WIDE R20, R7, 0x4, R18 ; /* 0x0000000407147825 */ /* 0x000fc600078e0212 */ /*07a0*/ LDG.E R14, [R14.64] ; /* 0x000000040e0e7981 */ /* 0x000368000c1e1900 */ /*07b0*/ LDG.E R29, [R10.64+0xc] ; /* 0x00000c040a1d7981 */ /* 0x000f62000c1e1900 */ /*07c0*/ IMAD.WIDE R8, R7, 0x4, R20 ; /* 0x0000000407087825 */ /* 0x001fc600078e0214 */ /*07d0*/ LDG.E R18, [R18.64] ; /* 0x0000000412127981 */ /* 0x000168000c1e1900 */ /*07e0*/ LDG.E R28, [R10.64+0x10] ; /* 0x000010040a1c7981 */ /* 0x000f62000c1e1900 */ /*07f0*/ IMAD.WIDE R12, R7, 0x4, R8 ; /* 0x00000004070c7825 */ /* 0x010fc600078e0208 */ /*0800*/ LDG.E R20, [R20.64] ; /* 0x0000000414147981 */ /* 0x000968000c1e1900 */ /*0810*/ LDG.E R15, [R10.64+0x14] ; /* 0x000014040a0f7981 */ /* 0x002f68000c1e1900 */ /*0820*/ LDG.E R17, [R8.64] ; /* 0x0000000408117981 */ /* 0x000368000c1e1900 */ /*0830*/ LDG.E R21, [R10.64+0x1c] ; /* 0x00001c040a157981 */ /* 0x010f28000c1e1900 */ /*0840*/ LDG.E R19, [R12.64] ; /* 0x000000040c137981 */ /* 0x001f28000c1e1900 */ /*0850*/ LDG.E R8, [R10.64+0x18] ; /* 0x000018040a087981 */ /* 0x002f22000c1e1900 */ /*0860*/ UIADD3 UR6, UP0, UR6, 0x20, URZ ; /* 0x0000002006067890 */ /* 0x000fe2000ff1e03f */ /*0870*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fc40003f0e170 */ /*0880*/ IADD3 R2, R2, 0x8, RZ ; /* 0x0000000802027810 */ /* 0x000fe40007ffe0ff */ /*0890*/ IADD3 R5, R5, -0x8, RZ ; /* 0xfffffff805057810 */ /* 0x000fe20007ffe0ff */ /*08a0*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*08b0*/ FFMA R22, R23, R22, R24 ; /* 0x0000001617167223 */ /* 0x004fc80000000018 */ /*08c0*/ FFMA R16, R16, R25, R22 ; /* 0x0000001910107223 */ /* 0x008fc80000000016 */ /*08d0*/ FFMA R16, R26, R27, R16 ; /* 0x0000001b1a107223 */ /* 0x020fc80000000010 */ /*08e0*/ FFMA R29, R14, R29, R16 ; /* 0x0000001d0e1d7223 */ /* 0x000fc80000000010 */ /*08f0*/ FFMA R18, R18, R28, R29 ; /* 0x0000001c12127223 */ /* 0x000fc8000000001d */ /*0900*/ FFMA R15, R20, R15, R18 ; /* 0x0000000f140f7223 */ /* 0x000fc80000000012 */ /*0910*/ FFMA R24, R17, R8, R15 ; /* 0x0000000811187223 */ /* 0x010fe4000000000f */ /*0920*/ IMAD.WIDE R8, R7, 0x4, R12 ; /* 0x0000000407087825 */ /* 0x000fc800078e020c */ /*0930*/ FFMA R24, R19, R21, R24 ; /* 0x0000001513187223 */ /* 0x000fe40000000018 */ /*0940*/ ISETP.NE.OR P0, PT, R5, RZ, P0 ; /* 0x000000ff0500720c */ /* 0x000fda0000705670 */ /*0950*/ @!P0 BRA 0xb00 ; /* 0x000001a000008947 */ /* 0x000fea0003800000 */ /*0960*/ MOV R10, UR6 ; /* 0x00000006000a7c02 */ /* 0x000fe40008000f00 */ /*0970*/ MOV R11, UR7 ; /* 0x00000007000b7c02 */ /* 0x000fe40008000f00 */ /*0980*/ MOV R7, c[0x0][0x170] ; /* 0x00005c0000077a02 */ /* 0x000fc60000000f00 */ /*0990*/ IMAD.WIDE R10, R6, 0x4, R10 ; /* 0x00000004060a7825 */ /* 0x000fc800078e020a */ /*09a0*/ IMAD.WIDE R16, R7.reuse, 0x4, R8 ; /* 0x0000000407107825 */ /* 0x040fe200078e0208 */ /*09b0*/ LDG.E R18, [R10.64] ; /* 0x000000040a127981 */ /* 0x000ea8000c1e1900 */ /*09c0*/ LDG.E R9, [R8.64] ; /* 0x0000000408097981 */ /* 0x000ea2000c1e1900 */ /*09d0*/ IMAD.WIDE R12, R7, 0x4, R16 ; /* 0x00000004070c7825 */ /* 0x000fc600078e0210 */ /*09e0*/ LDG.E R17, [R16.64] ; /* 0x0000000410117981 */ /* 0x000ee8000c1e1900 */ /*09f0*/ LDG.E R19, [R10.64+0x4] ; /* 0x000004040a137981 */ /* 0x000ee2000c1e1900 */ /*0a00*/ IMAD.WIDE R14, R7, 0x4, R12 ; /* 0x00000004070e7825 */ /* 0x000fc600078e020c */ /*0a10*/ LDG.E R21, [R12.64] ; /* 0x000000040c157981 */ /* 0x000f28000c1e1900 */ /*0a20*/ LDG.E R20, [R10.64+0x8] ; /* 0x000008040a147981 */ /* 0x000f28000c1e1900 */ /*0a30*/ LDG.E R22, [R10.64+0xc] ; /* 0x00000c040a167981 */ /* 0x000f68000c1e1900 */ /*0a40*/ LDG.E R23, [R14.64] ; /* 0x000000040e177981 */ /* 0x000f62000c1e1900 */ /*0a50*/ IADD3 R5, R5, -0x4, RZ ; /* 0xfffffffc05057810 */ /* 0x000fc80007ffe0ff */ /*0a60*/ ISETP.NE.AND P0, PT, R5, RZ, PT ; /* 0x000000ff0500720c */ /* 0x000fe20003f05270 */ /*0a70*/ UIADD3 UR6, UP0, UR6, 0x10, URZ ; /* 0x0000001006067890 */ /* 0x000fe2000ff1e03f */ /*0a80*/ IADD3 R2, R2, 0x4, RZ ; /* 0x0000000402027810 */ /* 0x000fc60007ffe0ff */ /*0a90*/ UIADD3.X UR7, URZ, UR7, URZ, UP0, !UPT ; /* 0x000000073f077290 */ /* 0x000fe200087fe43f */ /*0aa0*/ FFMA R18, R9, R18, R24 ; /* 0x0000001209127223 */ /* 0x004fc80000000018 */ /*0ab0*/ FFMA R18, R17, R19, R18 ; /* 0x0000001311127223 */ /* 0x008fe40000000012 */ /*0ac0*/ IMAD.WIDE R8, R7, 0x4, R14 ; /* 0x0000000407087825 */ /* 0x000fc800078e020e */ /*0ad0*/ FFMA R18, R21, R20, R18 ; /* 0x0000001415127223 */ /* 0x010fc80000000012 */ /*0ae0*/ FFMA R24, R23, R22, R18 ; /* 0x0000001617187223 */ /* 0x020fe20000000012 */ /*0af0*/ @P0 BRA 0x960 ; /* 0xfffffe6000000947 */ /* 0x000fea000383ffff */ /*0b00*/ ISETP.NE.AND P0, PT, R4, RZ, PT ; /* 0x000000ff0400720c */ /* 0x000fda0003f05270 */ /*0b10*/ @!P0 BRA 0xc10 ; /* 0x000000f000008947 */ /* 0x000fea0003800000 */ /*0b20*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0b30*/ IMAD R6, R0, c[0x0][0x160], R2 ; /* 0x0000580000067a24 */ /* 0x000fe400078e0202 */ /*0b40*/ IMAD R2, R2, c[0x0][0x170], R3 ; /* 0x00005c0002027a24 */ /* 0x000fce00078e0203 */ /*0b50*/ IMAD.WIDE R6, R6, R9, c[0x0][0x168] ; /* 0x00005a0006067625 */ /* 0x000fc800078e0209 */ /*0b60*/ IMAD.WIDE R8, R2, R9, c[0x0][0x178] ; /* 0x00005e0002087625 */ /* 0x000fca00078e0209 */ /*0b70*/ LDG.E R5, [R8.64] ; /* 0x0000000408057981 */ /* 0x0000a8000c1e1900 */ /*0b80*/ LDG.E R2, [R6.64] ; /* 0x0000000406027981 */ /* 0x0002a2000c1e1900 */ /*0b90*/ IADD3 R4, R4, -0x1, RZ ; /* 0xffffffff04047810 */ /* 0x000fe40007ffe0ff */ /*0ba0*/ MOV R11, c[0x0][0x170] ; /* 0x00005c00000b7a02 */ /* 0x000fe40000000f00 */ /*0bb0*/ ISETP.NE.AND P0, PT, R4, RZ, PT ; /* 0x000000ff0400720c */ /* 0x000fc60003f05270 */ /*0bc0*/ IMAD.WIDE R8, R11, 0x4, R8 ; /* 0x000000040b087825 */ /* 0x001fe200078e0208 */ /*0bd0*/ IADD3 R6, P1, R6, 0x4, RZ ; /* 0x0000000406067810 */ /* 0x002fc80007f3e0ff */ /*0be0*/ IADD3.X R7, RZ, R7, RZ, P1, !PT ; /* 0x00000007ff077210 */ /* 0x000fe20000ffe4ff */ /*0bf0*/ FFMA R24, R5, R2, R24 ; /* 0x0000000205187223 */ /* 0x004fc80000000018 */ /*0c00*/ @P0 BRA 0xb70 ; /* 0xffffff6000000947 */ /* 0x000fea000383ffff */ /*0c10*/ MOV R2, 0x4 ; /* 0x0000000400027802 */ /* 0x000fe20000000f00 */ /*0c20*/ IMAD R3, R0, c[0x0][0x180], R3 ; /* 0x0000600000037a24 */ /* 0x000fc800078e0203 */ /*0c30*/ IMAD.WIDE R2, R3, R2, c[0x0][0x188] ; /* 0x0000620003027625 */ /* 0x000fca00078e0202 */ /*0c40*/ STG.E [R2.64], R24 ; /* 0x0000001802007986 */ /* 0x000fe2000c101904 */ /*0c50*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0c60*/ BRA 0xc60; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0c70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c80*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c90*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ca0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0ce0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0cf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z12MatMulKernel6MatrixS_S_ .globl _Z12MatMulKernel6MatrixS_S_ .p2align 8 .type _Z12MatMulKernel6MatrixS_S_,@function _Z12MatMulKernel6MatrixS_S_: s_clause 0x2 s_load_b32 s4, s[0:1], 0x3c s_load_b32 s6, s[0:1], 0x0 s_load_b64 s[2:3], s[0:1], 0x28 v_bfe_u32 v2, v0, 10, 10 v_and_b32_e32 v3, 0x3ff, v0 s_waitcnt lgkmcnt(0) s_lshr_b32 s5, s4, 16 s_and_b32 s4, s4, 0xffff s_delay_alu instid0(VALU_DEP_1) v_mad_u64_u32 v[0:1], null, s15, s5, v[2:3] v_mad_u64_u32 v[1:2], null, s14, s4, v[3:4] s_cmp_lt_i32 s6, 1 s_cbranch_scc1 .LBB0_3 s_load_b64 s[8:9], s[0:1], 0x8 s_delay_alu instid0(VALU_DEP_2) v_mul_lo_u32 v2, v0, s6 s_clause 0x1 s_load_b32 s7, s[0:1], 0x10 s_load_b64 s[4:5], s[0:1], 0x18 v_mov_b32_e32 v6, 0 v_mov_b32_e32 v4, v1 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v3, 31, v2 v_lshlrev_b64 v[2:3], 2, v[2:3] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_add_co_u32 v2, vcc_lo, s8, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s9, v3, vcc_lo .p2align 6 .LBB0_2: v_ashrrev_i32_e32 v5, 31, v4 s_add_i32 s6, s6, -1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_1) s_cmp_eq_u32 s6, 0 v_lshlrev_b64 v[7:8], 2, v[4:5] v_add_nc_u32_e32 v4, s7, v4 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_u32 v7, vcc_lo, s4, v7 v_add_co_ci_u32_e32 v8, vcc_lo, s5, v8, vcc_lo global_load_b32 v5, v[2:3], off global_load_b32 v7, v[7:8], off v_add_co_u32 v2, vcc_lo, v2, 4 v_add_co_ci_u32_e32 v3, vcc_lo, 0, v3, vcc_lo s_waitcnt vmcnt(0) v_fmac_f32_e32 v6, v5, v7 s_cbranch_scc0 .LBB0_2 s_branch .LBB0_4 .LBB0_3: v_mov_b32_e32 v6, 0 .LBB0_4: s_load_b32 s0, s[0:1], 0x20 s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[2:3], null, v0, s0, v[1:2] v_ashrrev_i32_e32 v3, 31, v2 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[2:3] v_add_co_u32 v0, vcc_lo, s2, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v1, vcc_lo, s3, v1, vcc_lo global_store_b32 v[0:1], v6, off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z12MatMulKernel6MatrixS_S_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 304 .amdhsa_user_sgpr_count 14 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 1 .amdhsa_user_sgpr_dispatch_id 0 .amdhsa_user_sgpr_private_segment_size 0 .amdhsa_wavefront_size32 1 .amdhsa_uses_dynamic_stack 0 .amdhsa_enable_private_segment 0 .amdhsa_system_sgpr_workgroup_id_x 1 .amdhsa_system_sgpr_workgroup_id_y 1 .amdhsa_system_sgpr_workgroup_id_z 0 .amdhsa_system_sgpr_workgroup_info 0 .amdhsa_system_vgpr_workitem_id 1 .amdhsa_next_free_vgpr 9 .amdhsa_next_free_sgpr 16 .amdhsa_float_round_mode_32 0 .amdhsa_float_round_mode_16_64 0 .amdhsa_float_denorm_mode_32 3 .amdhsa_float_denorm_mode_16_64 3 .amdhsa_dx10_clamp 1 .amdhsa_ieee_mode 1 .amdhsa_fp16_overflow 0 .amdhsa_workgroup_processor_mode 1 .amdhsa_memory_ordered 1 .amdhsa_forward_progress 0 .amdhsa_shared_vgpr_count 0 .amdhsa_exception_fp_ieee_invalid_op 0 .amdhsa_exception_fp_denorm_src 0 .amdhsa_exception_fp_ieee_div_zero 0 .amdhsa_exception_fp_ieee_overflow 0 .amdhsa_exception_fp_ieee_underflow 0 .amdhsa_exception_fp_ieee_inexact 0 .amdhsa_exception_int_div_zero 0 .end_amdhsa_kernel .text .Lfunc_end0: .size _Z12MatMulKernel6MatrixS_S_, .Lfunc_end0-_Z12MatMulKernel6MatrixS_S_ .section .AMDGPU.csdata,"",@progbits .text .p2alignl 7, 3214868480 .fill 96, 4, 3214868480 .type __hip_cuid_,@object .section .bss,"aw",@nobits .globl __hip_cuid_ __hip_cuid_: .byte 0 .size __hip_cuid_, 1 .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym __hip_cuid_ .amdgpu_metadata --- amdhsa.kernels: - .args: - .offset: 0 .size: 16 .value_kind: by_value - .offset: 16 .size: 16 .value_kind: by_value - .offset: 32 .size: 16 .value_kind: by_value - .offset: 48 .size: 4 .value_kind: hidden_block_count_x - .offset: 52 .size: 4 .value_kind: hidden_block_count_y - .offset: 56 .size: 4 .value_kind: hidden_block_count_z - .offset: 60 .size: 2 .value_kind: hidden_group_size_x - .offset: 62 .size: 2 .value_kind: hidden_group_size_y - .offset: 64 .size: 2 .value_kind: hidden_group_size_z - .offset: 66 .size: 2 .value_kind: hidden_remainder_x - .offset: 68 .size: 2 .value_kind: hidden_remainder_y - .offset: 70 .size: 2 .value_kind: hidden_remainder_z - .offset: 88 .size: 8 .value_kind: hidden_global_offset_x - .offset: 96 .size: 8 .value_kind: hidden_global_offset_y - .offset: 104 .size: 8 .value_kind: hidden_global_offset_z - .offset: 112 .size: 2 .value_kind: hidden_grid_dims .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 304 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z12MatMulKernel6MatrixS_S_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z12MatMulKernel6MatrixS_S_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 9 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata
You are a helpful coding assistant specialized in converting from CUDA host assembly to AMD host assembly.
.file "tmpxft_001bbc4f_00000000-6_mm.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2062: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2062: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z6myrandv .type _Z6myrandv, @function _Z6myrandv: .LFB2058: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call rand@PLT movl %eax, %edx cltq imulq $-1610612725, %rax, %rax shrq $32, %rax addl %edx, %eax sarl $27, %eax sarl $31, %edx subl %edx, %eax addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2058: .size _Z6myrandv, .-_Z6myrandv .globl _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ .type _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_, @function _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_: .LFB2084: .cfi_startproc endbr64 subq $104, %rsp .cfi_def_cfa_offset 112 movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax movq %rdi, 64(%rsp) movq %rsi, 72(%rsp) movq %rdx, 80(%rsp) movl $1, 16(%rsp) movl $1, 20(%rsp) movl $1, 24(%rsp) movl $1, 28(%rsp) movl $1, 32(%rsp) movl $1, 36(%rsp) leaq 8(%rsp), %rcx movq %rsp, %rdx leaq 28(%rsp), %rsi leaq 16(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L9 .L5: movq 88(%rsp), %rax subq %fs:40, %rax jne .L10 addq $104, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L9: .cfi_restore_state pushq 8(%rsp) .cfi_def_cfa_offset 120 pushq 8(%rsp) .cfi_def_cfa_offset 128 leaq 80(%rsp), %r9 movq 44(%rsp), %rcx movl 52(%rsp), %r8d movq 32(%rsp), %rsi movl 40(%rsp), %edx leaq _Z12MatMulKernel6MatrixS_S_(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 112 jmp .L5 .L10: call __stack_chk_fail@PLT .cfi_endproc .LFE2084: .size _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_, .-_Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ .globl _Z12MatMulKernel6MatrixS_S_ .type _Z12MatMulKernel6MatrixS_S_, @function _Z12MatMulKernel6MatrixS_S_: .LFB2085: .cfi_startproc endbr64 subq $56, %rsp .cfi_def_cfa_offset 64 movq %rdi, 32(%rsp) movq %rsi, 40(%rsp) movq %rdx, 16(%rsp) movq %rcx, 24(%rsp) movq %r8, (%rsp) movq %r9, 8(%rsp) movq %rsp, %rdx leaq 16(%rsp), %rsi leaq 32(%rsp), %rdi call _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ addq $56, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _Z12MatMulKernel6MatrixS_S_, .-_Z12MatMulKernel6MatrixS_S_ .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "ERROR: allocation A %s\n" .LC1: .string "ERROR: allocation B %s\n" .LC2: .string "ERROR: allocation C %s\n" .LC3: .string "ERROR: calculation error %s\n" .text .globl _Z6MatMul6MatrixS_S_ .type _Z6MatMul6MatrixS_S_, @function _Z6MatMul6MatrixS_S_: .LFB2057: .cfi_startproc endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $200, %rsp .cfi_def_cfa_offset 256 movq %rdi, %rbx movq %rsi, 8(%rsp) movq %rdx, %r13 movq %rcx, 16(%rsp) movq %r8, %rbp movq %r9, 24(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax movq %rdi, %r14 sarq $32, %r14 movq %rdx, %r12 sarq $32, %r12 movq %r8, %r15 sarq $32, %r15 movl $0, %edi call cudaSetDevice@PLT call cudaDeviceSynchronize@PLT leaq 48(%rsp), %rsi leaq 40(%rsp), %rdi call cudaMemGetInfo@PLT movl %ebx, 80(%rsp) movl %r14d, 84(%rsp) imull %r14d, %ebx movslq %ebx, %rbx salq $2, %rbx leaq 88(%rsp), %rdi movq %rbx, %rsi call cudaMalloc@PLT call cudaGetLastError@PLT testl %eax, %eax jne .L21 movl $1, %ecx movq %rbx, %rdx movq 8(%rsp), %rsi movq 88(%rsp), %rdi call cudaMemcpy@PLT movl %r13d, 96(%rsp) movl %r12d, 100(%rsp) imull %r13d, %r12d movslq %r12d, %r12 salq $2, %r12 leaq 104(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT call cudaGetLastError@PLT testl %eax, %eax jne .L22 movl $1, %ecx movq %r12, %rdx movq 16(%rsp), %rsi movq 104(%rsp), %rdi call cudaMemcpy@PLT movl %ebp, 112(%rsp) movl %r15d, 116(%rsp) imull %r15d, %ebp movslq %ebp, %rbp salq $2, %rbp call cudaGetLastError@PLT movl %eax, %ebx leaq 120(%rsp), %rdi movq %rbp, %rsi call cudaMalloc@PLT testl %ebx, %ebx jne .L23 shrl $4, %r13d movl %r13d, 68(%rsp) shrl $4, %r14d movl %r14d, 72(%rsp) movl $16, 56(%rsp) movl $16, 60(%rsp) movl $0, %r9d movl $0, %r8d movq 56(%rsp), %rdx movl $1, %ecx movq 68(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L24 .L17: call cudaDeviceSynchronize@PLT call cudaGetLastError@PLT testl %eax, %eax jne .L25 movl $2, %ecx movq %rbp, %rdx movq 120(%rsp), %rsi movq 24(%rsp), %rdi call cudaMemcpy@PLT movq 88(%rsp), %rdi call cudaFree@PLT movq 104(%rsp), %rdi call cudaFree@PLT movq 120(%rsp), %rdi call cudaFree@PLT movq 184(%rsp), %rax subq %fs:40, %rax jne .L26 addq $200, %rsp .cfi_remember_state .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .L21: .cfi_restore_state movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC0(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L22: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC1(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L23: movl %ebx, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC2(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L24: movdqa 80(%rsp), %xmm0 movaps %xmm0, 128(%rsp) movdqa 96(%rsp), %xmm1 movaps %xmm1, 144(%rsp) movdqa 112(%rsp), %xmm2 movaps %xmm2, 160(%rsp) leaq 160(%rsp), %rdx leaq 144(%rsp), %rsi leaq 128(%rsp), %rdi call _Z41__device_stub__Z12MatMulKernel6MatrixS_S_R6MatrixS0_S0_ jmp .L17 .L25: movl %eax, %edi call cudaGetErrorString@PLT movq %rax, %rcx leaq .LC3(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $-1, %edi call exit@PLT .L26: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size _Z6MatMul6MatrixS_S_, .-_Z6MatMul6MatrixS_S_ .section .rodata.str1.1 .LC4: .string "Content of A: \n" .LC5: .string "%2d" .LC6: .string "\n" .LC7: .string "\n\nContent of B: \n" .LC8: .string "\n\nContent of C: \n" .LC9: .string "%4d" .text .globl main .type main, @function main: .LFB2059: .cfi_startproc endbr64 pushq %r15 .cfi_def_cfa_offset 16 .cfi_offset 15, -16 pushq %r14 .cfi_def_cfa_offset 24 .cfi_offset 14, -24 pushq %r13 .cfi_def_cfa_offset 32 .cfi_offset 13, -32 pushq %r12 .cfi_def_cfa_offset 40 .cfi_offset 12, -40 pushq %rbp .cfi_def_cfa_offset 48 .cfi_offset 6, -48 pushq %rbx .cfi_def_cfa_offset 56 .cfi_offset 3, -56 subq $24, %rsp .cfi_def_cfa_offset 80 movl $0, %edi call srand@PLT movl $1024, %edi call malloc@PLT movq %rax, %rbx movq %rax, (%rsp) movl $1024, %edi call malloc@PLT movq %rax, %r13 movl $1024, %edi call malloc@PLT movq %rax, 8(%rsp) leaq .LC4(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 64(%rbx), %rbp leaq 1088(%rbx), %r15 leaq .LC5(%rip), %r12 leaq .LC6(%rip), %r14 .L28: leaq -64(%rbp), %rbx .L29: call _Z6myrandv pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, (%rbx) cvttss2sil %xmm0, %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L29 movq %r14, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $64, %rbp cmpq %r15, %rbp jne .L28 leaq .LC7(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 64(%r13), %rbp leaq 1088(%r13), %r15 leaq .LC5(%rip), %r12 leaq .LC6(%rip), %r14 .L31: leaq -64(%rbp), %rbx .L32: call _Z6myrandv pxor %xmm0, %xmm0 cvtsi2ssl %eax, %xmm0 movss %xmm0, (%rbx) cvttss2sil %xmm0, %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L32 movq %r14, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $64, %rbp cmpq %r15, %rbp jne .L31 movabsq $68719476736, %r8 movq %r8, %rdi orq $16, %rdi movq %rdi, %rdx movq %rdi, %r8 movq 8(%rsp), %rbx movq %rbx, %r9 movq %r13, %rcx movq (%rsp), %rsi call _Z6MatMul6MatrixS_S_ leaq .LC8(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 64(%rbx), %rbp leaq 1088(%rbx), %r13 leaq .LC9(%rip), %r12 leaq .LC6(%rip), %r14 .L34: leaq -64(%rbp), %rbx .L35: cvttss2sil (%rbx), %edx movq %r12, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L35 movq %r14, %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addq $64, %rbp cmpq %r13, %rbp jne .L34 movl $0, %eax addq $24, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %rbp .cfi_def_cfa_offset 40 popq %r12 .cfi_def_cfa_offset 32 popq %r13 .cfi_def_cfa_offset 24 popq %r14 .cfi_def_cfa_offset 16 popq %r15 .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2059: .size main, .-main .section .rodata.str1.1 .LC10: .string "_Z12MatMulKernel6MatrixS_S_" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2087: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC10(%rip), %rdx movq %rdx, %rcx leaq _Z12MatMulKernel6MatrixS_S_(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2087: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4:
.text .file "mm.hip" .globl _Z6MatMul6MatrixS_S_ # -- Begin function _Z6MatMul6MatrixS_S_ .p2align 4, 0x90 .type _Z6MatMul6MatrixS_S_,@function _Z6MatMul6MatrixS_S_: # @_Z6MatMul6MatrixS_S_ .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 subq $200, %rsp .cfi_def_cfa_offset 256 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 movq %r9, 48(%rsp) # 8-byte Spill movq %r8, %r12 movq %rcx, 56(%rsp) # 8-byte Spill movq %rdx, %r13 movq %rsi, %rbp movq %rdi, %r15 movq %rdi, %rbx shrq $32, %rbx xorl %edi, %edi callq hipSetDevice callq hipDeviceSynchronize leaq 144(%rsp), %rdi leaq 136(%rsp), %rsi callq hipMemGetInfo movq %r15, 32(%rsp) imull %r15d, %ebx movslq %ebx, %rbx shlq $2, %rbx leaq 40(%rsp), %rdi movq %rbx, %rsi callq hipMalloc callq hipGetLastError testl %eax, %eax jne .LBB0_1 # %bb.4: movq %r13, %r14 shrq $32, %r14 movq 40(%rsp), %rdi movq %rbp, %rsi movq %rbx, %rdx movl $1, %ecx callq hipMemcpy movq %r13, 16(%rsp) imull %r13d, %r14d movslq %r14d, %rbx shlq $2, %rbx leaq 24(%rsp), %rdi movq %rbx, %rsi callq hipMalloc callq hipGetLastError testl %eax, %eax jne .LBB0_5 # %bb.6: movq %r12, %r14 shrq $32, %r14 movq 24(%rsp), %rdi movq 56(%rsp), %rsi # 8-byte Reload movq %rbx, %rdx movl $1, %ecx callq hipMemcpy movl %r12d, (%rsp) movl %r14d, 4(%rsp) imull %r12d, %r14d movslq %r14d, %r12 shlq $2, %r12 callq hipGetLastError movl %eax, %ebx leaq 8(%rsp), %rdi movq %r12, %rsi callq hipMalloc testl %ebx, %ebx jne .LBB0_7 # %bb.8: shrl $4, %r13d shrq $4, %r15 movabsq $1152921500311879680, %rdi # imm = 0xFFFFFFF00000000 andq %r15, %rdi orq %r13, %rdi movabsq $68719476752, %rdx # imm = 0x1000000010 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB0_10 # %bb.9: movups 32(%rsp), %xmm0 movups 16(%rsp), %xmm1 movups (%rsp), %xmm2 movups %xmm0, 184(%rsp) movups %xmm1, 168(%rsp) movups %xmm2, 152(%rsp) leaq 184(%rsp), %rax movq %rax, 112(%rsp) leaq 168(%rsp), %rax movq %rax, 120(%rsp) leaq 152(%rsp), %rax movq %rax, 128(%rsp) leaq 96(%rsp), %rdi leaq 80(%rsp), %rsi leaq 72(%rsp), %rdx leaq 64(%rsp), %rcx callq __hipPopCallConfiguration movq 96(%rsp), %rsi movl 104(%rsp), %edx movq 80(%rsp), %rcx movl 88(%rsp), %r8d leaq 112(%rsp), %r9 movl $_Z12MatMulKernel6MatrixS_S_, %edi pushq 64(%rsp) .cfi_adjust_cfa_offset 8 pushq 80(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB0_10: callq hipDeviceSynchronize callq hipGetLastError testl %eax, %eax jne .LBB0_11 # %bb.12: movq 8(%rsp), %rsi movq 48(%rsp), %rdi # 8-byte Reload movq %r12, %rdx movl $2, %ecx callq hipMemcpy movq 40(%rsp), %rdi callq hipFree movq 24(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree addq $200, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .LBB0_1: .cfi_def_cfa_offset 256 movq stderr(%rip), %rbx movl %eax, %edi callq hipGetErrorString movl $.L.str, %esi jmp .LBB0_2 .LBB0_5: movq stderr(%rip), %rbx movl %eax, %edi callq hipGetErrorString movl $.L.str.1, %esi jmp .LBB0_2 .LBB0_7: movq stderr(%rip), %r14 movl %ebx, %edi callq hipGetErrorString movl $.L.str.2, %esi movq %r14, %rdi jmp .LBB0_3 .LBB0_11: movq stderr(%rip), %rbx movl %eax, %edi callq hipGetErrorString movl $.L.str.3, %esi .LBB0_2: movq %rbx, %rdi .LBB0_3: movq %rax, %rdx xorl %eax, %eax callq fprintf movl $-1, %edi callq exit .Lfunc_end0: .size _Z6MatMul6MatrixS_S_, .Lfunc_end0-_Z6MatMul6MatrixS_S_ .cfi_endproc # -- End function .globl _Z27__device_stub__MatMulKernel6MatrixS_S_ # -- Begin function _Z27__device_stub__MatMulKernel6MatrixS_S_ .p2align 4, 0x90 .type _Z27__device_stub__MatMulKernel6MatrixS_S_,@function _Z27__device_stub__MatMulKernel6MatrixS_S_: # @_Z27__device_stub__MatMulKernel6MatrixS_S_ .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 80(%rsp) movq %rsi, 88(%rsp) movq %rdx, 64(%rsp) movq %rcx, 72(%rsp) movq %r8, 48(%rsp) movq %r9, 56(%rsp) leaq 80(%rsp), %rax movq %rax, 96(%rsp) leaq 64(%rsp), %rax movq %rax, 104(%rsp) leaq 48(%rsp), %rax movq %rax, 112(%rsp) leaq 32(%rsp), %rdi leaq 16(%rsp), %rsi leaq 8(%rsp), %rdx movq %rsp, %rcx callq __hipPopCallConfiguration movq 32(%rsp), %rsi movl 40(%rsp), %edx movq 16(%rsp), %rcx movl 24(%rsp), %r8d leaq 96(%rsp), %r9 movl $_Z12MatMulKernel6MatrixS_S_, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end1: .size _Z27__device_stub__MatMulKernel6MatrixS_S_, .Lfunc_end1-_Z27__device_stub__MatMulKernel6MatrixS_S_ .cfi_endproc # -- End function .globl _Z6myrandv # -- Begin function _Z6myrandv .p2align 4, 0x90 .type _Z6myrandv,@function _Z6myrandv: # @_Z6myrandv .cfi_startproc # %bb.0: pushq %rax .cfi_def_cfa_offset 16 callq rand cltq imulq $-1610612725, %rax, %rcx # imm = 0xA000000B shrq $32, %rcx addl %ecx, %eax movl %eax, %ecx shrl $31, %ecx sarl $27, %eax addl %ecx, %eax # kill: def $eax killed $eax killed $rax popq %rcx .cfi_def_cfa_offset 8 retq .Lfunc_end2: .size _Z6myrandv, .Lfunc_end2-_Z6myrandv .cfi_endproc # -- End function .globl main # -- Begin function main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %r15 .cfi_def_cfa_offset 24 pushq %r14 .cfi_def_cfa_offset 32 pushq %r13 .cfi_def_cfa_offset 40 pushq %r12 .cfi_def_cfa_offset 48 pushq %rbx .cfi_def_cfa_offset 56 pushq %rax .cfi_def_cfa_offset 64 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 xorl %edi, %edi callq srand movl $1024, %edi # imm = 0x400 callq malloc movq %rax, %r14 movl $1024, %edi # imm = 0x400 callq malloc movq %rax, %r15 movl $1024, %edi # imm = 0x400 callq malloc movq %rax, %rbx movl $.Lstr, %edi callq puts@PLT movq %r14, %r12 xorl %r13d, %r13d .p2align 4, 0x90 .LBB3_1: # %.preheader82 # =>This Loop Header: Depth=1 # Child Loop BB3_2 Depth 2 xorl %ebp, %ebp .p2align 4, 0x90 .LBB3_2: # Parent Loop BB3_1 Depth=1 # => This Inner Loop Header: Depth=2 callq rand cltq imulq $-1610612725, %rax, %rcx # imm = 0xA000000B shrq $32, %rcx addl %ecx, %eax movl %eax, %ecx shrl $31, %ecx sarl $27, %eax addl %ecx, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r12,%rbp,4) cvttss2si %xmm0, %esi movl $.L.str.6, %edi xorl %eax, %eax callq printf incq %rbp cmpq $16, %rbp jne .LBB3_2 # %bb.3: # in Loop: Header=BB3_1 Depth=1 movl $10, %edi callq putchar@PLT incq %r13 addq $64, %r12 cmpq $16, %r13 jne .LBB3_1 # %bb.4: movl $.Lstr.1, %edi callq puts@PLT movq %r15, %r12 xorl %r13d, %r13d .p2align 4, 0x90 .LBB3_5: # %.preheader81 # =>This Loop Header: Depth=1 # Child Loop BB3_6 Depth 2 xorl %ebp, %ebp .p2align 4, 0x90 .LBB3_6: # Parent Loop BB3_5 Depth=1 # => This Inner Loop Header: Depth=2 callq rand cltq imulq $-1610612725, %rax, %rcx # imm = 0xA000000B shrq $32, %rcx addl %ecx, %eax movl %eax, %ecx shrl $31, %ecx sarl $27, %eax addl %ecx, %eax xorps %xmm0, %xmm0 cvtsi2ss %eax, %xmm0 movss %xmm0, (%r12,%rbp,4) cvttss2si %xmm0, %esi movl $.L.str.6, %edi xorl %eax, %eax callq printf incq %rbp cmpq $16, %rbp jne .LBB3_6 # %bb.7: # in Loop: Header=BB3_5 Depth=1 movl $10, %edi callq putchar@PLT incq %r13 addq $64, %r12 cmpq $16, %r13 jne .LBB3_5 # %bb.8: movabsq $68719476752, %rdi # imm = 0x1000000010 movq %r14, %rsi movq %rdi, %rdx movq %r15, %rcx movq %rdi, %r8 movq %rbx, %r9 callq _Z6MatMul6MatrixS_S_ movl $.Lstr.2, %edi callq puts@PLT xorl %r14d, %r14d .p2align 4, 0x90 .LBB3_9: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB3_10 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB3_10: # Parent Loop BB3_9 Depth=1 # => This Inner Loop Header: Depth=2 cvttss2si (%rbx,%r15,4), %esi movl $.L.str.10, %edi xorl %eax, %eax callq printf incq %r15 cmpq $16, %r15 jne .LBB3_10 # %bb.11: # in Loop: Header=BB3_9 Depth=1 movl $10, %edi callq putchar@PLT incq %r14 addq $64, %rbx cmpq $16, %r14 jne .LBB3_9 # %bb.12: xorl %eax, %eax addq $8, %rsp .cfi_def_cfa_offset 56 popq %rbx .cfi_def_cfa_offset 48 popq %r12 .cfi_def_cfa_offset 40 popq %r13 .cfi_def_cfa_offset 32 popq %r14 .cfi_def_cfa_offset 24 popq %r15 .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end3: .size main, .Lfunc_end3-main .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_ctor .type __hip_module_ctor,@function __hip_module_ctor: # @__hip_module_ctor .cfi_startproc # %bb.0: subq $40, %rsp .cfi_def_cfa_offset 48 cmpq $0, __hip_gpubin_handle(%rip) jne .LBB4_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB4_2: movq __hip_gpubin_handle(%rip), %rdi xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z12MatMulKernel6MatrixS_S_, %esi movl $.L__unnamed_1, %edx movl $.L__unnamed_1, %ecx movl $-1, %r8d xorl %r9d, %r9d callq __hipRegisterFunction movl $__hip_module_dtor, %edi addq $40, %rsp .cfi_def_cfa_offset 8 jmp atexit # TAILCALL .Lfunc_end4: .size __hip_module_ctor, .Lfunc_end4-__hip_module_ctor .cfi_endproc # -- End function .p2align 4, 0x90 # -- Begin function __hip_module_dtor .type __hip_module_dtor,@function __hip_module_dtor: # @__hip_module_dtor .cfi_startproc # %bb.0: movq __hip_gpubin_handle(%rip), %rdi testq %rdi, %rdi je .LBB5_2 # %bb.1: pushq %rax .cfi_def_cfa_offset 16 callq __hipUnregisterFatBinary movq $0, __hip_gpubin_handle(%rip) addq $8, %rsp .cfi_def_cfa_offset 8 .LBB5_2: retq .Lfunc_end5: .size __hip_module_dtor, .Lfunc_end5-__hip_module_dtor .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "ERROR: allocation A %s\n" .size .L.str, 24 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "ERROR: allocation B %s\n" .size .L.str.1, 24 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "ERROR: allocation C %s\n" .size .L.str.2, 24 .type _Z12MatMulKernel6MatrixS_S_,@object # @_Z12MatMulKernel6MatrixS_S_ .section .rodata,"a",@progbits .globl _Z12MatMulKernel6MatrixS_S_ .p2align 3, 0x0 _Z12MatMulKernel6MatrixS_S_: .quad _Z27__device_stub__MatMulKernel6MatrixS_S_ .size _Z12MatMulKernel6MatrixS_S_, 8 .type .L.str.3,@object # @.str.3 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.3: .asciz "ERROR: calculation error %s\n" .size .L.str.3, 29 .type .L.str.6,@object # @.str.6 .L.str.6: .asciz "%2d" .size .L.str.6, 4 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "%4d" .size .L.str.10, 4 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z12MatMulKernel6MatrixS_S_" .size .L__unnamed_1, 28 .type __hip_fatbin_wrapper,@object # @__hip_fatbin_wrapper .section .hipFatBinSegment,"a",@progbits .p2align 3, 0x0 __hip_fatbin_wrapper: .long 1212764230 # 0x48495046 .long 1 # 0x1 .quad __hip_fatbin .quad 0 .size __hip_fatbin_wrapper, 24 .type __hip_gpubin_handle,@object # @__hip_gpubin_handle .local __hip_gpubin_handle .comm __hip_gpubin_handle,8,8 .section .init_array,"aw",@init_array .p2align 3, 0x0 .quad __hip_module_ctor .type __hip_cuid_,@object # @__hip_cuid_ .bss .globl __hip_cuid_ __hip_cuid_: .byte 0 # 0x0 .size __hip_cuid_, 1 .type .Lstr,@object # @str .section .rodata.str1.1,"aMS",@progbits,1 .Lstr: .asciz "Content of A: " .size .Lstr, 15 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "\n\nContent of B: " .size .Lstr.1, 17 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "\n\nContent of C: " .size .Lstr.2, 17 .section ".linker-options","e",@llvm_linker_options .ident "AMD clang version 18.0.0git (https://github.com/RadeonOpenCompute/llvm-project roc-6.3.2 25012 e5bf7e55c91490b07c49d8960fa7983d864936c4)" .section ".note.GNU-stack","",@progbits .addrsig .addrsig_sym _Z27__device_stub__MatMulKernel6MatrixS_S_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z12MatMulKernel6MatrixS_S_ .addrsig_sym __hip_fatbin .addrsig_sym __hip_fatbin_wrapper .addrsig_sym __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include "includes.h" __global__ void fourier_transform(float *in, float *out, int height, int width, int blockConfig) { // block elements and function variables int my_x, k, t; my_x = blockIdx.x * blockDim.x + threadIdx.x; // iterate through each element, going from frequency to time domain for (k = 0; k < height; k++) { // difference, which will be used to subtract off float realSum = 0.0; // iterate through the input element for (t = 0; t < width; t++) { // calculate the angle and update the sum float angle = 2 * M_PI * (my_x * height + t) * (my_x * width + k) / height; realSum += in[my_x * height + t] * cos(angle); } // each output element will be the current sum for that index out[my_x * height + k] = realSum; } }
.file "tmpxft_0002886e_00000000-6_fourier_transform.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2029: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaUnregisterFatBinary@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2029: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z42__device_stub__Z17fourier_transformPfS_iiiPfS_iii .type _Z42__device_stub__Z17fourier_transformPfS_iiiPfS_iii, @function _Z42__device_stub__Z17fourier_transformPfS_iiiPfS_iii: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movl %edx, 12(%rsp) movl %ecx, 8(%rsp) movl %r8d, 4(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 12(%rsp), %rax movq %rax, 112(%rsp) leaq 8(%rsp), %rax movq %rax, 120(%rsp) leaq 4(%rsp), %rax movq %rax, 128(%rsp) movl $1, 48(%rsp) movl $1, 52(%rsp) movl $1, 56(%rsp) movl $1, 60(%rsp) movl $1, 64(%rsp) movl $1, 68(%rsp) leaq 40(%rsp), %rcx leaq 32(%rsp), %rdx leaq 60(%rsp), %rsi leaq 48(%rsp), %rdi call __cudaPopCallConfiguration@PLT testl %eax, %eax je .L7 .L3: movq 136(%rsp), %rax subq %fs:40, %rax jne .L8 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 168 pushq 40(%rsp) .cfi_def_cfa_offset 176 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z17fourier_transformPfS_iii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z42__device_stub__Z17fourier_transformPfS_iiiPfS_iii, .-_Z42__device_stub__Z17fourier_transformPfS_iiiPfS_iii .globl _Z17fourier_transformPfS_iii .type _Z17fourier_transformPfS_iii, @function _Z17fourier_transformPfS_iii: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z42__device_stub__Z17fourier_transformPfS_iiiPfS_iii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z17fourier_transformPfS_iii, .-_Z17fourier_transformPfS_iii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z17fourier_transformPfS_iii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2054: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq _ZL15__fatDeviceText(%rip), %rdi call __cudaRegisterFatBinary@PLT movq %rax, %rdi movq %rax, _ZL20__cudaFatCubinHandle(%rip) pushq $0 .cfi_def_cfa_offset 24 pushq $0 .cfi_def_cfa_offset 32 pushq $0 .cfi_def_cfa_offset 40 pushq $0 .cfi_def_cfa_offset 48 movl $0, %r9d movl $-1, %r8d leaq .LC0(%rip), %rdx movq %rdx, %rcx leaq _Z17fourier_transformPfS_iii(%rip), %rsi call __cudaRegisterFunction@PLT addq $32, %rsp .cfi_def_cfa_offset 16 movq _ZL20__cudaFatCubinHandle(%rip), %rdi call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2054: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .nvFatBinSegment,"aw" .align 8 .type _ZL15__fatDeviceText, @object .size _ZL15__fatDeviceText, 24 _ZL15__fatDeviceText: .long 1180844977 .long 1 .quad fatbinData .quad 0 .local _ZL20__cudaFatCubinHandle .comm _ZL20__cudaFatCubinHandle,8,8 .ident "GCC: (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0" .section .note.GNU-stack,"",@progbits .section .note.gnu.property,"a" .align 8 .long 1f - 0f .long 4f - 1f .long 5 0: .string "GNU" 1: .align 8 .long 0xc0000002 .long 3f - 2f 2: .long 0x3 3: .align 8 4: