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 CUDA host assembly.
#include <stdlib.h> #include <stdio.h> void fill_matrix(double *mat, unsigned numRows, unsigned numCols) { for(unsigned i=0; i < numRows; i++) for(unsigned j=0; j < numCols; j++) { mat[i*numCols + j] = i*2.1f + j*3.2f; } } void print_matrix_to_file(double *mat, unsigned numRows, unsigned numCols) { const char *fname = "assignment2_2_out"; FILE *f = fopen(fname, "w"); for(unsigned i=0; i < numRows; i++) { for(unsigned j=0; j < numCols; j++) fprintf(f,"%4.4f ", mat[i*numCols + j]); fprintf(f,"\n"); } fclose(f); } __global__ void MatrixMulKernel_col_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.y; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.x; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } __global__ void MatrixMulKernel_row_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.x; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.y; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } int main(int argc,char **argv) { int N = 8192; size_t size = N *N* sizeof(double); int thread_dim_ll[8]; int thread_dim; double*h_matA = (double*)malloc(size); double*h_matB = (double*)malloc(size); double*h_matC = (double*)malloc(size); // result int loop, loop1, loop2; // loop variables float time_spent; fill_matrix(h_matA,N,N); fill_matrix(h_matB,N,N); printf("Thread dims\n"); for (loop=0;loop<8;loop++){ thread_dim_ll[loop]=pow(2,2+loop); } printf("\nMatrix A (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matA + N*loop1 + loop2)); } printf("\n\nMatrix B (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matB + N*loop1 + loop2)); } double* d_matA; cudaMalloc(&d_matA, size); double* d_matB; cudaMalloc(&d_matB, size); double* d_matC; cudaMalloc(&d_matC, size); //GPU timing cudaEvent_t start, stop; cudaEventCreate(&start); cudaEventCreate(&stop); // Copy vectors from host memory to device memory cudaMemcpy(d_matA, h_matA, size,cudaMemcpyHostToDevice); cudaMemcpy(d_matB, h_matB, size,cudaMemcpyHostToDevice); for (loop = 0; loop < 8; loop++){ //thread dim thread_dim=thread_dim_ll[loop]; // Invoke kernel dim3 threadsPerBlock = (thread_dim,thread_dim); dim3 blocksPerGrid ((N + threadsPerBlock.x - 1) /threadsPerBlock.x,(N + threadsPerBlock.y - 1) /threadsPerBlock.y); cudaEventRecord(start, 0); MatrixMulKernel_col_maj<<<blocksPerGrid, threadsPerBlock>>>(d_matA,d_matB, d_matC, N); cudaEventRecord(stop, 0); cudaEventSynchronize(stop); cudaEventElapsedTime(&time_spent, start, stop); printf("\nTime spent in col maj %f with threadsPerBlock %d \n",time_spent,thread_dim); } // h_C contains the result in host memory cudaMemcpy(h_matC, d_matC, size,cudaMemcpyDeviceToHost); printf("\n\nMatrix C (first 10*10 outputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matC + N*loop1 + loop2)); } // Log outputs printf("\nWritting to file assignment_2_1_out as Mat C"); print_matrix_to_file(h_matC,N,N); // Free device memory cudaFree(d_matA); cudaFree(d_matB); cudaFree(d_matC); // Free host memory free(h_matA); free(h_matB); free(h_matC); return 0; }
.file "tmpxft_0005130a_00000000-6_ee16b068_2.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 _Z11fill_matrixPdjj .type _Z11fill_matrixPdjj, @function _Z11fill_matrixPdjj: .LFB2057: .cfi_startproc endbr64 movq %rdi, %r9 movl $0, %r8d movl $0, %r10d movss .LC0(%rip), %xmm3 movss .LC1(%rip), %xmm2 testl %esi, %esi jne .L4 ret .L11: movl %r10d, %eax pxor %xmm1, %xmm1 cvtsi2ssq %rax, %xmm1 mulss %xmm3, %xmm1 movl $0, %eax .L10: leal (%r8,%rax), %ecx movl %eax, %edi pxor %xmm0, %xmm0 cvtsi2ssq %rdi, %xmm0 mulss %xmm2, %xmm0 addss %xmm1, %xmm0 cvtss2sd %xmm0, %xmm0 movsd %xmm0, (%r9,%rcx,8) addl $1, %eax cmpl %eax, %edx jne .L10 .L12: addl $1, %r10d addl %edx, %r8d cmpl %r10d, %esi je .L3 .L4: testl %edx, %edx jne .L11 jmp .L12 .L3: ret .cfi_endproc .LFE2057: .size _Z11fill_matrixPdjj, .-_Z11fill_matrixPdjj .section .rodata.str1.1,"aMS",@progbits,1 .LC2: .string "w" .LC3: .string "assignment2_2_out" .LC4: .string "%4.4f " .LC5: .string "\n" .text .globl _Z20print_matrix_to_filePdjj .type _Z20print_matrix_to_filePdjj, @function _Z20print_matrix_to_filePdjj: .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 subq $24, %rsp .cfi_def_cfa_offset 80 movq %rdi, %r13 movl %esi, %ebx movl %esi, 12(%rsp) movl %edx, %r15d leaq .LC2(%rip), %rsi leaq .LC3(%rip), %rdi call fopen@PLT movq %rax, %r12 movl %r15d, %ebp movl $0, 8(%rsp) leaq .LC4(%rip), %r14 testl %ebx, %ebx jne .L17 .L18: movq %r12, %rdi call fclose@PLT addq $24, %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 .L19: .cfi_restore_state movl %ebx, %eax movsd 0(%r13,%rax,8), %xmm0 movq %r14, %rdx movl $2, %esi movq %r12, %rdi movl $1, %eax call __fprintf_chk@PLT addl $1, %ebx cmpl %ebp, %ebx jne .L19 .L21: leaq .LC5(%rip), %rdx movl $2, %esi movq %r12, %rdi movl $0, %eax call __fprintf_chk@PLT addl $1, 8(%rsp) movl 8(%rsp), %eax addl %r15d, %ebp cmpl %eax, 12(%rsp) je .L18 .L17: movl %ebp, %ebx subl %r15d, %ebx testl %r15d, %r15d jne .L19 jmp .L21 .cfi_endproc .LFE2058: .size _Z20print_matrix_to_filePdjj, .-_Z20print_matrix_to_filePdjj .globl _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i .type _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i, @function _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i: .LFB2084: .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 .L31 .L27: movq 136(%rsp), %rax subq %fs:40, %rax jne .L32 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L31: .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 _Z23MatrixMulKernel_col_majPdS_S_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L27 .L32: call __stack_chk_fail@PLT .cfi_endproc .LFE2084: .size _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i, .-_Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i .globl _Z23MatrixMulKernel_col_majPdS_S_i .type _Z23MatrixMulKernel_col_majPdS_S_i, @function _Z23MatrixMulKernel_col_majPdS_S_i: .LFB2085: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _Z23MatrixMulKernel_col_majPdS_S_i, .-_Z23MatrixMulKernel_col_majPdS_S_i .section .rodata.str1.1 .LC6: .string "Thread dims\n" .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC8: .string "\nMatrix A (first 10*10 inputs)\n" .section .rodata.str1.1 .LC9: .string "%f " .section .rodata.str1.8 .align 8 .LC10: .string "\n\nMatrix B (first 10*10 inputs)\n" .align 8 .LC11: .string "\nTime spent in col maj %f with threadsPerBlock %d \n" .align 8 .LC12: .string "\n\nMatrix C (first 10*10 outputs)\n" .align 8 .LC13: .string "\nWritting to file assignment_2_1_out as Mat C" .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 $168, %rsp .cfi_def_cfa_offset 224 movq %fs:40, %rax movq %rax, 152(%rsp) xorl %eax, %eax movl $536870912, %edi call malloc@PLT movq %rax, %r15 movq %rax, 8(%rsp) movl $536870912, %edi call malloc@PLT movq %rax, %rbx movq %rax, 16(%rsp) movl $536870912, %edi call malloc@PLT movq %rax, 24(%rsp) movl $8192, %edx movl $8192, %esi movq %r15, %rdi call _Z11fill_matrixPdjj movl $8192, %edx movl $8192, %esi movq %rbx, %rdi call _Z11fill_matrixPdjj leaq .LC6(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 112(%rsp), %r12 leaq 144(%rsp), %r13 movq %r12, %rbx movl $2, %ebp .L36: pxor %xmm1, %xmm1 cvtsi2sdl %ebp, %xmm1 movsd .LC7(%rip), %xmm0 call pow@PLT cvttsd2sil %xmm0, %eax movl %eax, (%rbx) addl $1, %ebp addq $4, %rbx cmpq %r13, %rbx jne .L36 leaq .LC8(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 8(%rsp), %rax leaq 80(%rax), %rbp leaq 655440(%rax), %r15 leaq .LC9(%rip), %r14 .L37: leaq -80(%rbp), %rbx .L38: movsd (%rbx), %xmm0 movq %r14, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $8, %rbx cmpq %rbp, %rbx jne .L38 addq $65536, %rbp cmpq %r15, %rbp jne .L37 leaq .LC10(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 16(%rsp), %rax leaq 80(%rax), %rbp leaq 655440(%rax), %r15 leaq .LC9(%rip), %r14 .L40: leaq -80(%rbp), %rbx .L41: movsd (%rbx), %xmm0 movq %r14, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $8, %rbx cmpq %rbp, %rbx jne .L41 addq $65536, %rbp cmpq %r15, %rbp jne .L40 leaq 48(%rsp), %rdi movl $536870912, %esi call cudaMalloc@PLT leaq 56(%rsp), %rdi movl $536870912, %esi call cudaMalloc@PLT leaq 64(%rsp), %rdi movl $536870912, %esi call cudaMalloc@PLT leaq 72(%rsp), %rdi call cudaEventCreate@PLT leaq 80(%rsp), %rdi call cudaEventCreate@PLT movl $1, %ecx movl $536870912, %edx movq 8(%rsp), %rsi movq 48(%rsp), %rdi call cudaMemcpy@PLT movl $1, %ecx movl $536870912, %edx movq 16(%rsp), %rsi movq 56(%rsp), %rdi call cudaMemcpy@PLT leaq .LC11(%rip), %rbp jmp .L44 .L43: movl $0, %esi movq 80(%rsp), %rdi call cudaEventRecord@PLT movq 80(%rsp), %rdi call cudaEventSynchronize@PLT leaq 44(%rsp), %rdi movq 80(%rsp), %rdx movq 72(%rsp), %rsi call cudaEventElapsedTime@PLT pxor %xmm0, %xmm0 cvtss2sd 44(%rsp), %xmm0 movl %ebx, %edx movq %rbp, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $4, %r12 cmpq %r13, %r12 je .L55 .L44: movl (%r12), %ebx movl $1, 96(%rsp) leal 8191(%rbx), %eax movl $0, %edx divl %ebx movl %eax, 100(%rsp) movl $8192, 104(%rsp) movl $1, 108(%rsp) movl $0, %esi movq 72(%rsp), %rdi call cudaEventRecord@PLT movl %ebx, 88(%rsp) movl $1, 92(%rsp) movl 96(%rsp), %ecx movl $0, %r9d movl $0, %r8d movq 88(%rsp), %rdx movq 100(%rsp), %rdi movl 108(%rsp), %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax jne .L43 movl $8192, %ecx movq 64(%rsp), %rdx movq 56(%rsp), %rsi movq 48(%rsp), %rdi call _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i jmp .L43 .L55: movl $2, %ecx movl $536870912, %edx movq 64(%rsp), %rsi movq 24(%rsp), %rbx movq %rbx, %rdi call cudaMemcpy@PLT leaq .LC12(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 80(%rbx), %rbp leaq 655440(%rbx), %r13 leaq .LC9(%rip), %r12 .L45: leaq -80(%rbp), %rbx .L46: movsd (%rbx), %xmm0 movq %r12, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $8, %rbx cmpq %rbp, %rbx jne .L46 addq $65536, %rbp cmpq %r13, %rbp jne .L45 leaq .LC13(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movl $8192, %edx movl $8192, %esi movq 24(%rsp), %rbx movq %rbx, %rdi call _Z20print_matrix_to_filePdjj movq 48(%rsp), %rdi call cudaFree@PLT movq 56(%rsp), %rdi call cudaFree@PLT movq 64(%rsp), %rdi call cudaFree@PLT movq 8(%rsp), %rdi call free@PLT movq 16(%rsp), %rdi call free@PLT movq %rbx, %rdi call free@PLT movq 152(%rsp), %rax subq %fs:40, %rax jne .L56 movl $0, %eax addq $168, %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 .L56: .cfi_restore_state call __stack_chk_fail@PLT .cfi_endproc .LFE2059: .size main, .-main .globl _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i .type _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i, @function _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i: .LFB2086: .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 .L61 .L57: movq 136(%rsp), %rax subq %fs:40, %rax jne .L62 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L61: .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 _Z23MatrixMulKernel_row_majPdS_S_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L57 .L62: call __stack_chk_fail@PLT .cfi_endproc .LFE2086: .size _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i, .-_Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i .globl _Z23MatrixMulKernel_row_majPdS_S_i .type _Z23MatrixMulKernel_row_majPdS_S_i, @function _Z23MatrixMulKernel_row_majPdS_S_i: .LFB2087: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2087: .size _Z23MatrixMulKernel_row_majPdS_S_i, .-_Z23MatrixMulKernel_row_majPdS_S_i .section .rodata.str1.8 .align 8 .LC14: .string "_Z23MatrixMulKernel_row_majPdS_S_i" .align 8 .LC15: .string "_Z23MatrixMulKernel_col_majPdS_S_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2089: .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 .LC14(%rip), %rdx movq %rdx, %rcx leaq _Z23MatrixMulKernel_row_majPdS_S_i(%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 .LC15(%rip), %rdx movq %rdx, %rcx leaq _Z23MatrixMulKernel_col_majPdS_S_i(%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 .LFE2089: .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 1074161254 .align 4 .LC1: .long 1078774989 .section .rodata.cst8,"aM",@progbits,8 .align 8 .LC7: .long 0 .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 <stdlib.h> #include <stdio.h> void fill_matrix(double *mat, unsigned numRows, unsigned numCols) { for(unsigned i=0; i < numRows; i++) for(unsigned j=0; j < numCols; j++) { mat[i*numCols + j] = i*2.1f + j*3.2f; } } void print_matrix_to_file(double *mat, unsigned numRows, unsigned numCols) { const char *fname = "assignment2_2_out"; FILE *f = fopen(fname, "w"); for(unsigned i=0; i < numRows; i++) { for(unsigned j=0; j < numCols; j++) fprintf(f,"%4.4f ", mat[i*numCols + j]); fprintf(f,"\n"); } fclose(f); } __global__ void MatrixMulKernel_col_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.y; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.x; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } __global__ void MatrixMulKernel_row_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.x; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.y; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } int main(int argc,char **argv) { int N = 8192; size_t size = N *N* sizeof(double); int thread_dim_ll[8]; int thread_dim; double*h_matA = (double*)malloc(size); double*h_matB = (double*)malloc(size); double*h_matC = (double*)malloc(size); // result int loop, loop1, loop2; // loop variables float time_spent; fill_matrix(h_matA,N,N); fill_matrix(h_matB,N,N); printf("Thread dims\n"); for (loop=0;loop<8;loop++){ thread_dim_ll[loop]=pow(2,2+loop); } printf("\nMatrix A (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matA + N*loop1 + loop2)); } printf("\n\nMatrix B (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matB + N*loop1 + loop2)); } double* d_matA; cudaMalloc(&d_matA, size); double* d_matB; cudaMalloc(&d_matB, size); double* d_matC; cudaMalloc(&d_matC, size); //GPU timing cudaEvent_t start, stop; cudaEventCreate(&start); cudaEventCreate(&stop); // Copy vectors from host memory to device memory cudaMemcpy(d_matA, h_matA, size,cudaMemcpyHostToDevice); cudaMemcpy(d_matB, h_matB, size,cudaMemcpyHostToDevice); for (loop = 0; loop < 8; loop++){ //thread dim thread_dim=thread_dim_ll[loop]; // Invoke kernel dim3 threadsPerBlock = (thread_dim,thread_dim); dim3 blocksPerGrid ((N + threadsPerBlock.x - 1) /threadsPerBlock.x,(N + threadsPerBlock.y - 1) /threadsPerBlock.y); cudaEventRecord(start, 0); MatrixMulKernel_col_maj<<<blocksPerGrid, threadsPerBlock>>>(d_matA,d_matB, d_matC, N); cudaEventRecord(stop, 0); cudaEventSynchronize(stop); cudaEventElapsedTime(&time_spent, start, stop); printf("\nTime spent in col maj %f with threadsPerBlock %d \n",time_spent,thread_dim); } // h_C contains the result in host memory cudaMemcpy(h_matC, d_matC, size,cudaMemcpyDeviceToHost); printf("\n\nMatrix C (first 10*10 outputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matC + N*loop1 + loop2)); } // Log outputs printf("\nWritting to file assignment_2_1_out as Mat C"); print_matrix_to_file(h_matC,N,N); // Free device memory cudaFree(d_matA); cudaFree(d_matB); cudaFree(d_matC); // Free host memory free(h_matA); free(h_matB); free(h_matC); return 0; }
#include <hip/hip_runtime.h> #include <stdlib.h> #include <stdio.h> void fill_matrix(double *mat, unsigned numRows, unsigned numCols) { for(unsigned i=0; i < numRows; i++) for(unsigned j=0; j < numCols; j++) { mat[i*numCols + j] = i*2.1f + j*3.2f; } } void print_matrix_to_file(double *mat, unsigned numRows, unsigned numCols) { const char *fname = "assignment2_2_out"; FILE *f = fopen(fname, "w"); for(unsigned i=0; i < numRows; i++) { for(unsigned j=0; j < numCols; j++) fprintf(f,"%4.4f ", mat[i*numCols + j]); fprintf(f,"\n"); } fclose(f); } __global__ void MatrixMulKernel_col_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.y; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.x; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } __global__ void MatrixMulKernel_row_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.x; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.y; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } int main(int argc,char **argv) { int N = 8192; size_t size = N *N* sizeof(double); int thread_dim_ll[8]; int thread_dim; double*h_matA = (double*)malloc(size); double*h_matB = (double*)malloc(size); double*h_matC = (double*)malloc(size); // result int loop, loop1, loop2; // loop variables float time_spent; fill_matrix(h_matA,N,N); fill_matrix(h_matB,N,N); printf("Thread dims\n"); for (loop=0;loop<8;loop++){ thread_dim_ll[loop]=pow(2,2+loop); } printf("\nMatrix A (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matA + N*loop1 + loop2)); } printf("\n\nMatrix B (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matB + N*loop1 + loop2)); } double* d_matA; hipMalloc(&d_matA, size); double* d_matB; hipMalloc(&d_matB, size); double* d_matC; hipMalloc(&d_matC, size); //GPU timing hipEvent_t start, stop; hipEventCreate(&start); hipEventCreate(&stop); // Copy vectors from host memory to device memory hipMemcpy(d_matA, h_matA, size,hipMemcpyHostToDevice); hipMemcpy(d_matB, h_matB, size,hipMemcpyHostToDevice); for (loop = 0; loop < 8; loop++){ //thread dim thread_dim=thread_dim_ll[loop]; // Invoke kernel dim3 threadsPerBlock = (thread_dim,thread_dim); dim3 blocksPerGrid ((N + threadsPerBlock.x - 1) /threadsPerBlock.x,(N + threadsPerBlock.y - 1) /threadsPerBlock.y); hipEventRecord(start, 0); MatrixMulKernel_col_maj<<<blocksPerGrid, threadsPerBlock>>>(d_matA,d_matB, d_matC, N); hipEventRecord(stop, 0); hipEventSynchronize(stop); hipEventElapsedTime(&time_spent, start, stop); printf("\nTime spent in col maj %f with threadsPerBlock %d \n",time_spent,thread_dim); } // h_C contains the result in host memory hipMemcpy(h_matC, d_matC, size,hipMemcpyDeviceToHost); printf("\n\nMatrix C (first 10*10 outputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matC + N*loop1 + loop2)); } // Log outputs printf("\nWritting to file assignment_2_1_out as Mat C"); print_matrix_to_file(h_matC,N,N); // Free device memory hipFree(d_matA); hipFree(d_matB); hipFree(d_matC); // Free host memory free(h_matA); free(h_matB); free(h_matC); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include <stdlib.h> #include <stdio.h> void fill_matrix(double *mat, unsigned numRows, unsigned numCols) { for(unsigned i=0; i < numRows; i++) for(unsigned j=0; j < numCols; j++) { mat[i*numCols + j] = i*2.1f + j*3.2f; } } void print_matrix_to_file(double *mat, unsigned numRows, unsigned numCols) { const char *fname = "assignment2_2_out"; FILE *f = fopen(fname, "w"); for(unsigned i=0; i < numRows; i++) { for(unsigned j=0; j < numCols; j++) fprintf(f,"%4.4f ", mat[i*numCols + j]); fprintf(f,"\n"); } fclose(f); } __global__ void MatrixMulKernel_col_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.y; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.x; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } __global__ void MatrixMulKernel_row_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.x; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.y; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } int main(int argc,char **argv) { int N = 8192; size_t size = N *N* sizeof(double); int thread_dim_ll[8]; int thread_dim; double*h_matA = (double*)malloc(size); double*h_matB = (double*)malloc(size); double*h_matC = (double*)malloc(size); // result int loop, loop1, loop2; // loop variables float time_spent; fill_matrix(h_matA,N,N); fill_matrix(h_matB,N,N); printf("Thread dims\n"); for (loop=0;loop<8;loop++){ thread_dim_ll[loop]=pow(2,2+loop); } printf("\nMatrix A (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matA + N*loop1 + loop2)); } printf("\n\nMatrix B (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matB + N*loop1 + loop2)); } double* d_matA; hipMalloc(&d_matA, size); double* d_matB; hipMalloc(&d_matB, size); double* d_matC; hipMalloc(&d_matC, size); //GPU timing hipEvent_t start, stop; hipEventCreate(&start); hipEventCreate(&stop); // Copy vectors from host memory to device memory hipMemcpy(d_matA, h_matA, size,hipMemcpyHostToDevice); hipMemcpy(d_matB, h_matB, size,hipMemcpyHostToDevice); for (loop = 0; loop < 8; loop++){ //thread dim thread_dim=thread_dim_ll[loop]; // Invoke kernel dim3 threadsPerBlock = (thread_dim,thread_dim); dim3 blocksPerGrid ((N + threadsPerBlock.x - 1) /threadsPerBlock.x,(N + threadsPerBlock.y - 1) /threadsPerBlock.y); hipEventRecord(start, 0); MatrixMulKernel_col_maj<<<blocksPerGrid, threadsPerBlock>>>(d_matA,d_matB, d_matC, N); hipEventRecord(stop, 0); hipEventSynchronize(stop); hipEventElapsedTime(&time_spent, start, stop); printf("\nTime spent in col maj %f with threadsPerBlock %d \n",time_spent,thread_dim); } // h_C contains the result in host memory hipMemcpy(h_matC, d_matC, size,hipMemcpyDeviceToHost); printf("\n\nMatrix C (first 10*10 outputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matC + N*loop1 + loop2)); } // Log outputs printf("\nWritting to file assignment_2_1_out as Mat C"); print_matrix_to_file(h_matC,N,N); // Free device memory hipFree(d_matA); hipFree(d_matB); hipFree(d_matC); // Free host memory free(h_matA); free(h_matB); free(h_matC); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z23MatrixMulKernel_col_majPdS_S_i .globl _Z23MatrixMulKernel_col_majPdS_S_i .p2align 8 .type _Z23MatrixMulKernel_col_majPdS_S_i,@function _Z23MatrixMulKernel_col_majPdS_S_i: s_clause 0x1 s_load_b32 s3, s[0:1], 0x2c s_load_b32 s2, s[0:1], 0x18 v_bfe_u32 v2, v0, 10, 10 v_and_b32_e32 v3, 0x3ff, v0 s_waitcnt lgkmcnt(0) s_lshr_b32 s4, s3, 16 s_and_b32 s3, s3, 0xffff s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_1) v_mad_u64_u32 v[0:1], null, s15, s4, v[2:3] v_mad_u64_u32 v[1:2], null, s14, s3, v[3:4] s_mov_b32 s3, exec_lo v_max_i32_e32 v2, v0, v1 s_delay_alu instid0(VALU_DEP_1) v_cmpx_gt_i32_e64 s2, v2 s_cbranch_execz .LBB0_7 s_cmp_lt_i32 s2, 1 s_cbranch_scc1 .LBB0_5 s_load_b128 s[4:7], s[0:1], 0x0 v_mul_lo_u32 v2, v0, s2 v_mov_b32_e32 v6, 0 v_mov_b32_e32 v4, v1 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], 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, s4, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s5, v3, vcc_lo .p2align 6 .LBB0_3: 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], 3, v[4:5] s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_cvt_f64_f32_e32 v[5:6], v6 v_add_nc_u32_e32 v4, s2, v4 v_add_co_u32 v7, vcc_lo, s6, v7 s_delay_alu instid0(VALU_DEP_4) v_add_co_ci_u32_e32 v8, vcc_lo, s7, v8, vcc_lo global_load_b64 v[9:10], v[2:3], off global_load_b64 v[7:8], v[7:8], off v_add_co_u32 v2, vcc_lo, v2, 8 v_add_co_ci_u32_e32 v3, vcc_lo, 0, v3, vcc_lo s_waitcnt vmcnt(0) v_fma_f64 v[5:6], v[9:10], v[7:8], v[5:6] s_delay_alu instid0(VALU_DEP_1) v_cvt_f32_f64_e32 v6, v[5:6] s_cbranch_scc0 .LBB0_3 s_delay_alu instid0(VALU_DEP_1) v_cvt_f64_f32_e32 v[2:3], v6 s_branch .LBB0_6 .LBB0_5: v_mov_b32_e32 v2, 0 v_mov_b32_e32 v3, 0 .LBB0_6: s_load_b64 s[0:1], s[0:1], 0x10 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[4:5], null, v0, s2, v[1:2] v_ashrrev_i32_e32 v5, 31, v4 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 3, v[4:5] 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_b64 v[0:1], v[2:3], off .LBB0_7: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z23MatrixMulKernel_col_majPdS_S_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 288 .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 11 .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 _Z23MatrixMulKernel_col_majPdS_S_i, .Lfunc_end0-_Z23MatrixMulKernel_col_majPdS_S_i .section .AMDGPU.csdata,"",@progbits .text .protected _Z23MatrixMulKernel_row_majPdS_S_i .globl _Z23MatrixMulKernel_row_majPdS_S_i .p2align 8 .type _Z23MatrixMulKernel_row_majPdS_S_i,@function _Z23MatrixMulKernel_row_majPdS_S_i: s_clause 0x1 s_load_b32 s3, s[0:1], 0x2c s_load_b32 s2, s[0:1], 0x18 v_and_b32_e32 v2, 0x3ff, v0 v_bfe_u32 v3, v0, 10, 10 s_waitcnt lgkmcnt(0) s_lshr_b32 s4, s3, 16 s_and_b32 s3, s3, 0xffff s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_1) v_mad_u64_u32 v[0:1], null, s15, s4, v[2:3] v_mad_u64_u32 v[1:2], null, s14, s3, v[3:4] s_mov_b32 s3, exec_lo v_max_i32_e32 v2, v0, v1 s_delay_alu instid0(VALU_DEP_1) v_cmpx_gt_i32_e64 s2, v2 s_cbranch_execz .LBB1_7 s_cmp_lt_i32 s2, 1 s_cbranch_scc1 .LBB1_5 s_load_b128 s[4:7], s[0:1], 0x0 v_mul_lo_u32 v2, v0, s2 v_mov_b32_e32 v6, 0 v_mov_b32_e32 v4, v1 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], 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, s4, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s5, v3, vcc_lo .p2align 6 .LBB1_3: 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], 3, v[4:5] s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_cvt_f64_f32_e32 v[5:6], v6 v_add_nc_u32_e32 v4, s2, v4 v_add_co_u32 v7, vcc_lo, s6, v7 s_delay_alu instid0(VALU_DEP_4) v_add_co_ci_u32_e32 v8, vcc_lo, s7, v8, vcc_lo global_load_b64 v[9:10], v[2:3], off global_load_b64 v[7:8], v[7:8], off v_add_co_u32 v2, vcc_lo, v2, 8 v_add_co_ci_u32_e32 v3, vcc_lo, 0, v3, vcc_lo s_waitcnt vmcnt(0) v_fma_f64 v[5:6], v[9:10], v[7:8], v[5:6] s_delay_alu instid0(VALU_DEP_1) v_cvt_f32_f64_e32 v6, v[5:6] s_cbranch_scc0 .LBB1_3 s_delay_alu instid0(VALU_DEP_1) v_cvt_f64_f32_e32 v[2:3], v6 s_branch .LBB1_6 .LBB1_5: v_mov_b32_e32 v2, 0 v_mov_b32_e32 v3, 0 .LBB1_6: s_load_b64 s[0:1], s[0:1], 0x10 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[4:5], null, v0, s2, v[1:2] v_ashrrev_i32_e32 v5, 31, v4 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 3, v[4:5] 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_b64 v[0:1], v[2:3], off .LBB1_7: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z23MatrixMulKernel_row_majPdS_S_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 288 .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 11 .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_end1: .size _Z23MatrixMulKernel_row_majPdS_S_i, .Lfunc_end1-_Z23MatrixMulKernel_row_majPdS_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 - .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: _Z23MatrixMulKernel_col_majPdS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z23MatrixMulKernel_col_majPdS_S_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 11 .vgpr_spill_count: 0 .wavefront_size: 32 .workgroup_processor_mode: 1 - .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: 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: _Z23MatrixMulKernel_row_majPdS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z23MatrixMulKernel_row_majPdS_S_i.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 11 .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 <stdlib.h> #include <stdio.h> void fill_matrix(double *mat, unsigned numRows, unsigned numCols) { for(unsigned i=0; i < numRows; i++) for(unsigned j=0; j < numCols; j++) { mat[i*numCols + j] = i*2.1f + j*3.2f; } } void print_matrix_to_file(double *mat, unsigned numRows, unsigned numCols) { const char *fname = "assignment2_2_out"; FILE *f = fopen(fname, "w"); for(unsigned i=0; i < numRows; i++) { for(unsigned j=0; j < numCols; j++) fprintf(f,"%4.4f ", mat[i*numCols + j]); fprintf(f,"\n"); } fclose(f); } __global__ void MatrixMulKernel_col_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.y; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.x; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } __global__ void MatrixMulKernel_row_maj(double* M, double* N, double* P, int Width) { // Calculate the row index of the P element and M int Row = blockIdx.y*blockDim.y+threadIdx.x; // Calculate the column index of P and N int Col = blockIdx.x*blockDim.x+threadIdx.y; if ((Row < Width) && (Col < Width)) { float Pvalue = 0; for (int k = 0; k < Width; ++k) { Pvalue += M[Row*Width+k]*N[k*Width+Col]; } P[Row*Width+Col] = Pvalue; } } int main(int argc,char **argv) { int N = 8192; size_t size = N *N* sizeof(double); int thread_dim_ll[8]; int thread_dim; double*h_matA = (double*)malloc(size); double*h_matB = (double*)malloc(size); double*h_matC = (double*)malloc(size); // result int loop, loop1, loop2; // loop variables float time_spent; fill_matrix(h_matA,N,N); fill_matrix(h_matB,N,N); printf("Thread dims\n"); for (loop=0;loop<8;loop++){ thread_dim_ll[loop]=pow(2,2+loop); } printf("\nMatrix A (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matA + N*loop1 + loop2)); } printf("\n\nMatrix B (first 10*10 inputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matB + N*loop1 + loop2)); } double* d_matA; hipMalloc(&d_matA, size); double* d_matB; hipMalloc(&d_matB, size); double* d_matC; hipMalloc(&d_matC, size); //GPU timing hipEvent_t start, stop; hipEventCreate(&start); hipEventCreate(&stop); // Copy vectors from host memory to device memory hipMemcpy(d_matA, h_matA, size,hipMemcpyHostToDevice); hipMemcpy(d_matB, h_matB, size,hipMemcpyHostToDevice); for (loop = 0; loop < 8; loop++){ //thread dim thread_dim=thread_dim_ll[loop]; // Invoke kernel dim3 threadsPerBlock = (thread_dim,thread_dim); dim3 blocksPerGrid ((N + threadsPerBlock.x - 1) /threadsPerBlock.x,(N + threadsPerBlock.y - 1) /threadsPerBlock.y); hipEventRecord(start, 0); MatrixMulKernel_col_maj<<<blocksPerGrid, threadsPerBlock>>>(d_matA,d_matB, d_matC, N); hipEventRecord(stop, 0); hipEventSynchronize(stop); hipEventElapsedTime(&time_spent, start, stop); printf("\nTime spent in col maj %f with threadsPerBlock %d \n",time_spent,thread_dim); } // h_C contains the result in host memory hipMemcpy(h_matC, d_matC, size,hipMemcpyDeviceToHost); printf("\n\nMatrix C (first 10*10 outputs)\n"); for(loop1 = 0; loop1 < 10; loop1++){ for (loop2=0;loop2 < 10; loop2++) printf("%f ", *(h_matC + N*loop1 + loop2)); } // Log outputs printf("\nWritting to file assignment_2_1_out as Mat C"); print_matrix_to_file(h_matC,N,N); // Free device memory hipFree(d_matA); hipFree(d_matB); hipFree(d_matC); // Free host memory free(h_matA); free(h_matB); free(h_matC); return 0; }
.text .file "ee16b068_2.hip" .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function _Z11fill_matrixPdjj .LCPI0_0: .long 0x40066666 # float 2.0999999 .LCPI0_1: .long 0x404ccccd # float 3.20000005 .text .globl _Z11fill_matrixPdjj .p2align 4, 0x90 .type _Z11fill_matrixPdjj,@function _Z11fill_matrixPdjj: # @_Z11fill_matrixPdjj .cfi_startproc # %bb.0: testl %esi, %esi je .LBB0_6 # %bb.1: # %.preheader.lr.ph movl %edx, %eax xorl %ecx, %ecx movss .LCPI0_0(%rip), %xmm0 # xmm0 = mem[0],zero,zero,zero movss .LCPI0_1(%rip), %xmm1 # xmm1 = mem[0],zero,zero,zero xorl %r8d, %r8d jmp .LBB0_2 .p2align 4, 0x90 .LBB0_5: # %._crit_edge # in Loop: Header=BB0_2 Depth=1 incl %r8d addq %rax, %rcx cmpl %esi, %r8d je .LBB0_6 .LBB0_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB0_4 Depth 2 testl %edx, %edx je .LBB0_5 # %bb.3: # %.lr.ph # in Loop: Header=BB0_2 Depth=1 movl %r8d, %r9d xorps %xmm2, %xmm2 cvtsi2ss %r9, %xmm2 mulss %xmm0, %xmm2 xorl %r9d, %r9d .p2align 4, 0x90 .LBB0_4: # Parent Loop BB0_2 Depth=1 # => This Inner Loop Header: Depth=2 movl %r9d, %r10d xorps %xmm3, %xmm3 cvtsi2ss %r10, %xmm3 mulss %xmm1, %xmm3 addss %xmm2, %xmm3 cvtss2sd %xmm3, %xmm3 leal (%rcx,%r9), %r10d movsd %xmm3, (%rdi,%r10,8) incq %r9 cmpq %r9, %rax jne .LBB0_4 jmp .LBB0_5 .LBB0_6: # %._crit_edge15 retq .Lfunc_end0: .size _Z11fill_matrixPdjj, .Lfunc_end0-_Z11fill_matrixPdjj .cfi_endproc # -- End function .globl _Z20print_matrix_to_filePdjj # -- Begin function _Z20print_matrix_to_filePdjj .p2align 4, 0x90 .type _Z20print_matrix_to_filePdjj,@function _Z20print_matrix_to_filePdjj: # @_Z20print_matrix_to_filePdjj .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 $24, %rsp .cfi_def_cfa_offset 80 .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 %edx, %ebx movl %esi, %ebp movq %rdi, %r14 movl $.L.str, %edi movl $.L.str.1, %esi callq fopen movq %rax, %r15 movl %ebp, 12(%rsp) # 4-byte Spill testl %ebp, %ebp je .LBB1_4 # %bb.1: # %.preheader.lr.ph movl %ebx, %ecx movl %ebx, %eax movq %rax, 16(%rsp) # 8-byte Spill xorl %ebx, %ebx xorl %ebp, %ebp movl %ecx, 8(%rsp) # 4-byte Spill jmp .LBB1_2 .p2align 4, 0x90 .LBB1_3: # %._crit_edge # in Loop: Header=BB1_2 Depth=1 movl $10, %edi movq %r15, %rsi callq fputc@PLT incl %ebp movl 8(%rsp), %ecx # 4-byte Reload addl %ecx, %ebx cmpl 12(%rsp), %ebp # 4-byte Folded Reload je .LBB1_4 .LBB1_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB1_5 Depth 2 movq 16(%rsp), %r12 # 8-byte Reload movl %ebx, %r13d testl %ecx, %ecx je .LBB1_3 .p2align 4, 0x90 .LBB1_5: # Parent Loop BB1_2 Depth=1 # => This Inner Loop Header: Depth=2 movl %r13d, %eax movsd (%r14,%rax,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.2, %esi movq %r15, %rdi movb $1, %al callq fprintf incl %r13d decq %r12 jne .LBB1_5 jmp .LBB1_3 .LBB1_4: # %._crit_edge17 movq %r15, %rdi addq $24, %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 jmp fclose # TAILCALL .Lfunc_end1: .size _Z20print_matrix_to_filePdjj, .Lfunc_end1-_Z20print_matrix_to_filePdjj .cfi_endproc # -- End function .globl _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i # -- Begin function _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .p2align 4, 0x90 .type _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i,@function _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i: # @_Z38__device_stub__MatrixMulKernel_col_majPdS_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 $_Z23MatrixMulKernel_col_majPdS_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_end2: .size _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i, .Lfunc_end2-_Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .cfi_endproc # -- End function .globl _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i # -- Begin function _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .p2align 4, 0x90 .type _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i,@function _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i: # @_Z38__device_stub__MatrixMulKernel_row_majPdS_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 $_Z23MatrixMulKernel_row_majPdS_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_end3: .size _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i, .Lfunc_end3-_Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function main .LCPI4_0: .long 0x40066666 # float 2.0999999 .LCPI4_1: .long 0x404ccccd # float 3.20000005 .section .rodata.cst8,"aM",@progbits,8 .p2align 3, 0x0 .LCPI4_2: .quad 0x3ff0000000000000 # double 1 .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 $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 $536870912, %edi # imm = 0x20000000 callq malloc movq %rax, %r12 movl $536870912, %edi # imm = 0x20000000 callq malloc movq %rax, %r13 movl $536870912, %edi # imm = 0x20000000 callq malloc movq %rax, %r14 xorl %eax, %eax movss .LCPI4_0(%rip), %xmm0 # xmm0 = mem[0],zero,zero,zero movss .LCPI4_1(%rip), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %r12, %rcx .p2align 4, 0x90 .LBB4_1: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB4_2 Depth 2 movl %eax, %edx xorps %xmm2, %xmm2 cvtsi2ss %rdx, %xmm2 mulss %xmm0, %xmm2 xorl %edx, %edx .p2align 4, 0x90 .LBB4_2: # Parent Loop BB4_1 Depth=1 # => This Inner Loop Header: Depth=2 movl %edx, %esi xorps %xmm3, %xmm3 cvtsi2ss %rsi, %xmm3 mulss %xmm1, %xmm3 addss %xmm2, %xmm3 cvtss2sd %xmm3, %xmm3 movsd %xmm3, (%rcx,%rdx,8) incq %rdx cmpq $8192, %rdx # imm = 0x2000 jne .LBB4_2 # %bb.3: # %._crit_edge.i # in Loop: Header=BB4_1 Depth=1 incl %eax addq $65536, %rcx # imm = 0x10000 cmpl $8192, %eax # imm = 0x2000 jne .LBB4_1 # %bb.4: # %.preheader.i75.preheader xorl %eax, %eax movq %r13, %rcx .p2align 4, 0x90 .LBB4_5: # %.preheader.i75 # =>This Loop Header: Depth=1 # Child Loop BB4_6 Depth 2 movl %eax, %edx xorps %xmm2, %xmm2 cvtsi2ss %rdx, %xmm2 mulss %xmm0, %xmm2 xorl %edx, %edx .p2align 4, 0x90 .LBB4_6: # Parent Loop BB4_5 Depth=1 # => This Inner Loop Header: Depth=2 movl %edx, %esi xorps %xmm3, %xmm3 cvtsi2ss %rsi, %xmm3 mulss %xmm1, %xmm3 addss %xmm2, %xmm3 cvtss2sd %xmm3, %xmm3 movsd %xmm3, (%rcx,%rdx,8) incq %rdx cmpq $8192, %rdx # imm = 0x2000 jne .LBB4_6 # %bb.7: # %._crit_edge.i80 # in Loop: Header=BB4_5 Depth=1 incl %eax addq $65536, %rcx # imm = 0x10000 cmpl $8192, %eax # imm = 0x2000 jne .LBB4_5 # %bb.8: # %_Z11fill_matrixPdjj.exit82 movl $.Lstr, %edi callq puts@PLT xorl %ebx, %ebx .p2align 4, 0x90 .LBB4_9: # =>This Inner Loop Header: Depth=1 leal 2(%rbx), %edi movsd .LCPI4_2(%rip), %xmm0 # xmm0 = mem[0],zero callq ldexp@PLT cvttsd2si %xmm0, %eax movl %eax, 176(%rsp,%rbx,4) incq %rbx cmpq $8, %rbx jne .LBB4_9 # %bb.10: movq %r14, 48(%rsp) # 8-byte Spill movl $.Lstr.1, %edi callq puts@PLT xorl %ebx, %ebx movq %r12, %r14 .p2align 4, 0x90 .LBB4_11: # %.preheader91 # =>This Loop Header: Depth=1 # Child Loop BB4_12 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB4_12: # Parent Loop BB4_11 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r14,%r15,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.6, %edi movb $1, %al callq printf incq %r15 cmpq $10, %r15 jne .LBB4_12 # %bb.13: # in Loop: Header=BB4_11 Depth=1 incq %rbx addq $65536, %r14 # imm = 0x10000 cmpq $10, %rbx jne .LBB4_11 # %bb.14: movl $.Lstr.2, %edi callq puts@PLT xorl %ebx, %ebx movq %r13, %r14 .p2align 4, 0x90 .LBB4_15: # %.preheader90 # =>This Loop Header: Depth=1 # Child Loop BB4_16 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB4_16: # Parent Loop BB4_15 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r14,%r15,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.6, %edi movb $1, %al callq printf incq %r15 cmpq $10, %r15 jne .LBB4_16 # %bb.17: # in Loop: Header=BB4_15 Depth=1 incq %rbx addq $65536, %r14 # imm = 0x10000 cmpq $10, %rbx jne .LBB4_15 # %bb.18: leaq 24(%rsp), %rdi movl $536870912, %esi # imm = 0x20000000 callq hipMalloc leaq 16(%rsp), %rdi movl $536870912, %esi # imm = 0x20000000 callq hipMalloc leaq 8(%rsp), %rdi movl $536870912, %esi # imm = 0x20000000 callq hipMalloc leaq 40(%rsp), %rdi callq hipEventCreate movq %rsp, %rdi callq hipEventCreate movq 24(%rsp), %rdi movl $536870912, %edx # imm = 0x20000000 movq %r12, 64(%rsp) # 8-byte Spill movq %r12, %rsi movl $1, %ecx callq hipMemcpy movq 16(%rsp), %rdi movl $536870912, %edx # imm = 0x20000000 movq %r13, 56(%rsp) # 8-byte Spill movq %r13, %rsi movl $1, %ecx callq hipMemcpy xorl %ebx, %ebx movabsq $35184372088832, %r14 # imm = 0x200000000000 leaq 32(%rsp), %r15 jmp .LBB4_19 .p2align 4, 0x90 .LBB4_21: # in Loop: Header=BB4_19 Depth=1 movq (%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq (%rsp), %rdi callq hipEventSynchronize movq 40(%rsp), %rsi movq (%rsp), %rdx movq %r15, %rdi callq hipEventElapsedTime movss 32(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.8, %edi movl %r13d, %esi movb $1, %al callq printf incq %rbx cmpq $8, %rbx je .LBB4_22 .LBB4_19: # =>This Inner Loop Header: Depth=1 movl 176(%rsp,%rbx,4), %r13d movq %r13, %r12 movabsq $4294967296, %rax # imm = 0x100000000 orq %rax, %r12 leal 8191(%r13), %eax xorl %edx, %edx divl %r13d # kill: def $eax killed $eax def $rax leaq (%rax,%r14), %rbp movq 40(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq %rbp, %rdi movl $1, %esi movq %r12, %rdx movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB4_21 # %bb.20: # in Loop: Header=BB4_19 Depth=1 movq 24(%rsp), %rax movq 16(%rsp), %rcx movq 8(%rsp), %rdx movq %rax, 136(%rsp) movq %rcx, 128(%rsp) movq %rdx, 120(%rsp) movl $8192, 36(%rsp) # imm = 0x2000 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 36(%rsp), %rax movq %rax, 168(%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 movl $_Z23MatrixMulKernel_col_majPdS_S_i, %edi leaq 144(%rsp), %r9 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 jmp .LBB4_21 .LBB4_22: movq 8(%rsp), %rsi movl $536870912, %edx # imm = 0x20000000 movq 48(%rsp), %rbp # 8-byte Reload movq %rbp, %rdi movl $2, %ecx callq hipMemcpy movl $.Lstr.3, %edi callq puts@PLT xorl %ebx, %ebx movq %rbp, %r14 .p2align 4, 0x90 .LBB4_23: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB4_24 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB4_24: # Parent Loop BB4_23 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r14,%r15,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.6, %edi movb $1, %al callq printf incq %r15 cmpq $10, %r15 jne .LBB4_24 # %bb.25: # in Loop: Header=BB4_23 Depth=1 incq %rbx addq $65536, %r14 # imm = 0x10000 cmpq $10, %rbx jne .LBB4_23 # %bb.26: xorl %r14d, %r14d movl $.L.str.10, %edi xorl %eax, %eax callq printf movl $.L.str, %edi movl $.L.str.1, %esi callq fopen movq %rax, %rbx movq %rbp, %r15 movq 56(%rsp), %r13 # 8-byte Reload .p2align 4, 0x90 .LBB4_27: # %.preheader.i83 # =>This Loop Header: Depth=1 # Child Loop BB4_28 Depth 2 xorl %r12d, %r12d .p2align 4, 0x90 .LBB4_28: # Parent Loop BB4_27 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r15,%r12,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.2, %esi movq %rbx, %rdi movb $1, %al callq fprintf incq %r12 cmpq $8192, %r12 # imm = 0x2000 jne .LBB4_28 # %bb.29: # %._crit_edge.i87 # in Loop: Header=BB4_27 Depth=1 movl $10, %edi movq %rbx, %rsi callq fputc@PLT incl %r14d addq $65536, %r15 # imm = 0x10000 cmpl $8192, %r14d # imm = 0x2000 jne .LBB4_27 # %bb.30: # %_Z20print_matrix_to_filePdjj.exit movq %rbx, %rdi callq fclose movq 24(%rsp), %rdi callq hipFree movq 16(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree movq 64(%rsp), %rdi # 8-byte Reload callq free movq %r13, %rdi callq free movq %rbp, %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_end4: .size main, .Lfunc_end4-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: 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 .LBB5_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB5_2: movq __hip_gpubin_handle(%rip), %rbx xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z23MatrixMulKernel_col_majPdS_S_i, %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 $_Z23MatrixMulKernel_row_majPdS_S_i, %esi movl $.L__unnamed_2, %edx movl $.L__unnamed_2, %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_end5: .size __hip_module_ctor, .Lfunc_end5-__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 .LBB6_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 .LBB6_2: retq .Lfunc_end6: .size __hip_module_dtor, .Lfunc_end6-__hip_module_dtor .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "assignment2_2_out" .size .L.str, 18 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "w" .size .L.str.1, 2 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "%4.4f " .size .L.str.2, 7 .type _Z23MatrixMulKernel_col_majPdS_S_i,@object # @_Z23MatrixMulKernel_col_majPdS_S_i .section .rodata,"a",@progbits .globl _Z23MatrixMulKernel_col_majPdS_S_i .p2align 3, 0x0 _Z23MatrixMulKernel_col_majPdS_S_i: .quad _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .size _Z23MatrixMulKernel_col_majPdS_S_i, 8 .type _Z23MatrixMulKernel_row_majPdS_S_i,@object # @_Z23MatrixMulKernel_row_majPdS_S_i .globl _Z23MatrixMulKernel_row_majPdS_S_i .p2align 3, 0x0 _Z23MatrixMulKernel_row_majPdS_S_i: .quad _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .size _Z23MatrixMulKernel_row_majPdS_S_i, 8 .type .L.str.6,@object # @.str.6 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.6: .asciz "%f " .size .L.str.6, 4 .type .L.str.8,@object # @.str.8 .L.str.8: .asciz "\nTime spent in col maj %f with threadsPerBlock %d \n" .size .L.str.8, 52 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "\nWritting to file assignment_2_1_out as Mat C" .size .L.str.10, 46 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z23MatrixMulKernel_col_majPdS_S_i" .size .L__unnamed_1, 35 .type .L__unnamed_2,@object # @1 .L__unnamed_2: .asciz "_Z23MatrixMulKernel_row_majPdS_S_i" .size .L__unnamed_2, 35 .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 "Thread dims" .size .Lstr, 12 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "\nMatrix A (first 10*10 inputs)" .size .Lstr.1, 31 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "\n\nMatrix B (first 10*10 inputs)" .size .Lstr.2, 32 .type .Lstr.3,@object # @str.3 .Lstr.3: .asciz "\n\nMatrix C (first 10*10 outputs)" .size .Lstr.3, 33 .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 _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .addrsig_sym _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z23MatrixMulKernel_col_majPdS_S_i .addrsig_sym _Z23MatrixMulKernel_row_majPdS_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 host assembly to AMD host assembly.
.file "tmpxft_0005130a_00000000-6_ee16b068_2.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 _Z11fill_matrixPdjj .type _Z11fill_matrixPdjj, @function _Z11fill_matrixPdjj: .LFB2057: .cfi_startproc endbr64 movq %rdi, %r9 movl $0, %r8d movl $0, %r10d movss .LC0(%rip), %xmm3 movss .LC1(%rip), %xmm2 testl %esi, %esi jne .L4 ret .L11: movl %r10d, %eax pxor %xmm1, %xmm1 cvtsi2ssq %rax, %xmm1 mulss %xmm3, %xmm1 movl $0, %eax .L10: leal (%r8,%rax), %ecx movl %eax, %edi pxor %xmm0, %xmm0 cvtsi2ssq %rdi, %xmm0 mulss %xmm2, %xmm0 addss %xmm1, %xmm0 cvtss2sd %xmm0, %xmm0 movsd %xmm0, (%r9,%rcx,8) addl $1, %eax cmpl %eax, %edx jne .L10 .L12: addl $1, %r10d addl %edx, %r8d cmpl %r10d, %esi je .L3 .L4: testl %edx, %edx jne .L11 jmp .L12 .L3: ret .cfi_endproc .LFE2057: .size _Z11fill_matrixPdjj, .-_Z11fill_matrixPdjj .section .rodata.str1.1,"aMS",@progbits,1 .LC2: .string "w" .LC3: .string "assignment2_2_out" .LC4: .string "%4.4f " .LC5: .string "\n" .text .globl _Z20print_matrix_to_filePdjj .type _Z20print_matrix_to_filePdjj, @function _Z20print_matrix_to_filePdjj: .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 subq $24, %rsp .cfi_def_cfa_offset 80 movq %rdi, %r13 movl %esi, %ebx movl %esi, 12(%rsp) movl %edx, %r15d leaq .LC2(%rip), %rsi leaq .LC3(%rip), %rdi call fopen@PLT movq %rax, %r12 movl %r15d, %ebp movl $0, 8(%rsp) leaq .LC4(%rip), %r14 testl %ebx, %ebx jne .L17 .L18: movq %r12, %rdi call fclose@PLT addq $24, %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 .L19: .cfi_restore_state movl %ebx, %eax movsd 0(%r13,%rax,8), %xmm0 movq %r14, %rdx movl $2, %esi movq %r12, %rdi movl $1, %eax call __fprintf_chk@PLT addl $1, %ebx cmpl %ebp, %ebx jne .L19 .L21: leaq .LC5(%rip), %rdx movl $2, %esi movq %r12, %rdi movl $0, %eax call __fprintf_chk@PLT addl $1, 8(%rsp) movl 8(%rsp), %eax addl %r15d, %ebp cmpl %eax, 12(%rsp) je .L18 .L17: movl %ebp, %ebx subl %r15d, %ebx testl %r15d, %r15d jne .L19 jmp .L21 .cfi_endproc .LFE2058: .size _Z20print_matrix_to_filePdjj, .-_Z20print_matrix_to_filePdjj .globl _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i .type _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i, @function _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i: .LFB2084: .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 .L31 .L27: movq 136(%rsp), %rax subq %fs:40, %rax jne .L32 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L31: .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 _Z23MatrixMulKernel_col_majPdS_S_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L27 .L32: call __stack_chk_fail@PLT .cfi_endproc .LFE2084: .size _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i, .-_Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i .globl _Z23MatrixMulKernel_col_majPdS_S_i .type _Z23MatrixMulKernel_col_majPdS_S_i, @function _Z23MatrixMulKernel_col_majPdS_S_i: .LFB2085: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2085: .size _Z23MatrixMulKernel_col_majPdS_S_i, .-_Z23MatrixMulKernel_col_majPdS_S_i .section .rodata.str1.1 .LC6: .string "Thread dims\n" .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC8: .string "\nMatrix A (first 10*10 inputs)\n" .section .rodata.str1.1 .LC9: .string "%f " .section .rodata.str1.8 .align 8 .LC10: .string "\n\nMatrix B (first 10*10 inputs)\n" .align 8 .LC11: .string "\nTime spent in col maj %f with threadsPerBlock %d \n" .align 8 .LC12: .string "\n\nMatrix C (first 10*10 outputs)\n" .align 8 .LC13: .string "\nWritting to file assignment_2_1_out as Mat C" .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 $168, %rsp .cfi_def_cfa_offset 224 movq %fs:40, %rax movq %rax, 152(%rsp) xorl %eax, %eax movl $536870912, %edi call malloc@PLT movq %rax, %r15 movq %rax, 8(%rsp) movl $536870912, %edi call malloc@PLT movq %rax, %rbx movq %rax, 16(%rsp) movl $536870912, %edi call malloc@PLT movq %rax, 24(%rsp) movl $8192, %edx movl $8192, %esi movq %r15, %rdi call _Z11fill_matrixPdjj movl $8192, %edx movl $8192, %esi movq %rbx, %rdi call _Z11fill_matrixPdjj leaq .LC6(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 112(%rsp), %r12 leaq 144(%rsp), %r13 movq %r12, %rbx movl $2, %ebp .L36: pxor %xmm1, %xmm1 cvtsi2sdl %ebp, %xmm1 movsd .LC7(%rip), %xmm0 call pow@PLT cvttsd2sil %xmm0, %eax movl %eax, (%rbx) addl $1, %ebp addq $4, %rbx cmpq %r13, %rbx jne .L36 leaq .LC8(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 8(%rsp), %rax leaq 80(%rax), %rbp leaq 655440(%rax), %r15 leaq .LC9(%rip), %r14 .L37: leaq -80(%rbp), %rbx .L38: movsd (%rbx), %xmm0 movq %r14, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $8, %rbx cmpq %rbp, %rbx jne .L38 addq $65536, %rbp cmpq %r15, %rbp jne .L37 leaq .LC10(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movq 16(%rsp), %rax leaq 80(%rax), %rbp leaq 655440(%rax), %r15 leaq .LC9(%rip), %r14 .L40: leaq -80(%rbp), %rbx .L41: movsd (%rbx), %xmm0 movq %r14, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $8, %rbx cmpq %rbp, %rbx jne .L41 addq $65536, %rbp cmpq %r15, %rbp jne .L40 leaq 48(%rsp), %rdi movl $536870912, %esi call cudaMalloc@PLT leaq 56(%rsp), %rdi movl $536870912, %esi call cudaMalloc@PLT leaq 64(%rsp), %rdi movl $536870912, %esi call cudaMalloc@PLT leaq 72(%rsp), %rdi call cudaEventCreate@PLT leaq 80(%rsp), %rdi call cudaEventCreate@PLT movl $1, %ecx movl $536870912, %edx movq 8(%rsp), %rsi movq 48(%rsp), %rdi call cudaMemcpy@PLT movl $1, %ecx movl $536870912, %edx movq 16(%rsp), %rsi movq 56(%rsp), %rdi call cudaMemcpy@PLT leaq .LC11(%rip), %rbp jmp .L44 .L43: movl $0, %esi movq 80(%rsp), %rdi call cudaEventRecord@PLT movq 80(%rsp), %rdi call cudaEventSynchronize@PLT leaq 44(%rsp), %rdi movq 80(%rsp), %rdx movq 72(%rsp), %rsi call cudaEventElapsedTime@PLT pxor %xmm0, %xmm0 cvtss2sd 44(%rsp), %xmm0 movl %ebx, %edx movq %rbp, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $4, %r12 cmpq %r13, %r12 je .L55 .L44: movl (%r12), %ebx movl $1, 96(%rsp) leal 8191(%rbx), %eax movl $0, %edx divl %ebx movl %eax, 100(%rsp) movl $8192, 104(%rsp) movl $1, 108(%rsp) movl $0, %esi movq 72(%rsp), %rdi call cudaEventRecord@PLT movl %ebx, 88(%rsp) movl $1, 92(%rsp) movl 96(%rsp), %ecx movl $0, %r9d movl $0, %r8d movq 88(%rsp), %rdx movq 100(%rsp), %rdi movl 108(%rsp), %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax jne .L43 movl $8192, %ecx movq 64(%rsp), %rdx movq 56(%rsp), %rsi movq 48(%rsp), %rdi call _Z48__device_stub__Z23MatrixMulKernel_col_majPdS_S_iPdS_S_i jmp .L43 .L55: movl $2, %ecx movl $536870912, %edx movq 64(%rsp), %rsi movq 24(%rsp), %rbx movq %rbx, %rdi call cudaMemcpy@PLT leaq .LC12(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT leaq 80(%rbx), %rbp leaq 655440(%rbx), %r13 leaq .LC9(%rip), %r12 .L45: leaq -80(%rbp), %rbx .L46: movsd (%rbx), %xmm0 movq %r12, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $8, %rbx cmpq %rbp, %rbx jne .L46 addq $65536, %rbp cmpq %r13, %rbp jne .L45 leaq .LC13(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT movl $8192, %edx movl $8192, %esi movq 24(%rsp), %rbx movq %rbx, %rdi call _Z20print_matrix_to_filePdjj movq 48(%rsp), %rdi call cudaFree@PLT movq 56(%rsp), %rdi call cudaFree@PLT movq 64(%rsp), %rdi call cudaFree@PLT movq 8(%rsp), %rdi call free@PLT movq 16(%rsp), %rdi call free@PLT movq %rbx, %rdi call free@PLT movq 152(%rsp), %rax subq %fs:40, %rax jne .L56 movl $0, %eax addq $168, %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 .L56: .cfi_restore_state call __stack_chk_fail@PLT .cfi_endproc .LFE2059: .size main, .-main .globl _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i .type _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i, @function _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i: .LFB2086: .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 .L61 .L57: movq 136(%rsp), %rax subq %fs:40, %rax jne .L62 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L61: .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 _Z23MatrixMulKernel_row_majPdS_S_i(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L57 .L62: call __stack_chk_fail@PLT .cfi_endproc .LFE2086: .size _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i, .-_Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i .globl _Z23MatrixMulKernel_row_majPdS_S_i .type _Z23MatrixMulKernel_row_majPdS_S_i, @function _Z23MatrixMulKernel_row_majPdS_S_i: .LFB2087: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z48__device_stub__Z23MatrixMulKernel_row_majPdS_S_iPdS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2087: .size _Z23MatrixMulKernel_row_majPdS_S_i, .-_Z23MatrixMulKernel_row_majPdS_S_i .section .rodata.str1.8 .align 8 .LC14: .string "_Z23MatrixMulKernel_row_majPdS_S_i" .align 8 .LC15: .string "_Z23MatrixMulKernel_col_majPdS_S_i" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2089: .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 .LC14(%rip), %rdx movq %rdx, %rcx leaq _Z23MatrixMulKernel_row_majPdS_S_i(%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 .LC15(%rip), %rdx movq %rdx, %rcx leaq _Z23MatrixMulKernel_col_majPdS_S_i(%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 .LFE2089: .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 1074161254 .align 4 .LC1: .long 1078774989 .section .rodata.cst8,"aM",@progbits,8 .align 8 .LC7: .long 0 .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 "ee16b068_2.hip" .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function _Z11fill_matrixPdjj .LCPI0_0: .long 0x40066666 # float 2.0999999 .LCPI0_1: .long 0x404ccccd # float 3.20000005 .text .globl _Z11fill_matrixPdjj .p2align 4, 0x90 .type _Z11fill_matrixPdjj,@function _Z11fill_matrixPdjj: # @_Z11fill_matrixPdjj .cfi_startproc # %bb.0: testl %esi, %esi je .LBB0_6 # %bb.1: # %.preheader.lr.ph movl %edx, %eax xorl %ecx, %ecx movss .LCPI0_0(%rip), %xmm0 # xmm0 = mem[0],zero,zero,zero movss .LCPI0_1(%rip), %xmm1 # xmm1 = mem[0],zero,zero,zero xorl %r8d, %r8d jmp .LBB0_2 .p2align 4, 0x90 .LBB0_5: # %._crit_edge # in Loop: Header=BB0_2 Depth=1 incl %r8d addq %rax, %rcx cmpl %esi, %r8d je .LBB0_6 .LBB0_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB0_4 Depth 2 testl %edx, %edx je .LBB0_5 # %bb.3: # %.lr.ph # in Loop: Header=BB0_2 Depth=1 movl %r8d, %r9d xorps %xmm2, %xmm2 cvtsi2ss %r9, %xmm2 mulss %xmm0, %xmm2 xorl %r9d, %r9d .p2align 4, 0x90 .LBB0_4: # Parent Loop BB0_2 Depth=1 # => This Inner Loop Header: Depth=2 movl %r9d, %r10d xorps %xmm3, %xmm3 cvtsi2ss %r10, %xmm3 mulss %xmm1, %xmm3 addss %xmm2, %xmm3 cvtss2sd %xmm3, %xmm3 leal (%rcx,%r9), %r10d movsd %xmm3, (%rdi,%r10,8) incq %r9 cmpq %r9, %rax jne .LBB0_4 jmp .LBB0_5 .LBB0_6: # %._crit_edge15 retq .Lfunc_end0: .size _Z11fill_matrixPdjj, .Lfunc_end0-_Z11fill_matrixPdjj .cfi_endproc # -- End function .globl _Z20print_matrix_to_filePdjj # -- Begin function _Z20print_matrix_to_filePdjj .p2align 4, 0x90 .type _Z20print_matrix_to_filePdjj,@function _Z20print_matrix_to_filePdjj: # @_Z20print_matrix_to_filePdjj .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 $24, %rsp .cfi_def_cfa_offset 80 .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 %edx, %ebx movl %esi, %ebp movq %rdi, %r14 movl $.L.str, %edi movl $.L.str.1, %esi callq fopen movq %rax, %r15 movl %ebp, 12(%rsp) # 4-byte Spill testl %ebp, %ebp je .LBB1_4 # %bb.1: # %.preheader.lr.ph movl %ebx, %ecx movl %ebx, %eax movq %rax, 16(%rsp) # 8-byte Spill xorl %ebx, %ebx xorl %ebp, %ebp movl %ecx, 8(%rsp) # 4-byte Spill jmp .LBB1_2 .p2align 4, 0x90 .LBB1_3: # %._crit_edge # in Loop: Header=BB1_2 Depth=1 movl $10, %edi movq %r15, %rsi callq fputc@PLT incl %ebp movl 8(%rsp), %ecx # 4-byte Reload addl %ecx, %ebx cmpl 12(%rsp), %ebp # 4-byte Folded Reload je .LBB1_4 .LBB1_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB1_5 Depth 2 movq 16(%rsp), %r12 # 8-byte Reload movl %ebx, %r13d testl %ecx, %ecx je .LBB1_3 .p2align 4, 0x90 .LBB1_5: # Parent Loop BB1_2 Depth=1 # => This Inner Loop Header: Depth=2 movl %r13d, %eax movsd (%r14,%rax,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.2, %esi movq %r15, %rdi movb $1, %al callq fprintf incl %r13d decq %r12 jne .LBB1_5 jmp .LBB1_3 .LBB1_4: # %._crit_edge17 movq %r15, %rdi addq $24, %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 jmp fclose # TAILCALL .Lfunc_end1: .size _Z20print_matrix_to_filePdjj, .Lfunc_end1-_Z20print_matrix_to_filePdjj .cfi_endproc # -- End function .globl _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i # -- Begin function _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .p2align 4, 0x90 .type _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i,@function _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i: # @_Z38__device_stub__MatrixMulKernel_col_majPdS_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 $_Z23MatrixMulKernel_col_majPdS_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_end2: .size _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i, .Lfunc_end2-_Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .cfi_endproc # -- End function .globl _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i # -- Begin function _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .p2align 4, 0x90 .type _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i,@function _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i: # @_Z38__device_stub__MatrixMulKernel_row_majPdS_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 $_Z23MatrixMulKernel_row_majPdS_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_end3: .size _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i, .Lfunc_end3-_Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function main .LCPI4_0: .long 0x40066666 # float 2.0999999 .LCPI4_1: .long 0x404ccccd # float 3.20000005 .section .rodata.cst8,"aM",@progbits,8 .p2align 3, 0x0 .LCPI4_2: .quad 0x3ff0000000000000 # double 1 .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 $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 $536870912, %edi # imm = 0x20000000 callq malloc movq %rax, %r12 movl $536870912, %edi # imm = 0x20000000 callq malloc movq %rax, %r13 movl $536870912, %edi # imm = 0x20000000 callq malloc movq %rax, %r14 xorl %eax, %eax movss .LCPI4_0(%rip), %xmm0 # xmm0 = mem[0],zero,zero,zero movss .LCPI4_1(%rip), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %r12, %rcx .p2align 4, 0x90 .LBB4_1: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB4_2 Depth 2 movl %eax, %edx xorps %xmm2, %xmm2 cvtsi2ss %rdx, %xmm2 mulss %xmm0, %xmm2 xorl %edx, %edx .p2align 4, 0x90 .LBB4_2: # Parent Loop BB4_1 Depth=1 # => This Inner Loop Header: Depth=2 movl %edx, %esi xorps %xmm3, %xmm3 cvtsi2ss %rsi, %xmm3 mulss %xmm1, %xmm3 addss %xmm2, %xmm3 cvtss2sd %xmm3, %xmm3 movsd %xmm3, (%rcx,%rdx,8) incq %rdx cmpq $8192, %rdx # imm = 0x2000 jne .LBB4_2 # %bb.3: # %._crit_edge.i # in Loop: Header=BB4_1 Depth=1 incl %eax addq $65536, %rcx # imm = 0x10000 cmpl $8192, %eax # imm = 0x2000 jne .LBB4_1 # %bb.4: # %.preheader.i75.preheader xorl %eax, %eax movq %r13, %rcx .p2align 4, 0x90 .LBB4_5: # %.preheader.i75 # =>This Loop Header: Depth=1 # Child Loop BB4_6 Depth 2 movl %eax, %edx xorps %xmm2, %xmm2 cvtsi2ss %rdx, %xmm2 mulss %xmm0, %xmm2 xorl %edx, %edx .p2align 4, 0x90 .LBB4_6: # Parent Loop BB4_5 Depth=1 # => This Inner Loop Header: Depth=2 movl %edx, %esi xorps %xmm3, %xmm3 cvtsi2ss %rsi, %xmm3 mulss %xmm1, %xmm3 addss %xmm2, %xmm3 cvtss2sd %xmm3, %xmm3 movsd %xmm3, (%rcx,%rdx,8) incq %rdx cmpq $8192, %rdx # imm = 0x2000 jne .LBB4_6 # %bb.7: # %._crit_edge.i80 # in Loop: Header=BB4_5 Depth=1 incl %eax addq $65536, %rcx # imm = 0x10000 cmpl $8192, %eax # imm = 0x2000 jne .LBB4_5 # %bb.8: # %_Z11fill_matrixPdjj.exit82 movl $.Lstr, %edi callq puts@PLT xorl %ebx, %ebx .p2align 4, 0x90 .LBB4_9: # =>This Inner Loop Header: Depth=1 leal 2(%rbx), %edi movsd .LCPI4_2(%rip), %xmm0 # xmm0 = mem[0],zero callq ldexp@PLT cvttsd2si %xmm0, %eax movl %eax, 176(%rsp,%rbx,4) incq %rbx cmpq $8, %rbx jne .LBB4_9 # %bb.10: movq %r14, 48(%rsp) # 8-byte Spill movl $.Lstr.1, %edi callq puts@PLT xorl %ebx, %ebx movq %r12, %r14 .p2align 4, 0x90 .LBB4_11: # %.preheader91 # =>This Loop Header: Depth=1 # Child Loop BB4_12 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB4_12: # Parent Loop BB4_11 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r14,%r15,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.6, %edi movb $1, %al callq printf incq %r15 cmpq $10, %r15 jne .LBB4_12 # %bb.13: # in Loop: Header=BB4_11 Depth=1 incq %rbx addq $65536, %r14 # imm = 0x10000 cmpq $10, %rbx jne .LBB4_11 # %bb.14: movl $.Lstr.2, %edi callq puts@PLT xorl %ebx, %ebx movq %r13, %r14 .p2align 4, 0x90 .LBB4_15: # %.preheader90 # =>This Loop Header: Depth=1 # Child Loop BB4_16 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB4_16: # Parent Loop BB4_15 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r14,%r15,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.6, %edi movb $1, %al callq printf incq %r15 cmpq $10, %r15 jne .LBB4_16 # %bb.17: # in Loop: Header=BB4_15 Depth=1 incq %rbx addq $65536, %r14 # imm = 0x10000 cmpq $10, %rbx jne .LBB4_15 # %bb.18: leaq 24(%rsp), %rdi movl $536870912, %esi # imm = 0x20000000 callq hipMalloc leaq 16(%rsp), %rdi movl $536870912, %esi # imm = 0x20000000 callq hipMalloc leaq 8(%rsp), %rdi movl $536870912, %esi # imm = 0x20000000 callq hipMalloc leaq 40(%rsp), %rdi callq hipEventCreate movq %rsp, %rdi callq hipEventCreate movq 24(%rsp), %rdi movl $536870912, %edx # imm = 0x20000000 movq %r12, 64(%rsp) # 8-byte Spill movq %r12, %rsi movl $1, %ecx callq hipMemcpy movq 16(%rsp), %rdi movl $536870912, %edx # imm = 0x20000000 movq %r13, 56(%rsp) # 8-byte Spill movq %r13, %rsi movl $1, %ecx callq hipMemcpy xorl %ebx, %ebx movabsq $35184372088832, %r14 # imm = 0x200000000000 leaq 32(%rsp), %r15 jmp .LBB4_19 .p2align 4, 0x90 .LBB4_21: # in Loop: Header=BB4_19 Depth=1 movq (%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq (%rsp), %rdi callq hipEventSynchronize movq 40(%rsp), %rsi movq (%rsp), %rdx movq %r15, %rdi callq hipEventElapsedTime movss 32(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.8, %edi movl %r13d, %esi movb $1, %al callq printf incq %rbx cmpq $8, %rbx je .LBB4_22 .LBB4_19: # =>This Inner Loop Header: Depth=1 movl 176(%rsp,%rbx,4), %r13d movq %r13, %r12 movabsq $4294967296, %rax # imm = 0x100000000 orq %rax, %r12 leal 8191(%r13), %eax xorl %edx, %edx divl %r13d # kill: def $eax killed $eax def $rax leaq (%rax,%r14), %rbp movq 40(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq %rbp, %rdi movl $1, %esi movq %r12, %rdx movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB4_21 # %bb.20: # in Loop: Header=BB4_19 Depth=1 movq 24(%rsp), %rax movq 16(%rsp), %rcx movq 8(%rsp), %rdx movq %rax, 136(%rsp) movq %rcx, 128(%rsp) movq %rdx, 120(%rsp) movl $8192, 36(%rsp) # imm = 0x2000 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 36(%rsp), %rax movq %rax, 168(%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 movl $_Z23MatrixMulKernel_col_majPdS_S_i, %edi leaq 144(%rsp), %r9 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 jmp .LBB4_21 .LBB4_22: movq 8(%rsp), %rsi movl $536870912, %edx # imm = 0x20000000 movq 48(%rsp), %rbp # 8-byte Reload movq %rbp, %rdi movl $2, %ecx callq hipMemcpy movl $.Lstr.3, %edi callq puts@PLT xorl %ebx, %ebx movq %rbp, %r14 .p2align 4, 0x90 .LBB4_23: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB4_24 Depth 2 xorl %r15d, %r15d .p2align 4, 0x90 .LBB4_24: # Parent Loop BB4_23 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r14,%r15,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.6, %edi movb $1, %al callq printf incq %r15 cmpq $10, %r15 jne .LBB4_24 # %bb.25: # in Loop: Header=BB4_23 Depth=1 incq %rbx addq $65536, %r14 # imm = 0x10000 cmpq $10, %rbx jne .LBB4_23 # %bb.26: xorl %r14d, %r14d movl $.L.str.10, %edi xorl %eax, %eax callq printf movl $.L.str, %edi movl $.L.str.1, %esi callq fopen movq %rax, %rbx movq %rbp, %r15 movq 56(%rsp), %r13 # 8-byte Reload .p2align 4, 0x90 .LBB4_27: # %.preheader.i83 # =>This Loop Header: Depth=1 # Child Loop BB4_28 Depth 2 xorl %r12d, %r12d .p2align 4, 0x90 .LBB4_28: # Parent Loop BB4_27 Depth=1 # => This Inner Loop Header: Depth=2 movsd (%r15,%r12,8), %xmm0 # xmm0 = mem[0],zero movl $.L.str.2, %esi movq %rbx, %rdi movb $1, %al callq fprintf incq %r12 cmpq $8192, %r12 # imm = 0x2000 jne .LBB4_28 # %bb.29: # %._crit_edge.i87 # in Loop: Header=BB4_27 Depth=1 movl $10, %edi movq %rbx, %rsi callq fputc@PLT incl %r14d addq $65536, %r15 # imm = 0x10000 cmpl $8192, %r14d # imm = 0x2000 jne .LBB4_27 # %bb.30: # %_Z20print_matrix_to_filePdjj.exit movq %rbx, %rdi callq fclose movq 24(%rsp), %rdi callq hipFree movq 16(%rsp), %rdi callq hipFree movq 8(%rsp), %rdi callq hipFree movq 64(%rsp), %rdi # 8-byte Reload callq free movq %r13, %rdi callq free movq %rbp, %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_end4: .size main, .Lfunc_end4-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: 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 .LBB5_2 # %bb.1: movl $__hip_fatbin_wrapper, %edi callq __hipRegisterFatBinary movq %rax, __hip_gpubin_handle(%rip) .LBB5_2: movq __hip_gpubin_handle(%rip), %rbx xorps %xmm0, %xmm0 movups %xmm0, 16(%rsp) movups %xmm0, (%rsp) movl $_Z23MatrixMulKernel_col_majPdS_S_i, %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 $_Z23MatrixMulKernel_row_majPdS_S_i, %esi movl $.L__unnamed_2, %edx movl $.L__unnamed_2, %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_end5: .size __hip_module_ctor, .Lfunc_end5-__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 .LBB6_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 .LBB6_2: retq .Lfunc_end6: .size __hip_module_dtor, .Lfunc_end6-__hip_module_dtor .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "assignment2_2_out" .size .L.str, 18 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "w" .size .L.str.1, 2 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "%4.4f " .size .L.str.2, 7 .type _Z23MatrixMulKernel_col_majPdS_S_i,@object # @_Z23MatrixMulKernel_col_majPdS_S_i .section .rodata,"a",@progbits .globl _Z23MatrixMulKernel_col_majPdS_S_i .p2align 3, 0x0 _Z23MatrixMulKernel_col_majPdS_S_i: .quad _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .size _Z23MatrixMulKernel_col_majPdS_S_i, 8 .type _Z23MatrixMulKernel_row_majPdS_S_i,@object # @_Z23MatrixMulKernel_row_majPdS_S_i .globl _Z23MatrixMulKernel_row_majPdS_S_i .p2align 3, 0x0 _Z23MatrixMulKernel_row_majPdS_S_i: .quad _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .size _Z23MatrixMulKernel_row_majPdS_S_i, 8 .type .L.str.6,@object # @.str.6 .section .rodata.str1.1,"aMS",@progbits,1 .L.str.6: .asciz "%f " .size .L.str.6, 4 .type .L.str.8,@object # @.str.8 .L.str.8: .asciz "\nTime spent in col maj %f with threadsPerBlock %d \n" .size .L.str.8, 52 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "\nWritting to file assignment_2_1_out as Mat C" .size .L.str.10, 46 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z23MatrixMulKernel_col_majPdS_S_i" .size .L__unnamed_1, 35 .type .L__unnamed_2,@object # @1 .L__unnamed_2: .asciz "_Z23MatrixMulKernel_row_majPdS_S_i" .size .L__unnamed_2, 35 .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 "Thread dims" .size .Lstr, 12 .type .Lstr.1,@object # @str.1 .Lstr.1: .asciz "\nMatrix A (first 10*10 inputs)" .size .Lstr.1, 31 .type .Lstr.2,@object # @str.2 .Lstr.2: .asciz "\n\nMatrix B (first 10*10 inputs)" .size .Lstr.2, 32 .type .Lstr.3,@object # @str.3 .Lstr.3: .asciz "\n\nMatrix C (first 10*10 outputs)" .size .Lstr.3, 33 .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 _Z38__device_stub__MatrixMulKernel_col_majPdS_S_i .addrsig_sym _Z38__device_stub__MatrixMulKernel_row_majPdS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z23MatrixMulKernel_col_majPdS_S_i .addrsig_sym _Z23MatrixMulKernel_row_majPdS_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 "includes.h" __global__ void grayScale2(uchar3 *input, uchar3 *output,int width, int height) { int x = threadIdx.x + blockIdx.x * blockDim.x; int y = threadIdx.y + blockIdx.y * blockDim.y; int w = blockDim.x * gridDim.x; //if ((gridDim.x * gridDim.y) < width * height){ int tid = y*w + x; output[tid].x = (input[tid].x + input[tid].y + input[tid].z) / 3; output[tid].z = output[tid].y = output[tid].x; // } }
code for sm_80 Function : _Z10grayScale2P6uchar3S0_ii .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_TID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002200 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R7, SR_CTAID.Y ; /* 0x0000000000077919 */ /* 0x000e280000002600 */ /*0040*/ S2R R5, SR_CTAID.X ; /* 0x0000000000057919 */ /* 0x000e680000002500 */ /*0050*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000ea20000002100 */ /*0060*/ IMAD R0, R7, c[0x0][0x4], R0 ; /* 0x0000010007007a24 */ /* 0x001fe200078e0200 */ /*0070*/ HFMA2.MMA R7, -RZ, RZ, 0, 1.78813934326171875e-07 ; /* 0x00000003ff077435 */ /* 0x000fc600000001ff */ /*0080*/ IMAD R0, R0, c[0x0][0xc], R5 ; /* 0x0000030000007a24 */ /* 0x002fc800078e0205 */ /*0090*/ IMAD R0, R0, c[0x0][0x0], R3 ; /* 0x0000000000007a24 */ /* 0x004fc800078e0203 */ /*00a0*/ IMAD.WIDE R2, R0, R7, c[0x0][0x160] ; /* 0x0000580000027625 */ /* 0x000fca00078e0207 */ /*00b0*/ LDG.E.U8 R4, [R2.64] ; /* 0x0000000402047981 */ /* 0x000ea8000c1e1100 */ /*00c0*/ LDG.E.U8 R5, [R2.64+0x1] ; /* 0x0000010402057981 */ /* 0x000ea8000c1e1100 */ /*00d0*/ LDG.E.U8 R6, [R2.64+0x2] ; /* 0x0000020402067981 */ /* 0x000ea4000c1e1100 */ /*00e0*/ IADD3 R4, R6, R5, R4 ; /* 0x0000000506047210 */ /* 0x004fe20007ffe004 */ /*00f0*/ IMAD.WIDE R6, R0, R7, c[0x0][0x168] ; /* 0x00005a0000067625 */ /* 0x000fc800078e0207 */ /*0100*/ IMAD.WIDE.U32 R4, R4, -0x55555555, RZ ; /* 0xaaaaaaab04047825 */ /* 0x000fca00078e00ff */ /*0110*/ SHF.R.U32.HI R5, RZ, 0x1, R5 ; /* 0x00000001ff057819 */ /* 0x000fca0000011605 */ /*0120*/ STG.E.U8 [R6.64], R5 ; /* 0x0000000506007986 */ /* 0x000fe8000c101104 */ /*0130*/ STG.E.U8 [R6.64+0x1], R5 ; /* 0x0000010506007986 */ /* 0x000fe8000c101104 */ /*0140*/ STG.E.U8 [R6.64+0x2], R5 ; /* 0x0000020506007986 */ /* 0x000fe2000c101104 */ /*0150*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0160*/ BRA 0x160; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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 grayScale2(uchar3 *input, uchar3 *output,int width, int height) { int x = threadIdx.x + blockIdx.x * blockDim.x; int y = threadIdx.y + blockIdx.y * blockDim.y; int w = blockDim.x * gridDim.x; //if ((gridDim.x * gridDim.y) < width * height){ int tid = y*w + x; output[tid].x = (input[tid].x + input[tid].y + input[tid].z) / 3; output[tid].z = output[tid].y = output[tid].x; // } }
.file "tmpxft_00025f45_00000000-6_grayScale2.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__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii .type _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii, @function _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii: .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) 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) 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 _Z10grayScale2P6uchar3S0_ii(%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 _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii, .-_Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii .globl _Z10grayScale2P6uchar3S0_ii .type _Z10grayScale2P6uchar3S0_ii, @function _Z10grayScale2P6uchar3S0_ii: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z10grayScale2P6uchar3S0_ii, .-_Z10grayScale2P6uchar3S0_ii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z10grayScale2P6uchar3S0_ii" .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 _Z10grayScale2P6uchar3S0_ii(%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 grayScale2(uchar3 *input, uchar3 *output,int width, int height) { int x = threadIdx.x + blockIdx.x * blockDim.x; int y = threadIdx.y + blockIdx.y * blockDim.y; int w = blockDim.x * gridDim.x; //if ((gridDim.x * gridDim.y) < width * height){ int tid = y*w + x; output[tid].x = (input[tid].x + input[tid].y + input[tid].z) / 3; output[tid].z = output[tid].y = output[tid].x; // } }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void grayScale2(uchar3 *input, uchar3 *output,int width, int height) { int x = threadIdx.x + blockIdx.x * blockDim.x; int y = threadIdx.y + blockIdx.y * blockDim.y; int w = blockDim.x * gridDim.x; //if ((gridDim.x * gridDim.y) < width * height){ int tid = y*w + x; output[tid].x = (input[tid].x + input[tid].y + input[tid].z) / 3; output[tid].z = output[tid].y = output[tid].x; // } }
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 grayScale2(uchar3 *input, uchar3 *output,int width, int height) { int x = threadIdx.x + blockIdx.x * blockDim.x; int y = threadIdx.y + blockIdx.y * blockDim.y; int w = blockDim.x * gridDim.x; //if ((gridDim.x * gridDim.y) < width * height){ int tid = y*w + x; output[tid].x = (input[tid].x + input[tid].y + input[tid].z) / 3; output[tid].z = output[tid].y = output[tid].x; // } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .globl _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .p2align 8 .type _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii,@function _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii: s_clause 0x1 s_load_b32 s4, s[0:1], 0x24 s_load_b32 s5, s[0:1], 0x18 v_bfe_u32 v1, v0, 10, 10 v_and_b32_e32 v0, 0x3ff, v0 s_waitcnt lgkmcnt(0) s_lshr_b32 s2, s4, 16 s_and_b32 s4, s4, 0xffff v_mad_u64_u32 v[2:3], null, s15, s2, v[1:2] s_load_b128 s[0:3], s[0:1], 0x0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[3:4], null, v2, s5, s[14:15] v_mad_u64_u32 v[1:2], null, v3, s4, v[0:1] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) v_mad_i64_i32 v[2:3], null, v1, 3, s[0:1] s_clause 0x2 global_load_u8 v0, v[2:3], off global_load_u8 v4, v[2:3], off offset:1 global_load_u8 v2, v[2:3], off offset:2 s_waitcnt vmcnt(1) v_add_nc_u16 v0, v4, v0 s_waitcnt vmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_add_nc_u16 v0, v0, v2 v_mad_i64_i32 v[2:3], null, v1, 3, s[2:3] v_and_b32_e32 v0, 0xffff, v0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mul_u32_u24_e32 v0, 0x5556, v0 v_lshrrev_b32_e32 v0, 16, v0 s_clause 0x2 global_store_b8 v[2:3], v0, off global_store_b8 v[2:3], v0, off offset:1 global_store_b8 v[2:3], v0, off offset:2 s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .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 1 .amdhsa_next_free_vgpr 5 .amdhsa_next_free_sgpr 16 .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 _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, .Lfunc_end0-_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .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: 20 .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: _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .private_segment_fixed_size: 0 .sgpr_count: 16 .sgpr_spill_count: 0 .symbol: _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii.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 <hip/hip_runtime.h> #include "includes.h" __global__ void grayScale2(uchar3 *input, uchar3 *output,int width, int height) { int x = threadIdx.x + blockIdx.x * blockDim.x; int y = threadIdx.y + blockIdx.y * blockDim.y; int w = blockDim.x * gridDim.x; //if ((gridDim.x * gridDim.y) < width * height){ int tid = y*w + x; output[tid].x = (input[tid].x + input[tid].y + input[tid].z) / 3; output[tid].z = output[tid].y = output[tid].x; // } }
.text .file "grayScale2.hip" .globl _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii # -- Begin function _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .p2align 4, 0x90 .type _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii,@function _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii: # @_Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movl %edx, 12(%rsp) movl %ecx, 8(%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 8(%rsp), %rax movq %rax, 104(%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 $_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii, .Lfunc_end0-_Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .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 $_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, %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 _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii,@object # @_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .section .rodata,"a",@progbits .globl _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .p2align 3, 0x0 _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii: .quad _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .size _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii" .size .L__unnamed_1, 45 .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 _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .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 : _Z10grayScale2P6uchar3S0_ii .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_TID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002200 */ /*0020*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fc60000000a00 */ /*0030*/ S2R R7, SR_CTAID.Y ; /* 0x0000000000077919 */ /* 0x000e280000002600 */ /*0040*/ S2R R5, SR_CTAID.X ; /* 0x0000000000057919 */ /* 0x000e680000002500 */ /*0050*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000ea20000002100 */ /*0060*/ IMAD R0, R7, c[0x0][0x4], R0 ; /* 0x0000010007007a24 */ /* 0x001fe200078e0200 */ /*0070*/ HFMA2.MMA R7, -RZ, RZ, 0, 1.78813934326171875e-07 ; /* 0x00000003ff077435 */ /* 0x000fc600000001ff */ /*0080*/ IMAD R0, R0, c[0x0][0xc], R5 ; /* 0x0000030000007a24 */ /* 0x002fc800078e0205 */ /*0090*/ IMAD R0, R0, c[0x0][0x0], R3 ; /* 0x0000000000007a24 */ /* 0x004fc800078e0203 */ /*00a0*/ IMAD.WIDE R2, R0, R7, c[0x0][0x160] ; /* 0x0000580000027625 */ /* 0x000fca00078e0207 */ /*00b0*/ LDG.E.U8 R4, [R2.64] ; /* 0x0000000402047981 */ /* 0x000ea8000c1e1100 */ /*00c0*/ LDG.E.U8 R5, [R2.64+0x1] ; /* 0x0000010402057981 */ /* 0x000ea8000c1e1100 */ /*00d0*/ LDG.E.U8 R6, [R2.64+0x2] ; /* 0x0000020402067981 */ /* 0x000ea4000c1e1100 */ /*00e0*/ IADD3 R4, R6, R5, R4 ; /* 0x0000000506047210 */ /* 0x004fe20007ffe004 */ /*00f0*/ IMAD.WIDE R6, R0, R7, c[0x0][0x168] ; /* 0x00005a0000067625 */ /* 0x000fc800078e0207 */ /*0100*/ IMAD.WIDE.U32 R4, R4, -0x55555555, RZ ; /* 0xaaaaaaab04047825 */ /* 0x000fca00078e00ff */ /*0110*/ SHF.R.U32.HI R5, RZ, 0x1, R5 ; /* 0x00000001ff057819 */ /* 0x000fca0000011605 */ /*0120*/ STG.E.U8 [R6.64], R5 ; /* 0x0000000506007986 */ /* 0x000fe8000c101104 */ /*0130*/ STG.E.U8 [R6.64+0x1], R5 ; /* 0x0000010506007986 */ /* 0x000fe8000c101104 */ /*0140*/ STG.E.U8 [R6.64+0x2], R5 ; /* 0x0000020506007986 */ /* 0x000fe2000c101104 */ /*0150*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0160*/ BRA 0x160; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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 _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .globl _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .p2align 8 .type _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii,@function _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii: s_clause 0x1 s_load_b32 s4, s[0:1], 0x24 s_load_b32 s5, s[0:1], 0x18 v_bfe_u32 v1, v0, 10, 10 v_and_b32_e32 v0, 0x3ff, v0 s_waitcnt lgkmcnt(0) s_lshr_b32 s2, s4, 16 s_and_b32 s4, s4, 0xffff v_mad_u64_u32 v[2:3], null, s15, s2, v[1:2] s_load_b128 s[0:3], s[0:1], 0x0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mad_u64_u32 v[3:4], null, v2, s5, s[14:15] v_mad_u64_u32 v[1:2], null, v3, s4, v[0:1] s_waitcnt lgkmcnt(0) s_delay_alu instid0(VALU_DEP_1) v_mad_i64_i32 v[2:3], null, v1, 3, s[0:1] s_clause 0x2 global_load_u8 v0, v[2:3], off global_load_u8 v4, v[2:3], off offset:1 global_load_u8 v2, v[2:3], off offset:2 s_waitcnt vmcnt(1) v_add_nc_u16 v0, v4, v0 s_waitcnt vmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_add_nc_u16 v0, v0, v2 v_mad_i64_i32 v[2:3], null, v1, 3, s[2:3] v_and_b32_e32 v0, 0xffff, v0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mul_u32_u24_e32 v0, 0x5556, v0 v_lshrrev_b32_e32 v0, 16, v0 s_clause 0x2 global_store_b8 v[2:3], v0, off global_store_b8 v[2:3], v0, off offset:1 global_store_b8 v[2:3], v0, off offset:2 s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .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 1 .amdhsa_next_free_vgpr 5 .amdhsa_next_free_sgpr 16 .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 _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, .Lfunc_end0-_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .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: 20 .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: _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .private_segment_fixed_size: 0 .sgpr_count: 16 .sgpr_spill_count: 0 .symbol: _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii.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_00025f45_00000000-6_grayScale2.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__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii .type _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii, @function _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii: .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) 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) 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 _Z10grayScale2P6uchar3S0_ii(%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 _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii, .-_Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii .globl _Z10grayScale2P6uchar3S0_ii .type _Z10grayScale2P6uchar3S0_ii, @function _Z10grayScale2P6uchar3S0_ii: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z41__device_stub__Z10grayScale2P6uchar3S0_iiP6uchar3S0_ii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z10grayScale2P6uchar3S0_ii, .-_Z10grayScale2P6uchar3S0_ii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z10grayScale2P6uchar3S0_ii" .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 _Z10grayScale2P6uchar3S0_ii(%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 "grayScale2.hip" .globl _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii # -- Begin function _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .p2align 4, 0x90 .type _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii,@function _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii: # @_Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movl %edx, 12(%rsp) movl %ecx, 8(%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 8(%rsp), %rax movq %rax, 104(%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 $_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii, .Lfunc_end0-_Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .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 $_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, %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 _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii,@object # @_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .section .rodata,"a",@progbits .globl _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .p2align 3, 0x0 _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii: .quad _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .size _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii" .size .L__unnamed_1, 45 .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 _Z25__device_stub__grayScale2P15HIP_vector_typeIhLj3EES1_ii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z10grayScale2P15HIP_vector_typeIhLj3EES1_ii .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 sync_ndconv_groups() { }
code for sm_80 Function : _Z18sync_ndconv_groupsv .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.
#include "includes.h" __global__ void sync_ndconv_groups() { }
.file "tmpxft_000b925a_00000000-6_sync_ndconv_groups.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__Z18sync_ndconv_groupsvv .type _Z37__device_stub__Z18sync_ndconv_groupsvv, @function _Z37__device_stub__Z18sync_ndconv_groupsvv: .LFB2051: .cfi_startproc endbr64 subq $88, %rsp .cfi_def_cfa_offset 96 movq %fs:40, %rax movq %rax, 72(%rsp) xorl %eax, %eax 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 .L7 .L3: movq 72(%rsp), %rax subq %fs:40, %rax jne .L8 addq $88, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 8(%rsp) .cfi_def_cfa_offset 104 pushq 8(%rsp) .cfi_def_cfa_offset 112 leaq 80(%rsp), %r9 movq 44(%rsp), %rcx movl 52(%rsp), %r8d movq 32(%rsp), %rsi movl 40(%rsp), %edx leaq _Z18sync_ndconv_groupsv(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 96 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z37__device_stub__Z18sync_ndconv_groupsvv, .-_Z37__device_stub__Z18sync_ndconv_groupsvv .globl _Z18sync_ndconv_groupsv .type _Z18sync_ndconv_groupsv, @function _Z18sync_ndconv_groupsv: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z37__device_stub__Z18sync_ndconv_groupsvv addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z18sync_ndconv_groupsv, .-_Z18sync_ndconv_groupsv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z18sync_ndconv_groupsv" .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 _Z18sync_ndconv_groupsv(%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 sync_ndconv_groups() { }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void sync_ndconv_groups() { }
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 sync_ndconv_groups() { }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z18sync_ndconv_groupsv .globl _Z18sync_ndconv_groupsv .p2align 8 .type _Z18sync_ndconv_groupsv,@function _Z18sync_ndconv_groupsv: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z18sync_ndconv_groupsv .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 0 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 0 .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 1 .amdhsa_next_free_sgpr 1 .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 _Z18sync_ndconv_groupsv, .Lfunc_end0-_Z18sync_ndconv_groupsv .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: [] .group_segment_fixed_size: 0 .kernarg_segment_align: 4 .kernarg_segment_size: 0 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z18sync_ndconv_groupsv .private_segment_fixed_size: 0 .sgpr_count: 0 .sgpr_spill_count: 0 .symbol: _Z18sync_ndconv_groupsv.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 0 .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 sync_ndconv_groups() { }
.text .file "sync_ndconv_groups.hip" .globl _Z33__device_stub__sync_ndconv_groupsv # -- Begin function _Z33__device_stub__sync_ndconv_groupsv .p2align 4, 0x90 .type _Z33__device_stub__sync_ndconv_groupsv,@function _Z33__device_stub__sync_ndconv_groupsv: # @_Z33__device_stub__sync_ndconv_groupsv .cfi_startproc # %bb.0: subq $56, %rsp .cfi_def_cfa_offset 64 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 48(%rsp), %r9 movl $_Z18sync_ndconv_groupsv, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $72, %rsp .cfi_adjust_cfa_offset -72 retq .Lfunc_end0: .size _Z33__device_stub__sync_ndconv_groupsv, .Lfunc_end0-_Z33__device_stub__sync_ndconv_groupsv .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 $_Z18sync_ndconv_groupsv, %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 _Z18sync_ndconv_groupsv,@object # @_Z18sync_ndconv_groupsv .section .rodata,"a",@progbits .globl _Z18sync_ndconv_groupsv .p2align 3, 0x0 _Z18sync_ndconv_groupsv: .quad _Z33__device_stub__sync_ndconv_groupsv .size _Z18sync_ndconv_groupsv, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z18sync_ndconv_groupsv" .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 _Z33__device_stub__sync_ndconv_groupsv .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z18sync_ndconv_groupsv .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 : _Z18sync_ndconv_groupsv .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 _Z18sync_ndconv_groupsv .globl _Z18sync_ndconv_groupsv .p2align 8 .type _Z18sync_ndconv_groupsv,@function _Z18sync_ndconv_groupsv: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z18sync_ndconv_groupsv .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 0 .amdhsa_user_sgpr_count 15 .amdhsa_user_sgpr_dispatch_ptr 0 .amdhsa_user_sgpr_queue_ptr 0 .amdhsa_user_sgpr_kernarg_segment_ptr 0 .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 1 .amdhsa_next_free_sgpr 1 .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 _Z18sync_ndconv_groupsv, .Lfunc_end0-_Z18sync_ndconv_groupsv .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: [] .group_segment_fixed_size: 0 .kernarg_segment_align: 4 .kernarg_segment_size: 0 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z18sync_ndconv_groupsv .private_segment_fixed_size: 0 .sgpr_count: 0 .sgpr_spill_count: 0 .symbol: _Z18sync_ndconv_groupsv.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 0 .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_000b925a_00000000-6_sync_ndconv_groups.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__Z18sync_ndconv_groupsvv .type _Z37__device_stub__Z18sync_ndconv_groupsvv, @function _Z37__device_stub__Z18sync_ndconv_groupsvv: .LFB2051: .cfi_startproc endbr64 subq $88, %rsp .cfi_def_cfa_offset 96 movq %fs:40, %rax movq %rax, 72(%rsp) xorl %eax, %eax 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 .L7 .L3: movq 72(%rsp), %rax subq %fs:40, %rax jne .L8 addq $88, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L7: .cfi_restore_state pushq 8(%rsp) .cfi_def_cfa_offset 104 pushq 8(%rsp) .cfi_def_cfa_offset 112 leaq 80(%rsp), %r9 movq 44(%rsp), %rcx movl 52(%rsp), %r8d movq 32(%rsp), %rsi movl 40(%rsp), %edx leaq _Z18sync_ndconv_groupsv(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 96 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z37__device_stub__Z18sync_ndconv_groupsvv, .-_Z37__device_stub__Z18sync_ndconv_groupsvv .globl _Z18sync_ndconv_groupsv .type _Z18sync_ndconv_groupsv, @function _Z18sync_ndconv_groupsv: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z37__device_stub__Z18sync_ndconv_groupsvv addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z18sync_ndconv_groupsv, .-_Z18sync_ndconv_groupsv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z18sync_ndconv_groupsv" .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 _Z18sync_ndconv_groupsv(%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 "sync_ndconv_groups.hip" .globl _Z33__device_stub__sync_ndconv_groupsv # -- Begin function _Z33__device_stub__sync_ndconv_groupsv .p2align 4, 0x90 .type _Z33__device_stub__sync_ndconv_groupsv,@function _Z33__device_stub__sync_ndconv_groupsv: # @_Z33__device_stub__sync_ndconv_groupsv .cfi_startproc # %bb.0: subq $56, %rsp .cfi_def_cfa_offset 64 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 48(%rsp), %r9 movl $_Z18sync_ndconv_groupsv, %edi pushq (%rsp) .cfi_adjust_cfa_offset 8 pushq 16(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $72, %rsp .cfi_adjust_cfa_offset -72 retq .Lfunc_end0: .size _Z33__device_stub__sync_ndconv_groupsv, .Lfunc_end0-_Z33__device_stub__sync_ndconv_groupsv .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 $_Z18sync_ndconv_groupsv, %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 _Z18sync_ndconv_groupsv,@object # @_Z18sync_ndconv_groupsv .section .rodata,"a",@progbits .globl _Z18sync_ndconv_groupsv .p2align 3, 0x0 _Z18sync_ndconv_groupsv: .quad _Z33__device_stub__sync_ndconv_groupsv .size _Z18sync_ndconv_groupsv, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z18sync_ndconv_groupsv" .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 _Z33__device_stub__sync_ndconv_groupsv .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z18sync_ndconv_groupsv .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_runtime.h> #include <asm/unistd.h> #include <fcntl.h> #include <inttypes.h> #include <linux/kernel-page-flags.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string> #include <string.h> #include <sys/ioctl.h> #include <sys/mount.h> #include <sys/mman.h> #include <sys/stat.h> #include <sys/sysinfo.h> #include <sys/wait.h> #include <time.h> #include <unistd.h> #include <vector> #include <sys/time.h> #include <assert.h> #define gpuErrchk(ans) { gpuAssert((ans), __FILE__, __LINE__); } inline void gpuAssert(cudaError_t code, const char *file, int line, bool abort = true) { if (code != cudaSuccess) { fprintf(stderr, "GPUassert: %s %s %d\n", cudaGetErrorString(code), file, line); if (abort) exit(code); } } __global__ void kernel(int number_of_threads, float * dsum ,volatile int * d_mapping, int cnt, int fence_system_flag, int fence_block_flag) { int i; /*printf("D: i am [%d] \n", blockIdx.x * blockDim.x * blockDim.y * blockDim.z + threadIdx.z * blockDim.y * blockDim.x + threadIdx.y * blockDim.x + threadIdx.x); */ for(i=0; i < cnt ; i++){ *dsum = i; d_mapping[0] = *dsum; if(fence_system_flag){ __threadfence_system(); } if(fence_block_flag){ __threadfence_block(); } } } int main(int argc, char **argv) { int opt, BLOCKS = 1, THREADS = 1, cnt =10000, fence_system_flag =0, fence_block_flag = 0; cudaEvent_t start, stop; float elapsed_time =0; gpuErrchk(cudaEventCreate(&start)); gpuErrchk(cudaEventCreate(&stop)); while ((opt = getopt(argc, argv, "b:t:n:f:s:")) != -1) { switch (opt) { case 'b': BLOCKS = atoi(optarg); break; case 't': THREADS = atoi(optarg); break; case 'n': cnt = atoi(optarg); break; case 'f': fence_system_flag = atoi(optarg); break; case 's': fence_block_flag = atoi(optarg); break; default: fprintf(stderr, "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n", argv[0]); exit(EXIT_FAILURE); } } float * dsum; gpuErrchk(cudaMallocManaged((void **) &dsum, sizeof(uint64_t))); volatile int * h_mapping; gpuErrchk(cudaHostAlloc( (void**)&h_mapping, sizeof(volatile int), cudaHostAllocMapped)); volatile int * d_mapping; gpuErrchk(cudaHostGetDevicePointer((void**)&d_mapping,(void*)h_mapping,0)); *dsum = 0; cudaEventRecord(start, 0); kernel <<< BLOCKS, THREADS >>> (BLOCKS * THREADS,dsum,d_mapping, cnt, fence_system_flag, fence_block_flag); gpuErrchk(cudaDeviceSynchronize()); gpuErrchk(cudaEventRecord(stop, 0)); gpuErrchk(cudaEventSynchronize(stop)); gpuErrchk(cudaEventElapsedTime (&elapsed_time, start, stop)); assert(*dsum != 0); printf("H: elapsed_time is : %f \n", elapsed_time); cudaEventDestroy(start); cudaEventDestroy(stop); return 0; }
code for sm_80 Function : _Z6kerneliPfPViiii .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*/ IMAD.MOV.U32 R0, RZ, RZ, c[0x0][0x178] ; /* 0x00005e00ff007624 */ /* 0x000fca00078e00ff */ /*0020*/ ISETP.GE.AND P0, PT, R0, 0x1, PT ; /* 0x000000010000780c */ /* 0x000fda0003f06270 */ /*0030*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0040*/ ISETP.NE.AND P0, PT, RZ, c[0x0][0x17c], PT ; /* 0x00005f00ff007a0c */ /* 0x000fe20003f05270 */ /*0050*/ ULDC.64 UR8, c[0x0][0x118] ; /* 0x0000460000087ab9 */ /* 0x000fe20000000a00 */ /*0060*/ IADD3 R2, R0.reuse, -0x1, RZ ; /* 0xffffffff00027810 */ /* 0x040fe40007ffe0ff */ /*0070*/ LOP3.LUT R0, R0, 0x3, RZ, 0xc0, !PT ; /* 0x0000000300007812 */ /* 0x000fd200078ec0ff */ /*0080*/ @!P0 BRA 0x570 ; /* 0x000004e000008947 */ /* 0x000fea0003800000 */ /*0090*/ ISETP.GE.U32.AND P0, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe20003f06070 */ /*00a0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fd80008000000 */ /*00b0*/ @!P0 BRA 0x420 ; /* 0x0000036000008947 */ /* 0x000fea0003800000 */ /*00c0*/ IADD3 R6, -R0, c[0x0][0x178], RZ ; /* 0x00005e0000067a10 */ /* 0x000fe20007ffe1ff */ /*00d0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x000fe200078e00ff */ /*00e0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*00f0*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0100*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0110*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0120*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x009e220008201400 */ /*0130*/ UIADD3 UR5, UR4, 0x1, URZ ; /* 0x0000000104057890 */ /* 0x000fe2000fffe03f */ /*0140*/ ISETP.NE.AND P1, PT, RZ, c[0x0][0x180], PT ; /* 0x00006000ff007a0c */ /* 0x000fe20003f25270 */ /*0150*/ UIADD3 UR6, UR4, 0x2, URZ ; /* 0x0000000204067890 */ /* 0x000fe2000fffe03f */ /*0160*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fcc0003f0f070 */ /*0170*/ I2F R11, UR5 ; /* 0x00000005000b7d06 */ /* 0x004fe20008201400 */ /*0180*/ UIADD3 UR5, UR4, 0x3, URZ ; /* 0x0000000304057890 */ /* 0x000fe2000fffe03f */ /*0190*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0011ec000c101908 */ /*01a0*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x000e70000020f100 */ /*01b0*/ I2F R15, UR6 ; /* 0x00000006000f7d06 */ /* 0x000ea20008201400 */ /*01c0*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0021ee000c115908 */ /*01d0*/ I2F R17, UR5 ; /* 0x0000000500117d06 */ /* 0x000e700008201400 */ /*01e0*/ F2I.TRUNC.NTZ R13, R11 ; /* 0x0000000b000d7305 */ /* 0x0000e2000020f100 */ /*01f0*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0200*/ ERRBAR ; /* 0x00000000000079ab */ /* 0x004fc00000000000 */ /*0210*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*0220*/ @!P1 BRA 0x250 ; /* 0x0000002000009947 */ /* 0x000fea0003800000 */ /*0230*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0240*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*0250*/ STG.E [R2.64], R11 ; /* 0x0000000b02007986 */ /* 0x0005e2000c101908 */ /*0260*/ F2I.TRUNC.NTZ R7, R15 ; /* 0x0000000f00077305 */ /* 0x001426000020f100 */ /*0270*/ STG.E.STRONG.SYS [R4.64], R13 ; /* 0x0000000d04007986 */ /* 0x0085e2000c115908 */ /*0280*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0290*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*02a0*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*02b0*/ @P0 BRA 0x2d0 ; /* 0x0000001000000947 */ /* 0x000fea0003800000 */ /*02c0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*02d0*/ STG.E [R2.64], R15 ; /* 0x0000000f02007986 */ /* 0x0007e2000c101908 */ /*02e0*/ F2I.TRUNC.NTZ R9, R17 ; /* 0x0000001100097305 */ /* 0x002662000020f100 */ /*02f0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0f070 */ /*0300*/ STG.E.STRONG.SYS [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x0017e2000c115908 */ /*0310*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0320*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*0330*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*0340*/ @!P1 BRA 0x370 ; /* 0x0000002000009947 */ /* 0x000fea0003800000 */ /*0350*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0360*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*0370*/ STG.E [R2.64], R17 ; /* 0x0000001102007986 */ /* 0x0001e2000c101908 */ /*0380*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc60007ffe0ff */ /*0390*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0021e2000c115908 */ /*03a0*/ ISETP.NE.AND P1, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f25270 */ /*03b0*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*03c0*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*03d0*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*03e0*/ @P0 BRA 0x400 ; /* 0x0000001000000947 */ /* 0x000fea0003800000 */ /*03f0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0400*/ UIADD3 UR4, UR4, 0x4, URZ ; /* 0x0000000404047890 */ /* 0x000fe2000fffe03f */ /*0410*/ @P1 BRA 0x120 ; /* 0xfffffd0000001947 */ /* 0x000ff0000383ffff */ /*0420*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fda0003f05270 */ /*0430*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0440*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x00dfe400078e00ff */ /*0450*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0460*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0470*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0480*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x002e220008201400 */ /*0490*/ IADD3 R0, R0, -0x1, RZ ; /* 0xffffffff00007810 */ /* 0x000fe40007ffe0ff */ /*04a0*/ ISETP.NE.AND P1, PT, RZ, c[0x0][0x180], PT ; /* 0x00006000ff007a0c */ /* 0x000fe40003f25270 */ /*04b0*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fc60003f05270 */ /*04c0*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x001e22000020f100 */ /*04d0*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0003e8000c101908 */ /*04e0*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0013e2000c115908 */ /*04f0*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0500*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*0510*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*0520*/ @!P1 BRA 0x540 ; /* 0x0000001000009947 */ /* 0x000fea0003800000 */ /*0530*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0540*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0550*/ UIADD3 UR4, UR4, 0x1, URZ ; /* 0x0000000104047890 */ /* 0x000fe2000fffe03f */ /*0560*/ BRA 0x480 ; /* 0xffffff1000007947 */ /* 0x000ff0000383ffff */ /*0570*/ ISETP.NE.AND P0, PT, RZ, c[0x0][0x180], PT ; /* 0x00006000ff007a0c */ /* 0x000fda0003f05270 */ /*0580*/ @!P0 BRA 0x8d0 ; /* 0x0000034000008947 */ /* 0x000fea0003800000 */ /*0590*/ ISETP.GE.U32.AND P0, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe20003f06070 */ /*05a0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fd80008000000 */ /*05b0*/ @!P0 BRA 0x7d0 ; /* 0x0000021000008947 */ /* 0x000fea0003800000 */ /*05c0*/ IADD3 R6, -R0, c[0x0][0x178], RZ ; /* 0x00005e0000067a10 */ /* 0x000fe20007ffe1ff */ /*05d0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x000fe200078e00ff */ /*05e0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*05f0*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0600*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0610*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0620*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x004e220008201400 */ /*0630*/ UIADD3 UR5, UR4, 0x1, URZ ; /* 0x0000000104057890 */ /* 0x000fe4000fffe03f */ /*0640*/ UIADD3 UR6, UR4, 0x2, URZ ; /* 0x0000000204067890 */ /* 0x000fce000fffe03f */ /*0650*/ I2F R11, UR5 ; /* 0x00000005000b7d06 */ /* 0x002fe20008201400 */ /*0660*/ UIADD3 UR5, UR4, 0x3, URZ ; /* 0x0000000304057890 */ /* 0x000fe2000fffe03f */ /*0670*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0011ec000c101908 */ /*0680*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x000e70000020f100 */ /*0690*/ I2F R15, UR6 ; /* 0x00000006000f7d06 */ /* 0x008ea20008201400 */ /*06a0*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0021ee000c115908 */ /*06b0*/ I2F R19, UR5 ; /* 0x0000000500137d06 */ /* 0x000e700008201400 */ /*06c0*/ F2I.TRUNC.NTZ R13, R11 ; /* 0x0000000b000d7305 */ /* 0x0000f0000020f100 */ /*06d0*/ F2I.TRUNC.NTZ R17, R15 ; /* 0x0000000f00117305 */ /* 0x0040a2000020f100 */ /*06e0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*06f0*/ STG.E [R2.64], R11 ; /* 0x0000000b02007986 */ /* 0x0023e2000c101908 */ /*0700*/ F2I.TRUNC.NTZ R7, R19 ; /* 0x0000001300077305 */ /* 0x001226000020f100 */ /*0710*/ STG.E.STRONG.SYS [R4.64], R13 ; /* 0x0000000d04007986 */ /* 0x0083e2000c115908 */ /*0720*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0730*/ STG.E [R2.64], R15 ; /* 0x0000000f02007986 */ /* 0x0007e8000c101908 */ /*0740*/ STG.E.STRONG.SYS [R4.64], R17 ; /* 0x0000001104007986 */ /* 0x0047e2000c115908 */ /*0750*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0760*/ STG.E [R2.64], R19 ; /* 0x0000001302007986 */ /* 0x0005e2000c101908 */ /*0770*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc60007ffe0ff */ /*0780*/ STG.E.STRONG.SYS [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x0015e2000c115908 */ /*0790*/ ISETP.NE.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f05270 */ /*07a0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000ff60000000000 */ /*07b0*/ UIADD3 UR4, UR4, 0x4, URZ ; /* 0x0000000404047890 */ /* 0x000fe2000fffe03f */ /*07c0*/ @P0 BRA 0x620 ; /* 0xfffffe5000000947 */ /* 0x000ff0000383ffff */ /*07d0*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fda0003f05270 */ /*07e0*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*07f0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x00efe400078e00ff */ /*0800*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0810*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0820*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0830*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x002e220008201400 */ /*0840*/ IADD3 R0, R0, -0x1, RZ ; /* 0xffffffff00007810 */ /* 0x000fc80007ffe0ff */ /*0850*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fc60003f05270 */ /*0860*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x001e22000020f100 */ /*0870*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0003e8000c101908 */ /*0880*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0013e2000c115908 */ /*0890*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*08a0*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*08b0*/ UIADD3 UR4, UR4, 0x1, URZ ; /* 0x0000000104047890 */ /* 0x000fe2000fffe03f */ /*08c0*/ BRA 0x830 ; /* 0xffffff6000007947 */ /* 0x000ff0000383ffff */ /*08d0*/ ISETP.GE.U32.AND P1, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe20003f26070 */ /*08e0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*08f0*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fd60003f05270 */ /*0900*/ @!P1 BRA 0xab0 ; /* 0x000001a000009947 */ /* 0x000fea0003800000 */ /*0910*/ IADD3 R6, -R0, c[0x0][0x178], RZ ; /* 0x00005e0000067a10 */ /* 0x000fe20007ffe1ff */ /*0920*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff027624 */ /* 0x000fe200078e00ff */ /*0930*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*0940*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff037624 */ /* 0x000fe400078e00ff */ /*0950*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff047624 */ /* 0x000fe400078e00ff */ /*0960*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff057624 */ /* 0x000fe400078e00ff */ /*0970*/ UIADD3 UR5, UR4, 0x1, URZ ; /* 0x0000000104057890 */ /* 0x000fe2000fffe03f */ /*0980*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x001e220008201400 */ /*0990*/ UIADD3 UR6, UR4, 0x2, URZ ; /* 0x0000000204067890 */ /* 0x000fe2000fffe03f */ /*09a0*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc80007ffe0ff */ /*09b0*/ ISETP.NE.AND P1, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe40003f25270 */ /*09c0*/ I2F R8, UR5 ; /* 0x0000000500087d06 */ /* 0x000e620008201400 */ /*09d0*/ UIADD3 UR5, UR4, 0x3, URZ ; /* 0x0000000304057890 */ /* 0x000fe4000fffe03f */ /*09e0*/ UIADD3 UR4, UR4, 0x4, URZ ; /* 0x0000000404047890 */ /* 0x000fca000fffe03f */ /*09f0*/ I2F R10, UR6 ; /* 0x00000006000a7d06 */ /* 0x000eb00008201400 */ /*0a00*/ I2F R13, UR5 ; /* 0x00000005000d7d06 */ /* 0x000ff00008201400 */ /*0a10*/ F2I.TRUNC.NTZ R7, R7 ; /* 0x0000000700077305 */ /* 0x001e30000020f100 */ /*0a20*/ F2I.TRUNC.NTZ R9, R8 ; /* 0x0000000800097305 */ /* 0x002e70000020f100 */ /*0a30*/ F2I.TRUNC.NTZ R11, R10 ; /* 0x0000000a000b7305 */ /* 0x004ea2000020f100 */ /*0a40*/ STG.E.STRONG.SYS [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0011ee000c115908 */ /*0a50*/ F2I.TRUNC.NTZ R15, R13 ; /* 0x0000000d000f7305 */ /* 0x000ee2000020f100 */ /*0a60*/ STG.E.STRONG.SYS [R2.64], R9 ; /* 0x0000000902007986 */ /* 0x0021e8000c115908 */ /*0a70*/ STG.E.STRONG.SYS [R2.64], R11 ; /* 0x0000000b02007986 */ /* 0x0041e8000c115908 */ /*0a80*/ STG.E [R4.64], R13 ; /* 0x0000000d04007986 */ /* 0x0001e8000c101908 */ /*0a90*/ STG.E.STRONG.SYS [R2.64], R15 ; /* 0x0000000f02007986 */ /* 0x0081e2000c115908 */ /*0aa0*/ @P1 BRA 0x970 ; /* 0xfffffec000001947 */ /* 0x000fea000383ffff */ /*0ab0*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0ac0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x001fe400078e00ff */ /*0ad0*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fc400078e00ff */ /*0ae0*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0af0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0b00*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x002e220008201400 */ /*0b10*/ IADD3 R0, R0, -0x1, RZ ; /* 0xffffffff00007810 */ /* 0x000fe20007ffe0ff */ /*0b20*/ UIADD3 UR4, UR4, 0x1, URZ ; /* 0x0000000104047890 */ /* 0x000fc6000fffe03f */ /*0b30*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fc60003f05270 */ /*0b40*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x001e22000020f100 */ /*0b50*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0003e8000c101908 */ /*0b60*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0013ea000c115908 */ /*0b70*/ @P0 BRA 0xb00 ; /* 0xffffff8000000947 */ /* 0x000fea000383ffff */ /*0b80*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0b90*/ BRA 0xb90; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0ba0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0be0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c00*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c10*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c20*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c30*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c40*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c50*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <stdio.h> #include <cuda_runtime.h> #include <asm/unistd.h> #include <fcntl.h> #include <inttypes.h> #include <linux/kernel-page-flags.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string> #include <string.h> #include <sys/ioctl.h> #include <sys/mount.h> #include <sys/mman.h> #include <sys/stat.h> #include <sys/sysinfo.h> #include <sys/wait.h> #include <time.h> #include <unistd.h> #include <vector> #include <sys/time.h> #include <assert.h> #define gpuErrchk(ans) { gpuAssert((ans), __FILE__, __LINE__); } inline void gpuAssert(cudaError_t code, const char *file, int line, bool abort = true) { if (code != cudaSuccess) { fprintf(stderr, "GPUassert: %s %s %d\n", cudaGetErrorString(code), file, line); if (abort) exit(code); } } __global__ void kernel(int number_of_threads, float * dsum ,volatile int * d_mapping, int cnt, int fence_system_flag, int fence_block_flag) { int i; /*printf("D: i am [%d] \n", blockIdx.x * blockDim.x * blockDim.y * blockDim.z + threadIdx.z * blockDim.y * blockDim.x + threadIdx.y * blockDim.x + threadIdx.x); */ for(i=0; i < cnt ; i++){ *dsum = i; d_mapping[0] = *dsum; if(fence_system_flag){ __threadfence_system(); } if(fence_block_flag){ __threadfence_block(); } } } int main(int argc, char **argv) { int opt, BLOCKS = 1, THREADS = 1, cnt =10000, fence_system_flag =0, fence_block_flag = 0; cudaEvent_t start, stop; float elapsed_time =0; gpuErrchk(cudaEventCreate(&start)); gpuErrchk(cudaEventCreate(&stop)); while ((opt = getopt(argc, argv, "b:t:n:f:s:")) != -1) { switch (opt) { case 'b': BLOCKS = atoi(optarg); break; case 't': THREADS = atoi(optarg); break; case 'n': cnt = atoi(optarg); break; case 'f': fence_system_flag = atoi(optarg); break; case 's': fence_block_flag = atoi(optarg); break; default: fprintf(stderr, "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n", argv[0]); exit(EXIT_FAILURE); } } float * dsum; gpuErrchk(cudaMallocManaged((void **) &dsum, sizeof(uint64_t))); volatile int * h_mapping; gpuErrchk(cudaHostAlloc( (void**)&h_mapping, sizeof(volatile int), cudaHostAllocMapped)); volatile int * d_mapping; gpuErrchk(cudaHostGetDevicePointer((void**)&d_mapping,(void*)h_mapping,0)); *dsum = 0; cudaEventRecord(start, 0); kernel <<< BLOCKS, THREADS >>> (BLOCKS * THREADS,dsum,d_mapping, cnt, fence_system_flag, fence_block_flag); gpuErrchk(cudaDeviceSynchronize()); gpuErrchk(cudaEventRecord(stop, 0)); gpuErrchk(cudaEventSynchronize(stop)); gpuErrchk(cudaEventElapsedTime (&elapsed_time, start, stop)); assert(*dsum != 0); printf("H: elapsed_time is : %f \n", elapsed_time); cudaEventDestroy(start); cudaEventDestroy(stop); return 0; }
.file "tmpxft_0002120d_00000000-6_fence.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB3440: .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 .LFE3440: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata._Z9gpuAssert9cudaErrorPKcib.str1.1,"aMS",@progbits,1 .LC0: .string "GPUassert: %s %s %d\n" .section .text._Z9gpuAssert9cudaErrorPKcib,"axG",@progbits,_Z9gpuAssert9cudaErrorPKcib,comdat .weak _Z9gpuAssert9cudaErrorPKcib .type _Z9gpuAssert9cudaErrorPKcib, @function _Z9gpuAssert9cudaErrorPKcib: .LFB3436: .cfi_startproc endbr64 testl %edi, %edi jne .L9 ret .L9: 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 $8, %rsp .cfi_def_cfa_offset 48 movl %edi, %ebx movq %rsi, %r13 movl %edx, %r12d movl %ecx, %ebp call cudaGetErrorString@PLT movq %rax, %rcx movl %r12d, %r9d movq %r13, %r8 leaq .LC0(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT testb %bpl, %bpl jne .L10 addq $8, %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 .L10: .cfi_restore_state movl %ebx, %edi call exit@PLT .cfi_endproc .LFE3436: .size _Z9gpuAssert9cudaErrorPKcib, .-_Z9gpuAssert9cudaErrorPKcib .text .globl _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii .type _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii, @function _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii: .LFB3462: .cfi_startproc endbr64 subq $168, %rsp .cfi_def_cfa_offset 176 movl %edi, 28(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movl %ecx, 24(%rsp) movl %r8d, 4(%rsp) movl %r9d, (%rsp) movq %fs:40, %rax movq %rax, 152(%rsp) xorl %eax, %eax leaq 28(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) leaq 24(%rsp), %rax movq %rax, 120(%rsp) leaq 4(%rsp), %rax movq %rax, 128(%rsp) movq %rsp, %rax movq %rax, 136(%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 .L15 .L11: movq 152(%rsp), %rax subq %fs:40, %rax jne .L16 addq $168, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L15: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 184 pushq 40(%rsp) .cfi_def_cfa_offset 192 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z6kerneliPfPViiii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 176 jmp .L11 .L16: call __stack_chk_fail@PLT .cfi_endproc .LFE3462: .size _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii, .-_Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii .globl _Z6kerneliPfPViiii .type _Z6kerneliPfPViiii, @function _Z6kerneliPfPViiii: .LFB3463: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3463: .size _Z6kerneliPfPViiii, .-_Z6kerneliPfPViiii .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC2: .string "/home/ubuntu/Datasets/stackv2/train-structured/gganagno/GHammer/master/2_indexing_and_synchronisation/fencing/fence_test/fence.cu" .align 8 .LC3: .string "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n" .section .rodata.str1.1,"aMS",@progbits,1 .LC4: .string "b:t:n:f:s:" .LC5: .string "H: elapsed_time is : %f \n" .text .globl main .type main, @function main: .LFB3437: .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 $104, %rsp .cfi_def_cfa_offset 160 movl %edi, %r12d movq %rsi, %rbp movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax movl $0x00000000, 20(%rsp) leaq 24(%rsp), %rdi call cudaEventCreate@PLT movl %eax, %edi movl $1, %ecx movl $69, %edx leaq .LC2(%rip), %rbx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 32(%rsp), %rdi call cudaEventCreate@PLT movl %eax, %edi movl $1, %ecx movl $70, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movl $0, 12(%rsp) movl $0, 8(%rsp) movl $10000, 4(%rsp) movl $1, %r15d movl $1, %r14d leaq .LC4(%rip), %r13 leaq .L23(%rip), %rbx jmp .L20 .L27: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, %r14d .L20: movq %r13, %rdx movq %rbp, %rsi movl %r12d, %edi call getopt@PLT cmpl $-1, %eax je .L33 subl $98, %eax cmpl $18, %eax ja .L21 movl %eax, %eax movslq (%rbx,%rax,4), %rax addq %rbx, %rax notrack jmp *%rax .section .rodata .align 4 .align 4 .L23: .long .L27-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L26-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L25-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L24-.L23 .long .L22-.L23 .text .L22: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, %r15d jmp .L20 .L25: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, 4(%rsp) jmp .L20 .L26: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, 8(%rsp) jmp .L20 .L24: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, 12(%rsp) jmp .L20 .L21: movq 0(%rbp), %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 .L33: leaq 40(%rsp), %rdi movl $1, %edx movl $8, %esi call cudaMallocManaged@PLT movl %eax, %edi movl $1, %ecx movl $102, %edx leaq .LC2(%rip), %rbx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 48(%rsp), %rdi movl $2, %edx movl $4, %esi call cudaHostAlloc@PLT movl %eax, %edi movl $1, %ecx movl $107, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 56(%rsp), %rdi movl $0, %edx movq 48(%rsp), %rsi call cudaHostGetDevicePointer@PLT movl %eax, %edi movl $1, %ecx movl $111, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movq 40(%rsp), %rax movl $0x00000000, (%rax) movl $0, %esi movq 24(%rsp), %rdi call cudaEventRecord@PLT movl %r15d, 76(%rsp) movl $1, 80(%rsp) movl %r14d, 64(%rsp) movl $1, 68(%rsp) movl $0, %r9d movl $0, %r8d movq 76(%rsp), %rdx movl $1, %ecx movq 64(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L34 .L30: call cudaDeviceSynchronize@PLT movl %eax, %edi movl $1, %ecx movl $121, %edx leaq .LC2(%rip), %rbx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movl $0, %esi movq 32(%rsp), %rdi call cudaEventRecord@PLT movl %eax, %edi movl $1, %ecx movl $123, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movq 32(%rsp), %rdi call cudaEventSynchronize@PLT movl %eax, %edi movl $1, %ecx movl $125, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 20(%rsp), %rdi movq 32(%rsp), %rdx movq 24(%rsp), %rsi call cudaEventElapsedTime@PLT movl %eax, %edi movl $1, %ecx movl $127, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib pxor %xmm0, %xmm0 cvtss2sd 20(%rsp), %xmm0 leaq .LC5(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movq 24(%rsp), %rdi call cudaEventDestroy@PLT movq 32(%rsp), %rdi call cudaEventDestroy@PLT movq 88(%rsp), %rax subq %fs:40, %rax jne .L35 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 .L34: .cfi_restore_state movl %r14d, %edi imull %r15d, %edi movl 12(%rsp), %r9d movl 8(%rsp), %r8d movl 4(%rsp), %ecx movq 56(%rsp), %rdx movq 40(%rsp), %rsi call _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii jmp .L30 .L35: call __stack_chk_fail@PLT .cfi_endproc .LFE3437: .size main, .-main .section .rodata.str1.1 .LC6: .string "_Z6kerneliPfPViiii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB3465: .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 .LC6(%rip), %rdx movq %rdx, %rcx leaq _Z6kerneliPfPViiii(%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 .LFE3465: .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_runtime.h> #include <asm/unistd.h> #include <fcntl.h> #include <inttypes.h> #include <linux/kernel-page-flags.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string> #include <string.h> #include <sys/ioctl.h> #include <sys/mount.h> #include <sys/mman.h> #include <sys/stat.h> #include <sys/sysinfo.h> #include <sys/wait.h> #include <time.h> #include <unistd.h> #include <vector> #include <sys/time.h> #include <assert.h> #define gpuErrchk(ans) { gpuAssert((ans), __FILE__, __LINE__); } inline void gpuAssert(cudaError_t code, const char *file, int line, bool abort = true) { if (code != cudaSuccess) { fprintf(stderr, "GPUassert: %s %s %d\n", cudaGetErrorString(code), file, line); if (abort) exit(code); } } __global__ void kernel(int number_of_threads, float * dsum ,volatile int * d_mapping, int cnt, int fence_system_flag, int fence_block_flag) { int i; /*printf("D: i am [%d] \n", blockIdx.x * blockDim.x * blockDim.y * blockDim.z + threadIdx.z * blockDim.y * blockDim.x + threadIdx.y * blockDim.x + threadIdx.x); */ for(i=0; i < cnt ; i++){ *dsum = i; d_mapping[0] = *dsum; if(fence_system_flag){ __threadfence_system(); } if(fence_block_flag){ __threadfence_block(); } } } int main(int argc, char **argv) { int opt, BLOCKS = 1, THREADS = 1, cnt =10000, fence_system_flag =0, fence_block_flag = 0; cudaEvent_t start, stop; float elapsed_time =0; gpuErrchk(cudaEventCreate(&start)); gpuErrchk(cudaEventCreate(&stop)); while ((opt = getopt(argc, argv, "b:t:n:f:s:")) != -1) { switch (opt) { case 'b': BLOCKS = atoi(optarg); break; case 't': THREADS = atoi(optarg); break; case 'n': cnt = atoi(optarg); break; case 'f': fence_system_flag = atoi(optarg); break; case 's': fence_block_flag = atoi(optarg); break; default: fprintf(stderr, "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n", argv[0]); exit(EXIT_FAILURE); } } float * dsum; gpuErrchk(cudaMallocManaged((void **) &dsum, sizeof(uint64_t))); volatile int * h_mapping; gpuErrchk(cudaHostAlloc( (void**)&h_mapping, sizeof(volatile int), cudaHostAllocMapped)); volatile int * d_mapping; gpuErrchk(cudaHostGetDevicePointer((void**)&d_mapping,(void*)h_mapping,0)); *dsum = 0; cudaEventRecord(start, 0); kernel <<< BLOCKS, THREADS >>> (BLOCKS * THREADS,dsum,d_mapping, cnt, fence_system_flag, fence_block_flag); gpuErrchk(cudaDeviceSynchronize()); gpuErrchk(cudaEventRecord(stop, 0)); gpuErrchk(cudaEventSynchronize(stop)); gpuErrchk(cudaEventElapsedTime (&elapsed_time, start, stop)); assert(*dsum != 0); printf("H: elapsed_time is : %f \n", elapsed_time); cudaEventDestroy(start); cudaEventDestroy(stop); return 0; }
#include <stdio.h> #include <hip/hip_runtime.h> #include <asm/unistd.h> #include <fcntl.h> #include <inttypes.h> #include <linux/kernel-page-flags.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string> #include <string.h> #include <sys/ioctl.h> #include <sys/mount.h> #include <sys/mman.h> #include <sys/stat.h> #include <sys/sysinfo.h> #include <sys/wait.h> #include <time.h> #include <unistd.h> #include <vector> #include <sys/time.h> #include <assert.h> #define gpuErrchk(ans) { gpuAssert((ans), __FILE__, __LINE__); } inline void gpuAssert(hipError_t code, const char *file, int line, bool abort = true) { if (code != hipSuccess) { fprintf(stderr, "GPUassert: %s %s %d\n", hipGetErrorString(code), file, line); if (abort) exit(code); } } __global__ void kernel(int number_of_threads, float * dsum ,volatile int * d_mapping, int cnt, int fence_system_flag, int fence_block_flag) { int i; /*printf("D: i am [%d] \n", blockIdx.x * blockDim.x * blockDim.y * blockDim.z + threadIdx.z * blockDim.y * blockDim.x + threadIdx.y * blockDim.x + threadIdx.x); */ for(i=0; i < cnt ; i++){ *dsum = i; d_mapping[0] = *dsum; if(fence_system_flag){ __threadfence_system(); } if(fence_block_flag){ __threadfence_block(); } } } int main(int argc, char **argv) { int opt, BLOCKS = 1, THREADS = 1, cnt =10000, fence_system_flag =0, fence_block_flag = 0; hipEvent_t start, stop; float elapsed_time =0; gpuErrchk(hipEventCreate(&start)); gpuErrchk(hipEventCreate(&stop)); while ((opt = getopt(argc, argv, "b:t:n:f:s:")) != -1) { switch (opt) { case 'b': BLOCKS = atoi(optarg); break; case 't': THREADS = atoi(optarg); break; case 'n': cnt = atoi(optarg); break; case 'f': fence_system_flag = atoi(optarg); break; case 's': fence_block_flag = atoi(optarg); break; default: fprintf(stderr, "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n", argv[0]); exit(EXIT_FAILURE); } } float * dsum; gpuErrchk(hipMallocManaged((void **) &dsum, sizeof(uint64_t))); volatile int * h_mapping; gpuErrchk(hipHostAlloc( (void**)&h_mapping, sizeof(volatile int), hipHostMallocMapped)); volatile int * d_mapping; gpuErrchk(hipHostGetDevicePointer((void**)&d_mapping,(void*)h_mapping,0)); *dsum = 0; hipEventRecord(start, 0); kernel <<< BLOCKS, THREADS >>> (BLOCKS * THREADS,dsum,d_mapping, cnt, fence_system_flag, fence_block_flag); gpuErrchk(hipDeviceSynchronize()); gpuErrchk(hipEventRecord(stop, 0)); gpuErrchk(hipEventSynchronize(stop)); gpuErrchk(hipEventElapsedTime (&elapsed_time, start, stop)); assert(*dsum != 0); printf("H: elapsed_time is : %f \n", elapsed_time); hipEventDestroy(start); hipEventDestroy(stop); 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 <asm/unistd.h> #include <fcntl.h> #include <inttypes.h> #include <linux/kernel-page-flags.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string> #include <string.h> #include <sys/ioctl.h> #include <sys/mount.h> #include <sys/mman.h> #include <sys/stat.h> #include <sys/sysinfo.h> #include <sys/wait.h> #include <time.h> #include <unistd.h> #include <vector> #include <sys/time.h> #include <assert.h> #define gpuErrchk(ans) { gpuAssert((ans), __FILE__, __LINE__); } inline void gpuAssert(hipError_t code, const char *file, int line, bool abort = true) { if (code != hipSuccess) { fprintf(stderr, "GPUassert: %s %s %d\n", hipGetErrorString(code), file, line); if (abort) exit(code); } } __global__ void kernel(int number_of_threads, float * dsum ,volatile int * d_mapping, int cnt, int fence_system_flag, int fence_block_flag) { int i; /*printf("D: i am [%d] \n", blockIdx.x * blockDim.x * blockDim.y * blockDim.z + threadIdx.z * blockDim.y * blockDim.x + threadIdx.y * blockDim.x + threadIdx.x); */ for(i=0; i < cnt ; i++){ *dsum = i; d_mapping[0] = *dsum; if(fence_system_flag){ __threadfence_system(); } if(fence_block_flag){ __threadfence_block(); } } } int main(int argc, char **argv) { int opt, BLOCKS = 1, THREADS = 1, cnt =10000, fence_system_flag =0, fence_block_flag = 0; hipEvent_t start, stop; float elapsed_time =0; gpuErrchk(hipEventCreate(&start)); gpuErrchk(hipEventCreate(&stop)); while ((opt = getopt(argc, argv, "b:t:n:f:s:")) != -1) { switch (opt) { case 'b': BLOCKS = atoi(optarg); break; case 't': THREADS = atoi(optarg); break; case 'n': cnt = atoi(optarg); break; case 'f': fence_system_flag = atoi(optarg); break; case 's': fence_block_flag = atoi(optarg); break; default: fprintf(stderr, "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n", argv[0]); exit(EXIT_FAILURE); } } float * dsum; gpuErrchk(hipMallocManaged((void **) &dsum, sizeof(uint64_t))); volatile int * h_mapping; gpuErrchk(hipHostAlloc( (void**)&h_mapping, sizeof(volatile int), hipHostMallocMapped)); volatile int * d_mapping; gpuErrchk(hipHostGetDevicePointer((void**)&d_mapping,(void*)h_mapping,0)); *dsum = 0; hipEventRecord(start, 0); kernel <<< BLOCKS, THREADS >>> (BLOCKS * THREADS,dsum,d_mapping, cnt, fence_system_flag, fence_block_flag); gpuErrchk(hipDeviceSynchronize()); gpuErrchk(hipEventRecord(stop, 0)); gpuErrchk(hipEventSynchronize(stop)); gpuErrchk(hipEventElapsedTime (&elapsed_time, start, stop)); assert(*dsum != 0); printf("H: elapsed_time is : %f \n", elapsed_time); hipEventDestroy(start); hipEventDestroy(stop); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z6kerneliPfPViiii .globl _Z6kerneliPfPViiii .p2align 8 .type _Z6kerneliPfPViiii,@function _Z6kerneliPfPViiii: s_load_b32 s2, s[0:1], 0x18 s_waitcnt lgkmcnt(0) s_cmp_lt_i32 s2, 1 s_cbranch_scc1 .LBB0_7 s_clause 0x1 s_load_b64 s[8:9], s[0:1], 0x1c s_load_b128 s[4:7], s[0:1], 0x8 v_mov_b32_e32 v2, 0 s_mov_b32 s3, 0 s_waitcnt lgkmcnt(0) s_cmp_lg_u32 s8, 0 v_dual_mov_b32 v0, s6 :: v_dual_mov_b32 v1, s7 s_cselect_b32 s0, -1, 0 s_cmp_lg_u32 s9, 0 v_cndmask_b32_e64 v3, 0, 1, s0 s_cselect_b32 s0, -1, 0 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_cndmask_b32_e64 v4, 0, 1, s0 v_cmp_ne_u32_e64 s0, 1, v3 s_delay_alu instid0(VALU_DEP_2) v_cmp_ne_u32_e64 s1, 1, v4 s_branch .LBB0_3 .LBB0_2: s_add_i32 s3, s3, 1 s_delay_alu instid0(SALU_CYCLE_1) s_cmp_lg_u32 s2, s3 s_cbranch_scc0 .LBB0_7 .LBB0_3: v_cvt_f32_i32_e32 v3, s3 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) s_and_b32 vcc_lo, exec_lo, s0 v_cvt_i32_f32_e32 v4, v3 global_store_b32 v2, v3, s[4:5] flat_store_b32 v[0:1], v4 dlc s_waitcnt_vscnt null, 0x0 s_cbranch_vccnz .LBB0_5 s_waitcnt lgkmcnt(0) buffer_gl1_inv buffer_gl0_inv .LBB0_5: s_and_b32 vcc_lo, exec_lo, s1 s_cbranch_vccnz .LBB0_2 s_waitcnt lgkmcnt(0) buffer_gl0_inv s_branch .LBB0_2 .LBB0_7: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z6kerneliPfPViiii .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 36 .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 5 .amdhsa_next_free_sgpr 10 .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 _Z6kerneliPfPViiii, .Lfunc_end0-_Z6kerneliPfPViiii .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 - .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 .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 36 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z6kerneliPfPViiii .private_segment_fixed_size: 0 .sgpr_count: 12 .sgpr_spill_count: 0 .symbol: _Z6kerneliPfPViiii.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> #include <asm/unistd.h> #include <fcntl.h> #include <inttypes.h> #include <linux/kernel-page-flags.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string> #include <string.h> #include <sys/ioctl.h> #include <sys/mount.h> #include <sys/mman.h> #include <sys/stat.h> #include <sys/sysinfo.h> #include <sys/wait.h> #include <time.h> #include <unistd.h> #include <vector> #include <sys/time.h> #include <assert.h> #define gpuErrchk(ans) { gpuAssert((ans), __FILE__, __LINE__); } inline void gpuAssert(hipError_t code, const char *file, int line, bool abort = true) { if (code != hipSuccess) { fprintf(stderr, "GPUassert: %s %s %d\n", hipGetErrorString(code), file, line); if (abort) exit(code); } } __global__ void kernel(int number_of_threads, float * dsum ,volatile int * d_mapping, int cnt, int fence_system_flag, int fence_block_flag) { int i; /*printf("D: i am [%d] \n", blockIdx.x * blockDim.x * blockDim.y * blockDim.z + threadIdx.z * blockDim.y * blockDim.x + threadIdx.y * blockDim.x + threadIdx.x); */ for(i=0; i < cnt ; i++){ *dsum = i; d_mapping[0] = *dsum; if(fence_system_flag){ __threadfence_system(); } if(fence_block_flag){ __threadfence_block(); } } } int main(int argc, char **argv) { int opt, BLOCKS = 1, THREADS = 1, cnt =10000, fence_system_flag =0, fence_block_flag = 0; hipEvent_t start, stop; float elapsed_time =0; gpuErrchk(hipEventCreate(&start)); gpuErrchk(hipEventCreate(&stop)); while ((opt = getopt(argc, argv, "b:t:n:f:s:")) != -1) { switch (opt) { case 'b': BLOCKS = atoi(optarg); break; case 't': THREADS = atoi(optarg); break; case 'n': cnt = atoi(optarg); break; case 'f': fence_system_flag = atoi(optarg); break; case 's': fence_block_flag = atoi(optarg); break; default: fprintf(stderr, "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n", argv[0]); exit(EXIT_FAILURE); } } float * dsum; gpuErrchk(hipMallocManaged((void **) &dsum, sizeof(uint64_t))); volatile int * h_mapping; gpuErrchk(hipHostAlloc( (void**)&h_mapping, sizeof(volatile int), hipHostMallocMapped)); volatile int * d_mapping; gpuErrchk(hipHostGetDevicePointer((void**)&d_mapping,(void*)h_mapping,0)); *dsum = 0; hipEventRecord(start, 0); kernel <<< BLOCKS, THREADS >>> (BLOCKS * THREADS,dsum,d_mapping, cnt, fence_system_flag, fence_block_flag); gpuErrchk(hipDeviceSynchronize()); gpuErrchk(hipEventRecord(stop, 0)); gpuErrchk(hipEventSynchronize(stop)); gpuErrchk(hipEventElapsedTime (&elapsed_time, start, stop)); assert(*dsum != 0); printf("H: elapsed_time is : %f \n", elapsed_time); hipEventDestroy(start); hipEventDestroy(stop); return 0; }
.text .file "fence.hip" .globl _Z21__device_stub__kerneliPfPViiii # -- Begin function _Z21__device_stub__kerneliPfPViiii .p2align 4, 0x90 .type _Z21__device_stub__kerneliPfPViiii,@function _Z21__device_stub__kerneliPfPViiii: # @_Z21__device_stub__kerneliPfPViiii .cfi_startproc # %bb.0: subq $136, %rsp .cfi_def_cfa_offset 144 movl %edi, 12(%rsp) movq %rsi, 72(%rsp) movq %rdx, 64(%rsp) movl %ecx, 8(%rsp) movl %r8d, 4(%rsp) movl %r9d, (%rsp) leaq 12(%rsp), %rax movq %rax, 80(%rsp) leaq 72(%rsp), %rax movq %rax, 88(%rsp) leaq 64(%rsp), %rax movq %rax, 96(%rsp) leaq 8(%rsp), %rax movq %rax, 104(%rsp) leaq 4(%rsp), %rax movq %rax, 112(%rsp) movq %rsp, %rax movq %rax, 120(%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 $_Z6kerneliPfPViiii, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $152, %rsp .cfi_adjust_cfa_offset -152 retq .Lfunc_end0: .size _Z21__device_stub__kerneliPfPViiii, .Lfunc_end0-_Z21__device_stub__kerneliPfPViiii .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 $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 %rsi, %rbp movl %edi, %ebx movl $0, 12(%rsp) leaq 24(%rsp), %rdi callq hipEventCreate testl %eax, %eax jne .LBB1_1 # %bb.3: # %_Z9gpuAssert10hipError_tPKcib.exit leaq 16(%rsp), %rdi callq hipEventCreate testl %eax, %eax jne .LBB1_32 # %bb.4: # %_Z9gpuAssert10hipError_tPKcib.exit23.outer.preheader movl $1, %r12d movl $10000, %r14d # imm = 0x2710 xorl %eax, %eax movq %rax, 48(%rsp) # 8-byte Spill xorl %r15d, %r15d movl $1, %r13d jmp .LBB1_5 .p2align 4, 0x90 .LBB1_10: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r13 .LBB1_5: # %_Z9gpuAssert10hipError_tPKcib.exit23 # =>This Inner Loop Header: Depth=1 movl $.L.str.1, %edx movl %ebx, %edi movq %rbp, %rsi callq getopt # kill: def $eax killed $eax def $rax cmpl $98, %eax je .LBB1_10 # %bb.6: # %_Z9gpuAssert10hipError_tPKcib.exit23 # in Loop: Header=BB1_5 Depth=1 leal -102(%rax), %ecx cmpl $14, %ecx ja .LBB1_7 # %bb.9: # %_Z9gpuAssert10hipError_tPKcib.exit23 # in Loop: Header=BB1_5 Depth=1 jmpq *.LJTI1_0(,%rcx,8) .LBB1_13: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r15 jmp .LBB1_5 .LBB1_11: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r12 jmp .LBB1_5 .LBB1_12: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r14 jmp .LBB1_5 .LBB1_14: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, 48(%rsp) # 8-byte Spill jmp .LBB1_5 .LBB1_7: # %_Z9gpuAssert10hipError_tPKcib.exit23 cmpl $-1, %eax jne .LBB1_8 # %bb.15: leaq 56(%rsp), %rdi movl $8, %esi movl $1, %edx callq hipMallocManaged testl %eax, %eax jne .LBB1_16 # %bb.17: # %_Z9gpuAssert10hipError_tPKcib.exit25 leaq 72(%rsp), %rdi movl $4, %esi movl $2, %edx callq hipHostAlloc testl %eax, %eax jne .LBB1_18 # %bb.19: # %_Z9gpuAssert10hipError_tPKcib.exit27 movq 72(%rsp), %rsi leaq 64(%rsp), %rdi xorl %edx, %edx callq hipHostGetDevicePointer testl %eax, %eax jne .LBB1_20 # %bb.21: # %_Z9gpuAssert10hipError_tPKcib.exit29 movq 56(%rsp), %rax movl $0, (%rax) movq 24(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movl %r13d, %edi movabsq $4294967296, %rax # imm = 0x100000000 orq %rax, %rdi movl %r12d, %edx orq %rax, %rdx movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_23 # %bb.22: imull %r12d, %r13d movq 56(%rsp), %rax movq 64(%rsp), %rcx movl %r13d, 44(%rsp) movq %rax, 136(%rsp) movq %rcx, 128(%rsp) movl %r14d, 40(%rsp) movl %r15d, 36(%rsp) movq 48(%rsp), %rax # 8-byte Reload movl %eax, 32(%rsp) leaq 44(%rsp), %rax movq %rax, 144(%rsp) leaq 136(%rsp), %rax movq %rax, 152(%rsp) leaq 128(%rsp), %rax movq %rax, 160(%rsp) leaq 40(%rsp), %rax movq %rax, 168(%rsp) leaq 36(%rsp), %rax movq %rax, 176(%rsp) leaq 32(%rsp), %rax movq %rax, 184(%rsp) leaq 112(%rsp), %rdi leaq 96(%rsp), %rsi leaq 88(%rsp), %rdx leaq 80(%rsp), %rcx callq __hipPopCallConfiguration movq 112(%rsp), %rsi movl 120(%rsp), %edx movq 96(%rsp), %rcx movl 104(%rsp), %r8d leaq 144(%rsp), %r9 movl $_Z6kerneliPfPViiii, %edi pushq 80(%rsp) .cfi_adjust_cfa_offset 8 pushq 96(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB1_23: callq hipDeviceSynchronize testl %eax, %eax jne .LBB1_24 # %bb.25: # %_Z9gpuAssert10hipError_tPKcib.exit31 movq 16(%rsp), %rdi xorl %esi, %esi callq hipEventRecord testl %eax, %eax jne .LBB1_26 # %bb.27: # %_Z9gpuAssert10hipError_tPKcib.exit33 movq 16(%rsp), %rdi callq hipEventSynchronize testl %eax, %eax jne .LBB1_28 # %bb.29: # %_Z9gpuAssert10hipError_tPKcib.exit35 movq 24(%rsp), %rsi movq 16(%rsp), %rdx leaq 12(%rsp), %rdi callq hipEventElapsedTime testl %eax, %eax jne .LBB1_30 # %bb.31: # %_Z9gpuAssert10hipError_tPKcib.exit37 movss 12(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.3, %edi movb $1, %al callq printf movq 24(%rsp), %rdi callq hipEventDestroy movq 16(%rsp), %rdi callq hipEventDestroy xorl %eax, %eax 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 .LBB1_8: .cfi_def_cfa_offset 256 movq stderr(%rip), %rdi movq (%rbp), %rdx movl $.L.str.2, %esi xorl %eax, %eax callq fprintf movl $1, %edi callq exit .LBB1_1: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $69, %r8d jmp .LBB1_2 .LBB1_32: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $70, %r8d jmp .LBB1_2 .LBB1_16: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $102, %r8d jmp .LBB1_2 .LBB1_18: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $107, %r8d jmp .LBB1_2 .LBB1_20: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $111, %r8d jmp .LBB1_2 .LBB1_24: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $121, %r8d jmp .LBB1_2 .LBB1_26: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $123, %r8d jmp .LBB1_2 .LBB1_28: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $125, %r8d jmp .LBB1_2 .LBB1_30: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $128, %r8d .LBB1_2: xorl %eax, %eax callq fprintf movl %ebp, %edi callq exit .Lfunc_end1: .size main, .Lfunc_end1-main .cfi_endproc .section .rodata,"a",@progbits .p2align 3, 0x0 .LJTI1_0: .quad .LBB1_13 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_12 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_14 .quad .LBB1_11 # -- 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 $_Z6kerneliPfPViiii, %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 _Z6kerneliPfPViiii,@object # @_Z6kerneliPfPViiii .section .rodata,"a",@progbits .globl _Z6kerneliPfPViiii .p2align 3, 0x0 _Z6kerneliPfPViiii: .quad _Z21__device_stub__kerneliPfPViiii .size _Z6kerneliPfPViiii, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "/home/ubuntu/Datasets/stackv2/train-structured-repos-hip/gganagno/GHammer/master/2_indexing_and_synchronisation/fencing/fence_test/fence.hip" .size .L.str, 141 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "b:t:n:f:s:" .size .L.str.1, 11 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n" .size .L.str.2, 96 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "H: elapsed_time is : %f \n" .size .L.str.3, 26 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "GPUassert: %s %s %d\n" .size .L.str.4, 21 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z6kerneliPfPViiii" .size .L__unnamed_1, 19 .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__kerneliPfPViiii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z6kerneliPfPViiii .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 : _Z6kerneliPfPViiii .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*/ IMAD.MOV.U32 R0, RZ, RZ, c[0x0][0x178] ; /* 0x00005e00ff007624 */ /* 0x000fca00078e00ff */ /*0020*/ ISETP.GE.AND P0, PT, R0, 0x1, PT ; /* 0x000000010000780c */ /* 0x000fda0003f06270 */ /*0030*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0040*/ ISETP.NE.AND P0, PT, RZ, c[0x0][0x17c], PT ; /* 0x00005f00ff007a0c */ /* 0x000fe20003f05270 */ /*0050*/ ULDC.64 UR8, c[0x0][0x118] ; /* 0x0000460000087ab9 */ /* 0x000fe20000000a00 */ /*0060*/ IADD3 R2, R0.reuse, -0x1, RZ ; /* 0xffffffff00027810 */ /* 0x040fe40007ffe0ff */ /*0070*/ LOP3.LUT R0, R0, 0x3, RZ, 0xc0, !PT ; /* 0x0000000300007812 */ /* 0x000fd200078ec0ff */ /*0080*/ @!P0 BRA 0x570 ; /* 0x000004e000008947 */ /* 0x000fea0003800000 */ /*0090*/ ISETP.GE.U32.AND P0, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe20003f06070 */ /*00a0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fd80008000000 */ /*00b0*/ @!P0 BRA 0x420 ; /* 0x0000036000008947 */ /* 0x000fea0003800000 */ /*00c0*/ IADD3 R6, -R0, c[0x0][0x178], RZ ; /* 0x00005e0000067a10 */ /* 0x000fe20007ffe1ff */ /*00d0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x000fe200078e00ff */ /*00e0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*00f0*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0100*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0110*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0120*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x009e220008201400 */ /*0130*/ UIADD3 UR5, UR4, 0x1, URZ ; /* 0x0000000104057890 */ /* 0x000fe2000fffe03f */ /*0140*/ ISETP.NE.AND P1, PT, RZ, c[0x0][0x180], PT ; /* 0x00006000ff007a0c */ /* 0x000fe20003f25270 */ /*0150*/ UIADD3 UR6, UR4, 0x2, URZ ; /* 0x0000000204067890 */ /* 0x000fe2000fffe03f */ /*0160*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fcc0003f0f070 */ /*0170*/ I2F R11, UR5 ; /* 0x00000005000b7d06 */ /* 0x004fe20008201400 */ /*0180*/ UIADD3 UR5, UR4, 0x3, URZ ; /* 0x0000000304057890 */ /* 0x000fe2000fffe03f */ /*0190*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0011ec000c101908 */ /*01a0*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x000e70000020f100 */ /*01b0*/ I2F R15, UR6 ; /* 0x00000006000f7d06 */ /* 0x000ea20008201400 */ /*01c0*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0021ee000c115908 */ /*01d0*/ I2F R17, UR5 ; /* 0x0000000500117d06 */ /* 0x000e700008201400 */ /*01e0*/ F2I.TRUNC.NTZ R13, R11 ; /* 0x0000000b000d7305 */ /* 0x0000e2000020f100 */ /*01f0*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0200*/ ERRBAR ; /* 0x00000000000079ab */ /* 0x004fc00000000000 */ /*0210*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*0220*/ @!P1 BRA 0x250 ; /* 0x0000002000009947 */ /* 0x000fea0003800000 */ /*0230*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0240*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*0250*/ STG.E [R2.64], R11 ; /* 0x0000000b02007986 */ /* 0x0005e2000c101908 */ /*0260*/ F2I.TRUNC.NTZ R7, R15 ; /* 0x0000000f00077305 */ /* 0x001426000020f100 */ /*0270*/ STG.E.STRONG.SYS [R4.64], R13 ; /* 0x0000000d04007986 */ /* 0x0085e2000c115908 */ /*0280*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0290*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*02a0*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*02b0*/ @P0 BRA 0x2d0 ; /* 0x0000001000000947 */ /* 0x000fea0003800000 */ /*02c0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*02d0*/ STG.E [R2.64], R15 ; /* 0x0000000f02007986 */ /* 0x0007e2000c101908 */ /*02e0*/ F2I.TRUNC.NTZ R9, R17 ; /* 0x0000001100097305 */ /* 0x002662000020f100 */ /*02f0*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x80, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0f070 */ /*0300*/ STG.E.STRONG.SYS [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x0017e2000c115908 */ /*0310*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0320*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*0330*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*0340*/ @!P1 BRA 0x370 ; /* 0x0000002000009947 */ /* 0x000fea0003800000 */ /*0350*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0360*/ PLOP3.LUT P0, PT, PT, PT, PT, 0x8, 0x0 ; /* 0x000000000000781c */ /* 0x000fe40003f0e170 */ /*0370*/ STG.E [R2.64], R17 ; /* 0x0000001102007986 */ /* 0x0001e2000c101908 */ /*0380*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc60007ffe0ff */ /*0390*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0021e2000c115908 */ /*03a0*/ ISETP.NE.AND P1, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f25270 */ /*03b0*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*03c0*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*03d0*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*03e0*/ @P0 BRA 0x400 ; /* 0x0000001000000947 */ /* 0x000fea0003800000 */ /*03f0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0400*/ UIADD3 UR4, UR4, 0x4, URZ ; /* 0x0000000404047890 */ /* 0x000fe2000fffe03f */ /*0410*/ @P1 BRA 0x120 ; /* 0xfffffd0000001947 */ /* 0x000ff0000383ffff */ /*0420*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fda0003f05270 */ /*0430*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0440*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x00dfe400078e00ff */ /*0450*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0460*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0470*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0480*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x002e220008201400 */ /*0490*/ IADD3 R0, R0, -0x1, RZ ; /* 0xffffffff00007810 */ /* 0x000fe40007ffe0ff */ /*04a0*/ ISETP.NE.AND P1, PT, RZ, c[0x0][0x180], PT ; /* 0x00006000ff007a0c */ /* 0x000fe40003f25270 */ /*04b0*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fc60003f05270 */ /*04c0*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x001e22000020f100 */ /*04d0*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0003e8000c101908 */ /*04e0*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0013e2000c115908 */ /*04f0*/ MEMBAR.SC.SYS ; /* 0x0000000000007992 */ /* 0x000fec0000003000 */ /*0500*/ ERRBAR; /* 0x00000000000079ab */ /* 0x000fc00000000000 */ /*0510*/ CCTL.IVALL ; /* 0x00000000ff00798f */ /* 0x000fca0002000000 */ /*0520*/ @!P1 BRA 0x540 ; /* 0x0000001000009947 */ /* 0x000fea0003800000 */ /*0530*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0540*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0550*/ UIADD3 UR4, UR4, 0x1, URZ ; /* 0x0000000104047890 */ /* 0x000fe2000fffe03f */ /*0560*/ BRA 0x480 ; /* 0xffffff1000007947 */ /* 0x000ff0000383ffff */ /*0570*/ ISETP.NE.AND P0, PT, RZ, c[0x0][0x180], PT ; /* 0x00006000ff007a0c */ /* 0x000fda0003f05270 */ /*0580*/ @!P0 BRA 0x8d0 ; /* 0x0000034000008947 */ /* 0x000fea0003800000 */ /*0590*/ ISETP.GE.U32.AND P0, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe20003f06070 */ /*05a0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fd80008000000 */ /*05b0*/ @!P0 BRA 0x7d0 ; /* 0x0000021000008947 */ /* 0x000fea0003800000 */ /*05c0*/ IADD3 R6, -R0, c[0x0][0x178], RZ ; /* 0x00005e0000067a10 */ /* 0x000fe20007ffe1ff */ /*05d0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x000fe200078e00ff */ /*05e0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*05f0*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0600*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0610*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0620*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x004e220008201400 */ /*0630*/ UIADD3 UR5, UR4, 0x1, URZ ; /* 0x0000000104057890 */ /* 0x000fe4000fffe03f */ /*0640*/ UIADD3 UR6, UR4, 0x2, URZ ; /* 0x0000000204067890 */ /* 0x000fce000fffe03f */ /*0650*/ I2F R11, UR5 ; /* 0x00000005000b7d06 */ /* 0x002fe20008201400 */ /*0660*/ UIADD3 UR5, UR4, 0x3, URZ ; /* 0x0000000304057890 */ /* 0x000fe2000fffe03f */ /*0670*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0011ec000c101908 */ /*0680*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x000e70000020f100 */ /*0690*/ I2F R15, UR6 ; /* 0x00000006000f7d06 */ /* 0x008ea20008201400 */ /*06a0*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0021ee000c115908 */ /*06b0*/ I2F R19, UR5 ; /* 0x0000000500137d06 */ /* 0x000e700008201400 */ /*06c0*/ F2I.TRUNC.NTZ R13, R11 ; /* 0x0000000b000d7305 */ /* 0x0000f0000020f100 */ /*06d0*/ F2I.TRUNC.NTZ R17, R15 ; /* 0x0000000f00117305 */ /* 0x0040a2000020f100 */ /*06e0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*06f0*/ STG.E [R2.64], R11 ; /* 0x0000000b02007986 */ /* 0x0023e2000c101908 */ /*0700*/ F2I.TRUNC.NTZ R7, R19 ; /* 0x0000001300077305 */ /* 0x001226000020f100 */ /*0710*/ STG.E.STRONG.SYS [R4.64], R13 ; /* 0x0000000d04007986 */ /* 0x0083e2000c115908 */ /*0720*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0730*/ STG.E [R2.64], R15 ; /* 0x0000000f02007986 */ /* 0x0007e8000c101908 */ /*0740*/ STG.E.STRONG.SYS [R4.64], R17 ; /* 0x0000001104007986 */ /* 0x0047e2000c115908 */ /*0750*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*0760*/ STG.E [R2.64], R19 ; /* 0x0000001302007986 */ /* 0x0005e2000c101908 */ /*0770*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc60007ffe0ff */ /*0780*/ STG.E.STRONG.SYS [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x0015e2000c115908 */ /*0790*/ ISETP.NE.AND P0, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f05270 */ /*07a0*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000ff60000000000 */ /*07b0*/ UIADD3 UR4, UR4, 0x4, URZ ; /* 0x0000000404047890 */ /* 0x000fe2000fffe03f */ /*07c0*/ @P0 BRA 0x620 ; /* 0xfffffe5000000947 */ /* 0x000ff0000383ffff */ /*07d0*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fda0003f05270 */ /*07e0*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*07f0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x00efe400078e00ff */ /*0800*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fe400078e00ff */ /*0810*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0820*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0830*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x002e220008201400 */ /*0840*/ IADD3 R0, R0, -0x1, RZ ; /* 0xffffffff00007810 */ /* 0x000fc80007ffe0ff */ /*0850*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fc60003f05270 */ /*0860*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x001e22000020f100 */ /*0870*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0003e8000c101908 */ /*0880*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0013e2000c115908 */ /*0890*/ MEMBAR.SC.CTA ; /* 0x0000000000007992 */ /* 0x000fec0000000000 */ /*08a0*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*08b0*/ UIADD3 UR4, UR4, 0x1, URZ ; /* 0x0000000104047890 */ /* 0x000fe2000fffe03f */ /*08c0*/ BRA 0x830 ; /* 0xffffff6000007947 */ /* 0x000ff0000383ffff */ /*08d0*/ ISETP.GE.U32.AND P1, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe20003f26070 */ /*08e0*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*08f0*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fd60003f05270 */ /*0900*/ @!P1 BRA 0xab0 ; /* 0x000001a000009947 */ /* 0x000fea0003800000 */ /*0910*/ IADD3 R6, -R0, c[0x0][0x178], RZ ; /* 0x00005e0000067a10 */ /* 0x000fe20007ffe1ff */ /*0920*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff027624 */ /* 0x000fe200078e00ff */ /*0930*/ UMOV UR4, URZ ; /* 0x0000003f00047c82 */ /* 0x000fe20008000000 */ /*0940*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff037624 */ /* 0x000fe400078e00ff */ /*0950*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff047624 */ /* 0x000fe400078e00ff */ /*0960*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff057624 */ /* 0x000fe400078e00ff */ /*0970*/ UIADD3 UR5, UR4, 0x1, URZ ; /* 0x0000000104057890 */ /* 0x000fe2000fffe03f */ /*0980*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x001e220008201400 */ /*0990*/ UIADD3 UR6, UR4, 0x2, URZ ; /* 0x0000000204067890 */ /* 0x000fe2000fffe03f */ /*09a0*/ IADD3 R6, R6, -0x4, RZ ; /* 0xfffffffc06067810 */ /* 0x000fc80007ffe0ff */ /*09b0*/ ISETP.NE.AND P1, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe40003f25270 */ /*09c0*/ I2F R8, UR5 ; /* 0x0000000500087d06 */ /* 0x000e620008201400 */ /*09d0*/ UIADD3 UR5, UR4, 0x3, URZ ; /* 0x0000000304057890 */ /* 0x000fe4000fffe03f */ /*09e0*/ UIADD3 UR4, UR4, 0x4, URZ ; /* 0x0000000404047890 */ /* 0x000fca000fffe03f */ /*09f0*/ I2F R10, UR6 ; /* 0x00000006000a7d06 */ /* 0x000eb00008201400 */ /*0a00*/ I2F R13, UR5 ; /* 0x00000005000d7d06 */ /* 0x000ff00008201400 */ /*0a10*/ F2I.TRUNC.NTZ R7, R7 ; /* 0x0000000700077305 */ /* 0x001e30000020f100 */ /*0a20*/ F2I.TRUNC.NTZ R9, R8 ; /* 0x0000000800097305 */ /* 0x002e70000020f100 */ /*0a30*/ F2I.TRUNC.NTZ R11, R10 ; /* 0x0000000a000b7305 */ /* 0x004ea2000020f100 */ /*0a40*/ STG.E.STRONG.SYS [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0011ee000c115908 */ /*0a50*/ F2I.TRUNC.NTZ R15, R13 ; /* 0x0000000d000f7305 */ /* 0x000ee2000020f100 */ /*0a60*/ STG.E.STRONG.SYS [R2.64], R9 ; /* 0x0000000902007986 */ /* 0x0021e8000c115908 */ /*0a70*/ STG.E.STRONG.SYS [R2.64], R11 ; /* 0x0000000b02007986 */ /* 0x0041e8000c115908 */ /*0a80*/ STG.E [R4.64], R13 ; /* 0x0000000d04007986 */ /* 0x0001e8000c101908 */ /*0a90*/ STG.E.STRONG.SYS [R2.64], R15 ; /* 0x0000000f02007986 */ /* 0x0081e2000c115908 */ /*0aa0*/ @P1 BRA 0x970 ; /* 0xfffffec000001947 */ /* 0x000fea000383ffff */ /*0ab0*/ @!P0 EXIT ; /* 0x000000000000894d */ /* 0x000fea0003800000 */ /*0ac0*/ IMAD.MOV.U32 R2, RZ, RZ, c[0x0][0x168] ; /* 0x00005a00ff027624 */ /* 0x001fe400078e00ff */ /*0ad0*/ IMAD.MOV.U32 R3, RZ, RZ, c[0x0][0x16c] ; /* 0x00005b00ff037624 */ /* 0x000fc400078e00ff */ /*0ae0*/ IMAD.MOV.U32 R4, RZ, RZ, c[0x0][0x170] ; /* 0x00005c00ff047624 */ /* 0x000fe400078e00ff */ /*0af0*/ IMAD.MOV.U32 R5, RZ, RZ, c[0x0][0x174] ; /* 0x00005d00ff057624 */ /* 0x000fe400078e00ff */ /*0b00*/ I2F R7, UR4 ; /* 0x0000000400077d06 */ /* 0x002e220008201400 */ /*0b10*/ IADD3 R0, R0, -0x1, RZ ; /* 0xffffffff00007810 */ /* 0x000fe20007ffe0ff */ /*0b20*/ UIADD3 UR4, UR4, 0x1, URZ ; /* 0x0000000104047890 */ /* 0x000fc6000fffe03f */ /*0b30*/ ISETP.NE.AND P0, PT, R0, RZ, PT ; /* 0x000000ff0000720c */ /* 0x000fc60003f05270 */ /*0b40*/ F2I.TRUNC.NTZ R9, R7 ; /* 0x0000000700097305 */ /* 0x001e22000020f100 */ /*0b50*/ STG.E [R2.64], R7 ; /* 0x0000000702007986 */ /* 0x0003e8000c101908 */ /*0b60*/ STG.E.STRONG.SYS [R4.64], R9 ; /* 0x0000000904007986 */ /* 0x0013ea000c115908 */ /*0b70*/ @P0 BRA 0xb00 ; /* 0xffffff8000000947 */ /* 0x000fea000383ffff */ /*0b80*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0b90*/ BRA 0xb90; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0ba0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bb0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bc0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bd0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0be0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0bf0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c00*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c10*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c20*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c30*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c40*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c50*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c60*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0c70*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z6kerneliPfPViiii .globl _Z6kerneliPfPViiii .p2align 8 .type _Z6kerneliPfPViiii,@function _Z6kerneliPfPViiii: s_load_b32 s2, s[0:1], 0x18 s_waitcnt lgkmcnt(0) s_cmp_lt_i32 s2, 1 s_cbranch_scc1 .LBB0_7 s_clause 0x1 s_load_b64 s[8:9], s[0:1], 0x1c s_load_b128 s[4:7], s[0:1], 0x8 v_mov_b32_e32 v2, 0 s_mov_b32 s3, 0 s_waitcnt lgkmcnt(0) s_cmp_lg_u32 s8, 0 v_dual_mov_b32 v0, s6 :: v_dual_mov_b32 v1, s7 s_cselect_b32 s0, -1, 0 s_cmp_lg_u32 s9, 0 v_cndmask_b32_e64 v3, 0, 1, s0 s_cselect_b32 s0, -1, 0 s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(VALU_DEP_2) v_cndmask_b32_e64 v4, 0, 1, s0 v_cmp_ne_u32_e64 s0, 1, v3 s_delay_alu instid0(VALU_DEP_2) v_cmp_ne_u32_e64 s1, 1, v4 s_branch .LBB0_3 .LBB0_2: s_add_i32 s3, s3, 1 s_delay_alu instid0(SALU_CYCLE_1) s_cmp_lg_u32 s2, s3 s_cbranch_scc0 .LBB0_7 .LBB0_3: v_cvt_f32_i32_e32 v3, s3 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) s_and_b32 vcc_lo, exec_lo, s0 v_cvt_i32_f32_e32 v4, v3 global_store_b32 v2, v3, s[4:5] flat_store_b32 v[0:1], v4 dlc s_waitcnt_vscnt null, 0x0 s_cbranch_vccnz .LBB0_5 s_waitcnt lgkmcnt(0) buffer_gl1_inv buffer_gl0_inv .LBB0_5: s_and_b32 vcc_lo, exec_lo, s1 s_cbranch_vccnz .LBB0_2 s_waitcnt lgkmcnt(0) buffer_gl0_inv s_branch .LBB0_2 .LBB0_7: s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z6kerneliPfPViiii .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 36 .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 5 .amdhsa_next_free_sgpr 10 .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 _Z6kerneliPfPViiii, .Lfunc_end0-_Z6kerneliPfPViiii .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 - .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 .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 36 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z6kerneliPfPViiii .private_segment_fixed_size: 0 .sgpr_count: 12 .sgpr_spill_count: 0 .symbol: _Z6kerneliPfPViiii.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_0002120d_00000000-6_fence.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB3440: .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 .LFE3440: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata._Z9gpuAssert9cudaErrorPKcib.str1.1,"aMS",@progbits,1 .LC0: .string "GPUassert: %s %s %d\n" .section .text._Z9gpuAssert9cudaErrorPKcib,"axG",@progbits,_Z9gpuAssert9cudaErrorPKcib,comdat .weak _Z9gpuAssert9cudaErrorPKcib .type _Z9gpuAssert9cudaErrorPKcib, @function _Z9gpuAssert9cudaErrorPKcib: .LFB3436: .cfi_startproc endbr64 testl %edi, %edi jne .L9 ret .L9: 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 $8, %rsp .cfi_def_cfa_offset 48 movl %edi, %ebx movq %rsi, %r13 movl %edx, %r12d movl %ecx, %ebp call cudaGetErrorString@PLT movq %rax, %rcx movl %r12d, %r9d movq %r13, %r8 leaq .LC0(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT testb %bpl, %bpl jne .L10 addq $8, %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 .L10: .cfi_restore_state movl %ebx, %edi call exit@PLT .cfi_endproc .LFE3436: .size _Z9gpuAssert9cudaErrorPKcib, .-_Z9gpuAssert9cudaErrorPKcib .text .globl _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii .type _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii, @function _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii: .LFB3462: .cfi_startproc endbr64 subq $168, %rsp .cfi_def_cfa_offset 176 movl %edi, 28(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movl %ecx, 24(%rsp) movl %r8d, 4(%rsp) movl %r9d, (%rsp) movq %fs:40, %rax movq %rax, 152(%rsp) xorl %eax, %eax leaq 28(%rsp), %rax movq %rax, 96(%rsp) leaq 16(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) leaq 24(%rsp), %rax movq %rax, 120(%rsp) leaq 4(%rsp), %rax movq %rax, 128(%rsp) movq %rsp, %rax movq %rax, 136(%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 .L15 .L11: movq 152(%rsp), %rax subq %fs:40, %rax jne .L16 addq $168, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L15: .cfi_restore_state pushq 40(%rsp) .cfi_def_cfa_offset 184 pushq 40(%rsp) .cfi_def_cfa_offset 192 leaq 112(%rsp), %r9 movq 76(%rsp), %rcx movl 84(%rsp), %r8d movq 64(%rsp), %rsi movl 72(%rsp), %edx leaq _Z6kerneliPfPViiii(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 176 jmp .L11 .L16: call __stack_chk_fail@PLT .cfi_endproc .LFE3462: .size _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii, .-_Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii .globl _Z6kerneliPfPViiii .type _Z6kerneliPfPViiii, @function _Z6kerneliPfPViiii: .LFB3463: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3463: .size _Z6kerneliPfPViiii, .-_Z6kerneliPfPViiii .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC2: .string "/home/ubuntu/Datasets/stackv2/train-structured/gganagno/GHammer/master/2_indexing_and_synchronisation/fencing/fence_test/fence.cu" .align 8 .LC3: .string "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n" .section .rodata.str1.1,"aMS",@progbits,1 .LC4: .string "b:t:n:f:s:" .LC5: .string "H: elapsed_time is : %f \n" .text .globl main .type main, @function main: .LFB3437: .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 $104, %rsp .cfi_def_cfa_offset 160 movl %edi, %r12d movq %rsi, %rbp movq %fs:40, %rax movq %rax, 88(%rsp) xorl %eax, %eax movl $0x00000000, 20(%rsp) leaq 24(%rsp), %rdi call cudaEventCreate@PLT movl %eax, %edi movl $1, %ecx movl $69, %edx leaq .LC2(%rip), %rbx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 32(%rsp), %rdi call cudaEventCreate@PLT movl %eax, %edi movl $1, %ecx movl $70, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movl $0, 12(%rsp) movl $0, 8(%rsp) movl $10000, 4(%rsp) movl $1, %r15d movl $1, %r14d leaq .LC4(%rip), %r13 leaq .L23(%rip), %rbx jmp .L20 .L27: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, %r14d .L20: movq %r13, %rdx movq %rbp, %rsi movl %r12d, %edi call getopt@PLT cmpl $-1, %eax je .L33 subl $98, %eax cmpl $18, %eax ja .L21 movl %eax, %eax movslq (%rbx,%rax,4), %rax addq %rbx, %rax notrack jmp *%rax .section .rodata .align 4 .align 4 .L23: .long .L27-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L26-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L25-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L21-.L23 .long .L24-.L23 .long .L22-.L23 .text .L22: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, %r15d jmp .L20 .L25: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, 4(%rsp) jmp .L20 .L26: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, 8(%rsp) jmp .L20 .L24: movl $10, %edx movl $0, %esi movq optarg(%rip), %rdi call __isoc23_strtol@PLT movl %eax, 12(%rsp) jmp .L20 .L21: movq 0(%rbp), %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 .L33: leaq 40(%rsp), %rdi movl $1, %edx movl $8, %esi call cudaMallocManaged@PLT movl %eax, %edi movl $1, %ecx movl $102, %edx leaq .LC2(%rip), %rbx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 48(%rsp), %rdi movl $2, %edx movl $4, %esi call cudaHostAlloc@PLT movl %eax, %edi movl $1, %ecx movl $107, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 56(%rsp), %rdi movl $0, %edx movq 48(%rsp), %rsi call cudaHostGetDevicePointer@PLT movl %eax, %edi movl $1, %ecx movl $111, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movq 40(%rsp), %rax movl $0x00000000, (%rax) movl $0, %esi movq 24(%rsp), %rdi call cudaEventRecord@PLT movl %r15d, 76(%rsp) movl $1, 80(%rsp) movl %r14d, 64(%rsp) movl $1, 68(%rsp) movl $0, %r9d movl $0, %r8d movq 76(%rsp), %rdx movl $1, %ecx movq 64(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L34 .L30: call cudaDeviceSynchronize@PLT movl %eax, %edi movl $1, %ecx movl $121, %edx leaq .LC2(%rip), %rbx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movl $0, %esi movq 32(%rsp), %rdi call cudaEventRecord@PLT movl %eax, %edi movl $1, %ecx movl $123, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib movq 32(%rsp), %rdi call cudaEventSynchronize@PLT movl %eax, %edi movl $1, %ecx movl $125, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib leaq 20(%rsp), %rdi movq 32(%rsp), %rdx movq 24(%rsp), %rsi call cudaEventElapsedTime@PLT movl %eax, %edi movl $1, %ecx movl $127, %edx movq %rbx, %rsi call _Z9gpuAssert9cudaErrorPKcib pxor %xmm0, %xmm0 cvtss2sd 20(%rsp), %xmm0 leaq .LC5(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movq 24(%rsp), %rdi call cudaEventDestroy@PLT movq 32(%rsp), %rdi call cudaEventDestroy@PLT movq 88(%rsp), %rax subq %fs:40, %rax jne .L35 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 .L34: .cfi_restore_state movl %r14d, %edi imull %r15d, %edi movl 12(%rsp), %r9d movl 8(%rsp), %r8d movl 4(%rsp), %ecx movq 56(%rsp), %rdx movq 40(%rsp), %rsi call _Z32__device_stub__Z6kerneliPfPViiiiiPfPViiii jmp .L30 .L35: call __stack_chk_fail@PLT .cfi_endproc .LFE3437: .size main, .-main .section .rodata.str1.1 .LC6: .string "_Z6kerneliPfPViiii" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB3465: .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 .LC6(%rip), %rdx movq %rdx, %rcx leaq _Z6kerneliPfPViiii(%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 .LFE3465: .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 "fence.hip" .globl _Z21__device_stub__kerneliPfPViiii # -- Begin function _Z21__device_stub__kerneliPfPViiii .p2align 4, 0x90 .type _Z21__device_stub__kerneliPfPViiii,@function _Z21__device_stub__kerneliPfPViiii: # @_Z21__device_stub__kerneliPfPViiii .cfi_startproc # %bb.0: subq $136, %rsp .cfi_def_cfa_offset 144 movl %edi, 12(%rsp) movq %rsi, 72(%rsp) movq %rdx, 64(%rsp) movl %ecx, 8(%rsp) movl %r8d, 4(%rsp) movl %r9d, (%rsp) leaq 12(%rsp), %rax movq %rax, 80(%rsp) leaq 72(%rsp), %rax movq %rax, 88(%rsp) leaq 64(%rsp), %rax movq %rax, 96(%rsp) leaq 8(%rsp), %rax movq %rax, 104(%rsp) leaq 4(%rsp), %rax movq %rax, 112(%rsp) movq %rsp, %rax movq %rax, 120(%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 $_Z6kerneliPfPViiii, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $152, %rsp .cfi_adjust_cfa_offset -152 retq .Lfunc_end0: .size _Z21__device_stub__kerneliPfPViiii, .Lfunc_end0-_Z21__device_stub__kerneliPfPViiii .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 $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 %rsi, %rbp movl %edi, %ebx movl $0, 12(%rsp) leaq 24(%rsp), %rdi callq hipEventCreate testl %eax, %eax jne .LBB1_1 # %bb.3: # %_Z9gpuAssert10hipError_tPKcib.exit leaq 16(%rsp), %rdi callq hipEventCreate testl %eax, %eax jne .LBB1_32 # %bb.4: # %_Z9gpuAssert10hipError_tPKcib.exit23.outer.preheader movl $1, %r12d movl $10000, %r14d # imm = 0x2710 xorl %eax, %eax movq %rax, 48(%rsp) # 8-byte Spill xorl %r15d, %r15d movl $1, %r13d jmp .LBB1_5 .p2align 4, 0x90 .LBB1_10: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r13 .LBB1_5: # %_Z9gpuAssert10hipError_tPKcib.exit23 # =>This Inner Loop Header: Depth=1 movl $.L.str.1, %edx movl %ebx, %edi movq %rbp, %rsi callq getopt # kill: def $eax killed $eax def $rax cmpl $98, %eax je .LBB1_10 # %bb.6: # %_Z9gpuAssert10hipError_tPKcib.exit23 # in Loop: Header=BB1_5 Depth=1 leal -102(%rax), %ecx cmpl $14, %ecx ja .LBB1_7 # %bb.9: # %_Z9gpuAssert10hipError_tPKcib.exit23 # in Loop: Header=BB1_5 Depth=1 jmpq *.LJTI1_0(,%rcx,8) .LBB1_13: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r15 jmp .LBB1_5 .LBB1_11: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r12 jmp .LBB1_5 .LBB1_12: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, %r14 jmp .LBB1_5 .LBB1_14: # in Loop: Header=BB1_5 Depth=1 movq optarg(%rip), %rdi xorl %esi, %esi movl $10, %edx callq __isoc23_strtol movq %rax, 48(%rsp) # 8-byte Spill jmp .LBB1_5 .LBB1_7: # %_Z9gpuAssert10hipError_tPKcib.exit23 cmpl $-1, %eax jne .LBB1_8 # %bb.15: leaq 56(%rsp), %rdi movl $8, %esi movl $1, %edx callq hipMallocManaged testl %eax, %eax jne .LBB1_16 # %bb.17: # %_Z9gpuAssert10hipError_tPKcib.exit25 leaq 72(%rsp), %rdi movl $4, %esi movl $2, %edx callq hipHostAlloc testl %eax, %eax jne .LBB1_18 # %bb.19: # %_Z9gpuAssert10hipError_tPKcib.exit27 movq 72(%rsp), %rsi leaq 64(%rsp), %rdi xorl %edx, %edx callq hipHostGetDevicePointer testl %eax, %eax jne .LBB1_20 # %bb.21: # %_Z9gpuAssert10hipError_tPKcib.exit29 movq 56(%rsp), %rax movl $0, (%rax) movq 24(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movl %r13d, %edi movabsq $4294967296, %rax # imm = 0x100000000 orq %rax, %rdi movl %r12d, %edx orq %rax, %rdx movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_23 # %bb.22: imull %r12d, %r13d movq 56(%rsp), %rax movq 64(%rsp), %rcx movl %r13d, 44(%rsp) movq %rax, 136(%rsp) movq %rcx, 128(%rsp) movl %r14d, 40(%rsp) movl %r15d, 36(%rsp) movq 48(%rsp), %rax # 8-byte Reload movl %eax, 32(%rsp) leaq 44(%rsp), %rax movq %rax, 144(%rsp) leaq 136(%rsp), %rax movq %rax, 152(%rsp) leaq 128(%rsp), %rax movq %rax, 160(%rsp) leaq 40(%rsp), %rax movq %rax, 168(%rsp) leaq 36(%rsp), %rax movq %rax, 176(%rsp) leaq 32(%rsp), %rax movq %rax, 184(%rsp) leaq 112(%rsp), %rdi leaq 96(%rsp), %rsi leaq 88(%rsp), %rdx leaq 80(%rsp), %rcx callq __hipPopCallConfiguration movq 112(%rsp), %rsi movl 120(%rsp), %edx movq 96(%rsp), %rcx movl 104(%rsp), %r8d leaq 144(%rsp), %r9 movl $_Z6kerneliPfPViiii, %edi pushq 80(%rsp) .cfi_adjust_cfa_offset 8 pushq 96(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB1_23: callq hipDeviceSynchronize testl %eax, %eax jne .LBB1_24 # %bb.25: # %_Z9gpuAssert10hipError_tPKcib.exit31 movq 16(%rsp), %rdi xorl %esi, %esi callq hipEventRecord testl %eax, %eax jne .LBB1_26 # %bb.27: # %_Z9gpuAssert10hipError_tPKcib.exit33 movq 16(%rsp), %rdi callq hipEventSynchronize testl %eax, %eax jne .LBB1_28 # %bb.29: # %_Z9gpuAssert10hipError_tPKcib.exit35 movq 24(%rsp), %rsi movq 16(%rsp), %rdx leaq 12(%rsp), %rdi callq hipEventElapsedTime testl %eax, %eax jne .LBB1_30 # %bb.31: # %_Z9gpuAssert10hipError_tPKcib.exit37 movss 12(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.3, %edi movb $1, %al callq printf movq 24(%rsp), %rdi callq hipEventDestroy movq 16(%rsp), %rdi callq hipEventDestroy xorl %eax, %eax 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 .LBB1_8: .cfi_def_cfa_offset 256 movq stderr(%rip), %rdi movq (%rbp), %rdx movl $.L.str.2, %esi xorl %eax, %eax callq fprintf movl $1, %edi callq exit .LBB1_1: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $69, %r8d jmp .LBB1_2 .LBB1_32: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $70, %r8d jmp .LBB1_2 .LBB1_16: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $102, %r8d jmp .LBB1_2 .LBB1_18: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $107, %r8d jmp .LBB1_2 .LBB1_20: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $111, %r8d jmp .LBB1_2 .LBB1_24: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $121, %r8d jmp .LBB1_2 .LBB1_26: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $123, %r8d jmp .LBB1_2 .LBB1_28: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $125, %r8d jmp .LBB1_2 .LBB1_30: movq stderr(%rip), %rbx movl %eax, %edi movl %eax, %ebp callq hipGetErrorString movl $.L.str.4, %esi movl $.L.str, %ecx movq %rbx, %rdi movq %rax, %rdx movl $128, %r8d .LBB1_2: xorl %eax, %eax callq fprintf movl %ebp, %edi callq exit .Lfunc_end1: .size main, .Lfunc_end1-main .cfi_endproc .section .rodata,"a",@progbits .p2align 3, 0x0 .LJTI1_0: .quad .LBB1_13 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_12 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_8 .quad .LBB1_14 .quad .LBB1_11 # -- 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 $_Z6kerneliPfPViiii, %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 _Z6kerneliPfPViiii,@object # @_Z6kerneliPfPViiii .section .rodata,"a",@progbits .globl _Z6kerneliPfPViiii .p2align 3, 0x0 _Z6kerneliPfPViiii: .quad _Z21__device_stub__kerneliPfPViiii .size _Z6kerneliPfPViiii, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "/home/ubuntu/Datasets/stackv2/train-structured-repos-hip/gganagno/GHammer/master/2_indexing_and_synchronisation/fencing/fence_test/fence.hip" .size .L.str, 141 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "b:t:n:f:s:" .size .L.str.1, 11 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "Usage: %s -b [blocks] -t [threads] -n [count of iterations] -f [fence_system] -s [fence_block]\n" .size .L.str.2, 96 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "H: elapsed_time is : %f \n" .size .L.str.3, 26 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "GPUassert: %s %s %d\n" .size .L.str.4, 21 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z6kerneliPfPViiii" .size .L__unnamed_1, 19 .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__kerneliPfPViiii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z6kerneliPfPViiii .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 saxpy_float4s_shmem_doublebuffer ( float* y, float* x, float a, clock_t * timer_vals) { volatile __shared__ float sdata_x0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_1 [COMPUTE_THREADS_PER_CTA]; int tid = threadIdx.x ; unsigned int idx0, idx1; idx0 = blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; idx1 = COMPUTE_THREADS_PER_CTA * CTA_COUNT + blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; float4 * x_as_float4 = (float4 *)x; float4 * y_as_float4 = (float4 *)y; float4 result_y; for (int i=0; i < NUM_ITERS/4; i+=2) { float4 tmp1_x, tmp1_y; __syncthreads(); tmp1_x = x_as_float4[idx0]; tmp1_y = y_as_float4[idx0]; if (i!=0) { result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; idx1 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; } sdata_x0_0[tid] = tmp1_x.x; sdata_x1_0[tid] = tmp1_x.y; sdata_x2_0[tid] = tmp1_x.z; sdata_x3_0[tid] = tmp1_x.w; sdata_y0_0[tid] = tmp1_y.x; sdata_y1_0[tid] = tmp1_y.y; sdata_y2_0[tid] = tmp1_y.z; sdata_y3_0[tid] = tmp1_y.w; __syncthreads(); tmp1_x = x_as_float4[idx1]; tmp1_y = y_as_float4[idx1]; result_y.x = a * sdata_x0_0[tid] + sdata_y0_0[tid]; result_y.y = a * sdata_x1_0[tid] + sdata_y1_0[tid]; result_y.z = a * sdata_x2_0[tid] + sdata_y2_0[tid]; result_y.w = a * sdata_x3_0[tid] + sdata_y3_0[tid]; y_as_float4[idx0] = result_y; idx0 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; sdata_x0_1[tid] = tmp1_x.x; sdata_x1_1[tid] = tmp1_x.y; sdata_x2_1[tid] = tmp1_x.z; sdata_x3_1[tid] = tmp1_x.w; sdata_y0_1[tid] = tmp1_y.x; sdata_y1_1[tid] = tmp1_y.y; sdata_y2_1[tid] = tmp1_y.z; sdata_y3_1[tid] = tmp1_y.w; } __syncthreads(); result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; }
code for sm_80 Function : _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .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 R17, SR_CTAID.X ; /* 0x0000000000117919 */ /* 0x000e220000002500 */ /*0020*/ HFMA2.MMA R3, -RZ, RZ, 0, 0 ; /* 0x00000000ff037435 */ /* 0x000fe200000001ff */ /*0030*/ MOV R2, 0xe00 ; /* 0x00000e0000027802 */ /* 0x000fe20000000f00 */ /*0040*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0050*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e240000002100 */ /*0060*/ LEA R17, R17, R0, 0x7 ; /* 0x0000000011117211 */ /* 0x001fc800078e38ff */ /*0070*/ IADD3 R16, R17, 0x700, RZ ; /* 0x0000070011107810 */ /* 0x000fe40007ffe0ff */ /*0080*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fe20000010000 */ /*0090*/ MOV R18, 0x10 ; /* 0x0000001000127802 */ /* 0x000fca0000000f00 */ /*00a0*/ IMAD.WIDE.U32 R4, R17, R18, c[0x0][0x168] ; /* 0x00005a0011047625 */ /* 0x001fc800078e0012 */ /*00b0*/ IMAD.WIDE.U32 R20, R17, R18, c[0x0][0x160] ; /* 0x0000580011147625 */ /* 0x000fe400078e0012 */ /*00c0*/ LDG.E.128 R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000ea8000c1e1d00 */ /*00d0*/ LDG.E.128 R8, [R20.64] ; /* 0x0000000414087981 */ /* 0x000ee2000c1e1d00 */ /*00e0*/ ISETP.NE.AND P0, PT, R2, 0xe00, PT ; /* 0x00000e000200780c */ /* 0x000fda0003f05270 */ /*00f0*/ @P0 LDS R12, [R0.X4+0x1000] ; /* 0x00100000000c0984 */ /* 0x000fe20000004800 */ /*0100*/ @P0 IMAD.WIDE.U32 R24, R16.reuse, R18, c[0x0][0x160] ; /* 0x0000580010180625 */ /* 0x040fe200078e0012 */ /*0110*/ @P0 IADD3 R16, R16, 0xe00, RZ ; /* 0x00000e0010100810 */ /* 0x000fe40007ffe0ff */ /*0120*/ @P0 LDS R13, [R0.X4+0x1800] ; /* 0x00180000000d0984 */ /* 0x000e280000004800 */ /*0130*/ @P0 LDS R14, [R0.X4+0x1200] ; /* 0x00120000000e0984 */ /* 0x000fe80000004800 */ /*0140*/ @P0 LDS R15, [R0.X4+0x1a00] ; /* 0x001a0000000f0984 */ /* 0x000e680000004800 */ /*0150*/ @P0 LDS R19, [R0.X4+0x1400] ; /* 0x0014000000130984 */ /* 0x000fe80000004800 */ /*0160*/ @P0 LDS R22, [R0.X4+0x1c00] ; /* 0x001c000000160984 */ /* 0x000f280000004800 */ /*0170*/ @P0 LDS R23, [R0.X4+0x1600] ; /* 0x0016000000170984 */ /* 0x000fe80000004800 */ /*0180*/ @P0 LDS R26, [R0.X4+0x1e00] ; /* 0x001e0000001a0984 */ /* 0x000f620000004800 */ /*0190*/ @P0 FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c0a23 */ /* 0x001fc4000000000d */ /*01a0*/ @P0 FFMA R13, R14, c[0x0][0x170], R15 ; /* 0x00005c000e0d0a23 */ /* 0x002fe4000000000f */ /*01b0*/ @P0 FFMA R14, R19, c[0x0][0x170], R22 ; /* 0x00005c00130e0a23 */ /* 0x010fe40000000016 */ /*01c0*/ @P0 FFMA R15, R23, c[0x0][0x170], R26 ; /* 0x00005c00170f0a23 */ /* 0x020fe4000000001a */ /*01d0*/ IMAD.WIDE.U32 R26, R16, R18, c[0x0][0x168] ; /* 0x00005a00101a7625 */ /* 0x000fc600078e0012 */ /*01e0*/ @P0 STG.E.128 [R24.64], R12 ; /* 0x0000000c18000986 */ /* 0x000fe2000c101d04 */ /*01f0*/ IMAD.WIDE.U32 R22, R16, R18, c[0x0][0x160] ; /* 0x0000580010167625 */ /* 0x000fc600078e0012 */ /*0200*/ STS [R0.X4], R4 ; /* 0x0000000400007388 */ /* 0x004fe80000004800 */ /*0210*/ STS [R0.X4+0x200], R5 ; /* 0x0002000500007388 */ /* 0x000fe80000004800 */ /*0220*/ STS [R0.X4+0x400], R6 ; /* 0x0004000600007388 */ /* 0x000fe80000004800 */ /*0230*/ STS [R0.X4+0x600], R7 ; /* 0x0006000700007388 */ /* 0x0001e80000004800 */ /*0240*/ STS [R0.X4+0x800], R8 ; /* 0x0008000800007388 */ /* 0x008fe80000004800 */ /*0250*/ STS [R0.X4+0xa00], R9 ; /* 0x000a000900007388 */ /* 0x000fe80000004800 */ /*0260*/ STS [R0.X4+0xc00], R10 ; /* 0x000c000a00007388 */ /* 0x000fe80000004800 */ /*0270*/ STS [R0.X4+0xe00], R11 ; /* 0x000e000b00007388 */ /* 0x0003e80000004800 */ /*0280*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0290*/ LDG.E.128 R4, [R26.64] ; /* 0x000000041a047981 */ /* 0x0010a8000c1e1d00 */ /*02a0*/ LDG.E.128 R8, [R22.64] ; /* 0x0000000416087981 */ /* 0x002ee8000c1e1d00 */ /*02b0*/ LDS R12, [R0.X4] ; /* 0x00000000000c7984 */ /* 0x000fe80000004800 */ /*02c0*/ LDS R13, [R0.X4+0x800] ; /* 0x00080000000d7984 */ /* 0x000e680000004800 */ /*02d0*/ LDS R14, [R0.X4+0x200] ; /* 0x00020000000e7984 */ /* 0x000fe80000004800 */ /*02e0*/ LDS R19, [R0.X4+0xa00] ; /* 0x000a000000137984 */ /* 0x000f280000004800 */ /*02f0*/ LDS R25, [R0.X4+0x400] ; /* 0x0004000000197984 */ /* 0x000fe80000004800 */ /*0300*/ LDS R28, [R0.X4+0xc00] ; /* 0x000c0000001c7984 */ /* 0x000f680000004800 */ /*0310*/ LDS R15, [R0.X4+0x600] ; /* 0x00060000000f7984 */ /* 0x000fe80000004800 */ /*0320*/ LDS R24, [R0.X4+0xe00] ; /* 0x000e000000187984 */ /* 0x000e220000004800 */ /*0330*/ FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c7a23 */ /* 0x002fc4000000000d */ /*0340*/ FFMA R13, R14, c[0x0][0x170], R19 ; /* 0x00005c000e0d7a23 */ /* 0x010fe40000000013 */ /*0350*/ FFMA R14, R25, c[0x0][0x170], R28 ; /* 0x00005c00190e7a23 */ /* 0x020fe2000000001c */ /*0360*/ IADD3 R25, R17, 0xe00, RZ ; /* 0x00000e0011197810 */ /* 0x000fca0007ffe0ff */ /*0370*/ IMAD.WIDE.U32 R26, R25, R18, c[0x0][0x168] ; /* 0x00005a00191a7625 */ /* 0x001fc800078e0012 */ /*0380*/ FFMA R15, R15, c[0x0][0x170], R24 ; /* 0x00005c000f0f7a23 */ /* 0x000fca0000000018 */ /*0390*/ STG.E.128 [R20.64], R12 ; /* 0x0000000c14007986 */ /* 0x000fe2000c101d04 */ /*03a0*/ IMAD.WIDE.U32 R24, R25, R18, c[0x0][0x160] ; /* 0x0000580019187625 */ /* 0x000fc600078e0012 */ /*03b0*/ STS [R0.X4+0x1000], R4 ; /* 0x0010000400007388 */ /* 0x004fe80000004800 */ /*03c0*/ STS [R0.X4+0x1200], R5 ; /* 0x0012000500007388 */ /* 0x000fe80000004800 */ /*03d0*/ STS [R0.X4+0x1400], R6 ; /* 0x0014000600007388 */ /* 0x000fe80000004800 */ /*03e0*/ STS [R0.X4+0x1600], R7 ; /* 0x0016000700007388 */ /* 0x0001e80000004800 */ /*03f0*/ STS [R0.X4+0x1800], R8 ; /* 0x0018000800007388 */ /* 0x008fe80000004800 */ /*0400*/ STS [R0.X4+0x1a00], R9 ; /* 0x001a000900007388 */ /* 0x000fe80000004800 */ /*0410*/ STS [R0.X4+0x1c00], R10 ; /* 0x001c000a00007388 */ /* 0x000fe80000004800 */ /*0420*/ STS [R0.X4+0x1e00], R11 ; /* 0x001e000b00007388 */ /* 0x0003e80000004800 */ /*0430*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0440*/ LDG.E.128 R4, [R26.64] ; /* 0x000000041a047981 */ /* 0x001ea8000c1e1d00 */ /*0450*/ LDG.E.128 R8, [R24.64] ; /* 0x0000000418087981 */ /* 0x002ee2000c1e1d00 */ /*0460*/ IADD3 R16, R16, 0xe00, RZ ; /* 0x00000e0010107810 */ /* 0x000fc60007ffe0ff */ /*0470*/ LDS R12, [R0.X4+0x1000] ; /* 0x00100000000c7984 */ /* 0x000fe80000004800 */ /*0480*/ LDS R13, [R0.X4+0x1800] ; /* 0x00180000000d7984 */ /* 0x000e280000004800 */ /*0490*/ LDS R14, [R0.X4+0x1200] ; /* 0x00120000000e7984 */ /* 0x000fe80000004800 */ /*04a0*/ LDS R15, [R0.X4+0x1a00] ; /* 0x001a0000000f7984 */ /* 0x000e680000004800 */ /*04b0*/ LDS R19, [R0.X4+0x1400] ; /* 0x0014000000137984 */ /* 0x000fe80000004800 */ /*04c0*/ LDS R20, [R0.X4+0x1c00] ; /* 0x001c000000147984 */ /* 0x000f280000004800 */ /*04d0*/ LDS R21, [R0.X4+0x1600] ; /* 0x0016000000157984 */ /* 0x000fe80000004800 */ /*04e0*/ LDS R28, [R0.X4+0x1e00] ; /* 0x001e0000001c7984 */ /* 0x000f620000004800 */ /*04f0*/ FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c7a23 */ /* 0x001fc4000000000d */ /*0500*/ FFMA R13, R14, c[0x0][0x170], R15 ; /* 0x00005c000e0d7a23 */ /* 0x002fe4000000000f */ /*0510*/ FFMA R14, R19, c[0x0][0x170], R20 ; /* 0x00005c00130e7a23 */ /* 0x010fe40000000014 */ /*0520*/ FFMA R15, R21, c[0x0][0x170], R28 ; /* 0x00005c00150f7a23 */ /* 0x020fe4000000001c */ /*0530*/ IMAD.WIDE.U32 R20, R16, R18, c[0x0][0x168] ; /* 0x00005a0010147625 */ /* 0x000fc600078e0012 */ /*0540*/ STG.E.128 [R22.64], R12 ; /* 0x0000000c16007986 */ /* 0x000fe2000c101d04 */ /*0550*/ IMAD.WIDE.U32 R18, R16, R18, c[0x0][0x160] ; /* 0x0000580010127625 */ /* 0x000fc600078e0012 */ /*0560*/ STS [R0.X4], R4 ; /* 0x0000000400007388 */ /* 0x004fe80000004800 */ /*0570*/ STS [R0.X4+0x200], R5 ; /* 0x0002000500007388 */ /* 0x000fe80000004800 */ /*0580*/ STS [R0.X4+0x400], R6 ; /* 0x0004000600007388 */ /* 0x000fe80000004800 */ /*0590*/ STS [R0.X4+0x600], R7 ; /* 0x0006000700007388 */ /* 0x0001e80000004800 */ /*05a0*/ STS [R0.X4+0x800], R8 ; /* 0x0008000800007388 */ /* 0x008fe80000004800 */ /*05b0*/ STS [R0.X4+0xa00], R9 ; /* 0x000a000900007388 */ /* 0x000fe80000004800 */ /*05c0*/ STS [R0.X4+0xc00], R10 ; /* 0x000c000a00007388 */ /* 0x000fe80000004800 */ /*05d0*/ STS [R0.X4+0xe00], R11 ; /* 0x000e000b00007388 */ /* 0x0003e80000004800 */ /*05e0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*05f0*/ LDG.E.128 R4, [R20.64] ; /* 0x0000000414047981 */ /* 0x001ea8000c1e1d00 */ /*0600*/ LDG.E.128 R8, [R18.64] ; /* 0x0000000412087981 */ /* 0x002ee2000c1e1d00 */ /*0610*/ IADD3 R3, R3, 0x4, RZ ; /* 0x0000000403037810 */ /* 0x000fc40007ffe0ff */ /*0620*/ IADD3 R2, R2, 0x1c00, RZ ; /* 0x00001c0002027810 */ /* 0x000fe20007ffe0ff */ /*0630*/ LDS R12, [R0.X4] ; /* 0x00000000000c7984 */ /* 0x000fe20000004800 */ /*0640*/ ISETP.GE.U32.AND P0, PT, R3, 0x200, PT ; /* 0x000002000300780c */ /* 0x000fe40003f06070 */ /*0650*/ IADD3 R17, R17, 0x1c00, RZ ; /* 0x00001c0011117810 */ /* 0x000fe20007ffe0ff */ /*0660*/ LDS R13, [R0.X4+0x800] ; /* 0x00080000000d7984 */ /* 0x000e280000004800 */ /*0670*/ LDS R14, [R0.X4+0x200] ; /* 0x00020000000e7984 */ /* 0x000fe80000004800 */ /*0680*/ LDS R15, [R0.X4+0xa00] ; /* 0x000a0000000f7984 */ /* 0x000e680000004800 */ /*0690*/ LDS R22, [R0.X4+0x400] ; /* 0x0004000000167984 */ /* 0x000fe80000004800 */ /*06a0*/ LDS R23, [R0.X4+0xc00] ; /* 0x000c000000177984 */ /* 0x000f280000004800 */ /*06b0*/ LDS R26, [R0.X4+0x600] ; /* 0x00060000001a7984 */ /* 0x000fe80000004800 */ /*06c0*/ LDS R27, [R0.X4+0xe00] ; /* 0x000e0000001b7984 */ /* 0x000f620000004800 */ /*06d0*/ FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c7a23 */ /* 0x001fc4000000000d */ /*06e0*/ FFMA R13, R14, c[0x0][0x170], R15 ; /* 0x00005c000e0d7a23 */ /* 0x002fe4000000000f */ /*06f0*/ FFMA R14, R22, c[0x0][0x170], R23 ; /* 0x00005c00160e7a23 */ /* 0x010fe40000000017 */ /*0700*/ FFMA R15, R26, c[0x0][0x170], R27 ; /* 0x00005c001a0f7a23 */ /* 0x020fca000000001b */ /*0710*/ STG.E.128 [R24.64], R12 ; /* 0x0000000c18007986 */ /* 0x0001e8000c101d04 */ /*0720*/ STS [R0.X4+0x1000], R4 ; /* 0x0010000400007388 */ /* 0x0041e80000004800 */ /*0730*/ STS [R0.X4+0x1200], R5 ; /* 0x0012000500007388 */ /* 0x0001e80000004800 */ /*0740*/ STS [R0.X4+0x1400], R6 ; /* 0x0014000600007388 */ /* 0x0001e80000004800 */ /*0750*/ STS [R0.X4+0x1600], R7 ; /* 0x0016000700007388 */ /* 0x0001e80000004800 */ /*0760*/ STS [R0.X4+0x1800], R8 ; /* 0x0018000800007388 */ /* 0x0081e80000004800 */ /*0770*/ STS [R0.X4+0x1a00], R9 ; /* 0x001a000900007388 */ /* 0x0001e80000004800 */ /*0780*/ STS [R0.X4+0x1c00], R10 ; /* 0x001c000a00007388 */ /* 0x0001e80000004800 */ /*0790*/ STS [R0.X4+0x1e00], R11 ; /* 0x001e000b00007388 */ /* 0x0001e20000004800 */ /*07a0*/ @!P0 BRA 0x80 ; /* 0xfffff8d000008947 */ /* 0x000fea000383ffff */ /*07b0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*07c0*/ LDS R4, [R0.X4+0x1000] ; /* 0x0010000000047984 */ /* 0x001fe80000004800 */ /*07d0*/ LDS R3, [R0.X4+0x1800] ; /* 0x0018000000037984 */ /* 0x000e280000004800 */ /*07e0*/ LDS R5, [R0.X4+0x1200] ; /* 0x0012000000057984 */ /* 0x000fe80000004800 */ /*07f0*/ LDS R2, [R0.X4+0x1a00] ; /* 0x001a000000027984 */ /* 0x000e680000004800 */ /*0800*/ LDS R6, [R0.X4+0x1400] ; /* 0x0014000000067984 */ /* 0x000fe80000004800 */ /*0810*/ LDS R7, [R0.X4+0x1c00] ; /* 0x001c000000077984 */ /* 0x000ea80000004800 */ /*0820*/ LDS R8, [R0.X4+0x1600] ; /* 0x0016000000087984 */ /* 0x000fe80000004800 */ /*0830*/ LDS R9, [R0.X4+0x1e00] ; /* 0x001e000000097984 */ /* 0x000ee20000004800 */ /*0840*/ FFMA R4, R4, c[0x0][0x170], R3 ; /* 0x00005c0004047a23 */ /* 0x001fc40000000003 */ /*0850*/ FFMA R5, R5, c[0x0][0x170], R2 ; /* 0x00005c0005057a23 */ /* 0x002fe40000000002 */ /*0860*/ FFMA R6, R6, c[0x0][0x170], R7 ; /* 0x00005c0006067a23 */ /* 0x004fe40000000007 */ /*0870*/ FFMA R7, R8, c[0x0][0x170], R9 ; /* 0x00005c0008077a23 */ /* 0x008fca0000000009 */ /*0880*/ STG.E.128 [R18.64], R4 ; /* 0x0000000412007986 */ /* 0x000fe2000c101d04 */ /*0890*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*08a0*/ BRA 0x8a0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*08b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0900*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0910*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0920*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0930*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0940*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0950*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0960*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0970*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include "includes.h" __global__ void saxpy_float4s_shmem_doublebuffer ( float* y, float* x, float a, clock_t * timer_vals) { volatile __shared__ float sdata_x0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_1 [COMPUTE_THREADS_PER_CTA]; int tid = threadIdx.x ; unsigned int idx0, idx1; idx0 = blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; idx1 = COMPUTE_THREADS_PER_CTA * CTA_COUNT + blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; float4 * x_as_float4 = (float4 *)x; float4 * y_as_float4 = (float4 *)y; float4 result_y; for (int i=0; i < NUM_ITERS/4; i+=2) { float4 tmp1_x, tmp1_y; __syncthreads(); tmp1_x = x_as_float4[idx0]; tmp1_y = y_as_float4[idx0]; if (i!=0) { result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; idx1 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; } sdata_x0_0[tid] = tmp1_x.x; sdata_x1_0[tid] = tmp1_x.y; sdata_x2_0[tid] = tmp1_x.z; sdata_x3_0[tid] = tmp1_x.w; sdata_y0_0[tid] = tmp1_y.x; sdata_y1_0[tid] = tmp1_y.y; sdata_y2_0[tid] = tmp1_y.z; sdata_y3_0[tid] = tmp1_y.w; __syncthreads(); tmp1_x = x_as_float4[idx1]; tmp1_y = y_as_float4[idx1]; result_y.x = a * sdata_x0_0[tid] + sdata_y0_0[tid]; result_y.y = a * sdata_x1_0[tid] + sdata_y1_0[tid]; result_y.z = a * sdata_x2_0[tid] + sdata_y2_0[tid]; result_y.w = a * sdata_x3_0[tid] + sdata_y3_0[tid]; y_as_float4[idx0] = result_y; idx0 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; sdata_x0_1[tid] = tmp1_x.x; sdata_x1_1[tid] = tmp1_x.y; sdata_x2_1[tid] = tmp1_x.z; sdata_x3_1[tid] = tmp1_x.w; sdata_y0_1[tid] = tmp1_y.x; sdata_y1_1[tid] = tmp1_y.y; sdata_y2_1[tid] = tmp1_y.z; sdata_y3_1[tid] = tmp1_y.w; } __syncthreads(); result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; }
.file "tmpxft_0012ed74_00000000-6_saxpy_float4s_shmem_doublebuffer.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 _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl .type _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl, @function _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movss %xmm0, 12(%rsp) movq %rdx, (%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) 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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl(%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 _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl, .-_Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl .globl _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .type _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, @function _Z32saxpy_float4s_shmem_doublebufferPfS_fPl: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, .-_Z32saxpy_float4s_shmem_doublebufferPfS_fPl .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z32saxpy_float4s_shmem_doublebufferPfS_fPl" .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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl(%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 saxpy_float4s_shmem_doublebuffer ( float* y, float* x, float a, clock_t * timer_vals) { volatile __shared__ float sdata_x0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_1 [COMPUTE_THREADS_PER_CTA]; int tid = threadIdx.x ; unsigned int idx0, idx1; idx0 = blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; idx1 = COMPUTE_THREADS_PER_CTA * CTA_COUNT + blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; float4 * x_as_float4 = (float4 *)x; float4 * y_as_float4 = (float4 *)y; float4 result_y; for (int i=0; i < NUM_ITERS/4; i+=2) { float4 tmp1_x, tmp1_y; __syncthreads(); tmp1_x = x_as_float4[idx0]; tmp1_y = y_as_float4[idx0]; if (i!=0) { result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; idx1 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; } sdata_x0_0[tid] = tmp1_x.x; sdata_x1_0[tid] = tmp1_x.y; sdata_x2_0[tid] = tmp1_x.z; sdata_x3_0[tid] = tmp1_x.w; sdata_y0_0[tid] = tmp1_y.x; sdata_y1_0[tid] = tmp1_y.y; sdata_y2_0[tid] = tmp1_y.z; sdata_y3_0[tid] = tmp1_y.w; __syncthreads(); tmp1_x = x_as_float4[idx1]; tmp1_y = y_as_float4[idx1]; result_y.x = a * sdata_x0_0[tid] + sdata_y0_0[tid]; result_y.y = a * sdata_x1_0[tid] + sdata_y1_0[tid]; result_y.z = a * sdata_x2_0[tid] + sdata_y2_0[tid]; result_y.w = a * sdata_x3_0[tid] + sdata_y3_0[tid]; y_as_float4[idx0] = result_y; idx0 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; sdata_x0_1[tid] = tmp1_x.x; sdata_x1_1[tid] = tmp1_x.y; sdata_x2_1[tid] = tmp1_x.z; sdata_x3_1[tid] = tmp1_x.w; sdata_y0_1[tid] = tmp1_y.x; sdata_y1_1[tid] = tmp1_y.y; sdata_y2_1[tid] = tmp1_y.z; sdata_y3_1[tid] = tmp1_y.w; } __syncthreads(); result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void saxpy_float4s_shmem_doublebuffer ( float* y, float* x, float a, clock_t * timer_vals) { volatile __shared__ float sdata_x0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_1 [COMPUTE_THREADS_PER_CTA]; int tid = threadIdx.x ; unsigned int idx0, idx1; idx0 = blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; idx1 = COMPUTE_THREADS_PER_CTA * CTA_COUNT + blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; float4 * x_as_float4 = (float4 *)x; float4 * y_as_float4 = (float4 *)y; float4 result_y; for (int i=0; i < NUM_ITERS/4; i+=2) { float4 tmp1_x, tmp1_y; __syncthreads(); tmp1_x = x_as_float4[idx0]; tmp1_y = y_as_float4[idx0]; if (i!=0) { result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; idx1 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; } sdata_x0_0[tid] = tmp1_x.x; sdata_x1_0[tid] = tmp1_x.y; sdata_x2_0[tid] = tmp1_x.z; sdata_x3_0[tid] = tmp1_x.w; sdata_y0_0[tid] = tmp1_y.x; sdata_y1_0[tid] = tmp1_y.y; sdata_y2_0[tid] = tmp1_y.z; sdata_y3_0[tid] = tmp1_y.w; __syncthreads(); tmp1_x = x_as_float4[idx1]; tmp1_y = y_as_float4[idx1]; result_y.x = a * sdata_x0_0[tid] + sdata_y0_0[tid]; result_y.y = a * sdata_x1_0[tid] + sdata_y1_0[tid]; result_y.z = a * sdata_x2_0[tid] + sdata_y2_0[tid]; result_y.w = a * sdata_x3_0[tid] + sdata_y3_0[tid]; y_as_float4[idx0] = result_y; idx0 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; sdata_x0_1[tid] = tmp1_x.x; sdata_x1_1[tid] = tmp1_x.y; sdata_x2_1[tid] = tmp1_x.z; sdata_x3_1[tid] = tmp1_x.w; sdata_y0_1[tid] = tmp1_y.x; sdata_y1_1[tid] = tmp1_y.y; sdata_y2_1[tid] = tmp1_y.z; sdata_y3_1[tid] = tmp1_y.w; } __syncthreads(); result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; }
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 saxpy_float4s_shmem_doublebuffer ( float* y, float* x, float a, clock_t * timer_vals) { volatile __shared__ float sdata_x0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_1 [COMPUTE_THREADS_PER_CTA]; int tid = threadIdx.x ; unsigned int idx0, idx1; idx0 = blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; idx1 = COMPUTE_THREADS_PER_CTA * CTA_COUNT + blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; float4 * x_as_float4 = (float4 *)x; float4 * y_as_float4 = (float4 *)y; float4 result_y; for (int i=0; i < NUM_ITERS/4; i+=2) { float4 tmp1_x, tmp1_y; __syncthreads(); tmp1_x = x_as_float4[idx0]; tmp1_y = y_as_float4[idx0]; if (i!=0) { result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; idx1 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; } sdata_x0_0[tid] = tmp1_x.x; sdata_x1_0[tid] = tmp1_x.y; sdata_x2_0[tid] = tmp1_x.z; sdata_x3_0[tid] = tmp1_x.w; sdata_y0_0[tid] = tmp1_y.x; sdata_y1_0[tid] = tmp1_y.y; sdata_y2_0[tid] = tmp1_y.z; sdata_y3_0[tid] = tmp1_y.w; __syncthreads(); tmp1_x = x_as_float4[idx1]; tmp1_y = y_as_float4[idx1]; result_y.x = a * sdata_x0_0[tid] + sdata_y0_0[tid]; result_y.y = a * sdata_x1_0[tid] + sdata_y1_0[tid]; result_y.z = a * sdata_x2_0[tid] + sdata_y2_0[tid]; result_y.w = a * sdata_x3_0[tid] + sdata_y3_0[tid]; y_as_float4[idx0] = result_y; idx0 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; sdata_x0_1[tid] = tmp1_x.x; sdata_x1_1[tid] = tmp1_x.y; sdata_x2_1[tid] = tmp1_x.z; sdata_x3_1[tid] = tmp1_x.w; sdata_y0_1[tid] = tmp1_y.x; sdata_y1_1[tid] = tmp1_y.y; sdata_y2_1[tid] = tmp1_y.z; sdata_y3_1[tid] = tmp1_y.w; } __syncthreads(); result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .globl _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .p2align 8 .type _Z32saxpy_float4s_shmem_doublebufferPfS_fPl,@function _Z32saxpy_float4s_shmem_doublebufferPfS_fPl: v_dual_mov_b32 v12, 0 :: v_dual_lshlrev_b32 v1, 2, v0 s_mov_b64 s[2:3], src_shared_base s_clause 0x1 s_load_b128 s[4:7], s[0:1], 0x0 s_load_b32 s2, s[0:1], 0x10 v_lshl_add_u32 v11, s15, 7, v0 v_cmp_ne_u32_e32 vcc_lo, -1, v1 v_add_nc_u32_e32 v3, 0x1200, v1 v_add_nc_u32_e32 v5, 0x1600, v1 s_delay_alu instid0(VALU_DEP_4) v_add_nc_u32_e32 v9, 0x700, v11 v_cndmask_b32_e32 v13, 0, v1, vcc_lo v_cndmask_b32_e64 v14, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v3 v_dual_cndmask_b32 v17, 0, v3 :: v_dual_add_nc_u32 v2, 0x200, v1 v_cndmask_b32_e64 v18, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v5 v_add_nc_u32_e32 v3, 0x1a00, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(SKIP_2) | instid1(VALU_DEP_4) v_cmp_ne_u32_e64 s0, -1, v2 v_cndmask_b32_e32 v21, 0, v5, vcc_lo v_cndmask_b32_e64 v22, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v3 v_add_nc_u32_e32 v4, 0x600, v1 v_cndmask_b32_e64 v15, 0, v2, s0 v_cndmask_b32_e64 v16, 0, s3, s0 v_dual_cndmask_b32 v25, 0, v3 :: v_dual_add_nc_u32 v2, 0xa00, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(SKIP_2) | instid1(VALU_DEP_3) v_cmp_ne_u32_e64 s0, -1, v4 v_cndmask_b32_e64 v26, 0, s3, vcc_lo v_add_nc_u32_e32 v3, 0x400, v1 v_cndmask_b32_e64 v19, 0, v4, s0 v_cndmask_b32_e64 v20, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v2 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2) v_cndmask_b32_e64 v23, 0, v2, s0 v_add_nc_u32_e32 v2, 0x1e00, v1 v_cndmask_b32_e64 v24, 0, s3, s0 v_cmp_ne_u32_e32 vcc_lo, -1, v2 v_dual_cndmask_b32 v29, 0, v2 :: v_dual_add_nc_u32 v4, 0xe00, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_3) v_cmp_ne_u32_e64 s0, -1, v4 v_add_nc_u32_e32 v2, 0x800, v1 v_cndmask_b32_e64 v30, 0, s3, vcc_lo v_cndmask_b32_e64 v27, 0, v4, s0 v_cndmask_b32_e64 v28, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v3 v_cmp_ne_u32_e32 vcc_lo, -1, v2 v_add_nc_u32_e32 v4, 0xc00, v1 s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_2) | instid1(VALU_DEP_4) v_cndmask_b32_e64 v31, 0, v3, s0 v_cndmask_b32_e32 v33, 0, v2, vcc_lo v_cndmask_b32_e64 v32, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v4 v_or_b32_e32 v2, 0x1000, v1 v_cndmask_b32_e64 v34, 0, s3, vcc_lo v_add_nc_u32_e32 v3, 0x1400, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4) v_cndmask_b32_e64 v35, 0, v4, s0 v_cmp_ne_u32_e32 vcc_lo, -1, v2 v_add_nc_u32_e32 v4, 0x1800, v1 v_add_nc_u32_e32 v1, 0x1c00, v1 v_cndmask_b32_e64 v36, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v3 v_cndmask_b32_e32 v37, 0, v2, vcc_lo v_cndmask_b32_e64 v38, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v4 v_cmp_ne_u32_e64 s1, -1, v1 v_cndmask_b32_e64 v39, 0, v3, s0 v_cndmask_b32_e64 v40, 0, s3, s0 s_mov_b32 s0, -2 v_cndmask_b32_e32 v41, 0, v4, vcc_lo v_cndmask_b32_e64 v42, 0, s3, vcc_lo v_cndmask_b32_e64 v43, 0, v1, s1 v_cndmask_b32_e64 v44, 0, s3, s1 s_branch .LBB0_2 .LBB0_1: v_mov_b32_e32 v10, v12 s_waitcnt vmcnt(1) flat_store_b32 v[13:14], v5 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[31:32], v6 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[33:34], v7 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[35:36], v8 dlc s_waitcnt_vscnt null, 0x0 s_waitcnt vmcnt(0) flat_store_b32 v[37:38], v1 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[39:40], v2 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[41:42], v3 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[43:44], v4 dlc s_waitcnt_vscnt null, 0x0 s_waitcnt lgkmcnt(0) s_barrier v_lshlrev_b64 v[5:6], 4, v[9:10] buffer_gl0_inv s_add_i32 s0, s0, 2 v_add_nc_u32_e32 v11, 0xe00, v11 s_cmpk_gt_u32 s0, 0x1fd v_add_co_u32 v1, vcc_lo, s6, v5 v_add_co_ci_u32_e32 v2, vcc_lo, s7, v6, vcc_lo v_add_co_u32 v5, vcc_lo, s4, v5 v_add_co_ci_u32_e32 v6, vcc_lo, s5, v6, vcc_lo global_load_b128 v[1:4], v[1:2], off global_load_b128 v[5:8], v[5:6], off flat_load_b32 v10, v[13:14] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v47, v[37:38] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v51, v[31:32] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v48, v[39:40] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v52, v[33:34] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v49, v[41:42] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v53, v[35:36] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v50, v[43:44] glc dlc s_waitcnt vmcnt(0) lgkmcnt(4) v_dual_fmac_f32 v47, s2, v10 :: v_dual_fmac_f32 v48, s2, v51 s_waitcnt lgkmcnt(0) v_dual_fmac_f32 v49, s2, v52 :: v_dual_fmac_f32 v50, s2, v53 global_store_b128 v[45:46], v[47:50], off flat_store_b32 v[15:16], v1 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[19:20], v2 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[23:24], v3 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[27:28], v4 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[17:18], v5 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[21:22], v6 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[25:26], v7 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[29:30], v8 dlc s_waitcnt_vscnt null, 0x0 s_cbranch_scc1 .LBB0_4 .LBB0_2: v_lshlrev_b64 v[1:2], 4, v[11:12] s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv s_cmp_eq_u32 s0, -2 v_add_co_u32 v3, vcc_lo, s6, v1 v_add_co_ci_u32_e32 v4, vcc_lo, s7, v2, vcc_lo v_add_co_u32 v45, vcc_lo, s4, v1 v_add_co_ci_u32_e32 v46, vcc_lo, s5, v2, vcc_lo global_load_b128 v[5:8], v[3:4], off global_load_b128 v[1:4], v[45:46], off s_cbranch_scc1 .LBB0_1 flat_load_b32 v53, v[15:16] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v47, v[17:18] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v54, v[19:20] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v48, v[21:22] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v55, v[23:24] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v49, v[25:26] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v56, v[27:28] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v50, v[29:30] glc dlc s_waitcnt vmcnt(0) lgkmcnt(6) v_dual_mov_b32 v10, v12 :: v_dual_fmac_f32 v47, s2, v53 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_4) | instid1(VALU_DEP_3) v_lshlrev_b64 v[51:52], 4, v[9:10] s_waitcnt lgkmcnt(4) v_dual_fmac_f32 v48, s2, v54 :: v_dual_add_nc_u32 v9, 0xe00, v9 s_waitcnt lgkmcnt(0) v_dual_fmac_f32 v49, s2, v55 :: v_dual_fmac_f32 v50, s2, v56 v_add_co_u32 v51, vcc_lo, s4, v51 v_add_co_ci_u32_e32 v52, vcc_lo, s5, v52, vcc_lo global_store_b128 v[51:52], v[47:50], off s_branch .LBB0_1 .LBB0_4: v_lshlrev_b32_e32 v6, 2, v0 s_mov_b64 s[6:7], src_shared_base s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv v_add_nc_u32_e32 v0, 0x200, v6 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2) v_cmp_ne_u32_e32 vcc_lo, -1, v0 v_dual_cndmask_b32 v0, 0, v0 :: v_dual_add_nc_u32 v7, 0x1600, v6 v_cndmask_b32_e64 v1, 0, s7, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v7 flat_load_b32 v12, v[0:1] glc dlc s_waitcnt vmcnt(0) v_cndmask_b32_e32 v1, 0, v7, vcc_lo v_add_nc_u32_e32 v7, 0xe00, v6 v_add_nc_u32_e32 v2, 0x1200, v6 v_add_nc_u32_e32 v4, 0x600, v6 v_add_nc_u32_e32 v10, 0x1e00, v6 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3) v_cmp_ne_u32_e64 s0, -1, v2 v_cmp_ne_u32_e64 s1, -1, v4 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_cndmask_b32_e64 v2, 0, v2, s0 v_cndmask_b32_e64 v3, 0, s7, s0 v_cndmask_b32_e64 v4, 0, v4, s1 v_cndmask_b32_e64 v5, 0, s7, s1 v_cmp_ne_u32_e64 s1, -1, v10 flat_load_b32 v0, v[2:3] glc dlc s_waitcnt vmcnt(0) v_add_nc_u32_e32 v3, 0xa00, v6 flat_load_b32 v13, v[4:5] glc dlc s_waitcnt vmcnt(0) v_add_nc_u32_e32 v5, 0x1a00, v6 v_cndmask_b32_e64 v2, 0, s7, vcc_lo v_cndmask_b32_e64 v10, 0, v10, s1 v_cmp_ne_u32_e32 vcc_lo, -1, v3 v_cndmask_b32_e64 v11, 0, s7, s1 v_cndmask_b32_e32 v3, 0, v3, vcc_lo v_cndmask_b32_e64 v4, 0, s7, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v7 v_cndmask_b32_e32 v7, 0, v7, vcc_lo v_cndmask_b32_e64 v8, 0, s7, vcc_lo s_waitcnt lgkmcnt(1) v_fmac_f32_e32 v0, s2, v12 v_cmp_ne_u32_e64 s0, -1, v5 s_delay_alu instid0(VALU_DEP_1) v_cndmask_b32_e64 v5, 0, v5, s0 v_cndmask_b32_e64 v6, 0, s7, s0 flat_load_b32 v1, v[1:2] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v14, v[3:4] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v2, v[5:6] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v6, v[7:8] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v3, v[10:11] glc dlc s_waitcnt vmcnt(0) v_mov_b32_e32 v10, 0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[4:5], 4, v[9:10] v_add_co_u32 v4, vcc_lo, s4, v4 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v5, vcc_lo, s5, v5, vcc_lo s_waitcnt lgkmcnt(2) v_dual_fmac_f32 v1, s2, v13 :: v_dual_fmac_f32 v2, s2, v14 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v3, s2, v6 global_store_b128 v[4:5], v[0:3], off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .amdhsa_group_segment_fixed_size 8192 .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 57 .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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, .Lfunc_end0-_Z32saxpy_float4s_shmem_doublebufferPfS_fPl .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 - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer .group_segment_fixed_size: 8192 .kernarg_segment_align: 8 .kernarg_segment_size: 32 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z32saxpy_float4s_shmem_doublebufferPfS_fPl.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 57 .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 saxpy_float4s_shmem_doublebuffer ( float* y, float* x, float a, clock_t * timer_vals) { volatile __shared__ float sdata_x0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_0 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_x3_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y0_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y1_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y2_1 [COMPUTE_THREADS_PER_CTA]; volatile __shared__ float sdata_y3_1 [COMPUTE_THREADS_PER_CTA]; int tid = threadIdx.x ; unsigned int idx0, idx1; idx0 = blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; idx1 = COMPUTE_THREADS_PER_CTA * CTA_COUNT + blockIdx.x * COMPUTE_THREADS_PER_CTA + tid; float4 * x_as_float4 = (float4 *)x; float4 * y_as_float4 = (float4 *)y; float4 result_y; for (int i=0; i < NUM_ITERS/4; i+=2) { float4 tmp1_x, tmp1_y; __syncthreads(); tmp1_x = x_as_float4[idx0]; tmp1_y = y_as_float4[idx0]; if (i!=0) { result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; idx1 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; } sdata_x0_0[tid] = tmp1_x.x; sdata_x1_0[tid] = tmp1_x.y; sdata_x2_0[tid] = tmp1_x.z; sdata_x3_0[tid] = tmp1_x.w; sdata_y0_0[tid] = tmp1_y.x; sdata_y1_0[tid] = tmp1_y.y; sdata_y2_0[tid] = tmp1_y.z; sdata_y3_0[tid] = tmp1_y.w; __syncthreads(); tmp1_x = x_as_float4[idx1]; tmp1_y = y_as_float4[idx1]; result_y.x = a * sdata_x0_0[tid] + sdata_y0_0[tid]; result_y.y = a * sdata_x1_0[tid] + sdata_y1_0[tid]; result_y.z = a * sdata_x2_0[tid] + sdata_y2_0[tid]; result_y.w = a * sdata_x3_0[tid] + sdata_y3_0[tid]; y_as_float4[idx0] = result_y; idx0 += 2 * COMPUTE_THREADS_PER_CTA * CTA_COUNT ; sdata_x0_1[tid] = tmp1_x.x; sdata_x1_1[tid] = tmp1_x.y; sdata_x2_1[tid] = tmp1_x.z; sdata_x3_1[tid] = tmp1_x.w; sdata_y0_1[tid] = tmp1_y.x; sdata_y1_1[tid] = tmp1_y.y; sdata_y2_1[tid] = tmp1_y.z; sdata_y3_1[tid] = tmp1_y.w; } __syncthreads(); result_y.x = a * sdata_x0_1[tid] + sdata_y0_1[tid]; result_y.y = a * sdata_x1_1[tid] + sdata_y1_1[tid]; result_y.z = a * sdata_x2_1[tid] + sdata_y2_1[tid]; result_y.w = a * sdata_x3_1[tid] + sdata_y3_1[tid]; y_as_float4[idx1] = result_y; }
.text .file "saxpy_float4s_shmem_doublebuffer.hip" .globl _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl # -- Begin function _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .p2align 4, 0x90 .type _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl,@function _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl: # @_Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movss %xmm0, 4(%rsp) movq %rdx, 56(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 4(%rsp), %rax movq %rax, 96(%rsp) leaq 56(%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 $_Z32saxpy_float4s_shmem_doublebufferPfS_fPl, %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 _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl, .Lfunc_end0-_Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .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 $_Z32saxpy_float4s_shmem_doublebufferPfS_fPl, %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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl,@object # @_Z32saxpy_float4s_shmem_doublebufferPfS_fPl .section .rodata,"a",@progbits .globl _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .p2align 3, 0x0 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl: .quad _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .size _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z32saxpy_float4s_shmem_doublebufferPfS_fPl" .size .L__unnamed_1, 44 .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 _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .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 : _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .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 R17, SR_CTAID.X ; /* 0x0000000000117919 */ /* 0x000e220000002500 */ /*0020*/ HFMA2.MMA R3, -RZ, RZ, 0, 0 ; /* 0x00000000ff037435 */ /* 0x000fe200000001ff */ /*0030*/ MOV R2, 0xe00 ; /* 0x00000e0000027802 */ /* 0x000fe20000000f00 */ /*0040*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0050*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e240000002100 */ /*0060*/ LEA R17, R17, R0, 0x7 ; /* 0x0000000011117211 */ /* 0x001fc800078e38ff */ /*0070*/ IADD3 R16, R17, 0x700, RZ ; /* 0x0000070011107810 */ /* 0x000fe40007ffe0ff */ /*0080*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fe20000010000 */ /*0090*/ MOV R18, 0x10 ; /* 0x0000001000127802 */ /* 0x000fca0000000f00 */ /*00a0*/ IMAD.WIDE.U32 R4, R17, R18, c[0x0][0x168] ; /* 0x00005a0011047625 */ /* 0x001fc800078e0012 */ /*00b0*/ IMAD.WIDE.U32 R20, R17, R18, c[0x0][0x160] ; /* 0x0000580011147625 */ /* 0x000fe400078e0012 */ /*00c0*/ LDG.E.128 R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000ea8000c1e1d00 */ /*00d0*/ LDG.E.128 R8, [R20.64] ; /* 0x0000000414087981 */ /* 0x000ee2000c1e1d00 */ /*00e0*/ ISETP.NE.AND P0, PT, R2, 0xe00, PT ; /* 0x00000e000200780c */ /* 0x000fda0003f05270 */ /*00f0*/ @P0 LDS R12, [R0.X4+0x1000] ; /* 0x00100000000c0984 */ /* 0x000fe20000004800 */ /*0100*/ @P0 IMAD.WIDE.U32 R24, R16.reuse, R18, c[0x0][0x160] ; /* 0x0000580010180625 */ /* 0x040fe200078e0012 */ /*0110*/ @P0 IADD3 R16, R16, 0xe00, RZ ; /* 0x00000e0010100810 */ /* 0x000fe40007ffe0ff */ /*0120*/ @P0 LDS R13, [R0.X4+0x1800] ; /* 0x00180000000d0984 */ /* 0x000e280000004800 */ /*0130*/ @P0 LDS R14, [R0.X4+0x1200] ; /* 0x00120000000e0984 */ /* 0x000fe80000004800 */ /*0140*/ @P0 LDS R15, [R0.X4+0x1a00] ; /* 0x001a0000000f0984 */ /* 0x000e680000004800 */ /*0150*/ @P0 LDS R19, [R0.X4+0x1400] ; /* 0x0014000000130984 */ /* 0x000fe80000004800 */ /*0160*/ @P0 LDS R22, [R0.X4+0x1c00] ; /* 0x001c000000160984 */ /* 0x000f280000004800 */ /*0170*/ @P0 LDS R23, [R0.X4+0x1600] ; /* 0x0016000000170984 */ /* 0x000fe80000004800 */ /*0180*/ @P0 LDS R26, [R0.X4+0x1e00] ; /* 0x001e0000001a0984 */ /* 0x000f620000004800 */ /*0190*/ @P0 FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c0a23 */ /* 0x001fc4000000000d */ /*01a0*/ @P0 FFMA R13, R14, c[0x0][0x170], R15 ; /* 0x00005c000e0d0a23 */ /* 0x002fe4000000000f */ /*01b0*/ @P0 FFMA R14, R19, c[0x0][0x170], R22 ; /* 0x00005c00130e0a23 */ /* 0x010fe40000000016 */ /*01c0*/ @P0 FFMA R15, R23, c[0x0][0x170], R26 ; /* 0x00005c00170f0a23 */ /* 0x020fe4000000001a */ /*01d0*/ IMAD.WIDE.U32 R26, R16, R18, c[0x0][0x168] ; /* 0x00005a00101a7625 */ /* 0x000fc600078e0012 */ /*01e0*/ @P0 STG.E.128 [R24.64], R12 ; /* 0x0000000c18000986 */ /* 0x000fe2000c101d04 */ /*01f0*/ IMAD.WIDE.U32 R22, R16, R18, c[0x0][0x160] ; /* 0x0000580010167625 */ /* 0x000fc600078e0012 */ /*0200*/ STS [R0.X4], R4 ; /* 0x0000000400007388 */ /* 0x004fe80000004800 */ /*0210*/ STS [R0.X4+0x200], R5 ; /* 0x0002000500007388 */ /* 0x000fe80000004800 */ /*0220*/ STS [R0.X4+0x400], R6 ; /* 0x0004000600007388 */ /* 0x000fe80000004800 */ /*0230*/ STS [R0.X4+0x600], R7 ; /* 0x0006000700007388 */ /* 0x0001e80000004800 */ /*0240*/ STS [R0.X4+0x800], R8 ; /* 0x0008000800007388 */ /* 0x008fe80000004800 */ /*0250*/ STS [R0.X4+0xa00], R9 ; /* 0x000a000900007388 */ /* 0x000fe80000004800 */ /*0260*/ STS [R0.X4+0xc00], R10 ; /* 0x000c000a00007388 */ /* 0x000fe80000004800 */ /*0270*/ STS [R0.X4+0xe00], R11 ; /* 0x000e000b00007388 */ /* 0x0003e80000004800 */ /*0280*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0290*/ LDG.E.128 R4, [R26.64] ; /* 0x000000041a047981 */ /* 0x0010a8000c1e1d00 */ /*02a0*/ LDG.E.128 R8, [R22.64] ; /* 0x0000000416087981 */ /* 0x002ee8000c1e1d00 */ /*02b0*/ LDS R12, [R0.X4] ; /* 0x00000000000c7984 */ /* 0x000fe80000004800 */ /*02c0*/ LDS R13, [R0.X4+0x800] ; /* 0x00080000000d7984 */ /* 0x000e680000004800 */ /*02d0*/ LDS R14, [R0.X4+0x200] ; /* 0x00020000000e7984 */ /* 0x000fe80000004800 */ /*02e0*/ LDS R19, [R0.X4+0xa00] ; /* 0x000a000000137984 */ /* 0x000f280000004800 */ /*02f0*/ LDS R25, [R0.X4+0x400] ; /* 0x0004000000197984 */ /* 0x000fe80000004800 */ /*0300*/ LDS R28, [R0.X4+0xc00] ; /* 0x000c0000001c7984 */ /* 0x000f680000004800 */ /*0310*/ LDS R15, [R0.X4+0x600] ; /* 0x00060000000f7984 */ /* 0x000fe80000004800 */ /*0320*/ LDS R24, [R0.X4+0xe00] ; /* 0x000e000000187984 */ /* 0x000e220000004800 */ /*0330*/ FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c7a23 */ /* 0x002fc4000000000d */ /*0340*/ FFMA R13, R14, c[0x0][0x170], R19 ; /* 0x00005c000e0d7a23 */ /* 0x010fe40000000013 */ /*0350*/ FFMA R14, R25, c[0x0][0x170], R28 ; /* 0x00005c00190e7a23 */ /* 0x020fe2000000001c */ /*0360*/ IADD3 R25, R17, 0xe00, RZ ; /* 0x00000e0011197810 */ /* 0x000fca0007ffe0ff */ /*0370*/ IMAD.WIDE.U32 R26, R25, R18, c[0x0][0x168] ; /* 0x00005a00191a7625 */ /* 0x001fc800078e0012 */ /*0380*/ FFMA R15, R15, c[0x0][0x170], R24 ; /* 0x00005c000f0f7a23 */ /* 0x000fca0000000018 */ /*0390*/ STG.E.128 [R20.64], R12 ; /* 0x0000000c14007986 */ /* 0x000fe2000c101d04 */ /*03a0*/ IMAD.WIDE.U32 R24, R25, R18, c[0x0][0x160] ; /* 0x0000580019187625 */ /* 0x000fc600078e0012 */ /*03b0*/ STS [R0.X4+0x1000], R4 ; /* 0x0010000400007388 */ /* 0x004fe80000004800 */ /*03c0*/ STS [R0.X4+0x1200], R5 ; /* 0x0012000500007388 */ /* 0x000fe80000004800 */ /*03d0*/ STS [R0.X4+0x1400], R6 ; /* 0x0014000600007388 */ /* 0x000fe80000004800 */ /*03e0*/ STS [R0.X4+0x1600], R7 ; /* 0x0016000700007388 */ /* 0x0001e80000004800 */ /*03f0*/ STS [R0.X4+0x1800], R8 ; /* 0x0018000800007388 */ /* 0x008fe80000004800 */ /*0400*/ STS [R0.X4+0x1a00], R9 ; /* 0x001a000900007388 */ /* 0x000fe80000004800 */ /*0410*/ STS [R0.X4+0x1c00], R10 ; /* 0x001c000a00007388 */ /* 0x000fe80000004800 */ /*0420*/ STS [R0.X4+0x1e00], R11 ; /* 0x001e000b00007388 */ /* 0x0003e80000004800 */ /*0430*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0440*/ LDG.E.128 R4, [R26.64] ; /* 0x000000041a047981 */ /* 0x001ea8000c1e1d00 */ /*0450*/ LDG.E.128 R8, [R24.64] ; /* 0x0000000418087981 */ /* 0x002ee2000c1e1d00 */ /*0460*/ IADD3 R16, R16, 0xe00, RZ ; /* 0x00000e0010107810 */ /* 0x000fc60007ffe0ff */ /*0470*/ LDS R12, [R0.X4+0x1000] ; /* 0x00100000000c7984 */ /* 0x000fe80000004800 */ /*0480*/ LDS R13, [R0.X4+0x1800] ; /* 0x00180000000d7984 */ /* 0x000e280000004800 */ /*0490*/ LDS R14, [R0.X4+0x1200] ; /* 0x00120000000e7984 */ /* 0x000fe80000004800 */ /*04a0*/ LDS R15, [R0.X4+0x1a00] ; /* 0x001a0000000f7984 */ /* 0x000e680000004800 */ /*04b0*/ LDS R19, [R0.X4+0x1400] ; /* 0x0014000000137984 */ /* 0x000fe80000004800 */ /*04c0*/ LDS R20, [R0.X4+0x1c00] ; /* 0x001c000000147984 */ /* 0x000f280000004800 */ /*04d0*/ LDS R21, [R0.X4+0x1600] ; /* 0x0016000000157984 */ /* 0x000fe80000004800 */ /*04e0*/ LDS R28, [R0.X4+0x1e00] ; /* 0x001e0000001c7984 */ /* 0x000f620000004800 */ /*04f0*/ FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c7a23 */ /* 0x001fc4000000000d */ /*0500*/ FFMA R13, R14, c[0x0][0x170], R15 ; /* 0x00005c000e0d7a23 */ /* 0x002fe4000000000f */ /*0510*/ FFMA R14, R19, c[0x0][0x170], R20 ; /* 0x00005c00130e7a23 */ /* 0x010fe40000000014 */ /*0520*/ FFMA R15, R21, c[0x0][0x170], R28 ; /* 0x00005c00150f7a23 */ /* 0x020fe4000000001c */ /*0530*/ IMAD.WIDE.U32 R20, R16, R18, c[0x0][0x168] ; /* 0x00005a0010147625 */ /* 0x000fc600078e0012 */ /*0540*/ STG.E.128 [R22.64], R12 ; /* 0x0000000c16007986 */ /* 0x000fe2000c101d04 */ /*0550*/ IMAD.WIDE.U32 R18, R16, R18, c[0x0][0x160] ; /* 0x0000580010127625 */ /* 0x000fc600078e0012 */ /*0560*/ STS [R0.X4], R4 ; /* 0x0000000400007388 */ /* 0x004fe80000004800 */ /*0570*/ STS [R0.X4+0x200], R5 ; /* 0x0002000500007388 */ /* 0x000fe80000004800 */ /*0580*/ STS [R0.X4+0x400], R6 ; /* 0x0004000600007388 */ /* 0x000fe80000004800 */ /*0590*/ STS [R0.X4+0x600], R7 ; /* 0x0006000700007388 */ /* 0x0001e80000004800 */ /*05a0*/ STS [R0.X4+0x800], R8 ; /* 0x0008000800007388 */ /* 0x008fe80000004800 */ /*05b0*/ STS [R0.X4+0xa00], R9 ; /* 0x000a000900007388 */ /* 0x000fe80000004800 */ /*05c0*/ STS [R0.X4+0xc00], R10 ; /* 0x000c000a00007388 */ /* 0x000fe80000004800 */ /*05d0*/ STS [R0.X4+0xe00], R11 ; /* 0x000e000b00007388 */ /* 0x0003e80000004800 */ /*05e0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*05f0*/ LDG.E.128 R4, [R20.64] ; /* 0x0000000414047981 */ /* 0x001ea8000c1e1d00 */ /*0600*/ LDG.E.128 R8, [R18.64] ; /* 0x0000000412087981 */ /* 0x002ee2000c1e1d00 */ /*0610*/ IADD3 R3, R3, 0x4, RZ ; /* 0x0000000403037810 */ /* 0x000fc40007ffe0ff */ /*0620*/ IADD3 R2, R2, 0x1c00, RZ ; /* 0x00001c0002027810 */ /* 0x000fe20007ffe0ff */ /*0630*/ LDS R12, [R0.X4] ; /* 0x00000000000c7984 */ /* 0x000fe20000004800 */ /*0640*/ ISETP.GE.U32.AND P0, PT, R3, 0x200, PT ; /* 0x000002000300780c */ /* 0x000fe40003f06070 */ /*0650*/ IADD3 R17, R17, 0x1c00, RZ ; /* 0x00001c0011117810 */ /* 0x000fe20007ffe0ff */ /*0660*/ LDS R13, [R0.X4+0x800] ; /* 0x00080000000d7984 */ /* 0x000e280000004800 */ /*0670*/ LDS R14, [R0.X4+0x200] ; /* 0x00020000000e7984 */ /* 0x000fe80000004800 */ /*0680*/ LDS R15, [R0.X4+0xa00] ; /* 0x000a0000000f7984 */ /* 0x000e680000004800 */ /*0690*/ LDS R22, [R0.X4+0x400] ; /* 0x0004000000167984 */ /* 0x000fe80000004800 */ /*06a0*/ LDS R23, [R0.X4+0xc00] ; /* 0x000c000000177984 */ /* 0x000f280000004800 */ /*06b0*/ LDS R26, [R0.X4+0x600] ; /* 0x00060000001a7984 */ /* 0x000fe80000004800 */ /*06c0*/ LDS R27, [R0.X4+0xe00] ; /* 0x000e0000001b7984 */ /* 0x000f620000004800 */ /*06d0*/ FFMA R12, R12, c[0x0][0x170], R13 ; /* 0x00005c000c0c7a23 */ /* 0x001fc4000000000d */ /*06e0*/ FFMA R13, R14, c[0x0][0x170], R15 ; /* 0x00005c000e0d7a23 */ /* 0x002fe4000000000f */ /*06f0*/ FFMA R14, R22, c[0x0][0x170], R23 ; /* 0x00005c00160e7a23 */ /* 0x010fe40000000017 */ /*0700*/ FFMA R15, R26, c[0x0][0x170], R27 ; /* 0x00005c001a0f7a23 */ /* 0x020fca000000001b */ /*0710*/ STG.E.128 [R24.64], R12 ; /* 0x0000000c18007986 */ /* 0x0001e8000c101d04 */ /*0720*/ STS [R0.X4+0x1000], R4 ; /* 0x0010000400007388 */ /* 0x0041e80000004800 */ /*0730*/ STS [R0.X4+0x1200], R5 ; /* 0x0012000500007388 */ /* 0x0001e80000004800 */ /*0740*/ STS [R0.X4+0x1400], R6 ; /* 0x0014000600007388 */ /* 0x0001e80000004800 */ /*0750*/ STS [R0.X4+0x1600], R7 ; /* 0x0016000700007388 */ /* 0x0001e80000004800 */ /*0760*/ STS [R0.X4+0x1800], R8 ; /* 0x0018000800007388 */ /* 0x0081e80000004800 */ /*0770*/ STS [R0.X4+0x1a00], R9 ; /* 0x001a000900007388 */ /* 0x0001e80000004800 */ /*0780*/ STS [R0.X4+0x1c00], R10 ; /* 0x001c000a00007388 */ /* 0x0001e80000004800 */ /*0790*/ STS [R0.X4+0x1e00], R11 ; /* 0x001e000b00007388 */ /* 0x0001e20000004800 */ /*07a0*/ @!P0 BRA 0x80 ; /* 0xfffff8d000008947 */ /* 0x000fea000383ffff */ /*07b0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*07c0*/ LDS R4, [R0.X4+0x1000] ; /* 0x0010000000047984 */ /* 0x001fe80000004800 */ /*07d0*/ LDS R3, [R0.X4+0x1800] ; /* 0x0018000000037984 */ /* 0x000e280000004800 */ /*07e0*/ LDS R5, [R0.X4+0x1200] ; /* 0x0012000000057984 */ /* 0x000fe80000004800 */ /*07f0*/ LDS R2, [R0.X4+0x1a00] ; /* 0x001a000000027984 */ /* 0x000e680000004800 */ /*0800*/ LDS R6, [R0.X4+0x1400] ; /* 0x0014000000067984 */ /* 0x000fe80000004800 */ /*0810*/ LDS R7, [R0.X4+0x1c00] ; /* 0x001c000000077984 */ /* 0x000ea80000004800 */ /*0820*/ LDS R8, [R0.X4+0x1600] ; /* 0x0016000000087984 */ /* 0x000fe80000004800 */ /*0830*/ LDS R9, [R0.X4+0x1e00] ; /* 0x001e000000097984 */ /* 0x000ee20000004800 */ /*0840*/ FFMA R4, R4, c[0x0][0x170], R3 ; /* 0x00005c0004047a23 */ /* 0x001fc40000000003 */ /*0850*/ FFMA R5, R5, c[0x0][0x170], R2 ; /* 0x00005c0005057a23 */ /* 0x002fe40000000002 */ /*0860*/ FFMA R6, R6, c[0x0][0x170], R7 ; /* 0x00005c0006067a23 */ /* 0x004fe40000000007 */ /*0870*/ FFMA R7, R8, c[0x0][0x170], R9 ; /* 0x00005c0008077a23 */ /* 0x008fca0000000009 */ /*0880*/ STG.E.128 [R18.64], R4 ; /* 0x0000000412007986 */ /* 0x000fe2000c101d04 */ /*0890*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*08a0*/ BRA 0x8a0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*08b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*08f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0900*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0910*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0920*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0930*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0940*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0950*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0960*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0970*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .globl _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .p2align 8 .type _Z32saxpy_float4s_shmem_doublebufferPfS_fPl,@function _Z32saxpy_float4s_shmem_doublebufferPfS_fPl: v_dual_mov_b32 v12, 0 :: v_dual_lshlrev_b32 v1, 2, v0 s_mov_b64 s[2:3], src_shared_base s_clause 0x1 s_load_b128 s[4:7], s[0:1], 0x0 s_load_b32 s2, s[0:1], 0x10 v_lshl_add_u32 v11, s15, 7, v0 v_cmp_ne_u32_e32 vcc_lo, -1, v1 v_add_nc_u32_e32 v3, 0x1200, v1 v_add_nc_u32_e32 v5, 0x1600, v1 s_delay_alu instid0(VALU_DEP_4) v_add_nc_u32_e32 v9, 0x700, v11 v_cndmask_b32_e32 v13, 0, v1, vcc_lo v_cndmask_b32_e64 v14, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v3 v_dual_cndmask_b32 v17, 0, v3 :: v_dual_add_nc_u32 v2, 0x200, v1 v_cndmask_b32_e64 v18, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v5 v_add_nc_u32_e32 v3, 0x1a00, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(SKIP_2) | instid1(VALU_DEP_4) v_cmp_ne_u32_e64 s0, -1, v2 v_cndmask_b32_e32 v21, 0, v5, vcc_lo v_cndmask_b32_e64 v22, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v3 v_add_nc_u32_e32 v4, 0x600, v1 v_cndmask_b32_e64 v15, 0, v2, s0 v_cndmask_b32_e64 v16, 0, s3, s0 v_dual_cndmask_b32 v25, 0, v3 :: v_dual_add_nc_u32 v2, 0xa00, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(SKIP_2) | instid1(VALU_DEP_3) v_cmp_ne_u32_e64 s0, -1, v4 v_cndmask_b32_e64 v26, 0, s3, vcc_lo v_add_nc_u32_e32 v3, 0x400, v1 v_cndmask_b32_e64 v19, 0, v4, s0 v_cndmask_b32_e64 v20, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v2 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2) v_cndmask_b32_e64 v23, 0, v2, s0 v_add_nc_u32_e32 v2, 0x1e00, v1 v_cndmask_b32_e64 v24, 0, s3, s0 v_cmp_ne_u32_e32 vcc_lo, -1, v2 v_dual_cndmask_b32 v29, 0, v2 :: v_dual_add_nc_u32 v4, 0xe00, v1 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_3) v_cmp_ne_u32_e64 s0, -1, v4 v_add_nc_u32_e32 v2, 0x800, v1 v_cndmask_b32_e64 v30, 0, s3, vcc_lo v_cndmask_b32_e64 v27, 0, v4, s0 v_cndmask_b32_e64 v28, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v3 v_cmp_ne_u32_e32 vcc_lo, -1, v2 v_add_nc_u32_e32 v4, 0xc00, v1 s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_2) | instid1(VALU_DEP_4) v_cndmask_b32_e64 v31, 0, v3, s0 v_cndmask_b32_e32 v33, 0, v2, vcc_lo v_cndmask_b32_e64 v32, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v4 v_or_b32_e32 v2, 0x1000, v1 v_cndmask_b32_e64 v34, 0, s3, vcc_lo v_add_nc_u32_e32 v3, 0x1400, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4) v_cndmask_b32_e64 v35, 0, v4, s0 v_cmp_ne_u32_e32 vcc_lo, -1, v2 v_add_nc_u32_e32 v4, 0x1800, v1 v_add_nc_u32_e32 v1, 0x1c00, v1 v_cndmask_b32_e64 v36, 0, s3, s0 v_cmp_ne_u32_e64 s0, -1, v3 v_cndmask_b32_e32 v37, 0, v2, vcc_lo v_cndmask_b32_e64 v38, 0, s3, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v4 v_cmp_ne_u32_e64 s1, -1, v1 v_cndmask_b32_e64 v39, 0, v3, s0 v_cndmask_b32_e64 v40, 0, s3, s0 s_mov_b32 s0, -2 v_cndmask_b32_e32 v41, 0, v4, vcc_lo v_cndmask_b32_e64 v42, 0, s3, vcc_lo v_cndmask_b32_e64 v43, 0, v1, s1 v_cndmask_b32_e64 v44, 0, s3, s1 s_branch .LBB0_2 .LBB0_1: v_mov_b32_e32 v10, v12 s_waitcnt vmcnt(1) flat_store_b32 v[13:14], v5 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[31:32], v6 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[33:34], v7 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[35:36], v8 dlc s_waitcnt_vscnt null, 0x0 s_waitcnt vmcnt(0) flat_store_b32 v[37:38], v1 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[39:40], v2 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[41:42], v3 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[43:44], v4 dlc s_waitcnt_vscnt null, 0x0 s_waitcnt lgkmcnt(0) s_barrier v_lshlrev_b64 v[5:6], 4, v[9:10] buffer_gl0_inv s_add_i32 s0, s0, 2 v_add_nc_u32_e32 v11, 0xe00, v11 s_cmpk_gt_u32 s0, 0x1fd v_add_co_u32 v1, vcc_lo, s6, v5 v_add_co_ci_u32_e32 v2, vcc_lo, s7, v6, vcc_lo v_add_co_u32 v5, vcc_lo, s4, v5 v_add_co_ci_u32_e32 v6, vcc_lo, s5, v6, vcc_lo global_load_b128 v[1:4], v[1:2], off global_load_b128 v[5:8], v[5:6], off flat_load_b32 v10, v[13:14] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v47, v[37:38] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v51, v[31:32] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v48, v[39:40] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v52, v[33:34] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v49, v[41:42] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v53, v[35:36] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v50, v[43:44] glc dlc s_waitcnt vmcnt(0) lgkmcnt(4) v_dual_fmac_f32 v47, s2, v10 :: v_dual_fmac_f32 v48, s2, v51 s_waitcnt lgkmcnt(0) v_dual_fmac_f32 v49, s2, v52 :: v_dual_fmac_f32 v50, s2, v53 global_store_b128 v[45:46], v[47:50], off flat_store_b32 v[15:16], v1 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[19:20], v2 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[23:24], v3 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[27:28], v4 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[17:18], v5 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[21:22], v6 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[25:26], v7 dlc s_waitcnt_vscnt null, 0x0 flat_store_b32 v[29:30], v8 dlc s_waitcnt_vscnt null, 0x0 s_cbranch_scc1 .LBB0_4 .LBB0_2: v_lshlrev_b64 v[1:2], 4, v[11:12] s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv s_cmp_eq_u32 s0, -2 v_add_co_u32 v3, vcc_lo, s6, v1 v_add_co_ci_u32_e32 v4, vcc_lo, s7, v2, vcc_lo v_add_co_u32 v45, vcc_lo, s4, v1 v_add_co_ci_u32_e32 v46, vcc_lo, s5, v2, vcc_lo global_load_b128 v[5:8], v[3:4], off global_load_b128 v[1:4], v[45:46], off s_cbranch_scc1 .LBB0_1 flat_load_b32 v53, v[15:16] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v47, v[17:18] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v54, v[19:20] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v48, v[21:22] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v55, v[23:24] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v49, v[25:26] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v56, v[27:28] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v50, v[29:30] glc dlc s_waitcnt vmcnt(0) lgkmcnt(6) v_dual_mov_b32 v10, v12 :: v_dual_fmac_f32 v47, s2, v53 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_4) | instid1(VALU_DEP_3) v_lshlrev_b64 v[51:52], 4, v[9:10] s_waitcnt lgkmcnt(4) v_dual_fmac_f32 v48, s2, v54 :: v_dual_add_nc_u32 v9, 0xe00, v9 s_waitcnt lgkmcnt(0) v_dual_fmac_f32 v49, s2, v55 :: v_dual_fmac_f32 v50, s2, v56 v_add_co_u32 v51, vcc_lo, s4, v51 v_add_co_ci_u32_e32 v52, vcc_lo, s5, v52, vcc_lo global_store_b128 v[51:52], v[47:50], off s_branch .LBB0_1 .LBB0_4: v_lshlrev_b32_e32 v6, 2, v0 s_mov_b64 s[6:7], src_shared_base s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv v_add_nc_u32_e32 v0, 0x200, v6 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_2) | instid1(VALU_DEP_2) v_cmp_ne_u32_e32 vcc_lo, -1, v0 v_dual_cndmask_b32 v0, 0, v0 :: v_dual_add_nc_u32 v7, 0x1600, v6 v_cndmask_b32_e64 v1, 0, s7, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v7 flat_load_b32 v12, v[0:1] glc dlc s_waitcnt vmcnt(0) v_cndmask_b32_e32 v1, 0, v7, vcc_lo v_add_nc_u32_e32 v7, 0xe00, v6 v_add_nc_u32_e32 v2, 0x1200, v6 v_add_nc_u32_e32 v4, 0x600, v6 v_add_nc_u32_e32 v10, 0x1e00, v6 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3) v_cmp_ne_u32_e64 s0, -1, v2 v_cmp_ne_u32_e64 s1, -1, v4 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_3) v_cndmask_b32_e64 v2, 0, v2, s0 v_cndmask_b32_e64 v3, 0, s7, s0 v_cndmask_b32_e64 v4, 0, v4, s1 v_cndmask_b32_e64 v5, 0, s7, s1 v_cmp_ne_u32_e64 s1, -1, v10 flat_load_b32 v0, v[2:3] glc dlc s_waitcnt vmcnt(0) v_add_nc_u32_e32 v3, 0xa00, v6 flat_load_b32 v13, v[4:5] glc dlc s_waitcnt vmcnt(0) v_add_nc_u32_e32 v5, 0x1a00, v6 v_cndmask_b32_e64 v2, 0, s7, vcc_lo v_cndmask_b32_e64 v10, 0, v10, s1 v_cmp_ne_u32_e32 vcc_lo, -1, v3 v_cndmask_b32_e64 v11, 0, s7, s1 v_cndmask_b32_e32 v3, 0, v3, vcc_lo v_cndmask_b32_e64 v4, 0, s7, vcc_lo v_cmp_ne_u32_e32 vcc_lo, -1, v7 v_cndmask_b32_e32 v7, 0, v7, vcc_lo v_cndmask_b32_e64 v8, 0, s7, vcc_lo s_waitcnt lgkmcnt(1) v_fmac_f32_e32 v0, s2, v12 v_cmp_ne_u32_e64 s0, -1, v5 s_delay_alu instid0(VALU_DEP_1) v_cndmask_b32_e64 v5, 0, v5, s0 v_cndmask_b32_e64 v6, 0, s7, s0 flat_load_b32 v1, v[1:2] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v14, v[3:4] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v2, v[5:6] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v6, v[7:8] glc dlc s_waitcnt vmcnt(0) flat_load_b32 v3, v[10:11] glc dlc s_waitcnt vmcnt(0) v_mov_b32_e32 v10, 0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[4:5], 4, v[9:10] v_add_co_u32 v4, vcc_lo, s4, v4 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v5, vcc_lo, s5, v5, vcc_lo s_waitcnt lgkmcnt(2) v_dual_fmac_f32 v1, s2, v13 :: v_dual_fmac_f32 v2, s2, v14 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v3, s2, v6 global_store_b128 v[4:5], v[0:3], off s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .amdhsa_group_segment_fixed_size 8192 .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 57 .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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, .Lfunc_end0-_Z32saxpy_float4s_shmem_doublebufferPfS_fPl .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 - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer .group_segment_fixed_size: 8192 .kernarg_segment_align: 8 .kernarg_segment_size: 32 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z32saxpy_float4s_shmem_doublebufferPfS_fPl.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 57 .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_0012ed74_00000000-6_saxpy_float4s_shmem_doublebuffer.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 _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl .type _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl, @function _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movq %rsi, 16(%rsp) movss %xmm0, 12(%rsp) movq %rdx, (%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) 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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl(%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 _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl, .-_Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl .globl _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .type _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, @function _Z32saxpy_float4s_shmem_doublebufferPfS_fPl: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z57__device_stub__Z32saxpy_float4s_shmem_doublebufferPfS_fPlPfS_fPl addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, .-_Z32saxpy_float4s_shmem_doublebufferPfS_fPl .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z32saxpy_float4s_shmem_doublebufferPfS_fPl" .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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl(%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 "saxpy_float4s_shmem_doublebuffer.hip" .globl _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl # -- Begin function _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .p2align 4, 0x90 .type _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl,@function _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl: # @_Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movq %rsi, 64(%rsp) movss %xmm0, 4(%rsp) movq %rdx, 56(%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 64(%rsp), %rax movq %rax, 88(%rsp) leaq 4(%rsp), %rax movq %rax, 96(%rsp) leaq 56(%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 $_Z32saxpy_float4s_shmem_doublebufferPfS_fPl, %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 _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl, .Lfunc_end0-_Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .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 $_Z32saxpy_float4s_shmem_doublebufferPfS_fPl, %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 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl,@object # @_Z32saxpy_float4s_shmem_doublebufferPfS_fPl .section .rodata,"a",@progbits .globl _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .p2align 3, 0x0 _Z32saxpy_float4s_shmem_doublebufferPfS_fPl: .quad _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .size _Z32saxpy_float4s_shmem_doublebufferPfS_fPl, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z32saxpy_float4s_shmem_doublebufferPfS_fPl" .size .L__unnamed_1, 44 .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 _Z47__device_stub__saxpy_float4s_shmem_doublebufferPfS_fPl .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z32saxpy_float4s_shmem_doublebufferPfS_fPl .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" extern "C" { } __global__ void sgd_with_momentum(float* w, const float* dw, float learning_rate, float momentum, float* v, unsigned int len) { int tid = blockIdx.x*blockDim.x + threadIdx.x; if (tid < len) { v[tid] = momentum * v[tid] + dw[tid]; w[tid] -= learning_rate * v[tid]; } }
code for sm_80 Function : _Z17sgd_with_momentumPfPKfffS_j .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 */ /* 0x000e280000002500 */ /*0020*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R6, R6, c[0x0][0x0], R3 ; /* 0x0000000006067a24 */ /* 0x001fca00078e0203 */ /*0040*/ ISETP.GE.U32.AND P0, PT, R6, c[0x0][0x180], PT ; /* 0x0000600006007a0c */ /* 0x000fda0003f06070 */ /*0050*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0060*/ HFMA2.MMA R7, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff077435 */ /* 0x000fe200000001ff */ /*0070*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd20000000a00 */ /*0080*/ IMAD.WIDE R4, R6, R7, c[0x0][0x168] ; /* 0x00005a0006047625 */ /* 0x000fc800078e0207 */ /*0090*/ IMAD.WIDE R2, R6.reuse, R7.reuse, c[0x0][0x178] ; /* 0x00005e0006027625 */ /* 0x0c0fe400078e0207 */ /*00a0*/ LDG.E R5, [R4.64] ; /* 0x0000000404057981 */ /* 0x000ea8000c1e1900 */ /*00b0*/ LDG.E R0, [R2.64] ; /* 0x0000000402007981 */ /* 0x000ea2000c1e1900 */ /*00c0*/ IMAD.WIDE R6, R6, R7, c[0x0][0x160] ; /* 0x0000580006067625 */ /* 0x000fc800078e0207 */ /*00d0*/ FFMA R9, R0, c[0x0][0x174], R5 ; /* 0x00005d0000097a23 */ /* 0x004fca0000000005 */ /*00e0*/ STG.E [R2.64], R9 ; /* 0x0000000902007986 */ /* 0x000fe8000c101904 */ /*00f0*/ LDG.E R0, [R6.64] ; /* 0x0000000406007981 */ /* 0x000ea4000c1e1900 */ /*0100*/ FFMA R11, -R9, c[0x0][0x170], R0 ; /* 0x00005c00090b7a23 */ /* 0x004fca0000000100 */ /*0110*/ STG.E [R6.64], R11 ; /* 0x0000000b06007986 */ /* 0x000fe2000c101904 */ /*0120*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0130*/ BRA 0x130; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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" extern "C" { } __global__ void sgd_with_momentum(float* w, const float* dw, float learning_rate, float momentum, float* v, unsigned int len) { int tid = blockIdx.x*blockDim.x + threadIdx.x; if (tid < len) { v[tid] = momentum * v[tid] + dw[tid]; w[tid] -= learning_rate * v[tid]; } }
.file "tmpxft_001ae999_00000000-6_sgd_with_momentum.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__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j .type _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j, @function _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j: .LFB2051: .cfi_startproc endbr64 subq $184, %rsp .cfi_def_cfa_offset 192 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movss %xmm0, 28(%rsp) movss %xmm1, 24(%rsp) movq %rdx, 16(%rsp) movl %ecx, 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 28(%rsp), %rax movq %rax, 128(%rsp) leaq 24(%rsp), %rax movq %rax, 136(%rsp) leaq 16(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%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 _Z17sgd_with_momentumPfPKfffS_j(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 192 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j, .-_Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j .globl _Z17sgd_with_momentumPfPKfffS_j .type _Z17sgd_with_momentumPfPKfffS_j, @function _Z17sgd_with_momentumPfPKfffS_j: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z17sgd_with_momentumPfPKfffS_j, .-_Z17sgd_with_momentumPfPKfffS_j .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z17sgd_with_momentumPfPKfffS_j" .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 _Z17sgd_with_momentumPfPKfffS_j(%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" extern "C" { } __global__ void sgd_with_momentum(float* w, const float* dw, float learning_rate, float momentum, float* v, unsigned int len) { int tid = blockIdx.x*blockDim.x + threadIdx.x; if (tid < len) { v[tid] = momentum * v[tid] + dw[tid]; w[tid] -= learning_rate * v[tid]; } }
#include <hip/hip_runtime.h> #include "includes.h" extern "C" { } __global__ void sgd_with_momentum(float* w, const float* dw, float learning_rate, float momentum, float* v, unsigned int len) { int tid = blockIdx.x*blockDim.x + threadIdx.x; if (tid < len) { v[tid] = momentum * v[tid] + dw[tid]; w[tid] -= learning_rate * v[tid]; } }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include "includes.h" extern "C" { } __global__ void sgd_with_momentum(float* w, const float* dw, float learning_rate, float momentum, float* v, unsigned int len) { int tid = blockIdx.x*blockDim.x + threadIdx.x; if (tid < len) { v[tid] = momentum * v[tid] + dw[tid]; w[tid] -= learning_rate * v[tid]; } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z17sgd_with_momentumPfPKfffS_j .globl _Z17sgd_with_momentumPfPKfffS_j .p2align 8 .type _Z17sgd_with_momentumPfPKfffS_j,@function _Z17sgd_with_momentumPfPKfffS_j: s_clause 0x1 s_load_b32 s2, s[0:1], 0x34 s_load_b32 s3, s[0:1], 0x20 s_waitcnt lgkmcnt(0) s_and_b32 s2, s2, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s2, v[0:1] s_mov_b32 s2, exec_lo v_cmpx_gt_u32_e64 s3, v1 s_cbranch_execz .LBB0_2 s_load_b256 s[0:7], 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, s6, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s7, v1, vcc_lo v_add_co_u32 v4, vcc_lo, s2, v0 v_add_co_ci_u32_e32 v5, vcc_lo, s3, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s0, v0 global_load_b32 v6, v[2:3], off global_load_b32 v4, v[4:5], off v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo s_waitcnt vmcnt(0) v_fmac_f32_e32 v4, s5, v6 global_store_b32 v[2:3], v4, off global_load_b32 v2, v[0:1], off s_waitcnt vmcnt(0) v_fma_f32 v2, -v4, s4, 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 _Z17sgd_with_momentumPfPKfffS_j .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 296 .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 7 .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 _Z17sgd_with_momentumPfPKfffS_j, .Lfunc_end0-_Z17sgd_with_momentumPfPKfffS_j .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: 20 .size: 4 .value_kind: by_value - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer - .offset: 32 .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: _Z17sgd_with_momentumPfPKfffS_j .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z17sgd_with_momentumPfPKfffS_j.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 7 .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" extern "C" { } __global__ void sgd_with_momentum(float* w, const float* dw, float learning_rate, float momentum, float* v, unsigned int len) { int tid = blockIdx.x*blockDim.x + threadIdx.x; if (tid < len) { v[tid] = momentum * v[tid] + dw[tid]; w[tid] -= learning_rate * v[tid]; } }
.text .file "sgd_with_momentum.hip" .globl _Z32__device_stub__sgd_with_momentumPfPKfffS_j # -- Begin function _Z32__device_stub__sgd_with_momentumPfPKfffS_j .p2align 4, 0x90 .type _Z32__device_stub__sgd_with_momentumPfPKfffS_j,@function _Z32__device_stub__sgd_with_momentumPfPKfffS_j: # @_Z32__device_stub__sgd_with_momentumPfPKfffS_j .cfi_startproc # %bb.0: subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movss %xmm0, 20(%rsp) movss %xmm1, 16(%rsp) movq %rdx, 72(%rsp) movl %ecx, 12(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 20(%rsp), %rax movq %rax, 112(%rsp) leaq 16(%rsp), %rax movq %rax, 120(%rsp) leaq 72(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%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 $_Z17sgd_with_momentumPfPKfffS_j, %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 _Z32__device_stub__sgd_with_momentumPfPKfffS_j, .Lfunc_end0-_Z32__device_stub__sgd_with_momentumPfPKfffS_j .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 $_Z17sgd_with_momentumPfPKfffS_j, %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 _Z17sgd_with_momentumPfPKfffS_j,@object # @_Z17sgd_with_momentumPfPKfffS_j .section .rodata,"a",@progbits .globl _Z17sgd_with_momentumPfPKfffS_j .p2align 3, 0x0 _Z17sgd_with_momentumPfPKfffS_j: .quad _Z32__device_stub__sgd_with_momentumPfPKfffS_j .size _Z17sgd_with_momentumPfPKfffS_j, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z17sgd_with_momentumPfPKfffS_j" .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 _Z32__device_stub__sgd_with_momentumPfPKfffS_j .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z17sgd_with_momentumPfPKfffS_j .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 : _Z17sgd_with_momentumPfPKfffS_j .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 */ /* 0x000e280000002500 */ /*0020*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R6, R6, c[0x0][0x0], R3 ; /* 0x0000000006067a24 */ /* 0x001fca00078e0203 */ /*0040*/ ISETP.GE.U32.AND P0, PT, R6, c[0x0][0x180], PT ; /* 0x0000600006007a0c */ /* 0x000fda0003f06070 */ /*0050*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0060*/ HFMA2.MMA R7, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff077435 */ /* 0x000fe200000001ff */ /*0070*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd20000000a00 */ /*0080*/ IMAD.WIDE R4, R6, R7, c[0x0][0x168] ; /* 0x00005a0006047625 */ /* 0x000fc800078e0207 */ /*0090*/ IMAD.WIDE R2, R6.reuse, R7.reuse, c[0x0][0x178] ; /* 0x00005e0006027625 */ /* 0x0c0fe400078e0207 */ /*00a0*/ LDG.E R5, [R4.64] ; /* 0x0000000404057981 */ /* 0x000ea8000c1e1900 */ /*00b0*/ LDG.E R0, [R2.64] ; /* 0x0000000402007981 */ /* 0x000ea2000c1e1900 */ /*00c0*/ IMAD.WIDE R6, R6, R7, c[0x0][0x160] ; /* 0x0000580006067625 */ /* 0x000fc800078e0207 */ /*00d0*/ FFMA R9, R0, c[0x0][0x174], R5 ; /* 0x00005d0000097a23 */ /* 0x004fca0000000005 */ /*00e0*/ STG.E [R2.64], R9 ; /* 0x0000000902007986 */ /* 0x000fe8000c101904 */ /*00f0*/ LDG.E R0, [R6.64] ; /* 0x0000000406007981 */ /* 0x000ea4000c1e1900 */ /*0100*/ FFMA R11, -R9, c[0x0][0x170], R0 ; /* 0x00005c00090b7a23 */ /* 0x004fca0000000100 */ /*0110*/ STG.E [R6.64], R11 ; /* 0x0000000b06007986 */ /* 0x000fe2000c101904 */ /*0120*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0130*/ BRA 0x130; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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 _Z17sgd_with_momentumPfPKfffS_j .globl _Z17sgd_with_momentumPfPKfffS_j .p2align 8 .type _Z17sgd_with_momentumPfPKfffS_j,@function _Z17sgd_with_momentumPfPKfffS_j: s_clause 0x1 s_load_b32 s2, s[0:1], 0x34 s_load_b32 s3, s[0:1], 0x20 s_waitcnt lgkmcnt(0) s_and_b32 s2, s2, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s2, v[0:1] s_mov_b32 s2, exec_lo v_cmpx_gt_u32_e64 s3, v1 s_cbranch_execz .LBB0_2 s_load_b256 s[0:7], 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, s6, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s7, v1, vcc_lo v_add_co_u32 v4, vcc_lo, s2, v0 v_add_co_ci_u32_e32 v5, vcc_lo, s3, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s0, v0 global_load_b32 v6, v[2:3], off global_load_b32 v4, v[4:5], off v_add_co_ci_u32_e32 v1, vcc_lo, s1, v1, vcc_lo s_waitcnt vmcnt(0) v_fmac_f32_e32 v4, s5, v6 global_store_b32 v[2:3], v4, off global_load_b32 v2, v[0:1], off s_waitcnt vmcnt(0) v_fma_f32 v2, -v4, s4, 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 _Z17sgd_with_momentumPfPKfffS_j .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 296 .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 7 .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 _Z17sgd_with_momentumPfPKfffS_j, .Lfunc_end0-_Z17sgd_with_momentumPfPKfffS_j .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: 20 .size: 4 .value_kind: by_value - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer - .offset: 32 .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: _Z17sgd_with_momentumPfPKfffS_j .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z17sgd_with_momentumPfPKfffS_j.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 7 .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_001ae999_00000000-6_sgd_with_momentum.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__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j .type _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j, @function _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j: .LFB2051: .cfi_startproc endbr64 subq $184, %rsp .cfi_def_cfa_offset 192 movq %rdi, 40(%rsp) movq %rsi, 32(%rsp) movss %xmm0, 28(%rsp) movss %xmm1, 24(%rsp) movq %rdx, 16(%rsp) movl %ecx, 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 28(%rsp), %rax movq %rax, 128(%rsp) leaq 24(%rsp), %rax movq %rax, 136(%rsp) leaq 16(%rsp), %rax movq %rax, 144(%rsp) leaq 12(%rsp), %rax movq %rax, 152(%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 _Z17sgd_with_momentumPfPKfffS_j(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 192 jmp .L3 .L8: call __stack_chk_fail@PLT .cfi_endproc .LFE2051: .size _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j, .-_Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j .globl _Z17sgd_with_momentumPfPKfffS_j .type _Z17sgd_with_momentumPfPKfffS_j, @function _Z17sgd_with_momentumPfPKfffS_j: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z45__device_stub__Z17sgd_with_momentumPfPKfffS_jPfPKfffS_j addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z17sgd_with_momentumPfPKfffS_j, .-_Z17sgd_with_momentumPfPKfffS_j .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC0: .string "_Z17sgd_with_momentumPfPKfffS_j" .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 _Z17sgd_with_momentumPfPKfffS_j(%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 "sgd_with_momentum.hip" .globl _Z32__device_stub__sgd_with_momentumPfPKfffS_j # -- Begin function _Z32__device_stub__sgd_with_momentumPfPKfffS_j .p2align 4, 0x90 .type _Z32__device_stub__sgd_with_momentumPfPKfffS_j,@function _Z32__device_stub__sgd_with_momentumPfPKfffS_j: # @_Z32__device_stub__sgd_with_momentumPfPKfffS_j .cfi_startproc # %bb.0: subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 88(%rsp) movq %rsi, 80(%rsp) movss %xmm0, 20(%rsp) movss %xmm1, 16(%rsp) movq %rdx, 72(%rsp) movl %ecx, 12(%rsp) leaq 88(%rsp), %rax movq %rax, 96(%rsp) leaq 80(%rsp), %rax movq %rax, 104(%rsp) leaq 20(%rsp), %rax movq %rax, 112(%rsp) leaq 16(%rsp), %rax movq %rax, 120(%rsp) leaq 72(%rsp), %rax movq %rax, 128(%rsp) leaq 12(%rsp), %rax movq %rax, 136(%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 $_Z17sgd_with_momentumPfPKfffS_j, %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 _Z32__device_stub__sgd_with_momentumPfPKfffS_j, .Lfunc_end0-_Z32__device_stub__sgd_with_momentumPfPKfffS_j .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 $_Z17sgd_with_momentumPfPKfffS_j, %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 _Z17sgd_with_momentumPfPKfffS_j,@object # @_Z17sgd_with_momentumPfPKfffS_j .section .rodata,"a",@progbits .globl _Z17sgd_with_momentumPfPKfffS_j .p2align 3, 0x0 _Z17sgd_with_momentumPfPKfffS_j: .quad _Z32__device_stub__sgd_with_momentumPfPKfffS_j .size _Z17sgd_with_momentumPfPKfffS_j, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z17sgd_with_momentumPfPKfffS_j" .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 _Z32__device_stub__sgd_with_momentumPfPKfffS_j .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z17sgd_with_momentumPfPKfffS_j .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.
// Copyright (c) 2018 John Biddiscombe // // Distributed under the Boost Software License, Version 1.0. (See accompanying // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) #include "cuda_runtime.h" __global__ void saxpy(int n, float a, float *x, float *y) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i < n) y[i] = a * x[i] + y[i]; }
code for sm_80 Function : _Z5saxpyifPfS_ .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 R4, SR_CTAID.X ; /* 0x0000000000047919 */ /* 0x000e280000002500 */ /*0020*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R4, R4, c[0x0][0x0], R3 ; /* 0x0000000004047a24 */ /* 0x001fca00078e0203 */ /*0040*/ ISETP.GE.AND P0, PT, R4, c[0x0][0x160], PT ; /* 0x0000580004007a0c */ /* 0x000fda0003f06270 */ /*0050*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0060*/ HFMA2.MMA R5, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff057435 */ /* 0x000fe200000001ff */ /*0070*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd20000000a00 */ /*0080*/ IMAD.WIDE R2, R4, R5, c[0x0][0x168] ; /* 0x00005a0004027625 */ /* 0x000fc800078e0205 */ /*0090*/ IMAD.WIDE R4, R4, R5, c[0x0][0x170] ; /* 0x00005c0004047625 */ /* 0x000fe400078e0205 */ /*00a0*/ LDG.E R2, [R2.64] ; /* 0x0000000402027981 */ /* 0x000ea8000c1e1900 */ /*00b0*/ LDG.E R7, [R4.64] ; /* 0x0000000404077981 */ /* 0x000ea4000c1e1900 */ /*00c0*/ FFMA R7, R2, c[0x0][0x164], R7 ; /* 0x0000590002077a23 */ /* 0x004fca0000000007 */ /*00d0*/ STG.E [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x000fe2000c101904 */ /*00e0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*00f0*/ BRA 0xf0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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.
// Copyright (c) 2018 John Biddiscombe // // Distributed under the Boost Software License, Version 1.0. (See accompanying // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) #include "cuda_runtime.h" __global__ void saxpy(int n, float a, float *x, float *y) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i < n) y[i] = a * x[i] + y[i]; }
.file "tmpxft_00088ded_00000000-6_saxpy.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 _Z28__device_stub__Z5saxpyifPfS_ifPfS_ .type _Z28__device_stub__Z5saxpyifPfS_ifPfS_, @function _Z28__device_stub__Z5saxpyifPfS_ifPfS_: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movl %edi, 28(%rsp) movss %xmm0, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 28(%rsp), %rax movq %rax, 96(%rsp) leaq 24(%rsp), %rax movq %rax, 104(%rsp) leaq 16(%rsp), %rax movq %rax, 112(%rsp) leaq 8(%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 _Z5saxpyifPfS_(%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 _Z28__device_stub__Z5saxpyifPfS_ifPfS_, .-_Z28__device_stub__Z5saxpyifPfS_ifPfS_ .globl _Z5saxpyifPfS_ .type _Z5saxpyifPfS_, @function _Z5saxpyifPfS_: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z28__device_stub__Z5saxpyifPfS_ifPfS_ addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z5saxpyifPfS_, .-_Z5saxpyifPfS_ .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z5saxpyifPfS_" .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 _Z5saxpyifPfS_(%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.
// Copyright (c) 2018 John Biddiscombe // // Distributed under the Boost Software License, Version 1.0. (See accompanying // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) #include "cuda_runtime.h" __global__ void saxpy(int n, float a, float *x, float *y) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i < n) y[i] = a * x[i] + y[i]; }
// Copyright (c) 2018 John Biddiscombe // // Distributed under the Boost Software License, Version 1.0. (See accompanying // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) #include "hip/hip_runtime.h" __global__ void saxpy(int n, float a, float *x, float *y) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i < n) y[i] = a * x[i] + y[i]; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
// Copyright (c) 2018 John Biddiscombe // // Distributed under the Boost Software License, Version 1.0. (See accompanying // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) #include "hip/hip_runtime.h" __global__ void saxpy(int n, float a, float *x, float *y) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i < n) y[i] = a * x[i] + y[i]; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z5saxpyifPfS_ .globl _Z5saxpyifPfS_ .p2align 8 .type _Z5saxpyifPfS_,@function _Z5saxpyifPfS_: s_clause 0x1 s_load_b32 s2, s[0:1], 0x24 s_load_b32 s3, s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_and_b32 s2, s2, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s2, v[0:1] s_mov_b32 s2, exec_lo v_cmpx_gt_i32_e64 s3, v1 s_cbranch_execz .LBB0_2 s_load_b128 s[4:7], s[0:1], 0x8 v_ashrrev_i32_e32 v2, 31, v1 s_load_b32 s0, s[0:1], 0x4 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, s4, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s5, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s6, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s7, v1, vcc_lo global_load_b32 v2, v[2:3], off global_load_b32 v3, v[0:1], off s_waitcnt vmcnt(0) v_fmac_f32_e32 v3, s0, v2 global_store_b32 v[0:1], v3, off .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z5saxpyifPfS_ .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 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 _Z5saxpyifPfS_, .Lfunc_end0-_Z5saxpyifPfS_ .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 - .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: _Z5saxpyifPfS_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z5saxpyifPfS_.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.
// Copyright (c) 2018 John Biddiscombe // // Distributed under the Boost Software License, Version 1.0. (See accompanying // file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt) #include "hip/hip_runtime.h" __global__ void saxpy(int n, float a, float *x, float *y) { int i = blockIdx.x * blockDim.x + threadIdx.x; if (i < n) y[i] = a * x[i] + y[i]; }
.text .file "saxpy.hip" .globl _Z20__device_stub__saxpyifPfS_ # -- Begin function _Z20__device_stub__saxpyifPfS_ .p2align 4, 0x90 .type _Z20__device_stub__saxpyifPfS_,@function _Z20__device_stub__saxpyifPfS_: # @_Z20__device_stub__saxpyifPfS_ .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movl %edi, 12(%rsp) movss %xmm0, 8(%rsp) movq %rsi, 72(%rsp) movq %rdx, 64(%rsp) leaq 12(%rsp), %rax movq %rax, 80(%rsp) leaq 8(%rsp), %rax movq %rax, 88(%rsp) leaq 72(%rsp), %rax movq %rax, 96(%rsp) leaq 64(%rsp), %rax movq %rax, 104(%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 $_Z5saxpyifPfS_, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z20__device_stub__saxpyifPfS_, .Lfunc_end0-_Z20__device_stub__saxpyifPfS_ .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 $_Z5saxpyifPfS_, %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 _Z5saxpyifPfS_,@object # @_Z5saxpyifPfS_ .section .rodata,"a",@progbits .globl _Z5saxpyifPfS_ .p2align 3, 0x0 _Z5saxpyifPfS_: .quad _Z20__device_stub__saxpyifPfS_ .size _Z5saxpyifPfS_, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z5saxpyifPfS_" .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 _Z20__device_stub__saxpyifPfS_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z5saxpyifPfS_ .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 : _Z5saxpyifPfS_ .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 R4, SR_CTAID.X ; /* 0x0000000000047919 */ /* 0x000e280000002500 */ /*0020*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e240000002100 */ /*0030*/ IMAD R4, R4, c[0x0][0x0], R3 ; /* 0x0000000004047a24 */ /* 0x001fca00078e0203 */ /*0040*/ ISETP.GE.AND P0, PT, R4, c[0x0][0x160], PT ; /* 0x0000580004007a0c */ /* 0x000fda0003f06270 */ /*0050*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*0060*/ HFMA2.MMA R5, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff057435 */ /* 0x000fe200000001ff */ /*0070*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd20000000a00 */ /*0080*/ IMAD.WIDE R2, R4, R5, c[0x0][0x168] ; /* 0x00005a0004027625 */ /* 0x000fc800078e0205 */ /*0090*/ IMAD.WIDE R4, R4, R5, c[0x0][0x170] ; /* 0x00005c0004047625 */ /* 0x000fe400078e0205 */ /*00a0*/ LDG.E R2, [R2.64] ; /* 0x0000000402027981 */ /* 0x000ea8000c1e1900 */ /*00b0*/ LDG.E R7, [R4.64] ; /* 0x0000000404077981 */ /* 0x000ea4000c1e1900 */ /*00c0*/ FFMA R7, R2, c[0x0][0x164], R7 ; /* 0x0000590002077a23 */ /* 0x004fca0000000007 */ /*00d0*/ STG.E [R4.64], R7 ; /* 0x0000000704007986 */ /* 0x000fe2000c101904 */ /*00e0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*00f0*/ BRA 0xf0; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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 _Z5saxpyifPfS_ .globl _Z5saxpyifPfS_ .p2align 8 .type _Z5saxpyifPfS_,@function _Z5saxpyifPfS_: s_clause 0x1 s_load_b32 s2, s[0:1], 0x24 s_load_b32 s3, s[0:1], 0x0 s_waitcnt lgkmcnt(0) s_and_b32 s2, s2, 0xffff s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_mad_u64_u32 v[1:2], null, s15, s2, v[0:1] s_mov_b32 s2, exec_lo v_cmpx_gt_i32_e64 s3, v1 s_cbranch_execz .LBB0_2 s_load_b128 s[4:7], s[0:1], 0x8 v_ashrrev_i32_e32 v2, 31, v1 s_load_b32 s0, s[0:1], 0x4 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, s4, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s5, v1, vcc_lo v_add_co_u32 v0, vcc_lo, s6, v0 v_add_co_ci_u32_e32 v1, vcc_lo, s7, v1, vcc_lo global_load_b32 v2, v[2:3], off global_load_b32 v3, v[0:1], off s_waitcnt vmcnt(0) v_fmac_f32_e32 v3, s0, v2 global_store_b32 v[0:1], v3, off .LBB0_2: s_nop 0 s_sendmsg sendmsg(MSG_DEALLOC_VGPRS) s_endpgm .section .rodata,"a",@progbits .p2align 6, 0x0 .amdhsa_kernel _Z5saxpyifPfS_ .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 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 _Z5saxpyifPfS_, .Lfunc_end0-_Z5saxpyifPfS_ .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 - .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: _Z5saxpyifPfS_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z5saxpyifPfS_.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_00088ded_00000000-6_saxpy.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 _Z28__device_stub__Z5saxpyifPfS_ifPfS_ .type _Z28__device_stub__Z5saxpyifPfS_ifPfS_, @function _Z28__device_stub__Z5saxpyifPfS_ifPfS_: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movl %edi, 28(%rsp) movss %xmm0, 24(%rsp) movq %rsi, 16(%rsp) movq %rdx, 8(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 28(%rsp), %rax movq %rax, 96(%rsp) leaq 24(%rsp), %rax movq %rax, 104(%rsp) leaq 16(%rsp), %rax movq %rax, 112(%rsp) leaq 8(%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 _Z5saxpyifPfS_(%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 _Z28__device_stub__Z5saxpyifPfS_ifPfS_, .-_Z28__device_stub__Z5saxpyifPfS_ifPfS_ .globl _Z5saxpyifPfS_ .type _Z5saxpyifPfS_, @function _Z5saxpyifPfS_: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z28__device_stub__Z5saxpyifPfS_ifPfS_ addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z5saxpyifPfS_, .-_Z5saxpyifPfS_ .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z5saxpyifPfS_" .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 _Z5saxpyifPfS_(%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 "saxpy.hip" .globl _Z20__device_stub__saxpyifPfS_ # -- Begin function _Z20__device_stub__saxpyifPfS_ .p2align 4, 0x90 .type _Z20__device_stub__saxpyifPfS_,@function _Z20__device_stub__saxpyifPfS_: # @_Z20__device_stub__saxpyifPfS_ .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movl %edi, 12(%rsp) movss %xmm0, 8(%rsp) movq %rsi, 72(%rsp) movq %rdx, 64(%rsp) leaq 12(%rsp), %rax movq %rax, 80(%rsp) leaq 8(%rsp), %rax movq %rax, 88(%rsp) leaq 72(%rsp), %rax movq %rax, 96(%rsp) leaq 64(%rsp), %rax movq %rax, 104(%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 $_Z5saxpyifPfS_, %edi pushq 16(%rsp) .cfi_adjust_cfa_offset 8 pushq 32(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $136, %rsp .cfi_adjust_cfa_offset -136 retq .Lfunc_end0: .size _Z20__device_stub__saxpyifPfS_, .Lfunc_end0-_Z20__device_stub__saxpyifPfS_ .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 $_Z5saxpyifPfS_, %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 _Z5saxpyifPfS_,@object # @_Z5saxpyifPfS_ .section .rodata,"a",@progbits .globl _Z5saxpyifPfS_ .p2align 3, 0x0 _Z5saxpyifPfS_: .quad _Z20__device_stub__saxpyifPfS_ .size _Z5saxpyifPfS_, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z5saxpyifPfS_" .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 _Z20__device_stub__saxpyifPfS_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z5saxpyifPfS_ .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 sumMatrixOnGPU(float *MatA, float *MatB, float *MatC, int nx, int ny) { unsigned int ix = threadIdx.x + blockIdx.x * blockDim.x; unsigned int iy = threadIdx.y + blockIdx.y * blockDim.y; unsigned int idx = iy * nx + ix; //printf("nx: %d, ny: %d, ix: %d, iy: %d, idx: %d\n", nx, ny, ix, iy, idx); if (ix<nx && iy<ny) { MatC[idx] = MatA[idx] + MatB[idx]; //printf("GPU Add: %f + %f = %f.\n", MatA[idx], MatB[idx], MatC[idx]); } }
code for sm_80 Function : _Z14sumMatrixOnGPUPfS_S_ii .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 R3, SR_CTAID.Y ; /* 0x0000000000037919 */ /* 0x000e280000002600 */ /*0020*/ S2R R2, SR_TID.Y ; /* 0x0000000000027919 */ /* 0x000e280000002200 */ /*0030*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e680000002100 */ /*0040*/ S2R R5, SR_CTAID.X ; /* 0x0000000000057919 */ /* 0x000e620000002500 */ /*0050*/ IMAD R3, R3, c[0x0][0x4], R2 ; /* 0x0000010003037a24 */ /* 0x001fca00078e0202 */ /*0060*/ ISETP.GE.U32.AND P0, PT, R3, c[0x0][0x17c], PT ; /* 0x00005f0003007a0c */ /* 0x000fe20003f06070 */ /*0070*/ IMAD R0, R5, c[0x0][0x0], R0 ; /* 0x0000000005007a24 */ /* 0x002fca00078e0200 */ /*0080*/ ISETP.GE.U32.OR P0, PT, R0, c[0x0][0x178], P0 ; /* 0x00005e0000007a0c */ /* 0x000fda0000706470 */ /*0090*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*00a0*/ HFMA2.MMA R7, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff077435 */ /* 0x000fe200000001ff */ /*00b0*/ IMAD R0, R3, c[0x0][0x178], R0 ; /* 0x00005e0003007a24 */ /* 0x000fe200078e0200 */ /*00c0*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd00000000a00 */ /*00d0*/ IMAD.WIDE.U32 R4, R0, R7, c[0x0][0x168] ; /* 0x00005a0000047625 */ /* 0x000fc800078e0007 */ /*00e0*/ IMAD.WIDE.U32 R2, R0.reuse, R7.reuse, c[0x0][0x160] ; /* 0x0000580000027625 */ /* 0x0c0fe400078e0007 */ /*00f0*/ LDG.E R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000ea8000c1e1900 */ /*0100*/ LDG.E R3, [R2.64] ; /* 0x0000000402037981 */ /* 0x000ea2000c1e1900 */ /*0110*/ IMAD.WIDE.U32 R6, R0, R7, c[0x0][0x170] ; /* 0x00005c0000067625 */ /* 0x000fc800078e0007 */ /*0120*/ FADD R9, R4, R3 ; /* 0x0000000304097221 */ /* 0x004fca0000000000 */ /*0130*/ STG.E [R6.64], R9 ; /* 0x0000000906007986 */ /* 0x000fe2000c101904 */ /*0140*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0150*/ BRA 0x150; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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 sumMatrixOnGPU(float *MatA, float *MatB, float *MatC, int nx, int ny) { unsigned int ix = threadIdx.x + blockIdx.x * blockDim.x; unsigned int iy = threadIdx.y + blockIdx.y * blockDim.y; unsigned int idx = iy * nx + ix; //printf("nx: %d, ny: %d, ix: %d, iy: %d, idx: %d\n", nx, ny, ix, iy, idx); if (ix<nx && iy<ny) { MatC[idx] = MatA[idx] + MatB[idx]; //printf("GPU Add: %f + %f = %f.\n", MatA[idx], MatB[idx], MatC[idx]); } }
.file "tmpxft_001b0c1d_00000000-6_sumMatrixOnGPU.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__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii .type _Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii, @function _Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii: .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) movl %r8d, (%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) movq %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 _Z14sumMatrixOnGPUPfS_S_ii(%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__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii, .-_Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii .globl _Z14sumMatrixOnGPUPfS_S_ii .type _Z14sumMatrixOnGPUPfS_S_ii, @function _Z14sumMatrixOnGPUPfS_S_ii: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z14sumMatrixOnGPUPfS_S_ii, .-_Z14sumMatrixOnGPUPfS_S_ii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z14sumMatrixOnGPUPfS_S_ii" .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 _Z14sumMatrixOnGPUPfS_S_ii(%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 sumMatrixOnGPU(float *MatA, float *MatB, float *MatC, int nx, int ny) { unsigned int ix = threadIdx.x + blockIdx.x * blockDim.x; unsigned int iy = threadIdx.y + blockIdx.y * blockDim.y; unsigned int idx = iy * nx + ix; //printf("nx: %d, ny: %d, ix: %d, iy: %d, idx: %d\n", nx, ny, ix, iy, idx); if (ix<nx && iy<ny) { MatC[idx] = MatA[idx] + MatB[idx]; //printf("GPU Add: %f + %f = %f.\n", MatA[idx], MatB[idx], MatC[idx]); } }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void sumMatrixOnGPU(float *MatA, float *MatB, float *MatC, int nx, int ny) { unsigned int ix = threadIdx.x + blockIdx.x * blockDim.x; unsigned int iy = threadIdx.y + blockIdx.y * blockDim.y; unsigned int idx = iy * nx + ix; //printf("nx: %d, ny: %d, ix: %d, iy: %d, idx: %d\n", nx, ny, ix, iy, idx); if (ix<nx && iy<ny) { MatC[idx] = MatA[idx] + MatB[idx]; //printf("GPU Add: %f + %f = %f.\n", MatA[idx], MatB[idx], MatC[idx]); } }
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 sumMatrixOnGPU(float *MatA, float *MatB, float *MatC, int nx, int ny) { unsigned int ix = threadIdx.x + blockIdx.x * blockDim.x; unsigned int iy = threadIdx.y + blockIdx.y * blockDim.y; unsigned int idx = iy * nx + ix; //printf("nx: %d, ny: %d, ix: %d, iy: %d, idx: %d\n", nx, ny, ix, iy, idx); if (ix<nx && iy<ny) { MatC[idx] = MatA[idx] + MatB[idx]; //printf("GPU Add: %f + %f = %f.\n", MatA[idx], MatB[idx], MatC[idx]); } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z14sumMatrixOnGPUPfS_S_ii .globl _Z14sumMatrixOnGPUPfS_S_ii .p2align 8 .type _Z14sumMatrixOnGPUPfS_S_ii,@function _Z14sumMatrixOnGPUPfS_S_ii: s_clause 0x1 s_load_b32 s2, s[0:1], 0x2c s_load_b64 s[4:5], s[0:1], 0x18 v_and_b32_e32 v2, 0x3ff, v0 v_bfe_u32 v3, v0, 10, 10 s_waitcnt lgkmcnt(0) s_and_b32 s3, s2, 0xffff s_lshr_b32 s2, s2, 16 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_mad_u64_u32 v[0:1], null, s14, s3, v[2:3] v_mad_u64_u32 v[1:2], null, s15, s2, v[3:4] v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_u32_e64 s2, s5, v1 s_and_b32 s2, vcc_lo, s2 s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s3, s2 s_cbranch_execz .LBB0_2 s_load_b128 s[8:11], s[0:1], 0x0 v_mad_u64_u32 v[2:3], null, v1, s4, v[0:1] v_mov_b32_e32 v3, 0 s_load_b64 s[0:1], s[0:1], 0x10 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[2:3] s_waitcnt lgkmcnt(0) v_add_co_u32 v2, vcc_lo, s8, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s9, v1, vcc_lo v_add_co_u32 v4, vcc_lo, s10, v0 v_add_co_ci_u32_e32 v5, vcc_lo, s11, 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_f32_e32 v2, v2, v3 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 _Z14sumMatrixOnGPUPfS_S_ii .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 288 .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 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 _Z14sumMatrixOnGPUPfS_S_ii, .Lfunc_end0-_Z14sumMatrixOnGPUPfS_S_ii .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: 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: _Z14sumMatrixOnGPUPfS_S_ii .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z14sumMatrixOnGPUPfS_S_ii.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.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void sumMatrixOnGPU(float *MatA, float *MatB, float *MatC, int nx, int ny) { unsigned int ix = threadIdx.x + blockIdx.x * blockDim.x; unsigned int iy = threadIdx.y + blockIdx.y * blockDim.y; unsigned int idx = iy * nx + ix; //printf("nx: %d, ny: %d, ix: %d, iy: %d, idx: %d\n", nx, ny, ix, iy, idx); if (ix<nx && iy<ny) { MatC[idx] = MatA[idx] + MatB[idx]; //printf("GPU Add: %f + %f = %f.\n", MatA[idx], MatB[idx], MatC[idx]); } }
.text .file "sumMatrixOnGPU.hip" .globl _Z29__device_stub__sumMatrixOnGPUPfS_S_ii # -- Begin function _Z29__device_stub__sumMatrixOnGPUPfS_S_ii .p2align 4, 0x90 .type _Z29__device_stub__sumMatrixOnGPUPfS_S_ii,@function _Z29__device_stub__sumMatrixOnGPUPfS_S_ii: # @_Z29__device_stub__sumMatrixOnGPUPfS_S_ii .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) movl %r8d, (%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) movq %rsp, %rax movq %rax, 112(%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 $_Z14sumMatrixOnGPUPfS_S_ii, %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 _Z29__device_stub__sumMatrixOnGPUPfS_S_ii, .Lfunc_end0-_Z29__device_stub__sumMatrixOnGPUPfS_S_ii .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 $_Z14sumMatrixOnGPUPfS_S_ii, %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 _Z14sumMatrixOnGPUPfS_S_ii,@object # @_Z14sumMatrixOnGPUPfS_S_ii .section .rodata,"a",@progbits .globl _Z14sumMatrixOnGPUPfS_S_ii .p2align 3, 0x0 _Z14sumMatrixOnGPUPfS_S_ii: .quad _Z29__device_stub__sumMatrixOnGPUPfS_S_ii .size _Z14sumMatrixOnGPUPfS_S_ii, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z14sumMatrixOnGPUPfS_S_ii" .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 _Z29__device_stub__sumMatrixOnGPUPfS_S_ii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z14sumMatrixOnGPUPfS_S_ii .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 : _Z14sumMatrixOnGPUPfS_S_ii .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 R3, SR_CTAID.Y ; /* 0x0000000000037919 */ /* 0x000e280000002600 */ /*0020*/ S2R R2, SR_TID.Y ; /* 0x0000000000027919 */ /* 0x000e280000002200 */ /*0030*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e680000002100 */ /*0040*/ S2R R5, SR_CTAID.X ; /* 0x0000000000057919 */ /* 0x000e620000002500 */ /*0050*/ IMAD R3, R3, c[0x0][0x4], R2 ; /* 0x0000010003037a24 */ /* 0x001fca00078e0202 */ /*0060*/ ISETP.GE.U32.AND P0, PT, R3, c[0x0][0x17c], PT ; /* 0x00005f0003007a0c */ /* 0x000fe20003f06070 */ /*0070*/ IMAD R0, R5, c[0x0][0x0], R0 ; /* 0x0000000005007a24 */ /* 0x002fca00078e0200 */ /*0080*/ ISETP.GE.U32.OR P0, PT, R0, c[0x0][0x178], P0 ; /* 0x00005e0000007a0c */ /* 0x000fda0000706470 */ /*0090*/ @P0 EXIT ; /* 0x000000000000094d */ /* 0x000fea0003800000 */ /*00a0*/ HFMA2.MMA R7, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff077435 */ /* 0x000fe200000001ff */ /*00b0*/ IMAD R0, R3, c[0x0][0x178], R0 ; /* 0x00005e0003007a24 */ /* 0x000fe200078e0200 */ /*00c0*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fd00000000a00 */ /*00d0*/ IMAD.WIDE.U32 R4, R0, R7, c[0x0][0x168] ; /* 0x00005a0000047625 */ /* 0x000fc800078e0007 */ /*00e0*/ IMAD.WIDE.U32 R2, R0.reuse, R7.reuse, c[0x0][0x160] ; /* 0x0000580000027625 */ /* 0x0c0fe400078e0007 */ /*00f0*/ LDG.E R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000ea8000c1e1900 */ /*0100*/ LDG.E R3, [R2.64] ; /* 0x0000000402037981 */ /* 0x000ea2000c1e1900 */ /*0110*/ IMAD.WIDE.U32 R6, R0, R7, c[0x0][0x170] ; /* 0x00005c0000067625 */ /* 0x000fc800078e0007 */ /*0120*/ FADD R9, R4, R3 ; /* 0x0000000304097221 */ /* 0x004fca0000000000 */ /*0130*/ STG.E [R6.64], R9 ; /* 0x0000000906007986 */ /* 0x000fe2000c101904 */ /*0140*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0150*/ BRA 0x150; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*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 _Z14sumMatrixOnGPUPfS_S_ii .globl _Z14sumMatrixOnGPUPfS_S_ii .p2align 8 .type _Z14sumMatrixOnGPUPfS_S_ii,@function _Z14sumMatrixOnGPUPfS_S_ii: s_clause 0x1 s_load_b32 s2, s[0:1], 0x2c s_load_b64 s[4:5], s[0:1], 0x18 v_and_b32_e32 v2, 0x3ff, v0 v_bfe_u32 v3, v0, 10, 10 s_waitcnt lgkmcnt(0) s_and_b32 s3, s2, 0xffff s_lshr_b32 s2, s2, 16 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_mad_u64_u32 v[0:1], null, s14, s3, v[2:3] v_mad_u64_u32 v[1:2], null, s15, s2, v[3:4] v_cmp_gt_u32_e32 vcc_lo, s4, v0 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_gt_u32_e64 s2, s5, v1 s_and_b32 s2, vcc_lo, s2 s_delay_alu instid0(SALU_CYCLE_1) s_and_saveexec_b32 s3, s2 s_cbranch_execz .LBB0_2 s_load_b128 s[8:11], s[0:1], 0x0 v_mad_u64_u32 v[2:3], null, v1, s4, v[0:1] v_mov_b32_e32 v3, 0 s_load_b64 s[0:1], s[0:1], 0x10 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[2:3] s_waitcnt lgkmcnt(0) v_add_co_u32 v2, vcc_lo, s8, v0 s_delay_alu instid0(VALU_DEP_2) v_add_co_ci_u32_e32 v3, vcc_lo, s9, v1, vcc_lo v_add_co_u32 v4, vcc_lo, s10, v0 v_add_co_ci_u32_e32 v5, vcc_lo, s11, 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_f32_e32 v2, v2, v3 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 _Z14sumMatrixOnGPUPfS_S_ii .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 288 .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 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 _Z14sumMatrixOnGPUPfS_S_ii, .Lfunc_end0-_Z14sumMatrixOnGPUPfS_S_ii .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: 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: _Z14sumMatrixOnGPUPfS_S_ii .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z14sumMatrixOnGPUPfS_S_ii.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_001b0c1d_00000000-6_sumMatrixOnGPU.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__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii .type _Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii, @function _Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii: .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) movl %r8d, (%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) movq %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 _Z14sumMatrixOnGPUPfS_S_ii(%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__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii, .-_Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii .globl _Z14sumMatrixOnGPUPfS_S_ii .type _Z14sumMatrixOnGPUPfS_S_ii, @function _Z14sumMatrixOnGPUPfS_S_ii: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z40__device_stub__Z14sumMatrixOnGPUPfS_S_iiPfS_S_ii addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z14sumMatrixOnGPUPfS_S_ii, .-_Z14sumMatrixOnGPUPfS_S_ii .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z14sumMatrixOnGPUPfS_S_ii" .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 _Z14sumMatrixOnGPUPfS_S_ii(%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 "sumMatrixOnGPU.hip" .globl _Z29__device_stub__sumMatrixOnGPUPfS_S_ii # -- Begin function _Z29__device_stub__sumMatrixOnGPUPfS_S_ii .p2align 4, 0x90 .type _Z29__device_stub__sumMatrixOnGPUPfS_S_ii,@function _Z29__device_stub__sumMatrixOnGPUPfS_S_ii: # @_Z29__device_stub__sumMatrixOnGPUPfS_S_ii .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) movl %r8d, (%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) movq %rsp, %rax movq %rax, 112(%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 $_Z14sumMatrixOnGPUPfS_S_ii, %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 _Z29__device_stub__sumMatrixOnGPUPfS_S_ii, .Lfunc_end0-_Z29__device_stub__sumMatrixOnGPUPfS_S_ii .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 $_Z14sumMatrixOnGPUPfS_S_ii, %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 _Z14sumMatrixOnGPUPfS_S_ii,@object # @_Z14sumMatrixOnGPUPfS_S_ii .section .rodata,"a",@progbits .globl _Z14sumMatrixOnGPUPfS_S_ii .p2align 3, 0x0 _Z14sumMatrixOnGPUPfS_S_ii: .quad _Z29__device_stub__sumMatrixOnGPUPfS_S_ii .size _Z14sumMatrixOnGPUPfS_S_ii, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z14sumMatrixOnGPUPfS_S_ii" .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 _Z29__device_stub__sumMatrixOnGPUPfS_S_ii .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z14sumMatrixOnGPUPfS_S_ii .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.
/* * MSU CUDA Course Examples and Exercises. * * Copyright (c) 2011 Dmitry Mikushin * * This software is provided 'as-is', without any express or implied warranty. * In no event will the authors be held liable for any damages arising * from the use of this software. * Permission is granted to anyone to use this software for any purpose, * including commercial applications, and to alter it and redistribute it freely, * without any restrictons. */ #include <cuda_runtime.h> #include <malloc.h> #include <math.h> #include <stdio.h> #include <stdlib.h> // Maximum allowed results difference. #define EPS 1e-04 // Grid block size (see comment in pattern2d below). #define BLOCK_LENGTH 32 #define BLOCK_HEIGHT 16 #define IN(i,j) in[i + (j) * nx] #define OUT(i,j) out[i + (j) * nx] // GPU device kernel. __global__ void pattern2d_gpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { // Compute absolute (i,j) indexes for // the current GPU thread using grid mapping params. int i = blockIdx.x * BLOCK_LENGTH + threadIdx.x + bx; int j = blockIdx.y * BLOCK_HEIGHT + threadIdx.y + by; // Compute one data point - a piece of // work for the current GPU thread. OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // CPU control implementation. void pattern2d_cpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { for (int j = by; j < ny - ey; j++) for (int i = bx; i < nx - ex; i++) OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // Perform some dummy 2D field processing on GPU and CPU, // and compare results. int pattern2d(int nx, int ny, float* in, float* out, int pid, int step) { if ((nx <= 0) || (ny <= 0)) return -1; long np = nx * ny; size_t size = sizeof(float) * np; // Create GPU data array and copy input data to it. float* in_gpu; cudaError_t status = cudaMalloc((void**)&in_gpu, size); if (status != cudaSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc input memory on GPU by process", pid, status); return status; } status = cudaMemcpy(in_gpu, in, size, cudaMemcpyHostToDevice); if (status != cudaSuccess) { fprintf(stderr, "Cannot copy data from host to gpu by process %d, status = %d\n", pid, status); return status; } // Create CPU data output array and get // control result using CPU function. float* control = (float*)malloc(size); memset(control, 0, size); pattern2d_cpu( 1, nx, 1, 1, ny, 1, in, control); // Configure GPU computational grid: // nx = nblocks_x * block_length // ny = nblocks_y * block_height // // NOTE: we have degree of freedom in // selecting how real problem grid maps onto // computational grid. Usually these params // are tuned to get optimal performance. // // NOTE: chose of grid/block config is // also limited by device properties: // - Maximum number of threads per block (512) // - Maximum sizes of each dimension of a block (512 x 512 x 64) // - Maximum sizes of each dimension of a grid (65535 x 65535 x 1) int nblocks_x = (nx - 2) / BLOCK_LENGTH; int nblocks_y = (ny - 2) / BLOCK_HEIGHT; // Perform the same processing on GPU, // returning result to GPU array. float* out_gpu; status = cudaMalloc((void**)&out_gpu, size); if (status != cudaSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc output memory on GPU by process", pid, status); return status; } status = cudaMemset(out_gpu, 0, size); if (status != cudaSuccess) { fprintf(stderr, "Cannot erase output memory on GPU by process %d, status = %d\n", pid, status); return status; } pattern2d_gpu<<< dim3(nblocks_x, nblocks_y, 1), dim3(BLOCK_LENGTH, BLOCK_HEIGHT, 1)>>>( 1, nx, 1, 1, ny, 1, in_gpu, out_gpu); status = cudaGetLastError(); if (status != cudaSuccess) { fprintf(stderr, "Cannot execute CUDA kernel by process %d, status = %d\n", pid, status); return status; } status = cudaThreadSynchronize(); if (status != cudaSuccess) { fprintf(stderr, "Cannot synchronize thread by process %d, status = %d\n", pid, status); return status; } // Copy GPU result from GPU memory to CPU buffer. status = cudaMemcpy(out, out_gpu, size, cudaMemcpyDeviceToHost); if (status != cudaSuccess) { fprintf(stderr, "Cannot copy data from gpu to host by process %d, status = %d\n", pid, status); return status; } // Don't bother with processing the remainder // on GPU. Do it on CPU instead. pattern2d_cpu( 1, nx, 1, ny - (ny - 2) % BLOCK_HEIGHT - 2, ny, 1, in, out); pattern2d_cpu( nx - (nx - 2) % BLOCK_LENGTH - 2, nx, 1, 1, ny, 1, in, out); // Compare results and find the maximum abs difference. int maxi = 0, maxj = 0; float maxdiff = fabs(out[0] - control[0]); float* diffs = (float*)malloc(size); memset(diffs, 0, size); for (int j = 0; j < ny; j++) for (int i = 0; i < nx; i++) { float diff = fabs( out[i + j * nx] - control[i + j * nx]); if (diff > maxdiff) { maxdiff = diff; maxi = i; maxj = j; } diffs[i + j * nx] = diff; } // Release data arrays. status = cudaFree(in_gpu); if (status != cudaSuccess) { fprintf(stderr, "Cannot free device input memory by process %d, status = %d\n", pid, status); return status; } free(control); status = cudaFree(out_gpu); if (status != cudaSuccess) { fprintf(stderr, "Cannot free device output memory by process %d, status = %d\n", pid, status); return status; } free(diffs); printf("Step %d result abs max diff by process %d = %f @ (%d,%d)\n", step, pid, maxdiff, maxi, maxj); return 0; }
code for sm_80 Function : _Z13pattern2d_gpuiiiiiiPfS_ .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.Y ; /* 0x0000000000027919 */ /* 0x000e220000002600 */ /*0020*/ MOV R10, c[0x0][0x164] ; /* 0x00005900000a7a02 */ /* 0x000fe20000000f00 */ /*0030*/ HFMA2.MMA R7, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff077435 */ /* 0x000fe200000001ff */ /*0040*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0050*/ S2R R5, SR_TID.Y ; /* 0x0000000000057919 */ /* 0x000e280000002200 */ /*0060*/ S2R R0, SR_CTAID.X ; /* 0x0000000000007919 */ /* 0x000e680000002500 */ /*0070*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e620000002100 */ /*0080*/ LEA R2, R2, R5, 0x4 ; /* 0x0000000502027211 */ /* 0x001fc400078e20ff */ /*0090*/ LEA R0, R0, R3, 0x5 ; /* 0x0000000300007211 */ /* 0x002fe400078e28ff */ /*00a0*/ IADD3 R3, R2, c[0x0][0x16c], RZ ; /* 0x00005b0002037a10 */ /* 0x000fe40007ffe0ff */ /*00b0*/ IADD3 R4, R0, c[0x0][0x160], RZ ; /* 0x0000580000047a10 */ /* 0x000fc60007ffe0ff */ /*00c0*/ IMAD R5, R3.reuse, R10, -c[0x0][0x164] ; /* 0x8000590003057624 */ /* 0x040fe400078e020a */ /*00d0*/ IMAD R0, R3, c[0x0][0x164], R4 ; /* 0x0000590003007a24 */ /* 0x000fc600078e0204 */ /*00e0*/ IADD3 R6, R4, R5, RZ ; /* 0x0000000504067210 */ /* 0x000fe20007ffe0ff */ /*00f0*/ IMAD.WIDE R2, R0, R7, c[0x0][0x178] ; /* 0x00005e0000027625 */ /* 0x000fc800078e0207 */ /*0100*/ IMAD.WIDE R4, R6, R7, c[0x0][0x178] ; /* 0x00005e0006047625 */ /* 0x000fe200078e0207 */ /*0110*/ LDG.E R8, [R2.64] ; /* 0x0000000402087981 */ /* 0x000ea2000c1e1900 */ /*0120*/ LEA R6, R10, R6, 0x1 ; /* 0x000000060a067211 */ /* 0x000fc600078e08ff */ /*0130*/ LDG.E R9, [R2.64+-0x4] ; /* 0xfffffc0402097981 */ /* 0x000ea4000c1e1900 */ /*0140*/ IMAD.WIDE R6, R6, R7, c[0x0][0x178] ; /* 0x00005e0006067625 */ /* 0x000fe400078e0207 */ /*0150*/ LDG.E R11, [R2.64+0x4] ; /* 0x00000404020b7981 */ /* 0x000ee8000c1e1900 */ /*0160*/ LDG.E R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000f28000c1e1900 */ /*0170*/ LDG.E R7, [R6.64] ; /* 0x0000000406077981 */ /* 0x000f62000c1e1900 */ /*0180*/ BSSY B0, 0x2b0 ; /* 0x0000012000007945 */ /* 0x000fe20003800000 */ /*0190*/ FADD R8, R8, R9 ; /* 0x0000000908087221 */ /* 0x004fc80000000000 */ /*01a0*/ FADD R8, R8, R11 ; /* 0x0000000b08087221 */ /* 0x008fe40000000000 */ /*01b0*/ FADD R9, R4, R4 ; /* 0x0000000404097221 */ /* 0x010fc80000000000 */ /*01c0*/ FADD R8, R8, -R9 ; /* 0x8000000908087221 */ /* 0x000fc80000000000 */ /*01d0*/ FFMA R8, R7, 3, R8 ; /* 0x4040000007087823 */ /* 0x020fc80000000008 */ /*01e0*/ FADD R10, |R8|, -RZ ; /* 0x800000ff080a7221 */ /* 0x000fe20000000200 */ /*01f0*/ MUFU.RSQ R3, |R8| ; /* 0x4000000800037308 */ /* 0x0000680000001400 */ /*0200*/ IADD3 R9, R10, -0xd000000, RZ ; /* 0xf30000000a097810 */ /* 0x000fc80007ffe0ff */ /*0210*/ ISETP.GT.U32.AND P0, PT, R9, 0x727fffff, PT ; /* 0x727fffff0900780c */ /* 0x000fda0003f04070 */ /*0220*/ @!P0 BRA 0x260 ; /* 0x0000003000008947 */ /* 0x000fea0003800000 */ /*0230*/ MOV R7, 0x250 ; /* 0x0000025000077802 */ /* 0x003fe40000000f00 */ /*0240*/ CALL.REL.NOINC 0x2f0 ; /* 0x000000a000007944 */ /* 0x000fea0003c00000 */ /*0250*/ BRA 0x2a0 ; /* 0x0000004000007947 */ /* 0x000fea0003800000 */ /*0260*/ FMUL.FTZ R5, |R8|, R3 ; /* 0x0000000308057220 */ /* 0x003fe40000410200 */ /*0270*/ FMUL.FTZ R3, R3, 0.5 ; /* 0x3f00000003037820 */ /* 0x000fe40000410000 */ /*0280*/ FFMA R8, -R5, R5, |R8| ; /* 0x0000000505087223 */ /* 0x000fc80000000508 */ /*0290*/ FFMA R5, R8, R3, R5 ; /* 0x0000000308057223 */ /* 0x000fe40000000005 */ /*02a0*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*02b0*/ HFMA2.MMA R3, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff037435 */ /* 0x000fd400000001ff */ /*02c0*/ IMAD.WIDE R2, R0, R3, c[0x0][0x180] ; /* 0x0000600000027625 */ /* 0x000fca00078e0203 */ /*02d0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe2000c101904 */ /*02e0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*02f0*/ LOP3.LUT P0, RZ, R10, 0x7fffffff, RZ, 0xc0, !PT ; /* 0x7fffffff0aff7812 */ /* 0x000fda000780c0ff */ /*0300*/ @!P0 MOV R2, R10 ; /* 0x0000000a00028202 */ /* 0x000fe20000000f00 */ /*0310*/ @!P0 BRA 0x420 ; /* 0x0000010000008947 */ /* 0x000fea0003800000 */ /*0320*/ FSETP.GEU.FTZ.AND P0, PT, R10, RZ, PT ; /* 0x000000ff0a00720b */ /* 0x000fda0003f1e000 */ /*0330*/ @!P0 MOV R2, 0x7fffffff ; /* 0x7fffffff00028802 */ /* 0x000fe20000000f00 */ /*0340*/ @!P0 BRA 0x420 ; /* 0x000000d000008947 */ /* 0x000fea0003800000 */ /*0350*/ FSETP.GTU.FTZ.AND P0, PT, |R10|, +INF , PT ; /* 0x7f8000000a00780b */ /* 0x000fda0003f1c200 */ /*0360*/ @P0 FADD.FTZ R2, R10, 1 ; /* 0x3f8000000a020421 */ /* 0x000fe20000010000 */ /*0370*/ @P0 BRA 0x420 ; /* 0x000000a000000947 */ /* 0x000fea0003800000 */ /*0380*/ FSETP.NEU.FTZ.AND P0, PT, |R10|, +INF , PT ; /* 0x7f8000000a00780b */ /* 0x000fda0003f1d200 */ /*0390*/ @P0 FFMA R3, R10, 1.84467440737095516160e+19, RZ ; /* 0x5f8000000a030823 */ /* 0x000fc800000000ff */ /*03a0*/ @P0 MUFU.RSQ R2, R3 ; /* 0x0000000300020308 */ /* 0x000e240000001400 */ /*03b0*/ @P0 FMUL.FTZ R4, R3, R2 ; /* 0x0000000203040220 */ /* 0x001fe40000410000 */ /*03c0*/ @P0 FMUL.FTZ R6, R2, 0.5 ; /* 0x3f00000002060820 */ /* 0x000fe20000410000 */ /*03d0*/ @!P0 MOV R2, R10 ; /* 0x0000000a00028202 */ /* 0x000fe20000000f00 */ /*03e0*/ @P0 FADD.FTZ R5, -R4, -RZ ; /* 0x800000ff04050221 */ /* 0x000fc80000010100 */ /*03f0*/ @P0 FFMA R5, R4, R5, R3 ; /* 0x0000000504050223 */ /* 0x000fc80000000003 */ /*0400*/ @P0 FFMA R5, R5, R6, R4 ; /* 0x0000000605050223 */ /* 0x000fc80000000004 */ /*0410*/ @P0 FMUL.FTZ R2, R5, 2.3283064365386962891e-10 ; /* 0x2f80000005020820 */ /* 0x000fc80000410000 */ /*0420*/ HFMA2.MMA R3, -RZ, RZ, 0, 0 ; /* 0x00000000ff037435 */ /* 0x000fe200000001ff */ /*0430*/ MOV R5, R2 ; /* 0x0000000200057202 */ /* 0x000fe40000000f00 */ /*0440*/ MOV R2, R7 ; /* 0x0000000700027202 */ /* 0x000fc80000000f00 */ /*0450*/ RET.REL.NODEC R2 0x0 ; /* 0xfffffba002007950 */ /* 0x000fea0003c3ffff */ /*0460*/ BRA 0x460; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0470*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0480*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0490*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
/* * MSU CUDA Course Examples and Exercises. * * Copyright (c) 2011 Dmitry Mikushin * * This software is provided 'as-is', without any express or implied warranty. * In no event will the authors be held liable for any damages arising * from the use of this software. * Permission is granted to anyone to use this software for any purpose, * including commercial applications, and to alter it and redistribute it freely, * without any restrictons. */ #include <cuda_runtime.h> #include <malloc.h> #include <math.h> #include <stdio.h> #include <stdlib.h> // Maximum allowed results difference. #define EPS 1e-04 // Grid block size (see comment in pattern2d below). #define BLOCK_LENGTH 32 #define BLOCK_HEIGHT 16 #define IN(i,j) in[i + (j) * nx] #define OUT(i,j) out[i + (j) * nx] // GPU device kernel. __global__ void pattern2d_gpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { // Compute absolute (i,j) indexes for // the current GPU thread using grid mapping params. int i = blockIdx.x * BLOCK_LENGTH + threadIdx.x + bx; int j = blockIdx.y * BLOCK_HEIGHT + threadIdx.y + by; // Compute one data point - a piece of // work for the current GPU thread. OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // CPU control implementation. void pattern2d_cpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { for (int j = by; j < ny - ey; j++) for (int i = bx; i < nx - ex; i++) OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // Perform some dummy 2D field processing on GPU and CPU, // and compare results. int pattern2d(int nx, int ny, float* in, float* out, int pid, int step) { if ((nx <= 0) || (ny <= 0)) return -1; long np = nx * ny; size_t size = sizeof(float) * np; // Create GPU data array and copy input data to it. float* in_gpu; cudaError_t status = cudaMalloc((void**)&in_gpu, size); if (status != cudaSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc input memory on GPU by process", pid, status); return status; } status = cudaMemcpy(in_gpu, in, size, cudaMemcpyHostToDevice); if (status != cudaSuccess) { fprintf(stderr, "Cannot copy data from host to gpu by process %d, status = %d\n", pid, status); return status; } // Create CPU data output array and get // control result using CPU function. float* control = (float*)malloc(size); memset(control, 0, size); pattern2d_cpu( 1, nx, 1, 1, ny, 1, in, control); // Configure GPU computational grid: // nx = nblocks_x * block_length // ny = nblocks_y * block_height // // NOTE: we have degree of freedom in // selecting how real problem grid maps onto // computational grid. Usually these params // are tuned to get optimal performance. // // NOTE: chose of grid/block config is // also limited by device properties: // - Maximum number of threads per block (512) // - Maximum sizes of each dimension of a block (512 x 512 x 64) // - Maximum sizes of each dimension of a grid (65535 x 65535 x 1) int nblocks_x = (nx - 2) / BLOCK_LENGTH; int nblocks_y = (ny - 2) / BLOCK_HEIGHT; // Perform the same processing on GPU, // returning result to GPU array. float* out_gpu; status = cudaMalloc((void**)&out_gpu, size); if (status != cudaSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc output memory on GPU by process", pid, status); return status; } status = cudaMemset(out_gpu, 0, size); if (status != cudaSuccess) { fprintf(stderr, "Cannot erase output memory on GPU by process %d, status = %d\n", pid, status); return status; } pattern2d_gpu<<< dim3(nblocks_x, nblocks_y, 1), dim3(BLOCK_LENGTH, BLOCK_HEIGHT, 1)>>>( 1, nx, 1, 1, ny, 1, in_gpu, out_gpu); status = cudaGetLastError(); if (status != cudaSuccess) { fprintf(stderr, "Cannot execute CUDA kernel by process %d, status = %d\n", pid, status); return status; } status = cudaThreadSynchronize(); if (status != cudaSuccess) { fprintf(stderr, "Cannot synchronize thread by process %d, status = %d\n", pid, status); return status; } // Copy GPU result from GPU memory to CPU buffer. status = cudaMemcpy(out, out_gpu, size, cudaMemcpyDeviceToHost); if (status != cudaSuccess) { fprintf(stderr, "Cannot copy data from gpu to host by process %d, status = %d\n", pid, status); return status; } // Don't bother with processing the remainder // on GPU. Do it on CPU instead. pattern2d_cpu( 1, nx, 1, ny - (ny - 2) % BLOCK_HEIGHT - 2, ny, 1, in, out); pattern2d_cpu( nx - (nx - 2) % BLOCK_LENGTH - 2, nx, 1, 1, ny, 1, in, out); // Compare results and find the maximum abs difference. int maxi = 0, maxj = 0; float maxdiff = fabs(out[0] - control[0]); float* diffs = (float*)malloc(size); memset(diffs, 0, size); for (int j = 0; j < ny; j++) for (int i = 0; i < nx; i++) { float diff = fabs( out[i + j * nx] - control[i + j * nx]); if (diff > maxdiff) { maxdiff = diff; maxi = i; maxj = j; } diffs[i + j * nx] = diff; } // Release data arrays. status = cudaFree(in_gpu); if (status != cudaSuccess) { fprintf(stderr, "Cannot free device input memory by process %d, status = %d\n", pid, status); return status; } free(control); status = cudaFree(out_gpu); if (status != cudaSuccess) { fprintf(stderr, "Cannot free device output memory by process %d, status = %d\n", pid, status); return status; } free(diffs); printf("Step %d result abs max diff by process %d = %f @ (%d,%d)\n", step, pid, maxdiff, maxi, maxj); return 0; }
.file "tmpxft_000a59a6_00000000-6_pattern2d.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2061: .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 .LFE2061: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z13pattern2d_cpuiiiiiiPfS_ .type _Z13pattern2d_cpuiiiiiiPfS_, @function _Z13pattern2d_cpuiiiiiiPfS_: .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 movl %edi, %ebp movl %ecx, %ebx movq 56(%rsp), %rcx movq 64(%rsp), %rdi subl %r9d, %r8d cmpl %r8d, %ebx jge .L3 movl %esi, %r12d movl %ebx, %r11d imull %esi, %r11d movl %esi, %r13d subl %edx, %r13d movslq %ebp, %r14 movl %r13d, %eax subl %ebp, %eax addq %r14, %rax movq %rax, -8(%rsp) leal (%rsi,%rsi), %r15d movss .LC0(%rip), %xmm3 movss .LC1(%rip), %xmm2 jmp .L5 .L7: movl %r11d, %r10d subl %r12d, %r10d movslq %r11d, %r9 leaq (%r9,%r14), %rax salq $2, %rax movq -8(%rsp), %rsi addq %r9, %rsi salq $2, %rsi negq %r9 salq $2, %r9 movslq %r10d, %rdx leaq (%r9,%rdx,4), %rdx addq %rcx, %rdx addl %r15d, %r10d movslq %r10d, %r10 leaq (%r9,%r10,4), %r9 addq %rcx, %r9 .L6: movss (%rcx,%rax), %xmm0 addss -4(%rcx,%rax), %xmm0 addss 4(%rcx,%rax), %xmm0 movss (%rdx,%rax), %xmm1 addss %xmm1, %xmm1 subss %xmm1, %xmm0 movaps %xmm3, %xmm1 mulss (%r9,%rax), %xmm1 addss %xmm1, %xmm0 andps %xmm2, %xmm0 sqrtss %xmm0, %xmm0 movss %xmm0, (%rdi,%rax) addq $4, %rax cmpq %rsi, %rax jne .L6 .L8: addl $1, %ebx addl %r12d, %r11d cmpl %r8d, %ebx je .L3 .L5: cmpl %r13d, %ebp jl .L7 jmp .L8 .L3: 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 .LFE2057: .size _Z13pattern2d_cpuiiiiiiPfS_, .-_Z13pattern2d_cpuiiiiiiPfS_ .globl _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ .type _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_, @function _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_: .LFB2083: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movl %edi, 44(%rsp) movl %esi, 40(%rsp) movl %edx, 36(%rsp) movl %ecx, 32(%rsp) movl %r8d, 28(%rsp) movl %r9d, 24(%rsp) movq 208(%rsp), %rax movq %rax, 16(%rsp) movq 216(%rsp), %rax movq %rax, 8(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 44(%rsp), %rax movq %rax, 112(%rsp) leaq 40(%rsp), %rax movq %rax, 120(%rsp) leaq 36(%rsp), %rax movq %rax, 128(%rsp) leaq 32(%rsp), %rax movq %rax, 136(%rsp) leaq 28(%rsp), %rax movq %rax, 144(%rsp) leaq 24(%rsp), %rax movq %rax, 152(%rsp) leaq 16(%rsp), %rax movq %rax, 160(%rsp) leaq 8(%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 .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 _Z13pattern2d_gpuiiiiiiPfS_(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L11 .L16: call __stack_chk_fail@PLT .cfi_endproc .LFE2083: .size _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_, .-_Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ .globl _Z13pattern2d_gpuiiiiiiPfS_ .type _Z13pattern2d_gpuiiiiiiPfS_, @function _Z13pattern2d_gpuiiiiiiPfS_: .LFB2084: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 pushq 24(%rsp) .cfi_def_cfa_offset 24 pushq 24(%rsp) .cfi_def_cfa_offset 32 call _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2084: .size _Z13pattern2d_gpuiiiiiiPfS_, .-_Z13pattern2d_gpuiiiiiiPfS_ .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC2: .string "Cannot malloc input memory on GPU by process" .align 8 .LC3: .string "%s %d, status = %d\nInsufficient GPU memory?\n" .align 8 .LC4: .string "Cannot copy data from host to gpu by process %d, status = %d\n" .align 8 .LC5: .string "Cannot malloc output memory on GPU by process" .align 8 .LC6: .string "Cannot erase output memory on GPU by process %d, status = %d\n" .align 8 .LC7: .string "Cannot execute CUDA kernel by process %d, status = %d\n" .align 8 .LC8: .string "Cannot synchronize thread by process %d, status = %d\n" .align 8 .LC9: .string "Cannot copy data from gpu to host by process %d, status = %d\n" .align 8 .LC10: .string "Cannot free device input memory by process %d, status = %d\n" .align 8 .LC11: .string "Cannot free device output memory by process %d, status = %d\n" .align 8 .LC12: .string "Step %d result abs max diff by process %d = %f @ (%d,%d)\n" .text .globl _Z9pattern2diiPfS_ii .type _Z9pattern2diiPfS_ii, @function _Z9pattern2diiPfS_ii: .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 subq $88, %rsp .cfi_def_cfa_offset 144 movq %rcx, 8(%rsp) movl %r9d, 24(%rsp) movq %fs:40, %rax movq %rax, 72(%rsp) xorl %eax, %eax testl %edi, %edi jle .L37 movl %edi, %ebx movl %esi, %r12d movq %rdx, %r13 movl %r8d, %r14d testl %esi, %esi jle .L37 movl %edi, %ebp imull %esi, %ebp movslq %ebp, %rbp salq $2, %rbp leaq 32(%rsp), %rdi movq %rbp, %rsi call cudaMalloc@PLT testl %eax, %eax jne .L41 movl $1, %ecx movq %rbp, %rdx movq %r13, %rsi movq 32(%rsp), %rdi call cudaMemcpy@PLT testl %eax, %eax jne .L42 movq %rbp, %rdi call malloc@PLT movq %rax, %r15 movq %rbp, %rcx movq %rbp, %rdx movl $0, %esi movq %rax, %rdi call __memset_chk@PLT pushq %r15 .cfi_def_cfa_offset 152 pushq %r13 .cfi_def_cfa_offset 160 movl $1, %r9d movl %r12d, %r8d movl $1, %ecx movl $1, %edx movl %ebx, %esi movl $1, %edi call _Z13pattern2d_cpuiiiiiiPfS_ addq $16, %rsp .cfi_def_cfa_offset 144 leaq 40(%rsp), %rdi movq %rbp, %rsi call cudaMalloc@PLT testl %eax, %eax jne .L43 movq %rbp, %rdx movl $0, %esi movq 40(%rsp), %rdi call cudaMemset@PLT testl %eax, %eax jne .L44 movl $32, 60(%rsp) movl $16, 64(%rsp) movl $1, 68(%rsp) leal 29(%rbx), %eax movl %ebx, %ecx subl $2, %ecx movl %ecx, 16(%rsp) cmovns %ecx, %eax sarl $5, %eax movl %eax, 48(%rsp) leal 13(%r12), %eax movl %r12d, %esi subl $2, %esi movl %esi, 28(%rsp) cmovns %esi, %eax sarl $4, %eax movl %eax, 52(%rsp) movl $1, 56(%rsp) movl $0, %r9d movl $0, %r8d movq 60(%rsp), %rdx movl $1, %ecx movq 48(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L45 .L25: call cudaGetLastError@PLT testl %eax, %eax jne .L46 call cudaThreadSynchronize@PLT testl %eax, %eax jne .L47 movl $2, %ecx movq %rbp, %rdx movq 40(%rsp), %rsi movq 8(%rsp), %rdi call cudaMemcpy@PLT testl %eax, %eax jne .L48 movl 28(%rsp), %eax movl %eax, %edx sarl $31, %edx shrl $28, %edx addl %edx, %eax andl $15, %eax subl %edx, %eax movl %r12d, %ecx subl %eax, %ecx subl $2, %ecx pushq 8(%rsp) .cfi_def_cfa_offset 152 pushq %r13 .cfi_def_cfa_offset 160 movl $1, %r9d movl %r12d, %r8d movl $1, %edx movl %ebx, %esi movl $1, %edi call _Z13pattern2d_cpuiiiiiiPfS_ movl 32(%rsp), %eax movl %eax, %edx sarl $31, %edx shrl $27, %edx addl %edx, %eax andl $31, %eax subl %edx, %eax movl %ebx, %edi subl %eax, %edi subl $2, %edi pushq 24(%rsp) .cfi_def_cfa_offset 168 pushq %r13 .cfi_def_cfa_offset 176 movl $1, %r9d movl %r12d, %r8d movl $1, %ecx movl $1, %edx movl %ebx, %esi call _Z13pattern2d_cpuiiiiiiPfS_ movq 40(%rsp), %r13 movss 0(%r13), %xmm0 subss (%r15), %xmm0 andps .LC1(%rip), %xmm0 movss %xmm0, 40(%rsp) addq $32, %rsp .cfi_def_cfa_offset 144 movq %rbp, %rdi call malloc@PLT movq %rax, 16(%rsp) movq %rbp, %rcx movq %rbp, %rdx movl $0, %esi movq %rax, %rbp movq %rax, %rdi call __memset_chk@PLT movslq %ebx, %r8 salq $2, %r8 movq %rbp, %rdi movq %r13, %rdx movq %r15, %rsi movl $0, %ecx movl $0, %r13d movl $0, %ebp movss .LC1(%rip), %xmm1 .L29: movl $0, %eax .L32: movss (%rdx,%rax,4), %xmm0 subss (%rsi,%rax,4), %xmm0 andps %xmm1, %xmm0 movss 8(%rsp), %xmm2 comiss %xmm2, %xmm0 cmova %eax, %ebp movaps %xmm0, %xmm3 maxss %xmm2, %xmm3 movss %xmm3, 8(%rsp) cmova %ecx, %r13d movss %xmm0, (%rdi,%rax,4) addq $1, %rax cmpl %eax, %ebx jg .L32 addl $1, %ecx addq %r8, %rdi addq %r8, %rdx addq %r8, %rsi cmpl %ecx, %r12d jg .L29 movq 32(%rsp), %rdi call cudaFree@PLT testl %eax, %eax jne .L49 movq %r15, %rdi call free@PLT movq 40(%rsp), %rdi call cudaFree@PLT testl %eax, %eax jne .L50 movq 16(%rsp), %rdi call free@PLT pxor %xmm0, %xmm0 cvtss2sd 8(%rsp), %xmm0 movl %r13d, %r9d movl %ebp, %r8d movl %r14d, %ecx movl 24(%rsp), %edx leaq .LC12(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movl $0, %ebx jmp .L19 .L41: movl %eax, %ebx movl %eax, %r9d movl %r14d, %r8d leaq .LC2(%rip), %rcx leaq .LC3(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT .L19: movq 72(%rsp), %rax subq %fs:40, %rax jne .L51 movl %ebx, %eax addq $88, %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 .L42: .cfi_restore_state movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC4(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L43: movl %eax, %ebx movl %eax, %r9d movl %r14d, %r8d leaq .LC5(%rip), %rcx leaq .LC3(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L44: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC6(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L45: pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 movl $1, %r9d movl %r12d, %r8d movl $1, %ecx movl $1, %edx movl %ebx, %esi movl $1, %edi call _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L25 .L46: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC7(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L47: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC8(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L48: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC9(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L49: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC10(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L50: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC11(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L37: movl $-1, %ebx jmp .L19 .L51: call __stack_chk_fail@PLT .cfi_endproc .LFE2058: .size _Z9pattern2diiPfS_ii, .-_Z9pattern2diiPfS_ii .section .rodata.str1.1,"aMS",@progbits,1 .LC13: .string "_Z13pattern2d_gpuiiiiiiPfS_" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2086: .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 .LC13(%rip), %rdx movq %rdx, %rcx leaq _Z13pattern2d_gpuiiiiiiPfS_(%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 .LFE2086: .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 .section .rodata.cst16,"aM",@progbits,16 .align 16 .LC1: .long 2147483647 .long 0 .long 0 .long 0 .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.
/* * MSU CUDA Course Examples and Exercises. * * Copyright (c) 2011 Dmitry Mikushin * * This software is provided 'as-is', without any express or implied warranty. * In no event will the authors be held liable for any damages arising * from the use of this software. * Permission is granted to anyone to use this software for any purpose, * including commercial applications, and to alter it and redistribute it freely, * without any restrictons. */ #include <cuda_runtime.h> #include <malloc.h> #include <math.h> #include <stdio.h> #include <stdlib.h> // Maximum allowed results difference. #define EPS 1e-04 // Grid block size (see comment in pattern2d below). #define BLOCK_LENGTH 32 #define BLOCK_HEIGHT 16 #define IN(i,j) in[i + (j) * nx] #define OUT(i,j) out[i + (j) * nx] // GPU device kernel. __global__ void pattern2d_gpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { // Compute absolute (i,j) indexes for // the current GPU thread using grid mapping params. int i = blockIdx.x * BLOCK_LENGTH + threadIdx.x + bx; int j = blockIdx.y * BLOCK_HEIGHT + threadIdx.y + by; // Compute one data point - a piece of // work for the current GPU thread. OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // CPU control implementation. void pattern2d_cpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { for (int j = by; j < ny - ey; j++) for (int i = bx; i < nx - ex; i++) OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // Perform some dummy 2D field processing on GPU and CPU, // and compare results. int pattern2d(int nx, int ny, float* in, float* out, int pid, int step) { if ((nx <= 0) || (ny <= 0)) return -1; long np = nx * ny; size_t size = sizeof(float) * np; // Create GPU data array and copy input data to it. float* in_gpu; cudaError_t status = cudaMalloc((void**)&in_gpu, size); if (status != cudaSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc input memory on GPU by process", pid, status); return status; } status = cudaMemcpy(in_gpu, in, size, cudaMemcpyHostToDevice); if (status != cudaSuccess) { fprintf(stderr, "Cannot copy data from host to gpu by process %d, status = %d\n", pid, status); return status; } // Create CPU data output array and get // control result using CPU function. float* control = (float*)malloc(size); memset(control, 0, size); pattern2d_cpu( 1, nx, 1, 1, ny, 1, in, control); // Configure GPU computational grid: // nx = nblocks_x * block_length // ny = nblocks_y * block_height // // NOTE: we have degree of freedom in // selecting how real problem grid maps onto // computational grid. Usually these params // are tuned to get optimal performance. // // NOTE: chose of grid/block config is // also limited by device properties: // - Maximum number of threads per block (512) // - Maximum sizes of each dimension of a block (512 x 512 x 64) // - Maximum sizes of each dimension of a grid (65535 x 65535 x 1) int nblocks_x = (nx - 2) / BLOCK_LENGTH; int nblocks_y = (ny - 2) / BLOCK_HEIGHT; // Perform the same processing on GPU, // returning result to GPU array. float* out_gpu; status = cudaMalloc((void**)&out_gpu, size); if (status != cudaSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc output memory on GPU by process", pid, status); return status; } status = cudaMemset(out_gpu, 0, size); if (status != cudaSuccess) { fprintf(stderr, "Cannot erase output memory on GPU by process %d, status = %d\n", pid, status); return status; } pattern2d_gpu<<< dim3(nblocks_x, nblocks_y, 1), dim3(BLOCK_LENGTH, BLOCK_HEIGHT, 1)>>>( 1, nx, 1, 1, ny, 1, in_gpu, out_gpu); status = cudaGetLastError(); if (status != cudaSuccess) { fprintf(stderr, "Cannot execute CUDA kernel by process %d, status = %d\n", pid, status); return status; } status = cudaThreadSynchronize(); if (status != cudaSuccess) { fprintf(stderr, "Cannot synchronize thread by process %d, status = %d\n", pid, status); return status; } // Copy GPU result from GPU memory to CPU buffer. status = cudaMemcpy(out, out_gpu, size, cudaMemcpyDeviceToHost); if (status != cudaSuccess) { fprintf(stderr, "Cannot copy data from gpu to host by process %d, status = %d\n", pid, status); return status; } // Don't bother with processing the remainder // on GPU. Do it on CPU instead. pattern2d_cpu( 1, nx, 1, ny - (ny - 2) % BLOCK_HEIGHT - 2, ny, 1, in, out); pattern2d_cpu( nx - (nx - 2) % BLOCK_LENGTH - 2, nx, 1, 1, ny, 1, in, out); // Compare results and find the maximum abs difference. int maxi = 0, maxj = 0; float maxdiff = fabs(out[0] - control[0]); float* diffs = (float*)malloc(size); memset(diffs, 0, size); for (int j = 0; j < ny; j++) for (int i = 0; i < nx; i++) { float diff = fabs( out[i + j * nx] - control[i + j * nx]); if (diff > maxdiff) { maxdiff = diff; maxi = i; maxj = j; } diffs[i + j * nx] = diff; } // Release data arrays. status = cudaFree(in_gpu); if (status != cudaSuccess) { fprintf(stderr, "Cannot free device input memory by process %d, status = %d\n", pid, status); return status; } free(control); status = cudaFree(out_gpu); if (status != cudaSuccess) { fprintf(stderr, "Cannot free device output memory by process %d, status = %d\n", pid, status); return status; } free(diffs); printf("Step %d result abs max diff by process %d = %f @ (%d,%d)\n", step, pid, maxdiff, maxi, maxj); return 0; }
/* * MSU CUDA Course Examples and Exercises. * * Copyright (c) 2011 Dmitry Mikushin * * This software is provided 'as-is', without any express or implied warranty. * In no event will the authors be held liable for any damages arising * from the use of this software. * Permission is granted to anyone to use this software for any purpose, * including commercial applications, and to alter it and redistribute it freely, * without any restrictons. */ #include <hip/hip_runtime.h> #include <malloc.h> #include <math.h> #include <stdio.h> #include <stdlib.h> // Maximum allowed results difference. #define EPS 1e-04 // Grid block size (see comment in pattern2d below). #define BLOCK_LENGTH 32 #define BLOCK_HEIGHT 16 #define IN(i,j) in[i + (j) * nx] #define OUT(i,j) out[i + (j) * nx] // GPU device kernel. __global__ void pattern2d_gpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { // Compute absolute (i,j) indexes for // the current GPU thread using grid mapping params. int i = blockIdx.x * BLOCK_LENGTH + threadIdx.x + bx; int j = blockIdx.y * BLOCK_HEIGHT + threadIdx.y + by; // Compute one data point - a piece of // work for the current GPU thread. OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // CPU control implementation. void pattern2d_cpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { for (int j = by; j < ny - ey; j++) for (int i = bx; i < nx - ex; i++) OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // Perform some dummy 2D field processing on GPU and CPU, // and compare results. int pattern2d(int nx, int ny, float* in, float* out, int pid, int step) { if ((nx <= 0) || (ny <= 0)) return -1; long np = nx * ny; size_t size = sizeof(float) * np; // Create GPU data array and copy input data to it. float* in_gpu; hipError_t status = hipMalloc((void**)&in_gpu, size); if (status != hipSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc input memory on GPU by process", pid, status); return status; } status = hipMemcpy(in_gpu, in, size, hipMemcpyHostToDevice); if (status != hipSuccess) { fprintf(stderr, "Cannot copy data from host to gpu by process %d, status = %d\n", pid, status); return status; } // Create CPU data output array and get // control result using CPU function. float* control = (float*)malloc(size); memset(control, 0, size); pattern2d_cpu( 1, nx, 1, 1, ny, 1, in, control); // Configure GPU computational grid: // nx = nblocks_x * block_length // ny = nblocks_y * block_height // // NOTE: we have degree of freedom in // selecting how real problem grid maps onto // computational grid. Usually these params // are tuned to get optimal performance. // // NOTE: chose of grid/block config is // also limited by device properties: // - Maximum number of threads per block (512) // - Maximum sizes of each dimension of a block (512 x 512 x 64) // - Maximum sizes of each dimension of a grid (65535 x 65535 x 1) int nblocks_x = (nx - 2) / BLOCK_LENGTH; int nblocks_y = (ny - 2) / BLOCK_HEIGHT; // Perform the same processing on GPU, // returning result to GPU array. float* out_gpu; status = hipMalloc((void**)&out_gpu, size); if (status != hipSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc output memory on GPU by process", pid, status); return status; } status = hipMemset(out_gpu, 0, size); if (status != hipSuccess) { fprintf(stderr, "Cannot erase output memory on GPU by process %d, status = %d\n", pid, status); return status; } pattern2d_gpu<<< dim3(nblocks_x, nblocks_y, 1), dim3(BLOCK_LENGTH, BLOCK_HEIGHT, 1)>>>( 1, nx, 1, 1, ny, 1, in_gpu, out_gpu); status = hipGetLastError(); if (status != hipSuccess) { fprintf(stderr, "Cannot execute CUDA kernel by process %d, status = %d\n", pid, status); return status; } status = hipDeviceSynchronize(); if (status != hipSuccess) { fprintf(stderr, "Cannot synchronize thread by process %d, status = %d\n", pid, status); return status; } // Copy GPU result from GPU memory to CPU buffer. status = hipMemcpy(out, out_gpu, size, hipMemcpyDeviceToHost); if (status != hipSuccess) { fprintf(stderr, "Cannot copy data from gpu to host by process %d, status = %d\n", pid, status); return status; } // Don't bother with processing the remainder // on GPU. Do it on CPU instead. pattern2d_cpu( 1, nx, 1, ny - (ny - 2) % BLOCK_HEIGHT - 2, ny, 1, in, out); pattern2d_cpu( nx - (nx - 2) % BLOCK_LENGTH - 2, nx, 1, 1, ny, 1, in, out); // Compare results and find the maximum abs difference. int maxi = 0, maxj = 0; float maxdiff = fabs(out[0] - control[0]); float* diffs = (float*)malloc(size); memset(diffs, 0, size); for (int j = 0; j < ny; j++) for (int i = 0; i < nx; i++) { float diff = fabs( out[i + j * nx] - control[i + j * nx]); if (diff > maxdiff) { maxdiff = diff; maxi = i; maxj = j; } diffs[i + j * nx] = diff; } // Release data arrays. status = hipFree(in_gpu); if (status != hipSuccess) { fprintf(stderr, "Cannot free device input memory by process %d, status = %d\n", pid, status); return status; } free(control); status = hipFree(out_gpu); if (status != hipSuccess) { fprintf(stderr, "Cannot free device output memory by process %d, status = %d\n", pid, status); return status; } free(diffs); printf("Step %d result abs max diff by process %d = %f @ (%d,%d)\n", step, pid, maxdiff, maxi, maxj); return 0; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
/* * MSU CUDA Course Examples and Exercises. * * Copyright (c) 2011 Dmitry Mikushin * * This software is provided 'as-is', without any express or implied warranty. * In no event will the authors be held liable for any damages arising * from the use of this software. * Permission is granted to anyone to use this software for any purpose, * including commercial applications, and to alter it and redistribute it freely, * without any restrictons. */ #include <hip/hip_runtime.h> #include <malloc.h> #include <math.h> #include <stdio.h> #include <stdlib.h> // Maximum allowed results difference. #define EPS 1e-04 // Grid block size (see comment in pattern2d below). #define BLOCK_LENGTH 32 #define BLOCK_HEIGHT 16 #define IN(i,j) in[i + (j) * nx] #define OUT(i,j) out[i + (j) * nx] // GPU device kernel. __global__ void pattern2d_gpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { // Compute absolute (i,j) indexes for // the current GPU thread using grid mapping params. int i = blockIdx.x * BLOCK_LENGTH + threadIdx.x + bx; int j = blockIdx.y * BLOCK_HEIGHT + threadIdx.y + by; // Compute one data point - a piece of // work for the current GPU thread. OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // CPU control implementation. void pattern2d_cpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { for (int j = by; j < ny - ey; j++) for (int i = bx; i < nx - ex; i++) OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // Perform some dummy 2D field processing on GPU and CPU, // and compare results. int pattern2d(int nx, int ny, float* in, float* out, int pid, int step) { if ((nx <= 0) || (ny <= 0)) return -1; long np = nx * ny; size_t size = sizeof(float) * np; // Create GPU data array and copy input data to it. float* in_gpu; hipError_t status = hipMalloc((void**)&in_gpu, size); if (status != hipSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc input memory on GPU by process", pid, status); return status; } status = hipMemcpy(in_gpu, in, size, hipMemcpyHostToDevice); if (status != hipSuccess) { fprintf(stderr, "Cannot copy data from host to gpu by process %d, status = %d\n", pid, status); return status; } // Create CPU data output array and get // control result using CPU function. float* control = (float*)malloc(size); memset(control, 0, size); pattern2d_cpu( 1, nx, 1, 1, ny, 1, in, control); // Configure GPU computational grid: // nx = nblocks_x * block_length // ny = nblocks_y * block_height // // NOTE: we have degree of freedom in // selecting how real problem grid maps onto // computational grid. Usually these params // are tuned to get optimal performance. // // NOTE: chose of grid/block config is // also limited by device properties: // - Maximum number of threads per block (512) // - Maximum sizes of each dimension of a block (512 x 512 x 64) // - Maximum sizes of each dimension of a grid (65535 x 65535 x 1) int nblocks_x = (nx - 2) / BLOCK_LENGTH; int nblocks_y = (ny - 2) / BLOCK_HEIGHT; // Perform the same processing on GPU, // returning result to GPU array. float* out_gpu; status = hipMalloc((void**)&out_gpu, size); if (status != hipSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc output memory on GPU by process", pid, status); return status; } status = hipMemset(out_gpu, 0, size); if (status != hipSuccess) { fprintf(stderr, "Cannot erase output memory on GPU by process %d, status = %d\n", pid, status); return status; } pattern2d_gpu<<< dim3(nblocks_x, nblocks_y, 1), dim3(BLOCK_LENGTH, BLOCK_HEIGHT, 1)>>>( 1, nx, 1, 1, ny, 1, in_gpu, out_gpu); status = hipGetLastError(); if (status != hipSuccess) { fprintf(stderr, "Cannot execute CUDA kernel by process %d, status = %d\n", pid, status); return status; } status = hipDeviceSynchronize(); if (status != hipSuccess) { fprintf(stderr, "Cannot synchronize thread by process %d, status = %d\n", pid, status); return status; } // Copy GPU result from GPU memory to CPU buffer. status = hipMemcpy(out, out_gpu, size, hipMemcpyDeviceToHost); if (status != hipSuccess) { fprintf(stderr, "Cannot copy data from gpu to host by process %d, status = %d\n", pid, status); return status; } // Don't bother with processing the remainder // on GPU. Do it on CPU instead. pattern2d_cpu( 1, nx, 1, ny - (ny - 2) % BLOCK_HEIGHT - 2, ny, 1, in, out); pattern2d_cpu( nx - (nx - 2) % BLOCK_LENGTH - 2, nx, 1, 1, ny, 1, in, out); // Compare results and find the maximum abs difference. int maxi = 0, maxj = 0; float maxdiff = fabs(out[0] - control[0]); float* diffs = (float*)malloc(size); memset(diffs, 0, size); for (int j = 0; j < ny; j++) for (int i = 0; i < nx; i++) { float diff = fabs( out[i + j * nx] - control[i + j * nx]); if (diff > maxdiff) { maxdiff = diff; maxi = i; maxj = j; } diffs[i + j * nx] = diff; } // Release data arrays. status = hipFree(in_gpu); if (status != hipSuccess) { fprintf(stderr, "Cannot free device input memory by process %d, status = %d\n", pid, status); return status; } free(control); status = hipFree(out_gpu); if (status != hipSuccess) { fprintf(stderr, "Cannot free device output memory by process %d, status = %d\n", pid, status); return status; } free(diffs); printf("Step %d result abs max diff by process %d = %f @ (%d,%d)\n", step, pid, maxdiff, maxi, maxj); return 0; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z13pattern2d_gpuiiiiiiPfS_ .globl _Z13pattern2d_gpuiiiiiiPfS_ .p2align 8 .type _Z13pattern2d_gpuiiiiiiPfS_,@function _Z13pattern2d_gpuiiiiiiPfS_: s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x0 s_load_b32 s2, s[0:1], 0xc v_and_b32_e32 v1, 0x3ff, v0 v_bfe_u32 v2, v0, 10, 10 s_lshl_b32 s3, s14, 5 s_lshl_b32 s6, s15, 4 s_waitcnt lgkmcnt(0) v_add3_u32 v0, s4, s3, v1 v_add3_u32 v3, s2, s6, v2 s_load_b128 s[0:3], s[0:1], 0x18 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_mad_u64_u32 v[1:2], null, v3, s5, v[0:1] v_add_nc_u32_e32 v2, -1, v3 v_add_nc_u32_e32 v5, -1, v1 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_2) | instid1(VALU_DEP_4) v_mad_u64_u32 v[3:4], null, v2, s5, v[0:1] v_ashrrev_i32_e32 v2, 31, v1 v_add_nc_u32_e32 v7, 1, v1 v_ashrrev_i32_e32 v6, 31, v5 v_add_nc_u32_e32 v9, s5, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4) v_lshlrev_b64 v[0:1], 2, v[1:2] v_ashrrev_i32_e32 v8, 31, v7 s_delay_alu instid0(VALU_DEP_4) v_lshlrev_b64 v[5:6], 2, v[5:6] v_ashrrev_i32_e32 v4, 31, v3 v_ashrrev_i32_e32 v10, 31, v9 s_waitcnt lgkmcnt(0) v_add_co_u32 v11, vcc_lo, s0, v0 v_lshlrev_b64 v[7:8], 2, v[7:8] v_add_co_ci_u32_e32 v12, vcc_lo, s1, v1, vcc_lo v_lshlrev_b64 v[2:3], 2, v[3:4] v_add_co_u32 v4, vcc_lo, s0, v5 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v6, vcc_lo v_add_co_u32 v6, vcc_lo, s0, v7 v_lshlrev_b64 v[9:10], 2, v[9:10] v_add_co_ci_u32_e32 v7, vcc_lo, s1, v8, vcc_lo s_clause 0x1 global_load_b32 v8, v[11:12], off global_load_b32 v11, v[4:5], off v_add_co_u32 v2, vcc_lo, s0, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s1, v3, vcc_lo global_load_b32 v6, v[6:7], off v_add_co_u32 v4, vcc_lo, s0, v9 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v10, vcc_lo s_clause 0x1 global_load_b32 v2, v[2:3], off global_load_b32 v3, v[4:5], off s_waitcnt vmcnt(3) v_add_f32_e32 v4, v8, v11 s_waitcnt vmcnt(2) s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_add_f32_e32 v4, v4, v6 s_waitcnt vmcnt(1) v_fmac_f32_e32 v4, -2.0, v2 s_waitcnt vmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fmamk_f32 v2, v3, 0x40400000, v4 v_mul_f32_e64 v3, 0x4f800000, |v2| v_cmp_gt_f32_e64 vcc_lo, 0xf800000, |v2| s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cndmask_b32_e64 v2, |v2|, v3, vcc_lo v_sqrt_f32_e32 v3, v2 s_waitcnt_depctr 0xfff v_add_nc_u32_e32 v4, -1, v3 v_add_nc_u32_e32 v5, 1, v3 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2) v_fma_f32 v6, -v4, v3, v2 v_fma_f32 v7, -v5, v3, v2 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_ge_f32_e64 s0, 0, v6 v_cndmask_b32_e64 v3, v3, v4, s0 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_lt_f32_e64 s0, 0, v7 v_cndmask_b32_e64 v3, v3, v5, s0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mul_f32_e32 v4, 0x37800000, v3 v_cndmask_b32_e32 v3, v3, v4, vcc_lo v_cmp_class_f32_e64 vcc_lo, v2, 0x260 s_delay_alu instid0(VALU_DEP_2) v_cndmask_b32_e32 v2, v3, v2, 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_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 _Z13pattern2d_gpuiiiiiiPfS_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 40 .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 13 .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 _Z13pattern2d_gpuiiiiiiPfS_, .Lfunc_end0-_Z13pattern2d_gpuiiiiiiPfS_ .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 - .offset: 8 .size: 4 .value_kind: by_value - .offset: 12 .size: 4 .value_kind: by_value - .offset: 16 .size: 4 .value_kind: by_value - .offset: 20 .size: 4 .value_kind: by_value - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 32 .size: 8 .value_kind: global_buffer .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 40 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z13pattern2d_gpuiiiiiiPfS_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z13pattern2d_gpuiiiiiiPfS_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 13 .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.
/* * MSU CUDA Course Examples and Exercises. * * Copyright (c) 2011 Dmitry Mikushin * * This software is provided 'as-is', without any express or implied warranty. * In no event will the authors be held liable for any damages arising * from the use of this software. * Permission is granted to anyone to use this software for any purpose, * including commercial applications, and to alter it and redistribute it freely, * without any restrictons. */ #include <hip/hip_runtime.h> #include <malloc.h> #include <math.h> #include <stdio.h> #include <stdlib.h> // Maximum allowed results difference. #define EPS 1e-04 // Grid block size (see comment in pattern2d below). #define BLOCK_LENGTH 32 #define BLOCK_HEIGHT 16 #define IN(i,j) in[i + (j) * nx] #define OUT(i,j) out[i + (j) * nx] // GPU device kernel. __global__ void pattern2d_gpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { // Compute absolute (i,j) indexes for // the current GPU thread using grid mapping params. int i = blockIdx.x * BLOCK_LENGTH + threadIdx.x + bx; int j = blockIdx.y * BLOCK_HEIGHT + threadIdx.y + by; // Compute one data point - a piece of // work for the current GPU thread. OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // CPU control implementation. void pattern2d_cpu( int bx, int nx, int ex, int by, int ny, int ey, float* in, float* out) { for (int j = by; j < ny - ey; j++) for (int i = bx; i < nx - ex; i++) OUT(i,j) = sqrtf(fabs(IN(i,j) + IN(i-1,j) + IN(i+1,j) - 2.0f * IN(i,j-1) + 3.0f * IN(i,j+1))); } // Perform some dummy 2D field processing on GPU and CPU, // and compare results. int pattern2d(int nx, int ny, float* in, float* out, int pid, int step) { if ((nx <= 0) || (ny <= 0)) return -1; long np = nx * ny; size_t size = sizeof(float) * np; // Create GPU data array and copy input data to it. float* in_gpu; hipError_t status = hipMalloc((void**)&in_gpu, size); if (status != hipSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc input memory on GPU by process", pid, status); return status; } status = hipMemcpy(in_gpu, in, size, hipMemcpyHostToDevice); if (status != hipSuccess) { fprintf(stderr, "Cannot copy data from host to gpu by process %d, status = %d\n", pid, status); return status; } // Create CPU data output array and get // control result using CPU function. float* control = (float*)malloc(size); memset(control, 0, size); pattern2d_cpu( 1, nx, 1, 1, ny, 1, in, control); // Configure GPU computational grid: // nx = nblocks_x * block_length // ny = nblocks_y * block_height // // NOTE: we have degree of freedom in // selecting how real problem grid maps onto // computational grid. Usually these params // are tuned to get optimal performance. // // NOTE: chose of grid/block config is // also limited by device properties: // - Maximum number of threads per block (512) // - Maximum sizes of each dimension of a block (512 x 512 x 64) // - Maximum sizes of each dimension of a grid (65535 x 65535 x 1) int nblocks_x = (nx - 2) / BLOCK_LENGTH; int nblocks_y = (ny - 2) / BLOCK_HEIGHT; // Perform the same processing on GPU, // returning result to GPU array. float* out_gpu; status = hipMalloc((void**)&out_gpu, size); if (status != hipSuccess) { fprintf(stderr, "%s %d, status = %d\nInsufficient GPU memory?\n", "Cannot malloc output memory on GPU by process", pid, status); return status; } status = hipMemset(out_gpu, 0, size); if (status != hipSuccess) { fprintf(stderr, "Cannot erase output memory on GPU by process %d, status = %d\n", pid, status); return status; } pattern2d_gpu<<< dim3(nblocks_x, nblocks_y, 1), dim3(BLOCK_LENGTH, BLOCK_HEIGHT, 1)>>>( 1, nx, 1, 1, ny, 1, in_gpu, out_gpu); status = hipGetLastError(); if (status != hipSuccess) { fprintf(stderr, "Cannot execute CUDA kernel by process %d, status = %d\n", pid, status); return status; } status = hipDeviceSynchronize(); if (status != hipSuccess) { fprintf(stderr, "Cannot synchronize thread by process %d, status = %d\n", pid, status); return status; } // Copy GPU result from GPU memory to CPU buffer. status = hipMemcpy(out, out_gpu, size, hipMemcpyDeviceToHost); if (status != hipSuccess) { fprintf(stderr, "Cannot copy data from gpu to host by process %d, status = %d\n", pid, status); return status; } // Don't bother with processing the remainder // on GPU. Do it on CPU instead. pattern2d_cpu( 1, nx, 1, ny - (ny - 2) % BLOCK_HEIGHT - 2, ny, 1, in, out); pattern2d_cpu( nx - (nx - 2) % BLOCK_LENGTH - 2, nx, 1, 1, ny, 1, in, out); // Compare results and find the maximum abs difference. int maxi = 0, maxj = 0; float maxdiff = fabs(out[0] - control[0]); float* diffs = (float*)malloc(size); memset(diffs, 0, size); for (int j = 0; j < ny; j++) for (int i = 0; i < nx; i++) { float diff = fabs( out[i + j * nx] - control[i + j * nx]); if (diff > maxdiff) { maxdiff = diff; maxi = i; maxj = j; } diffs[i + j * nx] = diff; } // Release data arrays. status = hipFree(in_gpu); if (status != hipSuccess) { fprintf(stderr, "Cannot free device input memory by process %d, status = %d\n", pid, status); return status; } free(control); status = hipFree(out_gpu); if (status != hipSuccess) { fprintf(stderr, "Cannot free device output memory by process %d, status = %d\n", pid, status); return status; } free(diffs); printf("Step %d result abs max diff by process %d = %f @ (%d,%d)\n", step, pid, maxdiff, maxi, maxj); return 0; }
.text .file "pattern2d.hip" .globl _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ # -- Begin function _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .p2align 4, 0x90 .type _Z28__device_stub__pattern2d_gpuiiiiiiPfS_,@function _Z28__device_stub__pattern2d_gpuiiiiiiPfS_: # @_Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .cfi_startproc # %bb.0: subq $152, %rsp .cfi_def_cfa_offset 160 movl %edi, 28(%rsp) movl %esi, 24(%rsp) movl %edx, 20(%rsp) movl %ecx, 16(%rsp) movl %r8d, 12(%rsp) movl %r9d, 8(%rsp) leaq 28(%rsp), %rax movq %rax, 80(%rsp) leaq 24(%rsp), %rax movq %rax, 88(%rsp) leaq 20(%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 160(%rsp), %rax movq %rax, 128(%rsp) leaq 168(%rsp), %rax movq %rax, 136(%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 80(%rsp), %r9 movl $_Z13pattern2d_gpuiiiiiiPfS_, %edi pushq 32(%rsp) .cfi_adjust_cfa_offset 8 pushq 48(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $168, %rsp .cfi_adjust_cfa_offset -168 retq .Lfunc_end0: .size _Z28__device_stub__pattern2d_gpuiiiiiiPfS_, .Lfunc_end0-_Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function _Z13pattern2d_cpuiiiiiiPfS_ .LCPI1_0: .long 0x40400000 # float 3 .section .rodata.cst16,"aM",@progbits,16 .p2align 4, 0x0 .LCPI1_1: .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .text .globl _Z13pattern2d_cpuiiiiiiPfS_ .p2align 4, 0x90 .type _Z13pattern2d_cpuiiiiiiPfS_,@function _Z13pattern2d_cpuiiiiiiPfS_: # @_Z13pattern2d_cpuiiiiiiPfS_ .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 $88, %rsp .cfi_def_cfa_offset 144 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 # kill: def $edi killed $edi def $rdi movq %rdi, 16(%rsp) # 8-byte Spill subl %r9d, %r8d cmpl %ecx, %r8d jle .LBB1_4 # %bb.1: # %.preheader.lr.ph movq 152(%rsp), %rax movq 144(%rsp), %rdi movl $4294967295, %r9d # imm = 0xFFFFFFFF movslq %esi, %r10 subl %edx, %esi movq 16(%rsp), %rdx # 8-byte Reload movslq %edx, %r11 movq %r11, 40(%rsp) # 8-byte Spill movl %esi, 12(%rsp) # 4-byte Spill movslq %esi, %r13 movslq %ecx, %r11 movslq %r8d, %rcx movq %rcx, 32(%rsp) # 8-byte Spill leaq 1(%r11), %rcx imulq %r10, %rcx leaq (%rdi,%rcx,4), %r15 leaq (,%r10,4), %rcx movq %rcx, 24(%rsp) # 8-byte Spill leaq -1(%r11), %rcx imulq %r10, %rcx leaq (%rdi,%rcx,4), %rbx movq %r11, %rcx imulq %r10, %rcx leal (%r9,%rdx), %r14d addl %ecx, %r14d shlq $32, %r14 shlq $32, %r10 movq %r10, 48(%rsp) # 8-byte Spill leaq (%rax,%rcx,4), %rdx leaq (%rdi,%rcx,4), %r12 movss .LCPI1_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI1_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 jmp .LBB1_2 .p2align 4, 0x90 .LBB1_3: # %._crit_edge # in Loop: Header=BB1_2 Depth=1 movq 64(%rsp), %r11 # 8-byte Reload incq %r11 movq 24(%rsp), %rax # 8-byte Reload addq %rax, %r15 addq %rax, %rbx movq 56(%rsp), %r14 # 8-byte Reload addq 48(%rsp), %r14 # 8-byte Folded Reload addq %rax, %r12 addq %rax, %rdx cmpq 32(%rsp), %r11 # 8-byte Folded Reload jge .LBB1_4 .LBB1_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB1_5 Depth 2 movq %r11, 64(%rsp) # 8-byte Spill movq %r14, 56(%rsp) # 8-byte Spill movq 40(%rsp), %rbp # 8-byte Reload movq 16(%rsp), %rax # 8-byte Reload cmpl %eax, 12(%rsp) # 4-byte Folded Reload movq %rbx, 80(%rsp) # 8-byte Spill movq %rdx, 72(%rsp) # 8-byte Spill jg .LBB1_5 jmp .LBB1_3 .p2align 4, 0x90 .LBB1_7: # %call.sqrt # in Loop: Header=BB1_5 Depth=2 movq %rdi, %rbx callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI1_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI1_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %rdx # 8-byte Reload movl $4294967295, %r9d # imm = 0xFFFFFFFF movq %rbx, %rdi movq 80(%rsp), %rbx # 8-byte Reload .LBB1_8: # %.split # in Loop: Header=BB1_5 Depth=2 movss %xmm0, (%rdx,%rbp,4) incq %rbp addq %r9, %r14 incq %r14 cmpq %r13, %rbp jge .LBB1_3 .LBB1_5: # Parent Loop BB1_2 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r12,%rbp,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %r14, %rax sarq $30, %rax addss (%rdi,%rax), %xmm1 addss 4(%r12,%rbp,4), %xmm1 movss (%rbx,%rbp,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r15,%rbp,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB1_7 # %bb.6: # in Loop: Header=BB1_5 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB1_8 .LBB1_4: # %._crit_edge36 addq $88, %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_end1: .size _Z13pattern2d_cpuiiiiiiPfS_, .Lfunc_end1-_Z13pattern2d_cpuiiiiiiPfS_ .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function _Z9pattern2diiPfS_ii .LCPI2_0: .long 0x40400000 # float 3 .section .rodata.cst16,"aM",@progbits,16 .p2align 4, 0x0 .LCPI2_1: .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .text .globl _Z9pattern2diiPfS_ii .p2align 4, 0x90 .type _Z9pattern2diiPfS_ii,@function _Z9pattern2diiPfS_ii: # @_Z9pattern2diiPfS_ii .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 $312, %rsp # imm = 0x138 .cfi_def_cfa_offset 368 .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, 80(%rsp) # 8-byte Spill movl $-1, %r12d testl %edi, %edi jle .LBB2_62 # %bb.1: movl %esi, %r13d testl %esi, %esi jle .LBB2_62 # %bb.2: movl %r9d, %ebp movq %rcx, %r14 movl %edi, %r15d movl %r8d, 4(%rsp) # 4-byte Spill movl %r13d, %eax imull %edi, %eax movslq %eax, %rbx shlq $2, %rbx leaq 128(%rsp), %rdi movq %rbx, %rsi callq hipMalloc testl %eax, %eax jne .LBB2_3 # %bb.5: movq 128(%rsp), %rdi movq 80(%rsp), %rsi # 8-byte Reload movq %rbx, %rdx movl $1, %ecx callq hipMemcpy testl %eax, %eax jne .LBB2_6 # %bb.7: movq %r14, 104(%rsp) # 8-byte Spill movl %ebp, 136(%rsp) # 4-byte Spill movq %rbx, %rdi callq malloc movq %rax, 168(%rsp) # 8-byte Spill movq %rax, %rdi xorl %esi, %esi movq %rbx, 40(%rsp) # 8-byte Spill movq %rbx, %rdx callq memset@PLT leal -1(%r13), %ecx movq %r15, 16(%rsp) # 8-byte Spill movslq %r15d, %rax movq %rax, 48(%rsp) # 8-byte Spill movl %ecx, 140(%rsp) # 4-byte Spill movslq %ecx, %rax movq %rax, 120(%rsp) # 8-byte Spill movq %r13, 32(%rsp) # 8-byte Spill cmpl $3, %r13d jl .LBB2_16 # %bb.8: # %.preheader.lr.ph.i movq 80(%rsp), %rax # 8-byte Reload movq 48(%rsp), %rcx # 8-byte Reload leaq (%rax,%rcx,8), %r14 addq $4, %r14 leaq (,%rcx,4), %rdx movq %rdx, 64(%rsp) # 8-byte Spill leaq 4(%rax), %r15 leaq -2(%rcx), %rbx leaq (%rax,%rcx,4), %r10 addq $8, %r10 movq %rcx, %r8 shlq $32, %r8 movq 168(%rsp), %rax # 8-byte Reload leaq (%rax,%rcx,4), %r13 addq $4, %r13 movl $1, %ecx movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 movq %r8, 56(%rsp) # 8-byte Spill movq %r8, 8(%rsp) # 8-byte Spill jmp .LBB2_9 .p2align 4, 0x90 .LBB2_15: # %._crit_edge.i # in Loop: Header=BB2_9 Depth=1 movq 24(%rsp), %rcx # 8-byte Reload incq %rcx movq 64(%rsp), %rax # 8-byte Reload addq %rax, %r14 addq %rax, %r15 addq %rax, %r10 movq 8(%rsp), %r8 # 8-byte Reload addq 56(%rsp), %r8 # 8-byte Folded Reload movq %r8, 8(%rsp) # 8-byte Spill addq %rax, %r13 cmpq 120(%rsp), %rcx # 8-byte Folded Reload je .LBB2_16 .LBB2_9: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB2_11 Depth 2 movq %rcx, 24(%rsp) # 8-byte Spill cmpl $3, 16(%rsp) # 4-byte Folded Reload jl .LBB2_15 # %bb.10: # %.lr.ph.i # in Loop: Header=BB2_9 Depth=1 movq 8(%rsp), %rbp # 8-byte Reload xorl %r12d, %r12d movq %r10, 72(%rsp) # 8-byte Spill jmp .LBB2_11 .p2align 4, 0x90 .LBB2_13: # %call.sqrt # in Loop: Header=BB2_11 Depth=2 callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %r10 # 8-byte Reload .LBB2_14: # %.split # in Loop: Header=BB2_11 Depth=2 movss %xmm0, (%r13,%r12,4) incq %r12 movabsq $4294967296, %rax # imm = 0x100000000 addq %rax, %rbp cmpq %r12, %rbx je .LBB2_15 .LBB2_11: # Parent Loop BB2_9 Depth=1 # => This Inner Loop Header: Depth=2 movss -4(%r10,%r12,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %rbp, %rax sarq $30, %rax movq 80(%rsp), %rcx # 8-byte Reload addss (%rcx,%rax), %xmm1 addss (%r10,%r12,4), %xmm1 movss (%r15,%r12,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r14,%r12,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB2_13 # %bb.12: # in Loop: Header=BB2_11 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB2_14 .LBB2_16: # %_Z13pattern2d_cpuiiiiiiPfS_.exit movq 16(%rsp), %rbx # 8-byte Reload leal -2(%rbx), %eax leal 29(%rbx), %r15d testl %eax, %eax cmovnsl %eax, %r15d movq 32(%rsp), %rcx # 8-byte Reload leal -2(%rcx), %eax leal 13(%rcx), %r13d testl %eax, %eax cmovnsl %eax, %r13d leaq 112(%rsp), %rdi movq 40(%rsp), %r14 # 8-byte Reload movq %r14, %rsi callq hipMalloc testl %eax, %eax jne .LBB2_17 # %bb.18: movq 112(%rsp), %rdi xorl %esi, %esi movq %r14, %rdx callq hipMemset testl %eax, %eax jne .LBB2_19 # %bb.21: movl %r15d, %eax sarl $5, %eax movl %r13d, %ecx sarl $4, %ecx shlq $32, %rcx movl %eax, %edi orq %rcx, %rdi movabsq $68719476768, %rdx # imm = 0x1000000020 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB2_23 # %bb.22: movq 128(%rsp), %rax movq 112(%rsp), %rcx movl $1, 164(%rsp) movl %ebx, 160(%rsp) movl $1, 156(%rsp) movl $1, 152(%rsp) movq 32(%rsp), %rdx # 8-byte Reload movl %edx, 148(%rsp) movl $1, 144(%rsp) movq %rax, 232(%rsp) movq %rcx, 224(%rsp) leaq 164(%rsp), %rax movq %rax, 240(%rsp) leaq 160(%rsp), %rax movq %rax, 248(%rsp) leaq 156(%rsp), %rax movq %rax, 256(%rsp) leaq 152(%rsp), %rax movq %rax, 264(%rsp) leaq 148(%rsp), %rax movq %rax, 272(%rsp) leaq 144(%rsp), %rax movq %rax, 280(%rsp) leaq 232(%rsp), %rax movq %rax, 288(%rsp) leaq 224(%rsp), %rax movq %rax, 296(%rsp) leaq 208(%rsp), %rdi leaq 192(%rsp), %rsi leaq 184(%rsp), %rdx leaq 176(%rsp), %rcx callq __hipPopCallConfiguration movq 208(%rsp), %rsi movl 216(%rsp), %edx movq 192(%rsp), %rcx movl 200(%rsp), %r8d leaq 240(%rsp), %r9 movl $_Z13pattern2d_gpuiiiiiiPfS_, %edi pushq 176(%rsp) .cfi_adjust_cfa_offset 8 pushq 192(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB2_23: callq hipGetLastError testl %eax, %eax jne .LBB2_24 # %bb.25: callq hipDeviceSynchronize testl %eax, %eax jne .LBB2_26 # %bb.27: movq 112(%rsp), %rsi movq 104(%rsp), %rdi # 8-byte Reload movq %r14, %rdx movl $2, %ecx callq hipMemcpy testl %eax, %eax jne .LBB2_28 # %bb.29: movq %r15, 40(%rsp) # 8-byte Spill andl $-16, %r13d negl %r13d movq 32(%rsp), %rax # 8-byte Reload leal (%rax,%r13), %ecx addl $-2, %ecx # kill: def $eax killed $eax killed $rax subl %ecx, %eax addl $-2, %eax cmpl %eax, 140(%rsp) # 4-byte Folded Reload jle .LBB2_38 # %bb.30: # %.preheader.lr.ph.i144 movq 16(%rsp), %rcx # 8-byte Reload decl %ecx movslq %ecx, %r15 movslq %eax, %rdi leaq 1(%rdi), %rax movq 48(%rsp), %rdx # 8-byte Reload imulq %rdx, %rax movq 80(%rsp), %rcx # 8-byte Reload leaq (%rcx,%rax,4), %r13 addq $4, %r13 leaq (,%rdx,4), %rax movq %rax, 64(%rsp) # 8-byte Spill leaq -1(%rdi), %rax imulq %rdx, %rax leaq (%rcx,%rax,4), %r14 addq $4, %r14 decq %r15 movq %rdx, %r10 imulq %rdi, %r10 leaq (%rcx,%r10,4), %r11 addq $8, %r11 shlq $32, %rdx movq %rdx, 56(%rsp) # 8-byte Spill movq 104(%rsp), %rax # 8-byte Reload leaq (%rax,%r10,4), %r12 addq $4, %r12 shlq $32, %r10 movq %r10, 24(%rsp) # 8-byte Spill movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 jmp .LBB2_31 .p2align 4, 0x90 .LBB2_37: # %._crit_edge.i147 # in Loop: Header=BB2_31 Depth=1 movq 8(%rsp), %rdi # 8-byte Reload incq %rdi movq 64(%rsp), %rax # 8-byte Reload addq %rax, %r13 addq %rax, %r14 addq %rax, %r11 movq 56(%rsp), %rcx # 8-byte Reload addq %rcx, 24(%rsp) # 8-byte Folded Spill addq %rax, %r12 cmpq 120(%rsp), %rdi # 8-byte Folded Reload je .LBB2_38 .LBB2_31: # %.preheader.i145 # =>This Loop Header: Depth=1 # Child Loop BB2_33 Depth 2 movq %rdi, 8(%rsp) # 8-byte Spill cmpl $3, 16(%rsp) # 4-byte Folded Reload jl .LBB2_37 # %bb.32: # %.lr.ph.i149 # in Loop: Header=BB2_31 Depth=1 movq 24(%rsp), %rbp # 8-byte Reload xorl %ebx, %ebx movq %r11, 72(%rsp) # 8-byte Spill jmp .LBB2_33 .p2align 4, 0x90 .LBB2_35: # %call.sqrt273 # in Loop: Header=BB2_33 Depth=2 callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %r11 # 8-byte Reload .LBB2_36: # %.split272 # in Loop: Header=BB2_33 Depth=2 movss %xmm0, (%r12,%rbx,4) incq %rbx movabsq $4294967296, %rax # imm = 0x100000000 addq %rax, %rbp cmpq %rbx, %r15 je .LBB2_37 .LBB2_33: # Parent Loop BB2_31 Depth=1 # => This Inner Loop Header: Depth=2 movss -4(%r11,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %rbp, %rax sarq $30, %rax movq 80(%rsp), %rcx # 8-byte Reload addss (%rcx,%rax), %xmm1 addss (%r11,%rbx,4), %xmm1 movss (%r14,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r13,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB2_35 # %bb.34: # in Loop: Header=BB2_33 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB2_36 .LBB2_38: # %_Z13pattern2d_cpuiiiiiiPfS_.exit153 cmpl $3, 32(%rsp) # 4-byte Folded Reload jl .LBB2_47 # %bb.39: # %.preheader.lr.ph.i154 movq 40(%rsp), %rcx # 8-byte Reload andl $-32, %ecx negl %ecx movq 16(%rsp), %rdx # 8-byte Reload leal (%rdx,%rcx), %eax addl $-2, %eax movl %edx, %ecx subl %eax, %ecx movslq %ecx, %rax addl $-2, %ecx movl %ecx, 64(%rsp) # 4-byte Spill leaq -2(%rax), %rcx decl %edx movl %edx, 56(%rsp) # 4-byte Spill movslq %edx, %r13 subq %rcx, %r13 leaq -8(,%rax,4), %rcx movq 48(%rsp), %rsi # 8-byte Reload leaq (%rcx,%rsi,8), %r15 movq 80(%rsp), %rdx # 8-byte Reload addq %rdx, %r15 leaq (,%rsi,4), %r9 movq %r9, 40(%rsp) # 8-byte Spill leaq (%rdx,%rax,4), %r14 addq $-8, %r14 leaq (%rcx,%rsi,4), %r10 addq %rdx, %r10 addq %rsi, %rax addq $-2, %rax movabsq $4294967296, %rcx # imm = 0x100000000 addl %eax, %ecx decl %ecx shlq $32, %rcx movq %rcx, 8(%rsp) # 8-byte Spill shlq $32, %rsi movq %rsi, 48(%rsp) # 8-byte Spill movq 104(%rsp), %rcx # 8-byte Reload leaq (%rcx,%rax,4), %r12 movl $1, %esi movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 jmp .LBB2_40 .p2align 4, 0x90 .LBB2_46: # %._crit_edge.i157 # in Loop: Header=BB2_40 Depth=1 movq 24(%rsp), %rsi # 8-byte Reload incq %rsi movq 40(%rsp), %rax # 8-byte Reload addq %rax, %r15 addq %rax, %r14 addq %rax, %r10 movq 8(%rsp), %rcx # 8-byte Reload addq 48(%rsp), %rcx # 8-byte Folded Reload movq %rcx, 8(%rsp) # 8-byte Spill addq %rax, %r12 cmpq 120(%rsp), %rsi # 8-byte Folded Reload je .LBB2_47 .LBB2_40: # %.preheader.i155 # =>This Loop Header: Depth=1 # Child Loop BB2_42 Depth 2 movq %rsi, 24(%rsp) # 8-byte Spill movl 56(%rsp), %eax # 4-byte Reload cmpl 64(%rsp), %eax # 4-byte Folded Reload jle .LBB2_46 # %bb.41: # %.lr.ph.i159 # in Loop: Header=BB2_40 Depth=1 movq 8(%rsp), %rbp # 8-byte Reload xorl %ebx, %ebx movq %r10, 72(%rsp) # 8-byte Spill jmp .LBB2_42 .p2align 4, 0x90 .LBB2_44: # %call.sqrt275 # in Loop: Header=BB2_42 Depth=2 callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %r10 # 8-byte Reload .LBB2_45: # %.split274 # in Loop: Header=BB2_42 Depth=2 movss %xmm0, (%r12,%rbx,4) incq %rbx movabsq $4294967296, %rax # imm = 0x100000000 addq %rax, %rbp cmpq %rbx, %r13 je .LBB2_46 .LBB2_42: # Parent Loop BB2_40 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r10,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %rbp, %rax sarq $30, %rax movq 80(%rsp), %rcx # 8-byte Reload addss (%rcx,%rax), %xmm1 addss 4(%r10,%rbx,4), %xmm1 movss (%r14,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r15,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB2_44 # %bb.43: # in Loop: Header=BB2_42 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB2_45 .LBB2_47: # %_Z13pattern2d_cpuiiiiiiPfS_.exit163 movq 104(%rsp), %r10 # 8-byte Reload movss (%r10), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 168(%rsp), %r15 # 8-byte Reload subss (%r15), %xmm2 andps .LCPI2_1(%rip), %xmm2 movq 32(%rsp), %rax # 8-byte Reload testl %eax, %eax jle .LBB2_48 # %bb.49: # %.preheader.lr.ph movl %eax, %eax movq 16(%rsp), %r11 # 8-byte Reload movl %r11d, %ecx xorl %edx, %edx movaps .LCPI2_1(%rip), %xmm0 # xmm0 = [NaN,NaN,NaN,NaN] xorl %esi, %esi movaps %xmm2, %xmm1 xorl %ebp, %ebp xorl %ebx, %ebx movl 4(%rsp), %r14d # 4-byte Reload jmp .LBB2_50 .p2align 4, 0x90 .LBB2_51: # in Loop: Header=BB2_50 Depth=1 movaps %xmm1, %xmm2 .LBB2_52: # %._crit_edge # in Loop: Header=BB2_50 Depth=1 incq %rsi addl %r11d, %edx movaps %xmm2, %xmm1 cmpq %rax, %rsi je .LBB2_53 .LBB2_50: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB2_58 Depth 2 testl %r11d, %r11d jle .LBB2_51 # %bb.57: # %.lr.ph # in Loop: Header=BB2_50 Depth=1 movl %edx, %r8d leaq (%r15,%r8,4), %rdi leaq (%r10,%r8,4), %r8 xorl %r9d, %r9d .p2align 4, 0x90 .LBB2_58: # Parent Loop BB2_50 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r8,%r9,4), %xmm2 # xmm2 = mem[0],zero,zero,zero subss (%rdi,%r9,4), %xmm2 andps %xmm0, %xmm2 ucomiss %xmm1, %xmm2 cmoval %r9d, %ebx cmoval %esi, %ebp maxss %xmm1, %xmm2 incq %r9 movaps %xmm2, %xmm1 cmpq %r9, %rcx jne .LBB2_58 jmp .LBB2_52 .LBB2_48: xorl %ebx, %ebx xorl %ebp, %ebp movl 4(%rsp), %r14d # 4-byte Reload .LBB2_53: # %._crit_edge181 movaps %xmm2, 80(%rsp) # 16-byte Spill movq 128(%rsp), %rdi callq hipFree testl %eax, %eax jne .LBB2_54 # %bb.59: movq %r15, %rdi callq free movq 112(%rsp), %rdi callq hipFree testl %eax, %eax jne .LBB2_60 # %bb.61: movaps 80(%rsp), %xmm0 # 16-byte Reload cvtss2sd %xmm0, %xmm0 movl $.L.str.10, %edi movl 136(%rsp), %esi # 4-byte Reload movl %r14d, %edx movl %ebx, %ecx movl %ebp, %r8d movb $1, %al callq printf xorl %r12d, %r12d jmp .LBB2_62 .LBB2_3: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str, %esi movl $.L.str.1, %edx jmp .LBB2_4 .LBB2_6: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.2, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_17: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str, %esi movl $.L.str.3, %edx .LBB2_4: movl 4(%rsp), %ecx # 4-byte Reload movl %r12d, %r8d xorl %eax, %eax callq fprintf jmp .LBB2_62 .LBB2_19: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.4, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_24: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.5, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_26: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.6, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_28: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.7, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_54: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.8, %esi jmp .LBB2_55 .LBB2_60: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.9, %esi .LBB2_55: movl %r14d, %edx .LBB2_56: movl %r12d, %ecx xorl %eax, %eax callq fprintf .LBB2_62: movl %r12d, %eax addq $312, %rsp # imm = 0x138 .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_end2: .size _Z9pattern2diiPfS_ii, .Lfunc_end2-_Z9pattern2diiPfS_ii .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 .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 $_Z13pattern2d_gpuiiiiiiPfS_, %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 _Z13pattern2d_gpuiiiiiiPfS_,@object # @_Z13pattern2d_gpuiiiiiiPfS_ .section .rodata,"a",@progbits .globl _Z13pattern2d_gpuiiiiiiPfS_ .p2align 3, 0x0 _Z13pattern2d_gpuiiiiiiPfS_: .quad _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .size _Z13pattern2d_gpuiiiiiiPfS_, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%s %d, status = %d\nInsufficient GPU memory?\n" .size .L.str, 45 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "Cannot malloc input memory on GPU by process" .size .L.str.1, 45 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "Cannot copy data from host to gpu by process %d, status = %d\n" .size .L.str.2, 62 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "Cannot malloc output memory on GPU by process" .size .L.str.3, 46 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "Cannot erase output memory on GPU by process %d, status = %d\n" .size .L.str.4, 62 .type .L.str.5,@object # @.str.5 .L.str.5: .asciz "Cannot execute CUDA kernel by process %d, status = %d\n" .size .L.str.5, 55 .type .L.str.6,@object # @.str.6 .L.str.6: .asciz "Cannot synchronize thread by process %d, status = %d\n" .size .L.str.6, 54 .type .L.str.7,@object # @.str.7 .L.str.7: .asciz "Cannot copy data from gpu to host by process %d, status = %d\n" .size .L.str.7, 62 .type .L.str.8,@object # @.str.8 .L.str.8: .asciz "Cannot free device input memory by process %d, status = %d\n" .size .L.str.8, 60 .type .L.str.9,@object # @.str.9 .L.str.9: .asciz "Cannot free device output memory by process %d, status = %d\n" .size .L.str.9, 61 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "Step %d result abs max diff by process %d = %f @ (%d,%d)\n" .size .L.str.10, 58 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z13pattern2d_gpuiiiiiiPfS_" .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 _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z13pattern2d_gpuiiiiiiPfS_ .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 : _Z13pattern2d_gpuiiiiiiPfS_ .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.Y ; /* 0x0000000000027919 */ /* 0x000e220000002600 */ /*0020*/ MOV R10, c[0x0][0x164] ; /* 0x00005900000a7a02 */ /* 0x000fe20000000f00 */ /*0030*/ HFMA2.MMA R7, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff077435 */ /* 0x000fe200000001ff */ /*0040*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0050*/ S2R R5, SR_TID.Y ; /* 0x0000000000057919 */ /* 0x000e280000002200 */ /*0060*/ S2R R0, SR_CTAID.X ; /* 0x0000000000007919 */ /* 0x000e680000002500 */ /*0070*/ S2R R3, SR_TID.X ; /* 0x0000000000037919 */ /* 0x000e620000002100 */ /*0080*/ LEA R2, R2, R5, 0x4 ; /* 0x0000000502027211 */ /* 0x001fc400078e20ff */ /*0090*/ LEA R0, R0, R3, 0x5 ; /* 0x0000000300007211 */ /* 0x002fe400078e28ff */ /*00a0*/ IADD3 R3, R2, c[0x0][0x16c], RZ ; /* 0x00005b0002037a10 */ /* 0x000fe40007ffe0ff */ /*00b0*/ IADD3 R4, R0, c[0x0][0x160], RZ ; /* 0x0000580000047a10 */ /* 0x000fc60007ffe0ff */ /*00c0*/ IMAD R5, R3.reuse, R10, -c[0x0][0x164] ; /* 0x8000590003057624 */ /* 0x040fe400078e020a */ /*00d0*/ IMAD R0, R3, c[0x0][0x164], R4 ; /* 0x0000590003007a24 */ /* 0x000fc600078e0204 */ /*00e0*/ IADD3 R6, R4, R5, RZ ; /* 0x0000000504067210 */ /* 0x000fe20007ffe0ff */ /*00f0*/ IMAD.WIDE R2, R0, R7, c[0x0][0x178] ; /* 0x00005e0000027625 */ /* 0x000fc800078e0207 */ /*0100*/ IMAD.WIDE R4, R6, R7, c[0x0][0x178] ; /* 0x00005e0006047625 */ /* 0x000fe200078e0207 */ /*0110*/ LDG.E R8, [R2.64] ; /* 0x0000000402087981 */ /* 0x000ea2000c1e1900 */ /*0120*/ LEA R6, R10, R6, 0x1 ; /* 0x000000060a067211 */ /* 0x000fc600078e08ff */ /*0130*/ LDG.E R9, [R2.64+-0x4] ; /* 0xfffffc0402097981 */ /* 0x000ea4000c1e1900 */ /*0140*/ IMAD.WIDE R6, R6, R7, c[0x0][0x178] ; /* 0x00005e0006067625 */ /* 0x000fe400078e0207 */ /*0150*/ LDG.E R11, [R2.64+0x4] ; /* 0x00000404020b7981 */ /* 0x000ee8000c1e1900 */ /*0160*/ LDG.E R4, [R4.64] ; /* 0x0000000404047981 */ /* 0x000f28000c1e1900 */ /*0170*/ LDG.E R7, [R6.64] ; /* 0x0000000406077981 */ /* 0x000f62000c1e1900 */ /*0180*/ BSSY B0, 0x2b0 ; /* 0x0000012000007945 */ /* 0x000fe20003800000 */ /*0190*/ FADD R8, R8, R9 ; /* 0x0000000908087221 */ /* 0x004fc80000000000 */ /*01a0*/ FADD R8, R8, R11 ; /* 0x0000000b08087221 */ /* 0x008fe40000000000 */ /*01b0*/ FADD R9, R4, R4 ; /* 0x0000000404097221 */ /* 0x010fc80000000000 */ /*01c0*/ FADD R8, R8, -R9 ; /* 0x8000000908087221 */ /* 0x000fc80000000000 */ /*01d0*/ FFMA R8, R7, 3, R8 ; /* 0x4040000007087823 */ /* 0x020fc80000000008 */ /*01e0*/ FADD R10, |R8|, -RZ ; /* 0x800000ff080a7221 */ /* 0x000fe20000000200 */ /*01f0*/ MUFU.RSQ R3, |R8| ; /* 0x4000000800037308 */ /* 0x0000680000001400 */ /*0200*/ IADD3 R9, R10, -0xd000000, RZ ; /* 0xf30000000a097810 */ /* 0x000fc80007ffe0ff */ /*0210*/ ISETP.GT.U32.AND P0, PT, R9, 0x727fffff, PT ; /* 0x727fffff0900780c */ /* 0x000fda0003f04070 */ /*0220*/ @!P0 BRA 0x260 ; /* 0x0000003000008947 */ /* 0x000fea0003800000 */ /*0230*/ MOV R7, 0x250 ; /* 0x0000025000077802 */ /* 0x003fe40000000f00 */ /*0240*/ CALL.REL.NOINC 0x2f0 ; /* 0x000000a000007944 */ /* 0x000fea0003c00000 */ /*0250*/ BRA 0x2a0 ; /* 0x0000004000007947 */ /* 0x000fea0003800000 */ /*0260*/ FMUL.FTZ R5, |R8|, R3 ; /* 0x0000000308057220 */ /* 0x003fe40000410200 */ /*0270*/ FMUL.FTZ R3, R3, 0.5 ; /* 0x3f00000003037820 */ /* 0x000fe40000410000 */ /*0280*/ FFMA R8, -R5, R5, |R8| ; /* 0x0000000505087223 */ /* 0x000fc80000000508 */ /*0290*/ FFMA R5, R8, R3, R5 ; /* 0x0000000308057223 */ /* 0x000fe40000000005 */ /*02a0*/ BSYNC B0 ; /* 0x0000000000007941 */ /* 0x000fea0003800000 */ /*02b0*/ HFMA2.MMA R3, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff037435 */ /* 0x000fd400000001ff */ /*02c0*/ IMAD.WIDE R2, R0, R3, c[0x0][0x180] ; /* 0x0000600000027625 */ /* 0x000fca00078e0203 */ /*02d0*/ STG.E [R2.64], R5 ; /* 0x0000000502007986 */ /* 0x000fe2000c101904 */ /*02e0*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*02f0*/ LOP3.LUT P0, RZ, R10, 0x7fffffff, RZ, 0xc0, !PT ; /* 0x7fffffff0aff7812 */ /* 0x000fda000780c0ff */ /*0300*/ @!P0 MOV R2, R10 ; /* 0x0000000a00028202 */ /* 0x000fe20000000f00 */ /*0310*/ @!P0 BRA 0x420 ; /* 0x0000010000008947 */ /* 0x000fea0003800000 */ /*0320*/ FSETP.GEU.FTZ.AND P0, PT, R10, RZ, PT ; /* 0x000000ff0a00720b */ /* 0x000fda0003f1e000 */ /*0330*/ @!P0 MOV R2, 0x7fffffff ; /* 0x7fffffff00028802 */ /* 0x000fe20000000f00 */ /*0340*/ @!P0 BRA 0x420 ; /* 0x000000d000008947 */ /* 0x000fea0003800000 */ /*0350*/ FSETP.GTU.FTZ.AND P0, PT, |R10|, +INF , PT ; /* 0x7f8000000a00780b */ /* 0x000fda0003f1c200 */ /*0360*/ @P0 FADD.FTZ R2, R10, 1 ; /* 0x3f8000000a020421 */ /* 0x000fe20000010000 */ /*0370*/ @P0 BRA 0x420 ; /* 0x000000a000000947 */ /* 0x000fea0003800000 */ /*0380*/ FSETP.NEU.FTZ.AND P0, PT, |R10|, +INF , PT ; /* 0x7f8000000a00780b */ /* 0x000fda0003f1d200 */ /*0390*/ @P0 FFMA R3, R10, 1.84467440737095516160e+19, RZ ; /* 0x5f8000000a030823 */ /* 0x000fc800000000ff */ /*03a0*/ @P0 MUFU.RSQ R2, R3 ; /* 0x0000000300020308 */ /* 0x000e240000001400 */ /*03b0*/ @P0 FMUL.FTZ R4, R3, R2 ; /* 0x0000000203040220 */ /* 0x001fe40000410000 */ /*03c0*/ @P0 FMUL.FTZ R6, R2, 0.5 ; /* 0x3f00000002060820 */ /* 0x000fe20000410000 */ /*03d0*/ @!P0 MOV R2, R10 ; /* 0x0000000a00028202 */ /* 0x000fe20000000f00 */ /*03e0*/ @P0 FADD.FTZ R5, -R4, -RZ ; /* 0x800000ff04050221 */ /* 0x000fc80000010100 */ /*03f0*/ @P0 FFMA R5, R4, R5, R3 ; /* 0x0000000504050223 */ /* 0x000fc80000000003 */ /*0400*/ @P0 FFMA R5, R5, R6, R4 ; /* 0x0000000605050223 */ /* 0x000fc80000000004 */ /*0410*/ @P0 FMUL.FTZ R2, R5, 2.3283064365386962891e-10 ; /* 0x2f80000005020820 */ /* 0x000fc80000410000 */ /*0420*/ HFMA2.MMA R3, -RZ, RZ, 0, 0 ; /* 0x00000000ff037435 */ /* 0x000fe200000001ff */ /*0430*/ MOV R5, R2 ; /* 0x0000000200057202 */ /* 0x000fe40000000f00 */ /*0440*/ MOV R2, R7 ; /* 0x0000000700027202 */ /* 0x000fc80000000f00 */ /*0450*/ RET.REL.NODEC R2 0x0 ; /* 0xfffffba002007950 */ /* 0x000fea0003c3ffff */ /*0460*/ BRA 0x460; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*0470*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0480*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0490*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*04f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z13pattern2d_gpuiiiiiiPfS_ .globl _Z13pattern2d_gpuiiiiiiPfS_ .p2align 8 .type _Z13pattern2d_gpuiiiiiiPfS_,@function _Z13pattern2d_gpuiiiiiiPfS_: s_clause 0x1 s_load_b64 s[4:5], s[0:1], 0x0 s_load_b32 s2, s[0:1], 0xc v_and_b32_e32 v1, 0x3ff, v0 v_bfe_u32 v2, v0, 10, 10 s_lshl_b32 s3, s14, 5 s_lshl_b32 s6, s15, 4 s_waitcnt lgkmcnt(0) v_add3_u32 v0, s4, s3, v1 v_add3_u32 v3, s2, s6, v2 s_load_b128 s[0:3], s[0:1], 0x18 s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_mad_u64_u32 v[1:2], null, v3, s5, v[0:1] v_add_nc_u32_e32 v2, -1, v3 v_add_nc_u32_e32 v5, -1, v1 s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_2) | instid1(VALU_DEP_4) v_mad_u64_u32 v[3:4], null, v2, s5, v[0:1] v_ashrrev_i32_e32 v2, 31, v1 v_add_nc_u32_e32 v7, 1, v1 v_ashrrev_i32_e32 v6, 31, v5 v_add_nc_u32_e32 v9, s5, v1 s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4) v_lshlrev_b64 v[0:1], 2, v[1:2] v_ashrrev_i32_e32 v8, 31, v7 s_delay_alu instid0(VALU_DEP_4) v_lshlrev_b64 v[5:6], 2, v[5:6] v_ashrrev_i32_e32 v4, 31, v3 v_ashrrev_i32_e32 v10, 31, v9 s_waitcnt lgkmcnt(0) v_add_co_u32 v11, vcc_lo, s0, v0 v_lshlrev_b64 v[7:8], 2, v[7:8] v_add_co_ci_u32_e32 v12, vcc_lo, s1, v1, vcc_lo v_lshlrev_b64 v[2:3], 2, v[3:4] v_add_co_u32 v4, vcc_lo, s0, v5 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v6, vcc_lo v_add_co_u32 v6, vcc_lo, s0, v7 v_lshlrev_b64 v[9:10], 2, v[9:10] v_add_co_ci_u32_e32 v7, vcc_lo, s1, v8, vcc_lo s_clause 0x1 global_load_b32 v8, v[11:12], off global_load_b32 v11, v[4:5], off v_add_co_u32 v2, vcc_lo, s0, v2 v_add_co_ci_u32_e32 v3, vcc_lo, s1, v3, vcc_lo global_load_b32 v6, v[6:7], off v_add_co_u32 v4, vcc_lo, s0, v9 v_add_co_ci_u32_e32 v5, vcc_lo, s1, v10, vcc_lo s_clause 0x1 global_load_b32 v2, v[2:3], off global_load_b32 v3, v[4:5], off s_waitcnt vmcnt(3) v_add_f32_e32 v4, v8, v11 s_waitcnt vmcnt(2) s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_1) v_add_f32_e32 v4, v4, v6 s_waitcnt vmcnt(1) v_fmac_f32_e32 v4, -2.0, v2 s_waitcnt vmcnt(0) s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_fmamk_f32 v2, v3, 0x40400000, v4 v_mul_f32_e64 v3, 0x4f800000, |v2| v_cmp_gt_f32_e64 vcc_lo, 0xf800000, |v2| s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_cndmask_b32_e64 v2, |v2|, v3, vcc_lo v_sqrt_f32_e32 v3, v2 s_waitcnt_depctr 0xfff v_add_nc_u32_e32 v4, -1, v3 v_add_nc_u32_e32 v5, 1, v3 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2) v_fma_f32 v6, -v4, v3, v2 v_fma_f32 v7, -v5, v3, v2 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_ge_f32_e64 s0, 0, v6 v_cndmask_b32_e64 v3, v3, v4, s0 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1) v_cmp_lt_f32_e64 s0, 0, v7 v_cndmask_b32_e64 v3, v3, v5, s0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_mul_f32_e32 v4, 0x37800000, v3 v_cndmask_b32_e32 v3, v3, v4, vcc_lo v_cmp_class_f32_e64 vcc_lo, v2, 0x260 s_delay_alu instid0(VALU_DEP_2) v_cndmask_b32_e32 v2, v3, v2, 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_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 _Z13pattern2d_gpuiiiiiiPfS_ .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 40 .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 13 .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 _Z13pattern2d_gpuiiiiiiPfS_, .Lfunc_end0-_Z13pattern2d_gpuiiiiiiPfS_ .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 - .offset: 8 .size: 4 .value_kind: by_value - .offset: 12 .size: 4 .value_kind: by_value - .offset: 16 .size: 4 .value_kind: by_value - .offset: 20 .size: 4 .value_kind: by_value - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 32 .size: 8 .value_kind: global_buffer .group_segment_fixed_size: 0 .kernarg_segment_align: 8 .kernarg_segment_size: 40 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z13pattern2d_gpuiiiiiiPfS_ .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z13pattern2d_gpuiiiiiiPfS_.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 13 .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_000a59a6_00000000-6_pattern2d.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2061: .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 .LFE2061: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .globl _Z13pattern2d_cpuiiiiiiPfS_ .type _Z13pattern2d_cpuiiiiiiPfS_, @function _Z13pattern2d_cpuiiiiiiPfS_: .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 movl %edi, %ebp movl %ecx, %ebx movq 56(%rsp), %rcx movq 64(%rsp), %rdi subl %r9d, %r8d cmpl %r8d, %ebx jge .L3 movl %esi, %r12d movl %ebx, %r11d imull %esi, %r11d movl %esi, %r13d subl %edx, %r13d movslq %ebp, %r14 movl %r13d, %eax subl %ebp, %eax addq %r14, %rax movq %rax, -8(%rsp) leal (%rsi,%rsi), %r15d movss .LC0(%rip), %xmm3 movss .LC1(%rip), %xmm2 jmp .L5 .L7: movl %r11d, %r10d subl %r12d, %r10d movslq %r11d, %r9 leaq (%r9,%r14), %rax salq $2, %rax movq -8(%rsp), %rsi addq %r9, %rsi salq $2, %rsi negq %r9 salq $2, %r9 movslq %r10d, %rdx leaq (%r9,%rdx,4), %rdx addq %rcx, %rdx addl %r15d, %r10d movslq %r10d, %r10 leaq (%r9,%r10,4), %r9 addq %rcx, %r9 .L6: movss (%rcx,%rax), %xmm0 addss -4(%rcx,%rax), %xmm0 addss 4(%rcx,%rax), %xmm0 movss (%rdx,%rax), %xmm1 addss %xmm1, %xmm1 subss %xmm1, %xmm0 movaps %xmm3, %xmm1 mulss (%r9,%rax), %xmm1 addss %xmm1, %xmm0 andps %xmm2, %xmm0 sqrtss %xmm0, %xmm0 movss %xmm0, (%rdi,%rax) addq $4, %rax cmpq %rsi, %rax jne .L6 .L8: addl $1, %ebx addl %r12d, %r11d cmpl %r8d, %ebx je .L3 .L5: cmpl %r13d, %ebp jl .L7 jmp .L8 .L3: 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 .LFE2057: .size _Z13pattern2d_cpuiiiiiiPfS_, .-_Z13pattern2d_cpuiiiiiiPfS_ .globl _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ .type _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_, @function _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_: .LFB2083: .cfi_startproc endbr64 subq $200, %rsp .cfi_def_cfa_offset 208 movl %edi, 44(%rsp) movl %esi, 40(%rsp) movl %edx, 36(%rsp) movl %ecx, 32(%rsp) movl %r8d, 28(%rsp) movl %r9d, 24(%rsp) movq 208(%rsp), %rax movq %rax, 16(%rsp) movq 216(%rsp), %rax movq %rax, 8(%rsp) movq %fs:40, %rax movq %rax, 184(%rsp) xorl %eax, %eax leaq 44(%rsp), %rax movq %rax, 112(%rsp) leaq 40(%rsp), %rax movq %rax, 120(%rsp) leaq 36(%rsp), %rax movq %rax, 128(%rsp) leaq 32(%rsp), %rax movq %rax, 136(%rsp) leaq 28(%rsp), %rax movq %rax, 144(%rsp) leaq 24(%rsp), %rax movq %rax, 152(%rsp) leaq 16(%rsp), %rax movq %rax, 160(%rsp) leaq 8(%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 .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 _Z13pattern2d_gpuiiiiiiPfS_(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 208 jmp .L11 .L16: call __stack_chk_fail@PLT .cfi_endproc .LFE2083: .size _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_, .-_Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ .globl _Z13pattern2d_gpuiiiiiiPfS_ .type _Z13pattern2d_gpuiiiiiiPfS_, @function _Z13pattern2d_gpuiiiiiiPfS_: .LFB2084: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 pushq 24(%rsp) .cfi_def_cfa_offset 24 pushq 24(%rsp) .cfi_def_cfa_offset 32 call _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ addq $24, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2084: .size _Z13pattern2d_gpuiiiiiiPfS_, .-_Z13pattern2d_gpuiiiiiiPfS_ .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC2: .string "Cannot malloc input memory on GPU by process" .align 8 .LC3: .string "%s %d, status = %d\nInsufficient GPU memory?\n" .align 8 .LC4: .string "Cannot copy data from host to gpu by process %d, status = %d\n" .align 8 .LC5: .string "Cannot malloc output memory on GPU by process" .align 8 .LC6: .string "Cannot erase output memory on GPU by process %d, status = %d\n" .align 8 .LC7: .string "Cannot execute CUDA kernel by process %d, status = %d\n" .align 8 .LC8: .string "Cannot synchronize thread by process %d, status = %d\n" .align 8 .LC9: .string "Cannot copy data from gpu to host by process %d, status = %d\n" .align 8 .LC10: .string "Cannot free device input memory by process %d, status = %d\n" .align 8 .LC11: .string "Cannot free device output memory by process %d, status = %d\n" .align 8 .LC12: .string "Step %d result abs max diff by process %d = %f @ (%d,%d)\n" .text .globl _Z9pattern2diiPfS_ii .type _Z9pattern2diiPfS_ii, @function _Z9pattern2diiPfS_ii: .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 subq $88, %rsp .cfi_def_cfa_offset 144 movq %rcx, 8(%rsp) movl %r9d, 24(%rsp) movq %fs:40, %rax movq %rax, 72(%rsp) xorl %eax, %eax testl %edi, %edi jle .L37 movl %edi, %ebx movl %esi, %r12d movq %rdx, %r13 movl %r8d, %r14d testl %esi, %esi jle .L37 movl %edi, %ebp imull %esi, %ebp movslq %ebp, %rbp salq $2, %rbp leaq 32(%rsp), %rdi movq %rbp, %rsi call cudaMalloc@PLT testl %eax, %eax jne .L41 movl $1, %ecx movq %rbp, %rdx movq %r13, %rsi movq 32(%rsp), %rdi call cudaMemcpy@PLT testl %eax, %eax jne .L42 movq %rbp, %rdi call malloc@PLT movq %rax, %r15 movq %rbp, %rcx movq %rbp, %rdx movl $0, %esi movq %rax, %rdi call __memset_chk@PLT pushq %r15 .cfi_def_cfa_offset 152 pushq %r13 .cfi_def_cfa_offset 160 movl $1, %r9d movl %r12d, %r8d movl $1, %ecx movl $1, %edx movl %ebx, %esi movl $1, %edi call _Z13pattern2d_cpuiiiiiiPfS_ addq $16, %rsp .cfi_def_cfa_offset 144 leaq 40(%rsp), %rdi movq %rbp, %rsi call cudaMalloc@PLT testl %eax, %eax jne .L43 movq %rbp, %rdx movl $0, %esi movq 40(%rsp), %rdi call cudaMemset@PLT testl %eax, %eax jne .L44 movl $32, 60(%rsp) movl $16, 64(%rsp) movl $1, 68(%rsp) leal 29(%rbx), %eax movl %ebx, %ecx subl $2, %ecx movl %ecx, 16(%rsp) cmovns %ecx, %eax sarl $5, %eax movl %eax, 48(%rsp) leal 13(%r12), %eax movl %r12d, %esi subl $2, %esi movl %esi, 28(%rsp) cmovns %esi, %eax sarl $4, %eax movl %eax, 52(%rsp) movl $1, 56(%rsp) movl $0, %r9d movl $0, %r8d movq 60(%rsp), %rdx movl $1, %ecx movq 48(%rsp), %rdi movl $1, %esi call __cudaPushCallConfiguration@PLT testl %eax, %eax je .L45 .L25: call cudaGetLastError@PLT testl %eax, %eax jne .L46 call cudaThreadSynchronize@PLT testl %eax, %eax jne .L47 movl $2, %ecx movq %rbp, %rdx movq 40(%rsp), %rsi movq 8(%rsp), %rdi call cudaMemcpy@PLT testl %eax, %eax jne .L48 movl 28(%rsp), %eax movl %eax, %edx sarl $31, %edx shrl $28, %edx addl %edx, %eax andl $15, %eax subl %edx, %eax movl %r12d, %ecx subl %eax, %ecx subl $2, %ecx pushq 8(%rsp) .cfi_def_cfa_offset 152 pushq %r13 .cfi_def_cfa_offset 160 movl $1, %r9d movl %r12d, %r8d movl $1, %edx movl %ebx, %esi movl $1, %edi call _Z13pattern2d_cpuiiiiiiPfS_ movl 32(%rsp), %eax movl %eax, %edx sarl $31, %edx shrl $27, %edx addl %edx, %eax andl $31, %eax subl %edx, %eax movl %ebx, %edi subl %eax, %edi subl $2, %edi pushq 24(%rsp) .cfi_def_cfa_offset 168 pushq %r13 .cfi_def_cfa_offset 176 movl $1, %r9d movl %r12d, %r8d movl $1, %ecx movl $1, %edx movl %ebx, %esi call _Z13pattern2d_cpuiiiiiiPfS_ movq 40(%rsp), %r13 movss 0(%r13), %xmm0 subss (%r15), %xmm0 andps .LC1(%rip), %xmm0 movss %xmm0, 40(%rsp) addq $32, %rsp .cfi_def_cfa_offset 144 movq %rbp, %rdi call malloc@PLT movq %rax, 16(%rsp) movq %rbp, %rcx movq %rbp, %rdx movl $0, %esi movq %rax, %rbp movq %rax, %rdi call __memset_chk@PLT movslq %ebx, %r8 salq $2, %r8 movq %rbp, %rdi movq %r13, %rdx movq %r15, %rsi movl $0, %ecx movl $0, %r13d movl $0, %ebp movss .LC1(%rip), %xmm1 .L29: movl $0, %eax .L32: movss (%rdx,%rax,4), %xmm0 subss (%rsi,%rax,4), %xmm0 andps %xmm1, %xmm0 movss 8(%rsp), %xmm2 comiss %xmm2, %xmm0 cmova %eax, %ebp movaps %xmm0, %xmm3 maxss %xmm2, %xmm3 movss %xmm3, 8(%rsp) cmova %ecx, %r13d movss %xmm0, (%rdi,%rax,4) addq $1, %rax cmpl %eax, %ebx jg .L32 addl $1, %ecx addq %r8, %rdi addq %r8, %rdx addq %r8, %rsi cmpl %ecx, %r12d jg .L29 movq 32(%rsp), %rdi call cudaFree@PLT testl %eax, %eax jne .L49 movq %r15, %rdi call free@PLT movq 40(%rsp), %rdi call cudaFree@PLT testl %eax, %eax jne .L50 movq 16(%rsp), %rdi call free@PLT pxor %xmm0, %xmm0 cvtss2sd 8(%rsp), %xmm0 movl %r13d, %r9d movl %ebp, %r8d movl %r14d, %ecx movl 24(%rsp), %edx leaq .LC12(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movl $0, %ebx jmp .L19 .L41: movl %eax, %ebx movl %eax, %r9d movl %r14d, %r8d leaq .LC2(%rip), %rcx leaq .LC3(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT .L19: movq 72(%rsp), %rax subq %fs:40, %rax jne .L51 movl %ebx, %eax addq $88, %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 .L42: .cfi_restore_state movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC4(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L43: movl %eax, %ebx movl %eax, %r9d movl %r14d, %r8d leaq .LC5(%rip), %rcx leaq .LC3(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L44: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC6(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L45: pushq 40(%rsp) .cfi_def_cfa_offset 152 pushq 40(%rsp) .cfi_def_cfa_offset 160 movl $1, %r9d movl %r12d, %r8d movl $1, %ecx movl $1, %edx movl %ebx, %esi movl $1, %edi call _Z41__device_stub__Z13pattern2d_gpuiiiiiiPfS_iiiiiiPfS_ addq $16, %rsp .cfi_def_cfa_offset 144 jmp .L25 .L46: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC7(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L47: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC8(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L48: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC9(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L49: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC10(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L50: movl %eax, %ebx movl %eax, %r8d movl %r14d, %ecx leaq .LC11(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L19 .L37: movl $-1, %ebx jmp .L19 .L51: call __stack_chk_fail@PLT .cfi_endproc .LFE2058: .size _Z9pattern2diiPfS_ii, .-_Z9pattern2diiPfS_ii .section .rodata.str1.1,"aMS",@progbits,1 .LC13: .string "_Z13pattern2d_gpuiiiiiiPfS_" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2086: .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 .LC13(%rip), %rdx movq %rdx, %rcx leaq _Z13pattern2d_gpuiiiiiiPfS_(%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 .LFE2086: .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 .section .rodata.cst16,"aM",@progbits,16 .align 16 .LC1: .long 2147483647 .long 0 .long 0 .long 0 .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 "pattern2d.hip" .globl _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ # -- Begin function _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .p2align 4, 0x90 .type _Z28__device_stub__pattern2d_gpuiiiiiiPfS_,@function _Z28__device_stub__pattern2d_gpuiiiiiiPfS_: # @_Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .cfi_startproc # %bb.0: subq $152, %rsp .cfi_def_cfa_offset 160 movl %edi, 28(%rsp) movl %esi, 24(%rsp) movl %edx, 20(%rsp) movl %ecx, 16(%rsp) movl %r8d, 12(%rsp) movl %r9d, 8(%rsp) leaq 28(%rsp), %rax movq %rax, 80(%rsp) leaq 24(%rsp), %rax movq %rax, 88(%rsp) leaq 20(%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 160(%rsp), %rax movq %rax, 128(%rsp) leaq 168(%rsp), %rax movq %rax, 136(%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 80(%rsp), %r9 movl $_Z13pattern2d_gpuiiiiiiPfS_, %edi pushq 32(%rsp) .cfi_adjust_cfa_offset 8 pushq 48(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $168, %rsp .cfi_adjust_cfa_offset -168 retq .Lfunc_end0: .size _Z28__device_stub__pattern2d_gpuiiiiiiPfS_, .Lfunc_end0-_Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function _Z13pattern2d_cpuiiiiiiPfS_ .LCPI1_0: .long 0x40400000 # float 3 .section .rodata.cst16,"aM",@progbits,16 .p2align 4, 0x0 .LCPI1_1: .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .text .globl _Z13pattern2d_cpuiiiiiiPfS_ .p2align 4, 0x90 .type _Z13pattern2d_cpuiiiiiiPfS_,@function _Z13pattern2d_cpuiiiiiiPfS_: # @_Z13pattern2d_cpuiiiiiiPfS_ .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 $88, %rsp .cfi_def_cfa_offset 144 .cfi_offset %rbx, -56 .cfi_offset %r12, -48 .cfi_offset %r13, -40 .cfi_offset %r14, -32 .cfi_offset %r15, -24 .cfi_offset %rbp, -16 # kill: def $edi killed $edi def $rdi movq %rdi, 16(%rsp) # 8-byte Spill subl %r9d, %r8d cmpl %ecx, %r8d jle .LBB1_4 # %bb.1: # %.preheader.lr.ph movq 152(%rsp), %rax movq 144(%rsp), %rdi movl $4294967295, %r9d # imm = 0xFFFFFFFF movslq %esi, %r10 subl %edx, %esi movq 16(%rsp), %rdx # 8-byte Reload movslq %edx, %r11 movq %r11, 40(%rsp) # 8-byte Spill movl %esi, 12(%rsp) # 4-byte Spill movslq %esi, %r13 movslq %ecx, %r11 movslq %r8d, %rcx movq %rcx, 32(%rsp) # 8-byte Spill leaq 1(%r11), %rcx imulq %r10, %rcx leaq (%rdi,%rcx,4), %r15 leaq (,%r10,4), %rcx movq %rcx, 24(%rsp) # 8-byte Spill leaq -1(%r11), %rcx imulq %r10, %rcx leaq (%rdi,%rcx,4), %rbx movq %r11, %rcx imulq %r10, %rcx leal (%r9,%rdx), %r14d addl %ecx, %r14d shlq $32, %r14 shlq $32, %r10 movq %r10, 48(%rsp) # 8-byte Spill leaq (%rax,%rcx,4), %rdx leaq (%rdi,%rcx,4), %r12 movss .LCPI1_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI1_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 jmp .LBB1_2 .p2align 4, 0x90 .LBB1_3: # %._crit_edge # in Loop: Header=BB1_2 Depth=1 movq 64(%rsp), %r11 # 8-byte Reload incq %r11 movq 24(%rsp), %rax # 8-byte Reload addq %rax, %r15 addq %rax, %rbx movq 56(%rsp), %r14 # 8-byte Reload addq 48(%rsp), %r14 # 8-byte Folded Reload addq %rax, %r12 addq %rax, %rdx cmpq 32(%rsp), %r11 # 8-byte Folded Reload jge .LBB1_4 .LBB1_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB1_5 Depth 2 movq %r11, 64(%rsp) # 8-byte Spill movq %r14, 56(%rsp) # 8-byte Spill movq 40(%rsp), %rbp # 8-byte Reload movq 16(%rsp), %rax # 8-byte Reload cmpl %eax, 12(%rsp) # 4-byte Folded Reload movq %rbx, 80(%rsp) # 8-byte Spill movq %rdx, 72(%rsp) # 8-byte Spill jg .LBB1_5 jmp .LBB1_3 .p2align 4, 0x90 .LBB1_7: # %call.sqrt # in Loop: Header=BB1_5 Depth=2 movq %rdi, %rbx callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI1_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI1_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %rdx # 8-byte Reload movl $4294967295, %r9d # imm = 0xFFFFFFFF movq %rbx, %rdi movq 80(%rsp), %rbx # 8-byte Reload .LBB1_8: # %.split # in Loop: Header=BB1_5 Depth=2 movss %xmm0, (%rdx,%rbp,4) incq %rbp addq %r9, %r14 incq %r14 cmpq %r13, %rbp jge .LBB1_3 .LBB1_5: # Parent Loop BB1_2 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r12,%rbp,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %r14, %rax sarq $30, %rax addss (%rdi,%rax), %xmm1 addss 4(%r12,%rbp,4), %xmm1 movss (%rbx,%rbp,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r15,%rbp,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB1_7 # %bb.6: # in Loop: Header=BB1_5 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB1_8 .LBB1_4: # %._crit_edge36 addq $88, %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_end1: .size _Z13pattern2d_cpuiiiiiiPfS_, .Lfunc_end1-_Z13pattern2d_cpuiiiiiiPfS_ .cfi_endproc # -- End function .section .rodata.cst4,"aM",@progbits,4 .p2align 2, 0x0 # -- Begin function _Z9pattern2diiPfS_ii .LCPI2_0: .long 0x40400000 # float 3 .section .rodata.cst16,"aM",@progbits,16 .p2align 4, 0x0 .LCPI2_1: .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .long 0x7fffffff # float NaN .text .globl _Z9pattern2diiPfS_ii .p2align 4, 0x90 .type _Z9pattern2diiPfS_ii,@function _Z9pattern2diiPfS_ii: # @_Z9pattern2diiPfS_ii .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 $312, %rsp # imm = 0x138 .cfi_def_cfa_offset 368 .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, 80(%rsp) # 8-byte Spill movl $-1, %r12d testl %edi, %edi jle .LBB2_62 # %bb.1: movl %esi, %r13d testl %esi, %esi jle .LBB2_62 # %bb.2: movl %r9d, %ebp movq %rcx, %r14 movl %edi, %r15d movl %r8d, 4(%rsp) # 4-byte Spill movl %r13d, %eax imull %edi, %eax movslq %eax, %rbx shlq $2, %rbx leaq 128(%rsp), %rdi movq %rbx, %rsi callq hipMalloc testl %eax, %eax jne .LBB2_3 # %bb.5: movq 128(%rsp), %rdi movq 80(%rsp), %rsi # 8-byte Reload movq %rbx, %rdx movl $1, %ecx callq hipMemcpy testl %eax, %eax jne .LBB2_6 # %bb.7: movq %r14, 104(%rsp) # 8-byte Spill movl %ebp, 136(%rsp) # 4-byte Spill movq %rbx, %rdi callq malloc movq %rax, 168(%rsp) # 8-byte Spill movq %rax, %rdi xorl %esi, %esi movq %rbx, 40(%rsp) # 8-byte Spill movq %rbx, %rdx callq memset@PLT leal -1(%r13), %ecx movq %r15, 16(%rsp) # 8-byte Spill movslq %r15d, %rax movq %rax, 48(%rsp) # 8-byte Spill movl %ecx, 140(%rsp) # 4-byte Spill movslq %ecx, %rax movq %rax, 120(%rsp) # 8-byte Spill movq %r13, 32(%rsp) # 8-byte Spill cmpl $3, %r13d jl .LBB2_16 # %bb.8: # %.preheader.lr.ph.i movq 80(%rsp), %rax # 8-byte Reload movq 48(%rsp), %rcx # 8-byte Reload leaq (%rax,%rcx,8), %r14 addq $4, %r14 leaq (,%rcx,4), %rdx movq %rdx, 64(%rsp) # 8-byte Spill leaq 4(%rax), %r15 leaq -2(%rcx), %rbx leaq (%rax,%rcx,4), %r10 addq $8, %r10 movq %rcx, %r8 shlq $32, %r8 movq 168(%rsp), %rax # 8-byte Reload leaq (%rax,%rcx,4), %r13 addq $4, %r13 movl $1, %ecx movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 movq %r8, 56(%rsp) # 8-byte Spill movq %r8, 8(%rsp) # 8-byte Spill jmp .LBB2_9 .p2align 4, 0x90 .LBB2_15: # %._crit_edge.i # in Loop: Header=BB2_9 Depth=1 movq 24(%rsp), %rcx # 8-byte Reload incq %rcx movq 64(%rsp), %rax # 8-byte Reload addq %rax, %r14 addq %rax, %r15 addq %rax, %r10 movq 8(%rsp), %r8 # 8-byte Reload addq 56(%rsp), %r8 # 8-byte Folded Reload movq %r8, 8(%rsp) # 8-byte Spill addq %rax, %r13 cmpq 120(%rsp), %rcx # 8-byte Folded Reload je .LBB2_16 .LBB2_9: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB2_11 Depth 2 movq %rcx, 24(%rsp) # 8-byte Spill cmpl $3, 16(%rsp) # 4-byte Folded Reload jl .LBB2_15 # %bb.10: # %.lr.ph.i # in Loop: Header=BB2_9 Depth=1 movq 8(%rsp), %rbp # 8-byte Reload xorl %r12d, %r12d movq %r10, 72(%rsp) # 8-byte Spill jmp .LBB2_11 .p2align 4, 0x90 .LBB2_13: # %call.sqrt # in Loop: Header=BB2_11 Depth=2 callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %r10 # 8-byte Reload .LBB2_14: # %.split # in Loop: Header=BB2_11 Depth=2 movss %xmm0, (%r13,%r12,4) incq %r12 movabsq $4294967296, %rax # imm = 0x100000000 addq %rax, %rbp cmpq %r12, %rbx je .LBB2_15 .LBB2_11: # Parent Loop BB2_9 Depth=1 # => This Inner Loop Header: Depth=2 movss -4(%r10,%r12,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %rbp, %rax sarq $30, %rax movq 80(%rsp), %rcx # 8-byte Reload addss (%rcx,%rax), %xmm1 addss (%r10,%r12,4), %xmm1 movss (%r15,%r12,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r14,%r12,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB2_13 # %bb.12: # in Loop: Header=BB2_11 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB2_14 .LBB2_16: # %_Z13pattern2d_cpuiiiiiiPfS_.exit movq 16(%rsp), %rbx # 8-byte Reload leal -2(%rbx), %eax leal 29(%rbx), %r15d testl %eax, %eax cmovnsl %eax, %r15d movq 32(%rsp), %rcx # 8-byte Reload leal -2(%rcx), %eax leal 13(%rcx), %r13d testl %eax, %eax cmovnsl %eax, %r13d leaq 112(%rsp), %rdi movq 40(%rsp), %r14 # 8-byte Reload movq %r14, %rsi callq hipMalloc testl %eax, %eax jne .LBB2_17 # %bb.18: movq 112(%rsp), %rdi xorl %esi, %esi movq %r14, %rdx callq hipMemset testl %eax, %eax jne .LBB2_19 # %bb.21: movl %r15d, %eax sarl $5, %eax movl %r13d, %ecx sarl $4, %ecx shlq $32, %rcx movl %eax, %edi orq %rcx, %rdi movabsq $68719476768, %rdx # imm = 0x1000000020 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB2_23 # %bb.22: movq 128(%rsp), %rax movq 112(%rsp), %rcx movl $1, 164(%rsp) movl %ebx, 160(%rsp) movl $1, 156(%rsp) movl $1, 152(%rsp) movq 32(%rsp), %rdx # 8-byte Reload movl %edx, 148(%rsp) movl $1, 144(%rsp) movq %rax, 232(%rsp) movq %rcx, 224(%rsp) leaq 164(%rsp), %rax movq %rax, 240(%rsp) leaq 160(%rsp), %rax movq %rax, 248(%rsp) leaq 156(%rsp), %rax movq %rax, 256(%rsp) leaq 152(%rsp), %rax movq %rax, 264(%rsp) leaq 148(%rsp), %rax movq %rax, 272(%rsp) leaq 144(%rsp), %rax movq %rax, 280(%rsp) leaq 232(%rsp), %rax movq %rax, 288(%rsp) leaq 224(%rsp), %rax movq %rax, 296(%rsp) leaq 208(%rsp), %rdi leaq 192(%rsp), %rsi leaq 184(%rsp), %rdx leaq 176(%rsp), %rcx callq __hipPopCallConfiguration movq 208(%rsp), %rsi movl 216(%rsp), %edx movq 192(%rsp), %rcx movl 200(%rsp), %r8d leaq 240(%rsp), %r9 movl $_Z13pattern2d_gpuiiiiiiPfS_, %edi pushq 176(%rsp) .cfi_adjust_cfa_offset 8 pushq 192(%rsp) .cfi_adjust_cfa_offset 8 callq hipLaunchKernel addq $16, %rsp .cfi_adjust_cfa_offset -16 .LBB2_23: callq hipGetLastError testl %eax, %eax jne .LBB2_24 # %bb.25: callq hipDeviceSynchronize testl %eax, %eax jne .LBB2_26 # %bb.27: movq 112(%rsp), %rsi movq 104(%rsp), %rdi # 8-byte Reload movq %r14, %rdx movl $2, %ecx callq hipMemcpy testl %eax, %eax jne .LBB2_28 # %bb.29: movq %r15, 40(%rsp) # 8-byte Spill andl $-16, %r13d negl %r13d movq 32(%rsp), %rax # 8-byte Reload leal (%rax,%r13), %ecx addl $-2, %ecx # kill: def $eax killed $eax killed $rax subl %ecx, %eax addl $-2, %eax cmpl %eax, 140(%rsp) # 4-byte Folded Reload jle .LBB2_38 # %bb.30: # %.preheader.lr.ph.i144 movq 16(%rsp), %rcx # 8-byte Reload decl %ecx movslq %ecx, %r15 movslq %eax, %rdi leaq 1(%rdi), %rax movq 48(%rsp), %rdx # 8-byte Reload imulq %rdx, %rax movq 80(%rsp), %rcx # 8-byte Reload leaq (%rcx,%rax,4), %r13 addq $4, %r13 leaq (,%rdx,4), %rax movq %rax, 64(%rsp) # 8-byte Spill leaq -1(%rdi), %rax imulq %rdx, %rax leaq (%rcx,%rax,4), %r14 addq $4, %r14 decq %r15 movq %rdx, %r10 imulq %rdi, %r10 leaq (%rcx,%r10,4), %r11 addq $8, %r11 shlq $32, %rdx movq %rdx, 56(%rsp) # 8-byte Spill movq 104(%rsp), %rax # 8-byte Reload leaq (%rax,%r10,4), %r12 addq $4, %r12 shlq $32, %r10 movq %r10, 24(%rsp) # 8-byte Spill movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 jmp .LBB2_31 .p2align 4, 0x90 .LBB2_37: # %._crit_edge.i147 # in Loop: Header=BB2_31 Depth=1 movq 8(%rsp), %rdi # 8-byte Reload incq %rdi movq 64(%rsp), %rax # 8-byte Reload addq %rax, %r13 addq %rax, %r14 addq %rax, %r11 movq 56(%rsp), %rcx # 8-byte Reload addq %rcx, 24(%rsp) # 8-byte Folded Spill addq %rax, %r12 cmpq 120(%rsp), %rdi # 8-byte Folded Reload je .LBB2_38 .LBB2_31: # %.preheader.i145 # =>This Loop Header: Depth=1 # Child Loop BB2_33 Depth 2 movq %rdi, 8(%rsp) # 8-byte Spill cmpl $3, 16(%rsp) # 4-byte Folded Reload jl .LBB2_37 # %bb.32: # %.lr.ph.i149 # in Loop: Header=BB2_31 Depth=1 movq 24(%rsp), %rbp # 8-byte Reload xorl %ebx, %ebx movq %r11, 72(%rsp) # 8-byte Spill jmp .LBB2_33 .p2align 4, 0x90 .LBB2_35: # %call.sqrt273 # in Loop: Header=BB2_33 Depth=2 callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %r11 # 8-byte Reload .LBB2_36: # %.split272 # in Loop: Header=BB2_33 Depth=2 movss %xmm0, (%r12,%rbx,4) incq %rbx movabsq $4294967296, %rax # imm = 0x100000000 addq %rax, %rbp cmpq %rbx, %r15 je .LBB2_37 .LBB2_33: # Parent Loop BB2_31 Depth=1 # => This Inner Loop Header: Depth=2 movss -4(%r11,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %rbp, %rax sarq $30, %rax movq 80(%rsp), %rcx # 8-byte Reload addss (%rcx,%rax), %xmm1 addss (%r11,%rbx,4), %xmm1 movss (%r14,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r13,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB2_35 # %bb.34: # in Loop: Header=BB2_33 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB2_36 .LBB2_38: # %_Z13pattern2d_cpuiiiiiiPfS_.exit153 cmpl $3, 32(%rsp) # 4-byte Folded Reload jl .LBB2_47 # %bb.39: # %.preheader.lr.ph.i154 movq 40(%rsp), %rcx # 8-byte Reload andl $-32, %ecx negl %ecx movq 16(%rsp), %rdx # 8-byte Reload leal (%rdx,%rcx), %eax addl $-2, %eax movl %edx, %ecx subl %eax, %ecx movslq %ecx, %rax addl $-2, %ecx movl %ecx, 64(%rsp) # 4-byte Spill leaq -2(%rax), %rcx decl %edx movl %edx, 56(%rsp) # 4-byte Spill movslq %edx, %r13 subq %rcx, %r13 leaq -8(,%rax,4), %rcx movq 48(%rsp), %rsi # 8-byte Reload leaq (%rcx,%rsi,8), %r15 movq 80(%rsp), %rdx # 8-byte Reload addq %rdx, %r15 leaq (,%rsi,4), %r9 movq %r9, 40(%rsp) # 8-byte Spill leaq (%rdx,%rax,4), %r14 addq $-8, %r14 leaq (%rcx,%rsi,4), %r10 addq %rdx, %r10 addq %rsi, %rax addq $-2, %rax movabsq $4294967296, %rcx # imm = 0x100000000 addl %eax, %ecx decl %ecx shlq $32, %rcx movq %rcx, 8(%rsp) # 8-byte Spill shlq $32, %rsi movq %rsi, 48(%rsp) # 8-byte Spill movq 104(%rsp), %rcx # 8-byte Reload leaq (%rcx,%rax,4), %r12 movl $1, %esi movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] xorps %xmm4, %xmm4 jmp .LBB2_40 .p2align 4, 0x90 .LBB2_46: # %._crit_edge.i157 # in Loop: Header=BB2_40 Depth=1 movq 24(%rsp), %rsi # 8-byte Reload incq %rsi movq 40(%rsp), %rax # 8-byte Reload addq %rax, %r15 addq %rax, %r14 addq %rax, %r10 movq 8(%rsp), %rcx # 8-byte Reload addq 48(%rsp), %rcx # 8-byte Folded Reload movq %rcx, 8(%rsp) # 8-byte Spill addq %rax, %r12 cmpq 120(%rsp), %rsi # 8-byte Folded Reload je .LBB2_47 .LBB2_40: # %.preheader.i155 # =>This Loop Header: Depth=1 # Child Loop BB2_42 Depth 2 movq %rsi, 24(%rsp) # 8-byte Spill movl 56(%rsp), %eax # 4-byte Reload cmpl 64(%rsp), %eax # 4-byte Folded Reload jle .LBB2_46 # %bb.41: # %.lr.ph.i159 # in Loop: Header=BB2_40 Depth=1 movq 8(%rsp), %rbp # 8-byte Reload xorl %ebx, %ebx movq %r10, 72(%rsp) # 8-byte Spill jmp .LBB2_42 .p2align 4, 0x90 .LBB2_44: # %call.sqrt275 # in Loop: Header=BB2_42 Depth=2 callq sqrtf xorps %xmm4, %xmm4 movaps .LCPI2_1(%rip), %xmm3 # xmm3 = [NaN,NaN,NaN,NaN] movss .LCPI2_0(%rip), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 72(%rsp), %r10 # 8-byte Reload .LBB2_45: # %.split274 # in Loop: Header=BB2_42 Depth=2 movss %xmm0, (%r12,%rbx,4) incq %rbx movabsq $4294967296, %rax # imm = 0x100000000 addq %rax, %rbp cmpq %rbx, %r13 je .LBB2_46 .LBB2_42: # Parent Loop BB2_40 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r10,%rbx,4), %xmm1 # xmm1 = mem[0],zero,zero,zero movq %rbp, %rax sarq $30, %rax movq 80(%rsp), %rcx # 8-byte Reload addss (%rcx,%rax), %xmm1 addss 4(%r10,%rbx,4), %xmm1 movss (%r14,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero addss %xmm0, %xmm0 subss %xmm0, %xmm1 movss (%r15,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero mulss %xmm2, %xmm0 addss %xmm1, %xmm0 andps %xmm3, %xmm0 ucomiss %xmm4, %xmm0 jb .LBB2_44 # %bb.43: # in Loop: Header=BB2_42 Depth=2 sqrtss %xmm0, %xmm0 jmp .LBB2_45 .LBB2_47: # %_Z13pattern2d_cpuiiiiiiPfS_.exit163 movq 104(%rsp), %r10 # 8-byte Reload movss (%r10), %xmm2 # xmm2 = mem[0],zero,zero,zero movq 168(%rsp), %r15 # 8-byte Reload subss (%r15), %xmm2 andps .LCPI2_1(%rip), %xmm2 movq 32(%rsp), %rax # 8-byte Reload testl %eax, %eax jle .LBB2_48 # %bb.49: # %.preheader.lr.ph movl %eax, %eax movq 16(%rsp), %r11 # 8-byte Reload movl %r11d, %ecx xorl %edx, %edx movaps .LCPI2_1(%rip), %xmm0 # xmm0 = [NaN,NaN,NaN,NaN] xorl %esi, %esi movaps %xmm2, %xmm1 xorl %ebp, %ebp xorl %ebx, %ebx movl 4(%rsp), %r14d # 4-byte Reload jmp .LBB2_50 .p2align 4, 0x90 .LBB2_51: # in Loop: Header=BB2_50 Depth=1 movaps %xmm1, %xmm2 .LBB2_52: # %._crit_edge # in Loop: Header=BB2_50 Depth=1 incq %rsi addl %r11d, %edx movaps %xmm2, %xmm1 cmpq %rax, %rsi je .LBB2_53 .LBB2_50: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB2_58 Depth 2 testl %r11d, %r11d jle .LBB2_51 # %bb.57: # %.lr.ph # in Loop: Header=BB2_50 Depth=1 movl %edx, %r8d leaq (%r15,%r8,4), %rdi leaq (%r10,%r8,4), %r8 xorl %r9d, %r9d .p2align 4, 0x90 .LBB2_58: # Parent Loop BB2_50 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r8,%r9,4), %xmm2 # xmm2 = mem[0],zero,zero,zero subss (%rdi,%r9,4), %xmm2 andps %xmm0, %xmm2 ucomiss %xmm1, %xmm2 cmoval %r9d, %ebx cmoval %esi, %ebp maxss %xmm1, %xmm2 incq %r9 movaps %xmm2, %xmm1 cmpq %r9, %rcx jne .LBB2_58 jmp .LBB2_52 .LBB2_48: xorl %ebx, %ebx xorl %ebp, %ebp movl 4(%rsp), %r14d # 4-byte Reload .LBB2_53: # %._crit_edge181 movaps %xmm2, 80(%rsp) # 16-byte Spill movq 128(%rsp), %rdi callq hipFree testl %eax, %eax jne .LBB2_54 # %bb.59: movq %r15, %rdi callq free movq 112(%rsp), %rdi callq hipFree testl %eax, %eax jne .LBB2_60 # %bb.61: movaps 80(%rsp), %xmm0 # 16-byte Reload cvtss2sd %xmm0, %xmm0 movl $.L.str.10, %edi movl 136(%rsp), %esi # 4-byte Reload movl %r14d, %edx movl %ebx, %ecx movl %ebp, %r8d movb $1, %al callq printf xorl %r12d, %r12d jmp .LBB2_62 .LBB2_3: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str, %esi movl $.L.str.1, %edx jmp .LBB2_4 .LBB2_6: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.2, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_17: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str, %esi movl $.L.str.3, %edx .LBB2_4: movl 4(%rsp), %ecx # 4-byte Reload movl %r12d, %r8d xorl %eax, %eax callq fprintf jmp .LBB2_62 .LBB2_19: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.4, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_24: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.5, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_26: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.6, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_28: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.7, %esi movl 4(%rsp), %edx # 4-byte Reload jmp .LBB2_56 .LBB2_54: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.8, %esi jmp .LBB2_55 .LBB2_60: movl %eax, %r12d movq stderr(%rip), %rdi movl $.L.str.9, %esi .LBB2_55: movl %r14d, %edx .LBB2_56: movl %r12d, %ecx xorl %eax, %eax callq fprintf .LBB2_62: movl %r12d, %eax addq $312, %rsp # imm = 0x138 .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_end2: .size _Z9pattern2diiPfS_ii, .Lfunc_end2-_Z9pattern2diiPfS_ii .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 .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 $_Z13pattern2d_gpuiiiiiiPfS_, %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 _Z13pattern2d_gpuiiiiiiPfS_,@object # @_Z13pattern2d_gpuiiiiiiPfS_ .section .rodata,"a",@progbits .globl _Z13pattern2d_gpuiiiiiiPfS_ .p2align 3, 0x0 _Z13pattern2d_gpuiiiiiiPfS_: .quad _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .size _Z13pattern2d_gpuiiiiiiPfS_, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%s %d, status = %d\nInsufficient GPU memory?\n" .size .L.str, 45 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "Cannot malloc input memory on GPU by process" .size .L.str.1, 45 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "Cannot copy data from host to gpu by process %d, status = %d\n" .size .L.str.2, 62 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "Cannot malloc output memory on GPU by process" .size .L.str.3, 46 .type .L.str.4,@object # @.str.4 .L.str.4: .asciz "Cannot erase output memory on GPU by process %d, status = %d\n" .size .L.str.4, 62 .type .L.str.5,@object # @.str.5 .L.str.5: .asciz "Cannot execute CUDA kernel by process %d, status = %d\n" .size .L.str.5, 55 .type .L.str.6,@object # @.str.6 .L.str.6: .asciz "Cannot synchronize thread by process %d, status = %d\n" .size .L.str.6, 54 .type .L.str.7,@object # @.str.7 .L.str.7: .asciz "Cannot copy data from gpu to host by process %d, status = %d\n" .size .L.str.7, 62 .type .L.str.8,@object # @.str.8 .L.str.8: .asciz "Cannot free device input memory by process %d, status = %d\n" .size .L.str.8, 60 .type .L.str.9,@object # @.str.9 .L.str.9: .asciz "Cannot free device output memory by process %d, status = %d\n" .size .L.str.9, 61 .type .L.str.10,@object # @.str.10 .L.str.10: .asciz "Step %d result abs max diff by process %d = %f @ (%d,%d)\n" .size .L.str.10, 58 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z13pattern2d_gpuiiiiiiPfS_" .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 _Z28__device_stub__pattern2d_gpuiiiiiiPfS_ .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z13pattern2d_gpuiiiiiiPfS_ .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 <iostream> #include <cstdlib> #include<chrono> int main(void) { double gammaEulera = 0.; double N = 1000000;; auto start = std::chrono::high_resolution_clock::now(); for (int i = 1; i < N; i++) gammaEulera = gammaEulera + (1. / (double)i); gammaEulera = gammaEulera - log(N); auto finish = std::chrono::high_resolution_clock::now(); std::chrono::duration<double> elapsed = finish - start; printf("Przyblizona wartosc eulera CPU wynosi %f \n",gammaEulera ); printf("Czas wynosi %f",elapsed.count()); return 0; }
code for sm_80
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <stdio.h> #include <iostream> #include <cstdlib> #include<chrono> int main(void) { double gammaEulera = 0.; double N = 1000000;; auto start = std::chrono::high_resolution_clock::now(); for (int i = 1; i < N; i++) gammaEulera = gammaEulera + (1. / (double)i); gammaEulera = gammaEulera - log(N); auto finish = std::chrono::high_resolution_clock::now(); std::chrono::duration<double> elapsed = finish - start; printf("Przyblizona wartosc eulera CPU wynosi %f \n",gammaEulera ); printf("Czas wynosi %f",elapsed.count()); return 0; }
.file "tmpxft_0003cafb_00000000-6_euler-CPU.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB3775: .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 .LFE3775: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC4: .string "Przyblizona wartosc eulera CPU wynosi %f \n" .section .rodata.str1.1,"aMS",@progbits,1 .LC5: .string "Czas wynosi %f" .text .globl main .type main, @function main: .LFB3768: .cfi_startproc endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 subq $16, %rsp .cfi_def_cfa_offset 32 call _ZNSt6chrono3_V212system_clock3nowEv@PLT movq %rax, %rbx movl $1, %eax pxor %xmm0, %xmm0 movsd .LC1(%rip), %xmm3 .L4: pxor %xmm2, %xmm2 cvtsi2sdl %eax, %xmm2 movapd %xmm3, %xmm1 divsd %xmm2, %xmm1 addsd %xmm1, %xmm0 addl $1, %eax cmpl $1000000, %eax jne .L4 subsd .LC2(%rip), %xmm0 movsd %xmm0, (%rsp) call _ZNSt6chrono3_V212system_clock3nowEv@PLT subq %rbx, %rax pxor %xmm0, %xmm0 cvtsi2sdq %rax, %xmm0 movapd %xmm0, %xmm4 divsd .LC3(%rip), %xmm4 movsd %xmm4, 8(%rsp) movsd (%rsp), %xmm0 leaq .LC4(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movsd 8(%rsp), %xmm0 leaq .LC5(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movl $0, %eax addq $16, %rsp .cfi_def_cfa_offset 16 popq %rbx .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3768: .size main, .-main .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB3798: .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) call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3798: .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.cst8,"aM",@progbits,8 .align 8 .LC1: .long 0 .long 1072693248 .align 8 .LC2: .long -1718616160 .long 1076601226 .align 8 .LC3: .long 0 .long 1104006501 .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 <cstdlib> #include<chrono> int main(void) { double gammaEulera = 0.; double N = 1000000;; auto start = std::chrono::high_resolution_clock::now(); for (int i = 1; i < N; i++) gammaEulera = gammaEulera + (1. / (double)i); gammaEulera = gammaEulera - log(N); auto finish = std::chrono::high_resolution_clock::now(); std::chrono::duration<double> elapsed = finish - start; printf("Przyblizona wartosc eulera CPU wynosi %f \n",gammaEulera ); printf("Czas wynosi %f",elapsed.count()); return 0; }
#include <hip/hip_runtime.h> #include <stdio.h> #include <iostream> #include <cstdlib> #include<chrono> int main(void) { double gammaEulera = 0.; double N = 1000000;; auto start = std::chrono::high_resolution_clock::now(); for (int i = 1; i < N; i++) gammaEulera = gammaEulera + (1. / (double)i); gammaEulera = gammaEulera - log(N); auto finish = std::chrono::high_resolution_clock::now(); std::chrono::duration<double> elapsed = finish - start; printf("Przyblizona wartosc eulera CPU wynosi %f \n",gammaEulera ); printf("Czas wynosi %f",elapsed.count()); 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 <iostream> #include <cstdlib> #include<chrono> int main(void) { double gammaEulera = 0.; double N = 1000000;; auto start = std::chrono::high_resolution_clock::now(); for (int i = 1; i < N; i++) gammaEulera = gammaEulera + (1. / (double)i); gammaEulera = gammaEulera - log(N); auto finish = std::chrono::high_resolution_clock::now(); std::chrono::duration<double> elapsed = finish - start; printf("Przyblizona wartosc eulera CPU wynosi %f \n",gammaEulera ); printf("Czas wynosi %f",elapsed.count()); return 0; }
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] 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 <iostream> #include <cstdlib> #include<chrono> int main(void) { double gammaEulera = 0.; double N = 1000000;; auto start = std::chrono::high_resolution_clock::now(); for (int i = 1; i < N; i++) gammaEulera = gammaEulera + (1. / (double)i); gammaEulera = gammaEulera - log(N); auto finish = std::chrono::high_resolution_clock::now(); std::chrono::duration<double> elapsed = finish - start; printf("Przyblizona wartosc eulera CPU wynosi %f \n",gammaEulera ); printf("Czas wynosi %f",elapsed.count()); return 0; }
.text .file "euler-CPU.hip" # Start of file scope inline assembly .globl _ZSt21ios_base_library_initv # End of file scope inline assembly .section .rodata.cst8,"aM",@progbits,8 .p2align 3, 0x0 # -- Begin function main .LCPI0_0: .quad 0x3ff0000000000000 # double 1 .LCPI0_1: .quad 0xc02ba18a998fffa0 # double -13.815510557964274 .LCPI0_2: .quad 0x41cdcd6500000000 # double 1.0E+9 .text .globl main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %rbx .cfi_def_cfa_offset 24 subq $24, %rsp .cfi_def_cfa_offset 48 .cfi_offset %rbx, -24 .cfi_offset %rbp, -16 movl $999999, %ebp # imm = 0xF423F callq _ZNSt6chrono3_V212system_clock3nowEv movsd .LCPI0_0(%rip), %xmm3 # xmm3 = mem[0],zero xorpd %xmm2, %xmm2 movq %rax, %rbx movapd %xmm3, %xmm0 .p2align 4, 0x90 .LBB0_1: # =>This Inner Loop Header: Depth=1 movapd %xmm3, %xmm1 divsd %xmm0, %xmm1 addsd %xmm1, %xmm2 addsd %xmm3, %xmm0 decl %ebp jne .LBB0_1 # %bb.2: addsd .LCPI0_1(%rip), %xmm2 movsd %xmm2, 16(%rsp) # 8-byte Spill callq _ZNSt6chrono3_V212system_clock3nowEv subq %rbx, %rax xorps %xmm0, %xmm0 cvtsi2sd %rax, %xmm0 divsd .LCPI0_2(%rip), %xmm0 movsd %xmm0, 8(%rsp) # 8-byte Spill movl $.L.str, %edi movsd 16(%rsp), %xmm0 # 8-byte Reload # xmm0 = mem[0],zero movb $1, %al callq printf movl $.L.str.1, %edi movsd 8(%rsp), %xmm0 # 8-byte Reload # xmm0 = mem[0],zero movb $1, %al callq printf xorl %eax, %eax addq $24, %rsp .cfi_def_cfa_offset 24 popq %rbx .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end0: .size main, .Lfunc_end0-main .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "Przyblizona wartosc eulera CPU wynosi %f \n" .size .L.str, 43 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "Czas wynosi %f" .size .L.str.1, 15 .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 __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] 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_0003cafb_00000000-6_euler-CPU.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB3775: .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 .LFE3775: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC4: .string "Przyblizona wartosc eulera CPU wynosi %f \n" .section .rodata.str1.1,"aMS",@progbits,1 .LC5: .string "Czas wynosi %f" .text .globl main .type main, @function main: .LFB3768: .cfi_startproc endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 subq $16, %rsp .cfi_def_cfa_offset 32 call _ZNSt6chrono3_V212system_clock3nowEv@PLT movq %rax, %rbx movl $1, %eax pxor %xmm0, %xmm0 movsd .LC1(%rip), %xmm3 .L4: pxor %xmm2, %xmm2 cvtsi2sdl %eax, %xmm2 movapd %xmm3, %xmm1 divsd %xmm2, %xmm1 addsd %xmm1, %xmm0 addl $1, %eax cmpl $1000000, %eax jne .L4 subsd .LC2(%rip), %xmm0 movsd %xmm0, (%rsp) call _ZNSt6chrono3_V212system_clock3nowEv@PLT subq %rbx, %rax pxor %xmm0, %xmm0 cvtsi2sdq %rax, %xmm0 movapd %xmm0, %xmm4 divsd .LC3(%rip), %xmm4 movsd %xmm4, 8(%rsp) movsd (%rsp), %xmm0 leaq .LC4(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movsd 8(%rsp), %xmm0 leaq .LC5(%rip), %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT movl $0, %eax addq $16, %rsp .cfi_def_cfa_offset 16 popq %rbx .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3768: .size main, .-main .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB3798: .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) call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE3798: .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.cst8,"aM",@progbits,8 .align 8 .LC1: .long 0 .long 1072693248 .align 8 .LC2: .long -1718616160 .long 1076601226 .align 8 .LC3: .long 0 .long 1104006501 .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 "euler-CPU.hip" # Start of file scope inline assembly .globl _ZSt21ios_base_library_initv # End of file scope inline assembly .section .rodata.cst8,"aM",@progbits,8 .p2align 3, 0x0 # -- Begin function main .LCPI0_0: .quad 0x3ff0000000000000 # double 1 .LCPI0_1: .quad 0xc02ba18a998fffa0 # double -13.815510557964274 .LCPI0_2: .quad 0x41cdcd6500000000 # double 1.0E+9 .text .globl main .p2align 4, 0x90 .type main,@function main: # @main .cfi_startproc # %bb.0: pushq %rbp .cfi_def_cfa_offset 16 pushq %rbx .cfi_def_cfa_offset 24 subq $24, %rsp .cfi_def_cfa_offset 48 .cfi_offset %rbx, -24 .cfi_offset %rbp, -16 movl $999999, %ebp # imm = 0xF423F callq _ZNSt6chrono3_V212system_clock3nowEv movsd .LCPI0_0(%rip), %xmm3 # xmm3 = mem[0],zero xorpd %xmm2, %xmm2 movq %rax, %rbx movapd %xmm3, %xmm0 .p2align 4, 0x90 .LBB0_1: # =>This Inner Loop Header: Depth=1 movapd %xmm3, %xmm1 divsd %xmm0, %xmm1 addsd %xmm1, %xmm2 addsd %xmm3, %xmm0 decl %ebp jne .LBB0_1 # %bb.2: addsd .LCPI0_1(%rip), %xmm2 movsd %xmm2, 16(%rsp) # 8-byte Spill callq _ZNSt6chrono3_V212system_clock3nowEv subq %rbx, %rax xorps %xmm0, %xmm0 cvtsi2sd %rax, %xmm0 divsd .LCPI0_2(%rip), %xmm0 movsd %xmm0, 8(%rsp) # 8-byte Spill movl $.L.str, %edi movsd 16(%rsp), %xmm0 # 8-byte Reload # xmm0 = mem[0],zero movb $1, %al callq printf movl $.L.str.1, %edi movsd 8(%rsp), %xmm0 # 8-byte Reload # xmm0 = mem[0],zero movb $1, %al callq printf xorl %eax, %eax addq $24, %rsp .cfi_def_cfa_offset 24 popq %rbx .cfi_def_cfa_offset 16 popq %rbp .cfi_def_cfa_offset 8 retq .Lfunc_end0: .size main, .Lfunc_end0-main .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "Przyblizona wartosc eulera CPU wynosi %f \n" .size .L.str, 43 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "Czas wynosi %f" .size .L.str.1, 15 .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 __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
#include "includes.h" __global__ void kernelBackprop1(float *delta_nabla_w,int w_off,float *activations,float *delta_nabla_b,int b_off) { delta_nabla_w[w_off+(blockIdx.x*blockDim.x)+threadIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; //delta_nabla_w[w_off+(threadIdx.x*gridDim.x)+blockIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; }
code for sm_80 Function : _Z15kernelBackprop1PfiS_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 R7, SR_CTAID.X ; /* 0x0000000000077919 */ /* 0x000e220000002500 */ /*0020*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe40000000a00 */ /*0040*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e620000002100 */ /*0050*/ IADD3 R4, R7, c[0x0][0x180], RZ ; /* 0x0000600007047a10 */ /* 0x001fcc0007ffe0ff */ /*0060*/ IMAD.WIDE.U32 R2, R0, R9, c[0x0][0x170] ; /* 0x00005c0000027625 */ /* 0x002fc800078e0009 */ /*0070*/ IMAD.WIDE.U32 R4, R4, R9, c[0x0][0x178] ; /* 0x00005e0004047625 */ /* 0x000fe400078e0009 */ /*0080*/ LDG.E R2, [R2.64] ; /* 0x0000000402027981 */ /* 0x000ea8000c1e1900 */ /*0090*/ LDG.E R5, [R4.64] ; /* 0x0000000404057981 */ /* 0x000ea2000c1e1900 */ /*00a0*/ IMAD R0, R7, c[0x0][0x0], R0 ; /* 0x0000000007007a24 */ /* 0x000fca00078e0200 */ /*00b0*/ IADD3 R0, R0, c[0x0][0x168], RZ ; /* 0x00005a0000007a10 */ /* 0x000fca0007ffe0ff */ /*00c0*/ IMAD.WIDE.U32 R6, R0, R9, c[0x0][0x160] ; /* 0x0000580000067625 */ /* 0x000fc800078e0009 */ /*00d0*/ FMUL R9, R2, R5 ; /* 0x0000000502097220 */ /* 0x004fca0000400000 */ /*00e0*/ STG.E [R6.64], R9 ; /* 0x0000000906007986 */ /* 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 kernelBackprop1(float *delta_nabla_w,int w_off,float *activations,float *delta_nabla_b,int b_off) { delta_nabla_w[w_off+(blockIdx.x*blockDim.x)+threadIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; //delta_nabla_w[w_off+(threadIdx.x*gridDim.x)+blockIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; }
.file "tmpxft_00165639_00000000-6_kernelBackprop1.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__Z15kernelBackprop1PfiS_S_iPfiS_S_i .type _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i, @function _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movl %esi, 20(%rsp) movq %rdx, 8(%rsp) movq %rcx, (%rsp) movl %r8d, 16(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 20(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) movq %rsp, %rax movq %rax, 120(%rsp) leaq 16(%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 _Z15kernelBackprop1PfiS_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 _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i, .-_Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i .globl _Z15kernelBackprop1PfiS_S_i .type _Z15kernelBackprop1PfiS_S_i, @function _Z15kernelBackprop1PfiS_S_i: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z15kernelBackprop1PfiS_S_i, .-_Z15kernelBackprop1PfiS_S_i .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z15kernelBackprop1PfiS_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 _Z15kernelBackprop1PfiS_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 kernelBackprop1(float *delta_nabla_w,int w_off,float *activations,float *delta_nabla_b,int b_off) { delta_nabla_w[w_off+(blockIdx.x*blockDim.x)+threadIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; //delta_nabla_w[w_off+(threadIdx.x*gridDim.x)+blockIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; }
#include <hip/hip_runtime.h> #include "includes.h" __global__ void kernelBackprop1(float *delta_nabla_w,int w_off,float *activations,float *delta_nabla_b,int b_off) { delta_nabla_w[w_off+(blockIdx.x*blockDim.x)+threadIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; //delta_nabla_w[w_off+(threadIdx.x*gridDim.x)+blockIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; }
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 kernelBackprop1(float *delta_nabla_w,int w_off,float *activations,float *delta_nabla_b,int b_off) { delta_nabla_w[w_off+(blockIdx.x*blockDim.x)+threadIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; //delta_nabla_w[w_off+(threadIdx.x*gridDim.x)+blockIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z15kernelBackprop1PfiS_S_i .globl _Z15kernelBackprop1PfiS_S_i .p2align 8 .type _Z15kernelBackprop1PfiS_S_i,@function _Z15kernelBackprop1PfiS_S_i: s_load_b128 s[4:7], s[0:1], 0x10 v_lshlrev_b32_e32 v1, 2, v0 s_clause 0x1 s_load_b32 s2, s[0:1], 0x20 s_load_b32 s8, s[0:1], 0x34 s_mov_b32 s3, 0 s_waitcnt lgkmcnt(0) global_load_b32 v2, v1, s[4:5] s_load_b32 s4, s[0:1], 0x8 s_add_i32 s2, s15, s2 v_mov_b32_e32 v1, 0 s_lshl_b64 s[2:3], s[2:3], 2 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(SALU_CYCLE_1) s_add_u32 s2, s6, s2 s_addc_u32 s3, s7, s3 s_load_b32 s2, s[2:3], 0x0 s_load_b64 s[0:1], s[0:1], 0x0 s_and_b32 s3, s8, 0xffff s_mul_i32 s15, s15, s3 s_waitcnt lgkmcnt(0) v_add3_u32 v0, s15, s4, v0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[0:1] 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 s_waitcnt vmcnt(0) v_mul_f32_e32 v2, s2, 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 _Z15kernelBackprop1PfiS_S_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 296 .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 .text .Lfunc_end0: .size _Z15kernelBackprop1PfiS_S_i, .Lfunc_end0-_Z15kernelBackprop1PfiS_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 - .offset: 8 .size: 4 .value_kind: by_value - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer - .offset: 32 .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: _Z15kernelBackprop1PfiS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z15kernelBackprop1PfiS_S_i.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 HIP source code to AMD host assembly.
#include <hip/hip_runtime.h> #include "includes.h" __global__ void kernelBackprop1(float *delta_nabla_w,int w_off,float *activations,float *delta_nabla_b,int b_off) { delta_nabla_w[w_off+(blockIdx.x*blockDim.x)+threadIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; //delta_nabla_w[w_off+(threadIdx.x*gridDim.x)+blockIdx.x]=activations[threadIdx.x]*delta_nabla_b[b_off+blockIdx.x]; }
.text .file "kernelBackprop1.hip" .globl _Z30__device_stub__kernelBackprop1PfiS_S_i # -- Begin function _Z30__device_stub__kernelBackprop1PfiS_S_i .p2align 4, 0x90 .type _Z30__device_stub__kernelBackprop1PfiS_S_i,@function _Z30__device_stub__kernelBackprop1PfiS_S_i: # @_Z30__device_stub__kernelBackprop1PfiS_S_i .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movl %esi, 4(%rsp) movq %rdx, 64(%rsp) movq %rcx, 56(%rsp) movl %r8d, (%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 4(%rsp), %rax movq %rax, 88(%rsp) leaq 64(%rsp), %rax movq %rax, 96(%rsp) leaq 56(%rsp), %rax movq %rax, 104(%rsp) movq %rsp, %rax movq %rax, 112(%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 $_Z15kernelBackprop1PfiS_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__kernelBackprop1PfiS_S_i, .Lfunc_end0-_Z30__device_stub__kernelBackprop1PfiS_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 $_Z15kernelBackprop1PfiS_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 _Z15kernelBackprop1PfiS_S_i,@object # @_Z15kernelBackprop1PfiS_S_i .section .rodata,"a",@progbits .globl _Z15kernelBackprop1PfiS_S_i .p2align 3, 0x0 _Z15kernelBackprop1PfiS_S_i: .quad _Z30__device_stub__kernelBackprop1PfiS_S_i .size _Z15kernelBackprop1PfiS_S_i, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z15kernelBackprop1PfiS_S_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 _Z30__device_stub__kernelBackprop1PfiS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z15kernelBackprop1PfiS_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 : _Z15kernelBackprop1PfiS_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 R7, SR_CTAID.X ; /* 0x0000000000077919 */ /* 0x000e220000002500 */ /*0020*/ HFMA2.MMA R9, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff097435 */ /* 0x000fe200000001ff */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe40000000a00 */ /*0040*/ S2R R0, SR_TID.X ; /* 0x0000000000007919 */ /* 0x000e620000002100 */ /*0050*/ IADD3 R4, R7, c[0x0][0x180], RZ ; /* 0x0000600007047a10 */ /* 0x001fcc0007ffe0ff */ /*0060*/ IMAD.WIDE.U32 R2, R0, R9, c[0x0][0x170] ; /* 0x00005c0000027625 */ /* 0x002fc800078e0009 */ /*0070*/ IMAD.WIDE.U32 R4, R4, R9, c[0x0][0x178] ; /* 0x00005e0004047625 */ /* 0x000fe400078e0009 */ /*0080*/ LDG.E R2, [R2.64] ; /* 0x0000000402027981 */ /* 0x000ea8000c1e1900 */ /*0090*/ LDG.E R5, [R4.64] ; /* 0x0000000404057981 */ /* 0x000ea2000c1e1900 */ /*00a0*/ IMAD R0, R7, c[0x0][0x0], R0 ; /* 0x0000000007007a24 */ /* 0x000fca00078e0200 */ /*00b0*/ IADD3 R0, R0, c[0x0][0x168], RZ ; /* 0x00005a0000007a10 */ /* 0x000fca0007ffe0ff */ /*00c0*/ IMAD.WIDE.U32 R6, R0, R9, c[0x0][0x160] ; /* 0x0000580000067625 */ /* 0x000fc800078e0009 */ /*00d0*/ FMUL R9, R2, R5 ; /* 0x0000000502097220 */ /* 0x004fca0000400000 */ /*00e0*/ STG.E [R6.64], R9 ; /* 0x0000000906007986 */ /* 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 _Z15kernelBackprop1PfiS_S_i .globl _Z15kernelBackprop1PfiS_S_i .p2align 8 .type _Z15kernelBackprop1PfiS_S_i,@function _Z15kernelBackprop1PfiS_S_i: s_load_b128 s[4:7], s[0:1], 0x10 v_lshlrev_b32_e32 v1, 2, v0 s_clause 0x1 s_load_b32 s2, s[0:1], 0x20 s_load_b32 s8, s[0:1], 0x34 s_mov_b32 s3, 0 s_waitcnt lgkmcnt(0) global_load_b32 v2, v1, s[4:5] s_load_b32 s4, s[0:1], 0x8 s_add_i32 s2, s15, s2 v_mov_b32_e32 v1, 0 s_lshl_b64 s[2:3], s[2:3], 2 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_4) | instid1(SALU_CYCLE_1) s_add_u32 s2, s6, s2 s_addc_u32 s3, s7, s3 s_load_b32 s2, s[2:3], 0x0 s_load_b64 s[0:1], s[0:1], 0x0 s_and_b32 s3, s8, 0xffff s_mul_i32 s15, s15, s3 s_waitcnt lgkmcnt(0) v_add3_u32 v0, s15, s4, v0 s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1) v_lshlrev_b64 v[0:1], 2, v[0:1] 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 s_waitcnt vmcnt(0) v_mul_f32_e32 v2, s2, 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 _Z15kernelBackprop1PfiS_S_i .amdhsa_group_segment_fixed_size 0 .amdhsa_private_segment_fixed_size 0 .amdhsa_kernarg_size 296 .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 .text .Lfunc_end0: .size _Z15kernelBackprop1PfiS_S_i, .Lfunc_end0-_Z15kernelBackprop1PfiS_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 - .offset: 8 .size: 4 .value_kind: by_value - .address_space: global .offset: 16 .size: 8 .value_kind: global_buffer - .address_space: global .offset: 24 .size: 8 .value_kind: global_buffer - .offset: 32 .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: _Z15kernelBackprop1PfiS_S_i .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z15kernelBackprop1PfiS_S_i.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 host assembly to AMD host assembly.
.file "tmpxft_00165639_00000000-6_kernelBackprop1.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__Z15kernelBackprop1PfiS_S_iPfiS_S_i .type _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i, @function _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i: .LFB2051: .cfi_startproc endbr64 subq $152, %rsp .cfi_def_cfa_offset 160 movq %rdi, 24(%rsp) movl %esi, 20(%rsp) movq %rdx, 8(%rsp) movq %rcx, (%rsp) movl %r8d, 16(%rsp) movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 24(%rsp), %rax movq %rax, 96(%rsp) leaq 20(%rsp), %rax movq %rax, 104(%rsp) leaq 8(%rsp), %rax movq %rax, 112(%rsp) movq %rsp, %rax movq %rax, 120(%rsp) leaq 16(%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 _Z15kernelBackprop1PfiS_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 _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i, .-_Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i .globl _Z15kernelBackprop1PfiS_S_i .type _Z15kernelBackprop1PfiS_S_i, @function _Z15kernelBackprop1PfiS_S_i: .LFB2052: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z41__device_stub__Z15kernelBackprop1PfiS_S_iPfiS_S_i addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .size _Z15kernelBackprop1PfiS_S_i, .-_Z15kernelBackprop1PfiS_S_i .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "_Z15kernelBackprop1PfiS_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 _Z15kernelBackprop1PfiS_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 "kernelBackprop1.hip" .globl _Z30__device_stub__kernelBackprop1PfiS_S_i # -- Begin function _Z30__device_stub__kernelBackprop1PfiS_S_i .p2align 4, 0x90 .type _Z30__device_stub__kernelBackprop1PfiS_S_i,@function _Z30__device_stub__kernelBackprop1PfiS_S_i: # @_Z30__device_stub__kernelBackprop1PfiS_S_i .cfi_startproc # %bb.0: subq $120, %rsp .cfi_def_cfa_offset 128 movq %rdi, 72(%rsp) movl %esi, 4(%rsp) movq %rdx, 64(%rsp) movq %rcx, 56(%rsp) movl %r8d, (%rsp) leaq 72(%rsp), %rax movq %rax, 80(%rsp) leaq 4(%rsp), %rax movq %rax, 88(%rsp) leaq 64(%rsp), %rax movq %rax, 96(%rsp) leaq 56(%rsp), %rax movq %rax, 104(%rsp) movq %rsp, %rax movq %rax, 112(%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 $_Z15kernelBackprop1PfiS_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__kernelBackprop1PfiS_S_i, .Lfunc_end0-_Z30__device_stub__kernelBackprop1PfiS_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 $_Z15kernelBackprop1PfiS_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 _Z15kernelBackprop1PfiS_S_i,@object # @_Z15kernelBackprop1PfiS_S_i .section .rodata,"a",@progbits .globl _Z15kernelBackprop1PfiS_S_i .p2align 3, 0x0 _Z15kernelBackprop1PfiS_S_i: .quad _Z30__device_stub__kernelBackprop1PfiS_S_i .size _Z15kernelBackprop1PfiS_S_i, 8 .type .L__unnamed_1,@object # @0 .section .rodata.str1.1,"aMS",@progbits,1 .L__unnamed_1: .asciz "_Z15kernelBackprop1PfiS_S_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 _Z30__device_stub__kernelBackprop1PfiS_S_i .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z15kernelBackprop1PfiS_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 <iostream> #include "gMat.cuh" #include <vector> #include "mytime.h" #include <functional> #include <math.h> #include <stdlib.h> #include <string> void manyMult(gMat& A, gMat& B, gMat& C,int iter,int mem){ for (int i=0; i!=iter; ++i){ prod(A,B,C,mem); prod(A,C,B,mem); } } void multstat(int mem, int row, int col){ std::cout << "multiplying " << row << " by "<< col <<" matrices.\n"; cudaDeviceProp prop; cudaGetDeviceProperties(&prop,0); std::cout << "using "<< mem << " of "<<prop.sharedMemPerBlock << "shared memory\n"; int w=floor(sqrt(mem/2/sizeof(float))); std::cout << "Tile sizes will be: "<< w<< " by "<< w << "\n"; std::cout << "This amounts to: " << w*w << " threads \n"; std::cout << ceil(row/ (float) w)*ceil(col/ (float) w) << " tiles will be used "<<std::endl; } int main(int argc, char *argv[]){ int w; if ( argc >= 2) w=atoi(argv[1]); else w=20; int r=1000; int c=1000; gMat A=eye(r); gMat B=randgMat(r,c); gMat C=randgMat(r,c); A.name="A"; B.name="B"; C.name="C"; int mem=2*w*w*sizeof(float); auto mm=std::bind(manyMult,A,B,C,300,mem); multstat(mem,r,c); mm(); A.cleanup(); B.cleanup(); C.cleanup(); } void test1(){ std::vector<float> adat={1,2,3,4}; std::vector<float> bdat={2,0,0,2,0,2}; std::vector<float> cdat={0,0,0,0,0,0}; gMat A{adat,2,2}; gMat B{bdat,2,3}; gMat C{cdat,2,3}; std::cout <<"A:\n" << A << std::flush; std::cout <<"B:\n" << B << std::flush; int mem=8*sizeof(float); prod(A,B,C,mem); std::cout <<"C:\n" << C << std::flush; }
code for sm_80
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <iostream> #include "gMat.cuh" #include <vector> #include "mytime.h" #include <functional> #include <math.h> #include <stdlib.h> #include <string> void manyMult(gMat& A, gMat& B, gMat& C,int iter,int mem){ for (int i=0; i!=iter; ++i){ prod(A,B,C,mem); prod(A,C,B,mem); } } void multstat(int mem, int row, int col){ std::cout << "multiplying " << row << " by "<< col <<" matrices.\n"; cudaDeviceProp prop; cudaGetDeviceProperties(&prop,0); std::cout << "using "<< mem << " of "<<prop.sharedMemPerBlock << "shared memory\n"; int w=floor(sqrt(mem/2/sizeof(float))); std::cout << "Tile sizes will be: "<< w<< " by "<< w << "\n"; std::cout << "This amounts to: " << w*w << " threads \n"; std::cout << ceil(row/ (float) w)*ceil(col/ (float) w) << " tiles will be used "<<std::endl; } int main(int argc, char *argv[]){ int w; if ( argc >= 2) w=atoi(argv[1]); else w=20; int r=1000; int c=1000; gMat A=eye(r); gMat B=randgMat(r,c); gMat C=randgMat(r,c); A.name="A"; B.name="B"; C.name="C"; int mem=2*w*w*sizeof(float); auto mm=std::bind(manyMult,A,B,C,300,mem); multstat(mem,r,c); mm(); A.cleanup(); B.cleanup(); C.cleanup(); } void test1(){ std::vector<float> adat={1,2,3,4}; std::vector<float> bdat={2,0,0,2,0,2}; std::vector<float> cdat={0,0,0,0,0,0}; gMat A{adat,2,2}; gMat B{bdat,2,3}; gMat C{cdat,2,3}; std::cout <<"A:\n" << A << std::flush; std::cout <<"B:\n" << B << std::flush; int mem=8*sizeof(float); prod(A,B,C,mem); std::cout <<"C:\n" << C << std::flush; }
.file "tmpxft_001999a4_00000000-6_gMatTest.cudafe1.cpp" .text #APP .globl _ZSt21ios_base_library_initv #NO_APP .globl _Z8manyMultR4gMatS0_S0_ii .type _Z8manyMultR4gMatS0_S0_ii, @function _Z8manyMultR4gMatS0_S0_ii: .LFB5573: .cfi_startproc endbr64 testl %ecx, %ecx je .L6 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 movq %rdi, %rbp movq %rsi, %r12 movq %rdx, %r13 movl %r8d, %r14d movslq %ecx, %r15 movl $0, %ebx .L3: movl %r14d, %ecx movq %r13, %rdx movq %r12, %rsi movq %rbp, %rdi call _Z4prodRK4gMatS1_RS_i@PLT movl %r14d, %ecx movq %r12, %rdx movq %r13, %rsi movq %rbp, %rdi call _Z4prodRK4gMatS1_RS_i@PLT addq $1, %rbx cmpq %rbx, %r15 jne .L3 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 .L6: .cfi_restore 3 .cfi_restore 6 .cfi_restore 12 .cfi_restore 13 .cfi_restore 14 .cfi_restore 15 ret .cfi_endproc .LFE5573: .size _Z8manyMultR4gMatS0_S0_ii, .-_Z8manyMultR4gMatS0_S0_ii .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB5613: .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 .LFE5613: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "multiplying " .LC1: .string " by " .LC2: .string " matrices.\n" .LC3: .string "using " .LC4: .string " of " .LC5: .string "shared memory\n" .LC9: .string "Tile sizes will be: " .LC10: .string "\n" .LC11: .string "This amounts to: " .LC12: .string " threads \n" .LC16: .string " tiles will be used " .text .globl _Z8multstatiii .type _Z8multstatiii, @function _Z8multstatiii: .LFB5574: .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 $1040, %rsp .cfi_def_cfa_offset 1088 movl %edi, %r12d movl %esi, %ebp movl %edx, %ebx movq %fs:40, %rax movq %rax, 1032(%rsp) xorl %eax, %eax movl $12, %edx leaq .LC0(%rip), %rsi leaq _ZSt4cout(%rip), %r13 movq %r13, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %ebp, %esi movq %r13, %rdi call _ZNSolsEi@PLT movq %rax, %r14 movl $4, %edx leaq .LC1(%rip), %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %ebx, %esi movq %r14, %rdi call _ZNSolsEi@PLT movq %rax, %rdi movl $11, %edx leaq .LC2(%rip), %rsi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movq %rsp, %rdi movl $0, %esi call cudaGetDeviceProperties_v2@PLT movl $6, %edx leaq .LC3(%rip), %rsi movq %r13, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %r12d, %esi movq %r13, %rdi call _ZNSolsEi@PLT movq %rax, %r13 movl $4, %edx leaq .LC4(%rip), %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movq 296(%rsp), %rsi movq %r13, %rdi call _ZNSo9_M_insertImEERSoT_@PLT movq %rax, %rdi movl $14, %edx leaq .LC5(%rip), %rsi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %r12d, %eax shrl $31, %eax addl %r12d, %eax sarl %eax cltq shrq $2, %rax pxor %xmm0, %xmm0 cvtsi2sdq %rax, %xmm0 pxor %xmm1, %xmm1 ucomisd %xmm0, %xmm1 ja .L26 sqrtsd %xmm0, %xmm0 .L16: movapd %xmm0, %xmm3 movsd .LC18(%rip), %xmm1 movapd %xmm0, %xmm2 andpd %xmm1, %xmm2 movsd .LC7(%rip), %xmm4 ucomisd %xmm2, %xmm4 jbe .L17 cvttsd2siq %xmm0, %rax pxor %xmm2, %xmm2 cvtsi2sdq %rax, %xmm2 andnpd %xmm0, %xmm1 movapd %xmm2, %xmm3 orpd %xmm1, %xmm3 .L17: cvttsd2sil %xmm3, %r12d movl $20, %edx leaq .LC9(%rip), %rsi leaq _ZSt4cout(%rip), %r13 movq %r13, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %r12d, %esi movq %r13, %rdi call _ZNSolsEi@PLT movq %rax, %r14 movl $4, %edx leaq .LC1(%rip), %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %r12d, %esi movq %r14, %rdi call _ZNSolsEi@PLT movq %rax, %rdi movl $1, %edx leaq .LC10(%rip), %rsi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl $17, %edx leaq .LC11(%rip), %rsi movq %r13, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movl %r12d, %esi imull %r12d, %esi movq %r13, %rdi call _ZNSolsEi@PLT movq %rax, %rdi movl $10, %edx leaq .LC12(%rip), %rsi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT pxor %xmm4, %xmm4 cvtsi2ssl %r12d, %xmm4 pxor %xmm0, %xmm0 cvtsi2ssl %ebp, %xmm0 divss %xmm4, %xmm0 movaps %xmm0, %xmm1 movss .LC17(%rip), %xmm3 movaps %xmm0, %xmm2 andps %xmm3, %xmm2 movss .LC13(%rip), %xmm5 ucomiss %xmm2, %xmm5 jbe .L18 cvttss2sil %xmm0, %eax pxor %xmm2, %xmm2 cvtsi2ssl %eax, %xmm2 cmpnless %xmm2, %xmm1 movss .LC15(%rip), %xmm5 andps %xmm5, %xmm1 addss %xmm2, %xmm1 andnps %xmm0, %xmm3 orps %xmm3, %xmm1 .L18: movaps %xmm1, %xmm2 pxor %xmm1, %xmm1 cvtsi2ssl %ebx, %xmm1 divss %xmm4, %xmm1 movaps %xmm1, %xmm0 movss .LC17(%rip), %xmm4 movaps %xmm1, %xmm3 andps %xmm4, %xmm3 movss .LC13(%rip), %xmm5 ucomiss %xmm3, %xmm5 jbe .L19 cvttss2sil %xmm1, %eax pxor %xmm3, %xmm3 cvtsi2ssl %eax, %xmm3 cmpnless %xmm3, %xmm0 movss .LC15(%rip), %xmm5 andps %xmm5, %xmm0 addss %xmm3, %xmm0 andnps %xmm1, %xmm4 orps %xmm4, %xmm0 .L19: mulss %xmm2, %xmm0 cvtss2sd %xmm0, %xmm0 leaq _ZSt4cout(%rip), %rdi call _ZNSo9_M_insertIdEERSoT_@PLT movq %rax, %rbx movl $20, %edx leaq .LC16(%rip), %rsi movq %rax, %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT movq (%rbx), %rax movq -24(%rax), %rax movq 240(%rbx,%rax), %rbp testq %rbp, %rbp je .L28 cmpb $0, 56(%rbp) je .L22 movzbl 67(%rbp), %esi .L23: movsbl %sil, %esi movq %rbx, %rdi call _ZNSo3putEc@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT movq 1032(%rsp), %rax subq %fs:40, %rax jne .L29 addq $1040, %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 .L26: .cfi_restore_state call sqrt@PLT jmp .L16 .L28: movq 1032(%rsp), %rax subq %fs:40, %rax jne .L30 call _ZSt16__throw_bad_castv@PLT .L30: call __stack_chk_fail@PLT .L22: movq %rbp, %rdi call _ZNKSt5ctypeIcE13_M_widen_initEv@PLT movq 0(%rbp), %rax movl $10, %esi movq %rbp, %rdi call *48(%rax) movl %eax, %esi jmp .L23 .L29: call __stack_chk_fail@PLT .cfi_endproc .LFE5574: .size _Z8multstatiii, .-_Z8multstatiii .section .text._ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev,"axG",@progbits,_ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED5Ev,comdat .align 2 .weak _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev .type _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev, @function _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev: .LFB5604: .cfi_startproc endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 movq %rdi, %rbx movq 120(%rdi), %rdi leaq 136(%rbx), %rax cmpq %rax, %rdi je .L32 movq 136(%rbx), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L32: movq 72(%rbx), %rdi leaq 88(%rbx), %rax cmpq %rax, %rdi je .L33 movq 88(%rbx), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L33: movq 24(%rbx), %rdi leaq 40(%rbx), %rax cmpq %rax, %rdi je .L31 movq 40(%rbx), %rsi addq $1, %rsi call _ZdlPvm@PLT .L31: popq %rbx .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE5604: .size _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev, .-_ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev .weak _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED1Ev .set _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED1Ev,_ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB5636: .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) call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE5636: .size _ZL24__sti____cudaRegisterAllv, .-_ZL24__sti____cudaRegisterAllv .section .init_array,"aw" .align 8 .quad _ZL24__sti____cudaRegisterAllv .section .text._ZNSt6vectorIfSaIfEED2Ev,"axG",@progbits,_ZNSt6vectorIfSaIfEED5Ev,comdat .align 2 .weak _ZNSt6vectorIfSaIfEED2Ev .type _ZNSt6vectorIfSaIfEED2Ev, @function _ZNSt6vectorIfSaIfEED2Ev: .LFB5998: .cfi_startproc endbr64 movq (%rdi), %rax testq %rax, %rax je .L41 subq $8, %rsp .cfi_def_cfa_offset 16 movq 16(%rdi), %rsi subq %rax, %rsi movq %rax, %rdi call _ZdlPvm@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .L41: ret .cfi_endproc .LFE5998: .size _ZNSt6vectorIfSaIfEED2Ev, .-_ZNSt6vectorIfSaIfEED2Ev .weak _ZNSt6vectorIfSaIfEED1Ev .set _ZNSt6vectorIfSaIfEED1Ev,_ZNSt6vectorIfSaIfEED2Ev .section .text._ZNSt12_Vector_baseIfSaIfEED2Ev,"axG",@progbits,_ZNSt12_Vector_baseIfSaIfEED5Ev,comdat .align 2 .weak _ZNSt12_Vector_baseIfSaIfEED2Ev .type _ZNSt12_Vector_baseIfSaIfEED2Ev, @function _ZNSt12_Vector_baseIfSaIfEED2Ev: .LFB6202: .cfi_startproc endbr64 movq (%rdi), %rax testq %rax, %rax je .L47 subq $8, %rsp .cfi_def_cfa_offset 16 movq 16(%rdi), %rsi subq %rax, %rsi movq %rax, %rdi call _ZdlPvm@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .L47: ret .cfi_endproc .LFE6202: .size _ZNSt12_Vector_baseIfSaIfEED2Ev, .-_ZNSt12_Vector_baseIfSaIfEED2Ev .weak _ZNSt12_Vector_baseIfSaIfEED1Ev .set _ZNSt12_Vector_baseIfSaIfEED1Ev,_ZNSt12_Vector_baseIfSaIfEED2Ev .section .rodata.str1.1 .LC21: .string "A:\n" .LC22: .string "B:\n" .LC23: .string "C:\n" .text .globl _Z5test1v .type _Z5test1v, @function _Z5test1v: .LFB5610: .cfi_startproc .cfi_personality 0x9b,DW.ref.__gxx_personality_v0 .cfi_lsda 0x1b,.LLSDA5610 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 $288, %rsp .cfi_def_cfa_offset 320 movq %fs:40, %rax movq %rax, 280(%rsp) xorl %eax, %eax movq $0, (%rsp) movq $0, 8(%rsp) movq $0, 16(%rsp) movl $16, %edi .LEHB0: call _Znwm@PLT .LEHE0: movq %rax, %rbp movq %rax, (%rsp) leaq 16(%rax), %rax movq %rax, 16(%rsp) movabsq $4611686019492741120, %rcx movq %rcx, 0(%rbp) movabsq $4647714816524288000, %rbx movq %rbx, 8(%rbp) movq %rax, 8(%rsp) movss .LC19(%rip), %xmm0 movss %xmm0, 224(%rsp) movl $0x00000000, 228(%rsp) movl $0x00000000, 232(%rsp) movss %xmm0, 236(%rsp) movl $0x00000000, 240(%rsp) movss %xmm0, 244(%rsp) movq $0, 32(%rsp) movq $0, 40(%rsp) movq $0, 48(%rsp) movl $24, %edi .LEHB1: call _Znwm@PLT .LEHE1: movq %rax, %rbx movq %rax, 32(%rsp) leaq 24(%rax), %rax movq %rax, 48(%rsp) movdqa 224(%rsp), %xmm1 movups %xmm1, (%rbx) movq 240(%rsp), %rdx movq %rdx, 16(%rbx) movq %rax, 40(%rsp) movl $0x00000000, 224(%rsp) movl $0x00000000, 228(%rsp) movl $0x00000000, 232(%rsp) movl $0x00000000, 236(%rsp) movl $0x00000000, 240(%rsp) movl $0x00000000, 244(%rsp) movq $0, 64(%rsp) movq $0, 72(%rsp) movq $0, 80(%rsp) movl $24, %edi .LEHB2: call _Znwm@PLT .LEHE2: jmp .L90 .L83: endbr64 movq %rax, %rbx movq %rsp, %rdi call _ZNSt12_Vector_baseIfSaIfEED2Ev movq 280(%rsp), %rax subq %fs:40, %rax je .L53 call __stack_chk_fail@PLT .L53: movq %rbx, %rdi .LEHB3: call _Unwind_Resume@PLT .LEHE3: .L90: movq %rax, %r12 movq %rax, 64(%rsp) leaq 24(%rax), %rax movq %rax, 80(%rsp) movdqa 224(%rsp), %xmm2 movups %xmm2, (%r12) movq 240(%rsp), %rdx movq %rdx, 16(%r12) movq %rax, 72(%rsp) movq $0, 96(%rsp) movq $0, 104(%rsp) movq $0, 112(%rsp) movl $16, %edi .LEHB4: call _Znwm@PLT .LEHE4: jmp .L91 .L82: endbr64 movq %rax, %rbx leaq 32(%rsp), %rdi call _ZNSt12_Vector_baseIfSaIfEED2Ev .L56: movq %rsp, %rdi call _ZNSt6vectorIfSaIfEED1Ev movq 280(%rsp), %rax subq %fs:40, %rax je .L73 call __stack_chk_fail@PLT .L91: movq %rax, 96(%rsp) leaq 16(%rax), %rdx movq %rdx, 112(%rsp) movdqu 0(%rbp), %xmm3 movups %xmm3, (%rax) movq %rdx, 104(%rsp) leaq 96(%rsp), %rsi leaq 128(%rsp), %rdi movl $0, %r8d movl $2, %ecx movl $2, %edx .LEHB5: call _ZN4gMatC1ESt6vectorIfSaIfEEiii@PLT .LEHE5: jmp .L92 .L84: endbr64 movq %rax, %rbx leaq 64(%rsp), %rdi call _ZNSt12_Vector_baseIfSaIfEED2Ev .L59: leaq 32(%rsp), %rdi call _ZNSt6vectorIfSaIfEED1Ev jmp .L56 .L92: movq 96(%rsp), %rdi testq %rdi, %rdi je .L60 movq 112(%rsp), %rsi subq %rdi, %rsi call _ZdlPvm@PLT .L60: movq $0, 96(%rsp) movq $0, 104(%rsp) movq $0, 112(%rsp) movl $24, %edi .LEHB6: call _Znwm@PLT .LEHE6: movq %rax, 96(%rsp) leaq 24(%rax), %rdx movq %rdx, 112(%rsp) movdqu (%rbx), %xmm4 movups %xmm4, (%rax) movq 16(%rbx), %rcx movq %rcx, 16(%rax) movq %rdx, 104(%rsp) leaq 96(%rsp), %rsi leaq 176(%rsp), %rdi movl $0, %r8d movl $3, %ecx movl $2, %edx .LEHB7: call _ZN4gMatC1ESt6vectorIfSaIfEEiii@PLT .LEHE7: movq 96(%rsp), %rdi testq %rdi, %rdi je .L61 movq 112(%rsp), %rsi subq %rdi, %rsi call _ZdlPvm@PLT .L61: movq $0, 96(%rsp) movq $0, 104(%rsp) movq $0, 112(%rsp) movl $24, %edi .LEHB8: call _Znwm@PLT .LEHE8: movq %rax, 96(%rsp) leaq 24(%rax), %rdx movq %rdx, 112(%rsp) movdqu (%r12), %xmm5 movups %xmm5, (%rax) movq 16(%r12), %rcx movq %rcx, 16(%rax) movq %rdx, 104(%rsp) leaq 96(%rsp), %rsi leaq 224(%rsp), %rdi movl $0, %r8d movl $3, %ecx movl $2, %edx .LEHB9: call _ZN4gMatC1ESt6vectorIfSaIfEEiii@PLT .LEHE9: movq 96(%rsp), %rdi testq %rdi, %rdi je .L62 movq 112(%rsp), %rsi subq %rdi, %rsi call _ZdlPvm@PLT .L62: movl $3, %edx leaq .LC21(%rip), %rsi leaq _ZSt4cout(%rip), %rdi .LEHB10: call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT leaq 128(%rsp), %rsi leaq _ZSt4cout(%rip), %rdi call _ZlsRSoR4gMat@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT movl $3, %edx leaq .LC22(%rip), %rsi leaq _ZSt4cout(%rip), %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT leaq 176(%rsp), %rsi leaq _ZSt4cout(%rip), %rdi call _ZlsRSoR4gMat@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT leaq 224(%rsp), %rdx leaq 176(%rsp), %rsi leaq 128(%rsp), %rdi movl $32, %ecx call _Z4prodRK4gMatS1_RS_i@PLT movl $3, %edx leaq .LC23(%rip), %rsi leaq _ZSt4cout(%rip), %rdi call _ZSt16__ostream_insertIcSt11char_traitsIcEERSt13basic_ostreamIT_T0_ES6_PKS3_l@PLT leaq 224(%rsp), %rsi leaq _ZSt4cout(%rip), %rdi call _ZlsRSoR4gMat@PLT movq %rax, %rdi call _ZNSo5flushEv@PLT .LEHE10: movq 240(%rsp), %rdi leaq 256(%rsp), %rax cmpq %rax, %rdi je .L63 movq 256(%rsp), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L63: movq 192(%rsp), %rdi leaq 208(%rsp), %rax cmpq %rax, %rdi je .L64 movq 208(%rsp), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L64: movq 144(%rsp), %rdi leaq 160(%rsp), %rax cmpq %rax, %rdi je .L65 movq 160(%rsp), %rax leaq 1(%rax), %rsi call _ZdlPvm@PLT .L65: movl $24, %esi movq %r12, %rdi call _ZdlPvm@PLT movl $24, %esi movq %rbx, %rdi call _ZdlPvm@PLT movl $16, %esi movq %rbp, %rdi call _ZdlPvm@PLT movq 280(%rsp), %rax subq %fs:40, %rax jne .L93 addq $288, %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 .L75: .cfi_restore_state endbr64 movq %rax, %rbx leaq 96(%rsp), %rdi call _ZNSt6vectorIfSaIfEED1Ev .L67: leaq 64(%rsp), %rdi call _ZNSt6vectorIfSaIfEED1Ev jmp .L59 .L77: endbr64 movq %rax, %rbx leaq 96(%rsp), %rdi call _ZNSt6vectorIfSaIfEED1Ev .L69: leaq 144(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT jmp .L67 .L79: endbr64 movq %rax, %rbx leaq 96(%rsp), %rdi call _ZNSt6vectorIfSaIfEED1Ev .L71: leaq 192(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT jmp .L69 .L81: endbr64 movq %rax, %rbx leaq 240(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT jmp .L71 .L80: endbr64 movq %rax, %rbx jmp .L71 .L78: endbr64 movq %rax, %rbx jmp .L69 .L76: endbr64 movq %rax, %rbx jmp .L67 .L73: movq %rbx, %rdi .LEHB11: call _Unwind_Resume@PLT .LEHE11: .L93: call __stack_chk_fail@PLT .cfi_endproc .LFE5610: .globl __gxx_personality_v0 .section .gcc_except_table,"a",@progbits .LLSDA5610: .byte 0xff .byte 0xff .byte 0x1 .uleb128 .LLSDACSE5610-.LLSDACSB5610 .LLSDACSB5610: .uleb128 .LEHB0-.LFB5610 .uleb128 .LEHE0-.LEHB0 .uleb128 .L83-.LFB5610 .uleb128 0 .uleb128 .LEHB1-.LFB5610 .uleb128 .LEHE1-.LEHB1 .uleb128 .L82-.LFB5610 .uleb128 0 .uleb128 .LEHB2-.LFB5610 .uleb128 .LEHE2-.LEHB2 .uleb128 .L84-.LFB5610 .uleb128 0 .uleb128 .LEHB3-.LFB5610 .uleb128 .LEHE3-.LEHB3 .uleb128 0 .uleb128 0 .uleb128 .LEHB4-.LFB5610 .uleb128 .LEHE4-.LEHB4 .uleb128 .L76-.LFB5610 .uleb128 0 .uleb128 .LEHB5-.LFB5610 .uleb128 .LEHE5-.LEHB5 .uleb128 .L75-.LFB5610 .uleb128 0 .uleb128 .LEHB6-.LFB5610 .uleb128 .LEHE6-.LEHB6 .uleb128 .L78-.LFB5610 .uleb128 0 .uleb128 .LEHB7-.LFB5610 .uleb128 .LEHE7-.LEHB7 .uleb128 .L77-.LFB5610 .uleb128 0 .uleb128 .LEHB8-.LFB5610 .uleb128 .LEHE8-.LEHB8 .uleb128 .L80-.LFB5610 .uleb128 0 .uleb128 .LEHB9-.LFB5610 .uleb128 .LEHE9-.LEHB9 .uleb128 .L79-.LFB5610 .uleb128 0 .uleb128 .LEHB10-.LFB5610 .uleb128 .LEHE10-.LEHB10 .uleb128 .L81-.LFB5610 .uleb128 0 .uleb128 .LEHB11-.LFB5610 .uleb128 .LEHE11-.LEHB11 .uleb128 0 .uleb128 0 .LLSDACSE5610: .text .size _Z5test1v, .-_Z5test1v .section .text._ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag,"axG",@progbits,_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag,comdat .align 2 .weak _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag .type _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag, @function _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag: .LFB6246: .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 .L101 movq (%rdi), %rdi cmpq $1, %rdx jne .L97 movzbl (%rsi), %eax movb %al, (%rdi) .L98: movq (%rsp), %rax movq %rax, 8(%rbx) movq (%rbx), %rdx movb $0, (%rdx,%rax) movq 8(%rsp), %rax subq %fs:40, %rax jne .L102 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 .L101: .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) .L96: movq %rbp, %rdx movq %r12, %rsi call memcpy@PLT jmp .L98 .L97: testq %rdx, %rdx je .L98 jmp .L96 .L102: call __stack_chk_fail@PLT .cfi_endproc .LFE6246: .size _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag, .-_ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag .section .rodata.str1.1 .LC24: .string "A" .LC25: .string "B" .LC26: .string "C" .text .globl main .type main, @function main: .LFB5575: .cfi_startproc .cfi_personality 0x9b,DW.ref.__gxx_personality_v0 .cfi_lsda 0x1b,.LLSDA5575 endbr64 pushq %rbx .cfi_def_cfa_offset 16 .cfi_offset 3, -16 subq $320, %rsp .cfi_def_cfa_offset 336 movq %fs:40, %rax movq %rax, 312(%rsp) xorl %eax, %eax movl $20, %ebx cmpl $1, %edi jg .L126 .L104: movq %rsp, %rdi movl $1000, %esi .LEHB12: call _Z3eyei@PLT .LEHE12: leaq 48(%rsp), %rdi movl $1000, %edx movl $1000, %esi .LEHB13: call _Z8randgMatii@PLT .LEHE13: jmp .L127 .L126: movq 8(%rsi), %rdi movl $10, %edx movl $0, %esi call __isoc23_strtol@PLT movl %eax, %ebx jmp .L104 .L127: leaq 96(%rsp), %rdi movl $1000, %edx movl $1000, %esi .LEHB14: call _Z8randgMatii@PLT .LEHE14: leaq 16(%rsp), %rdi leaq .LC24(%rip), %rsi .LEHB15: call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE6assignEPKc@PLT leaq 64(%rsp), %rdi leaq .LC25(%rip), %rsi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE6assignEPKc@PLT leaq 112(%rsp), %rdi leaq .LC26(%rip), %rsi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE6assignEPKc@PLT imull %ebx, %ebx sall $3, %ebx leaq _Z8manyMultR4gMatS0_S0_ii(%rip), %rax movq %rax, 144(%rsp) movl %ebx, 152(%rsp) movl $300, 156(%rsp) movq 96(%rsp), %rax movq %rax, 160(%rsp) movl 104(%rsp), %eax movl %eax, 168(%rsp) movl 108(%rsp), %eax movl %eax, 172(%rsp) leaq 192(%rsp), %rax movq %rax, 176(%rsp) movq 112(%rsp), %rsi movq %rsi, %rdx addq 120(%rsp), %rdx leaq 176(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag .LEHE15: movq 48(%rsp), %rax movq %rax, 208(%rsp) movl 56(%rsp), %eax movl %eax, 216(%rsp) movl 60(%rsp), %eax movl %eax, 220(%rsp) leaq 240(%rsp), %rax movq %rax, 224(%rsp) movq 64(%rsp), %rsi movq %rsi, %rdx addq 72(%rsp), %rdx leaq 224(%rsp), %rdi .LEHB16: call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag .LEHE16: jmp .L128 .L121: endbr64 movq %rax, %rbx leaq 176(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT .L107: leaq 112(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT .L111: leaq 64(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT .L112: leaq 16(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 312(%rsp), %rax subq %fs:40, %rax je .L113 call __stack_chk_fail@PLT .L128: movq (%rsp), %rax movq %rax, 256(%rsp) movl 8(%rsp), %eax movl %eax, 264(%rsp) movl 12(%rsp), %eax movl %eax, 268(%rsp) leaq 288(%rsp), %rax movq %rax, 272(%rsp) movq 16(%rsp), %rsi movq %rsi, %rdx addq 24(%rsp), %rdx leaq 272(%rsp), %rdi .LEHB17: call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE12_M_constructIPcEEvT_S7_St20forward_iterator_tag .LEHE17: jmp .L129 .L120: endbr64 movq %rax, %rbx leaq 224(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT leaq 176(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT jmp .L107 .L129: movl $1000, %edx movl $1000, %esi movl %ebx, %edi .LEHB18: call _Z8multstatiii leaq 160(%rsp), %rdx leaq 208(%rsp), %rsi leaq 256(%rsp), %rdi movl 152(%rsp), %r8d movl 156(%rsp), %ecx call *144(%rsp) movq %rsp, %rdi call _ZN4gMat7cleanupEv@PLT leaq 48(%rsp), %rdi call _ZN4gMat7cleanupEv@PLT leaq 96(%rsp), %rdi call _ZN4gMat7cleanupEv@PLT .LEHE18: leaq 152(%rsp), %rdi call _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev leaq 112(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT leaq 64(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT leaq 16(%rsp), %rdi call _ZNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEE10_M_disposeEv@PLT movq 312(%rsp), %rax subq %fs:40, %rax jne .L130 movl $0, %eax addq $320, %rsp .cfi_remember_state .cfi_def_cfa_offset 16 popq %rbx .cfi_def_cfa_offset 8 ret .L119: .cfi_restore_state endbr64 movq %rax, %rbx leaq 152(%rsp), %rdi call _ZNSt11_Tuple_implILm0EJ4gMatS0_S0_iiEED2Ev jmp .L107 .L118: endbr64 movq %rax, %rbx jmp .L107 .L117: endbr64 movq %rax, %rbx jmp .L111 .L116: endbr64 movq %rax, %rbx jmp .L112 .L113: movq %rbx, %rdi .LEHB19: call _Unwind_Resume@PLT .LEHE19: .L130: call __stack_chk_fail@PLT .cfi_endproc .LFE5575: .section .gcc_except_table .LLSDA5575: .byte 0xff .byte 0xff .byte 0x1 .uleb128 .LLSDACSE5575-.LLSDACSB5575 .LLSDACSB5575: .uleb128 .LEHB12-.LFB5575 .uleb128 .LEHE12-.LEHB12 .uleb128 0 .uleb128 0 .uleb128 .LEHB13-.LFB5575 .uleb128 .LEHE13-.LEHB13 .uleb128 .L116-.LFB5575 .uleb128 0 .uleb128 .LEHB14-.LFB5575 .uleb128 .LEHE14-.LEHB14 .uleb128 .L117-.LFB5575 .uleb128 0 .uleb128 .LEHB15-.LFB5575 .uleb128 .LEHE15-.LEHB15 .uleb128 .L118-.LFB5575 .uleb128 0 .uleb128 .LEHB16-.LFB5575 .uleb128 .LEHE16-.LEHB16 .uleb128 .L121-.LFB5575 .uleb128 0 .uleb128 .LEHB17-.LFB5575 .uleb128 .LEHE17-.LEHB17 .uleb128 .L120-.LFB5575 .uleb128 0 .uleb128 .LEHB18-.LFB5575 .uleb128 .LEHE18-.LEHB18 .uleb128 .L119-.LFB5575 .uleb128 0 .uleb128 .LEHB19-.LFB5575 .uleb128 .LEHE19-.LEHB19 .uleb128 0 .uleb128 0 .LLSDACSE5575: .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 .section .rodata.cst8,"aM",@progbits,8 .align 8 .LC7: .long 0 .long 1127219200 .section .rodata.cst4,"aM",@progbits,4 .align 4 .LC13: .long 1258291200 .align 4 .LC15: .long 1065353216 .set .LC17,.LC18+4 .section .rodata.cst8 .align 8 .LC18: .long -1 .long 2147483647 .section .rodata.cst4 .align 4 .LC19: .long 1073741824 .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 <iostream> #include "gMat.cuh" #include <vector> #include "mytime.h" #include <functional> #include <math.h> #include <stdlib.h> #include <string> void manyMult(gMat& A, gMat& B, gMat& C,int iter,int mem){ for (int i=0; i!=iter; ++i){ prod(A,B,C,mem); prod(A,C,B,mem); } } void multstat(int mem, int row, int col){ std::cout << "multiplying " << row << " by "<< col <<" matrices.\n"; cudaDeviceProp prop; cudaGetDeviceProperties(&prop,0); std::cout << "using "<< mem << " of "<<prop.sharedMemPerBlock << "shared memory\n"; int w=floor(sqrt(mem/2/sizeof(float))); std::cout << "Tile sizes will be: "<< w<< " by "<< w << "\n"; std::cout << "This amounts to: " << w*w << " threads \n"; std::cout << ceil(row/ (float) w)*ceil(col/ (float) w) << " tiles will be used "<<std::endl; } int main(int argc, char *argv[]){ int w; if ( argc >= 2) w=atoi(argv[1]); else w=20; int r=1000; int c=1000; gMat A=eye(r); gMat B=randgMat(r,c); gMat C=randgMat(r,c); A.name="A"; B.name="B"; C.name="C"; int mem=2*w*w*sizeof(float); auto mm=std::bind(manyMult,A,B,C,300,mem); multstat(mem,r,c); mm(); A.cleanup(); B.cleanup(); C.cleanup(); } void test1(){ std::vector<float> adat={1,2,3,4}; std::vector<float> bdat={2,0,0,2,0,2}; std::vector<float> cdat={0,0,0,0,0,0}; gMat A{adat,2,2}; gMat B{bdat,2,3}; gMat C{cdat,2,3}; std::cout <<"A:\n" << A << std::flush; std::cout <<"B:\n" << B << std::flush; int mem=8*sizeof(float); prod(A,B,C,mem); std::cout <<"C:\n" << C << std::flush; }
#include <hip/hip_runtime.h> #include <iostream> #include "gMat.cuh" #include <vector> #include "mytime.h" #include <functional> #include <math.h> #include <stdlib.h> #include <string> void manyMult(gMat& A, gMat& B, gMat& C,int iter,int mem){ for (int i=0; i!=iter; ++i){ prod(A,B,C,mem); prod(A,C,B,mem); } } void multstat(int mem, int row, int col){ std::cout << "multiplying " << row << " by "<< col <<" matrices.\n"; hipDeviceProp_t prop; hipGetDeviceProperties(&prop,0); std::cout << "using "<< mem << " of "<<prop.sharedMemPerBlock << "shared memory\n"; int w=floor(sqrt(mem/2/sizeof(float))); std::cout << "Tile sizes will be: "<< w<< " by "<< w << "\n"; std::cout << "This amounts to: " << w*w << " threads \n"; std::cout << ceil(row/ (float) w)*ceil(col/ (float) w) << " tiles will be used "<<std::endl; } int main(int argc, char *argv[]){ int w; if ( argc >= 2) w=atoi(argv[1]); else w=20; int r=1000; int c=1000; gMat A=eye(r); gMat B=randgMat(r,c); gMat C=randgMat(r,c); A.name="A"; B.name="B"; C.name="C"; int mem=2*w*w*sizeof(float); auto mm=std::bind(manyMult,A,B,C,300,mem); multstat(mem,r,c); mm(); A.cleanup(); B.cleanup(); C.cleanup(); } void test1(){ std::vector<float> adat={1,2,3,4}; std::vector<float> bdat={2,0,0,2,0,2}; std::vector<float> cdat={0,0,0,0,0,0}; gMat A{adat,2,2}; gMat B{bdat,2,3}; gMat C{cdat,2,3}; std::cout <<"A:\n" << A << std::flush; std::cout <<"B:\n" << B << std::flush; int mem=8*sizeof(float); prod(A,B,C,mem); std::cout <<"C:\n" << C << std::flush; }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include <iostream> #include "gMat.cuh" #include <vector> #include "mytime.h" #include <functional> #include <math.h> #include <stdlib.h> #include <string> void manyMult(gMat& A, gMat& B, gMat& C,int iter,int mem){ for (int i=0; i!=iter; ++i){ prod(A,B,C,mem); prod(A,C,B,mem); } } void multstat(int mem, int row, int col){ std::cout << "multiplying " << row << " by "<< col <<" matrices.\n"; hipDeviceProp_t prop; hipGetDeviceProperties(&prop,0); std::cout << "using "<< mem << " of "<<prop.sharedMemPerBlock << "shared memory\n"; int w=floor(sqrt(mem/2/sizeof(float))); std::cout << "Tile sizes will be: "<< w<< " by "<< w << "\n"; std::cout << "This amounts to: " << w*w << " threads \n"; std::cout << ceil(row/ (float) w)*ceil(col/ (float) w) << " tiles will be used "<<std::endl; } int main(int argc, char *argv[]){ int w; if ( argc >= 2) w=atoi(argv[1]); else w=20; int r=1000; int c=1000; gMat A=eye(r); gMat B=randgMat(r,c); gMat C=randgMat(r,c); A.name="A"; B.name="B"; C.name="C"; int mem=2*w*w*sizeof(float); auto mm=std::bind(manyMult,A,B,C,300,mem); multstat(mem,r,c); mm(); A.cleanup(); B.cleanup(); C.cleanup(); } void test1(){ std::vector<float> adat={1,2,3,4}; std::vector<float> bdat={2,0,0,2,0,2}; std::vector<float> cdat={0,0,0,0,0,0}; gMat A{adat,2,2}; gMat B{bdat,2,3}; gMat C{cdat,2,3}; std::cout <<"A:\n" << A << std::flush; std::cout <<"B:\n" << B << std::flush; int mem=8*sizeof(float); prod(A,B,C,mem); std::cout <<"C:\n" << C << std::flush; }
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] 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
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] 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 <cuda.h> /* Matrices are stored in row-major order: */ /* M(row, col) = (M.width*row +col); */ typedef struct{ /* suppose we use only square matrices */ int width; float *elements; } Matrix; /* Thread block size */ #define TILE_WIDTH 2 /***********************/ /* TODO, write KERNEL */ /***********************/ __global__ void MatMul(const float* Md, const float* Nd, float* Pd, const int Width) { __shared__ float Mds[TILE_WIDTH][TILE_WIDTH]; __shared__ float Nds[TILE_WIDTH][TILE_WIDTH]; int bx = blockIdx.x; int by = blockIdx.y; int tx = threadIdx.x; int ty = threadIdx.y; // Identify the row and column of the Pd element to work on int Row = by * TILE_WIDTH + ty; int Col = bx * TILE_WIDTH + tx; float Pvalue = 0; //Loop over the Md and Nd tiles required to compute the Pd element for (int m = 0; m < Width/TILE_WIDTH; m++) { // Coolaborative loading of Md and Nd tiles into shared memory Mds[ty][tx] = Md[Row*Width + (m*TILE_WIDTH + tx)]; Nds[ty][tx] = Nd[Col + (m*TILE_WIDTH + ty)*Width]; __syncthreads(); for (int k = 0; k < TILE_WIDTH; k++) Pvalue += Mds[ty][k] * Nds[k][tx]; __syncthreads(); } Pd[Row*Width+Col] = Pvalue; } /**/ void test(const Matrix C); /**/ int main(int argc, char* argv[]) { int i; /* init matrices */ Matrix h_A, h_B, h_C; cudaEvent_t start; cudaEvent_t stop; cudaEventCreate(&start); cudaEventCreate(&stop); /*******************/ /** READING INPUT **/ /*******************/ int size = 0; //dimension of matrices scanf("%d", &size); int full_size = sizeof(float)*size*size; h_A.width = size; h_B.width = size; h_C.width = size; /* Allocate host memory */ h_A.elements = (float*)malloc(full_size); h_B.elements = (float*)malloc(full_size); h_C.elements = (float*)malloc(full_size); /**/ for(i=0;i<size*size;++i){ scanf("%f", &h_A.elements[i]);} for(i=0;i<size*size;++i){ scanf("%f", &h_B.elements[i]);} /********************/ /** FINISHED INPUT **/ /********************/ /*************************/ /* allocate device */ /* memory for A,B,C */ /*************************/ Matrix d_A, d_B, d_C; d_A.width = size; d_B.width = size; d_C.width = size; cudaMalloc(&d_A.elements, full_size); cudaMalloc(&d_B.elements, full_size); cudaMalloc(&d_C.elements, full_size); cudaEventRecord(start,0); /***********************************/ /* copy vectors A&B to device */ /***********************************/ cudaMemcpy(d_A.elements, h_A.elements, full_size, cudaMemcpyHostToDevice); cudaMemcpy(d_B.elements, h_B.elements, full_size, cudaMemcpyHostToDevice); /*********************/ /* call kernel */ /*********************/ dim3 dimBlock(TILE_WIDTH, TILE_WIDTH); dim3 dimGrid(h_B.width/dimBlock.x, h_A.width/dimBlock.y); MatMul<<<dimGrid,dimBlock>>>(d_A.elements, d_B.elements, d_C.elements,d_A.width); /**************************/ /* copy result back */ /**************************/ cudaMemcpy(h_C.elements, d_C.elements, full_size, cudaMemcpyDeviceToHost); cudaEventRecord(stop, 0); cudaEventSynchronize(stop); float elapsedTime; cudaEventElapsedTime(&elapsedTime, start, stop); fprintf(stderr,"Elapsed time = %f (s)\n",elapsedTime); cudaEventDestroy(start); cudaEventDestroy(stop); /*******************************************/ /** Testing output, don't change anything! */ /*******************************************/ test(h_C); /* free device memory */ cudaFree(d_A.elements); cudaFree(d_B.elements); cudaFree(d_C.elements); /* free host memory */ free(h_A.elements); free(h_B.elements); free(h_C.elements); return 0; } //function to test the input, don't change anything! void test(const Matrix C) { int i,j; //int size = C.width*C.width; for(i=0;i<C.width;++i) { for(j=0;j<C.width;++j) printf("%4.1f ", C.elements[i*C.width+j]); printf("\n"); } }
code for sm_80 Function : _Z6MatMulPKfS0_Pfi .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 R3, SR_CTAID.Y ; /* 0x0000000000037919 */ /* 0x000e220000002600 */ /*0020*/ MOV R14, c[0x0][0x178] ; /* 0x00005e00000e7a02 */ /* 0x000fe20000000f00 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0050*/ S2R R0, SR_TID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002200 */ /*0060*/ ISETP.GE.AND P0, PT, R14, 0x2, PT ; /* 0x000000020e00780c */ /* 0x000fc60003f06270 */ /*0070*/ S2R R12, SR_CTAID.X ; /* 0x00000000000c7919 */ /* 0x000e680000002500 */ /*0080*/ S2R R13, SR_TID.X ; /* 0x00000000000d7919 */ /* 0x000ea20000002100 */ /*0090*/ LEA R16, R3, R0, 0x1 ; /* 0x0000000003107211 */ /* 0x001fca00078e08ff */ /*00a0*/ @!P0 BRA 0x730 ; /* 0x0000068000008947 */ /* 0x000fea0003800000 */ /*00b0*/ LEA.HI R2, R14, c[0x0][0x178], RZ, 0x1 ; /* 0x00005e000e027a11 */ /* 0x000fe200078f08ff */ /*00c0*/ HFMA2.MMA R19, -RZ, RZ, 0, 0 ; /* 0x00000000ff137435 */ /* 0x000fe200000001ff */ /*00d0*/ SHF.L.U32 R18, R0, 0x3, RZ ; /* 0x0000000300127819 */ /* 0x000fe200000006ff */ /*00e0*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*00f0*/ SHF.R.S32.HI R22, RZ, 0x1, R2 ; /* 0x00000001ff167819 */ /* 0x000fe40000011402 */ /*0100*/ SHF.L.U32 R21, R14, 0x1, RZ ; /* 0x000000010e157819 */ /* 0x000fe200000006ff */ /*0110*/ IMAD R18, R13, 0x4, R18 ; /* 0x000000040d127824 */ /* 0x004fe200078e0212 */ /*0120*/ IADD3 R2, R22.reuse, -0x1, RZ ; /* 0xffffffff16027810 */ /* 0x040fe40007ffe0ff */ /*0130*/ LOP3.LUT R17, R22, 0x3, RZ, 0xc0, !PT ; /* 0x0000000316117812 */ /* 0x000fc400078ec0ff */ /*0140*/ ISETP.GE.U32.AND P1, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe40003f26070 */ /*0150*/ ISETP.NE.AND P0, PT, R17, RZ, PT ; /* 0x000000ff1100720c */ /* 0x000fd60003f05270 */ /*0160*/ @!P1 BRA 0x560 ; /* 0x000003f000009947 */ /* 0x000fea0003800000 */ /*0170*/ MOV R20, 0x4 ; /* 0x0000000400147802 */ /* 0x000fe20000000f00 */ /*0180*/ IMAD R15, R0, c[0x0][0x178], R13.reuse ; /* 0x00005e00000f7a24 */ /* 0x100fe200078e020d */ /*0190*/ IADD3 R22, -R22, R17, RZ ; /* 0x0000001116167210 */ /* 0x000fe20007ffe1ff */ /*01a0*/ IMAD R3, R16, c[0x0][0x178], R13 ; /* 0x00005e0010037a24 */ /* 0x000fe200078e020d */ /*01b0*/ MOV R19, RZ ; /* 0x000000ff00137202 */ /* 0x000fe20000000f00 */ /*01c0*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*01d0*/ LEA R15, R12, R15, 0x1 ; /* 0x0000000f0c0f7211 */ /* 0x002fe200078e08ff */ /*01e0*/ IMAD.WIDE R2, R3, R20, c[0x0][0x160] ; /* 0x0000580003027625 */ /* 0x000fc800078e0214 */ /*01f0*/ IMAD.WIDE R4, R15, R20, c[0x0][0x168] ; /* 0x00005a000f047625 */ /* 0x000fe200078e0214 */ /*0200*/ LDG.E R23, [R2.64] ; /* 0x0000000402177981 */ /* 0x000ea8000c1e1900 */ /*0210*/ LDG.E R25, [R4.64] ; /* 0x0000000404197981 */ /* 0x000ee2000c1e1900 */ /*0220*/ IMAD.WIDE R10, R21, 0x4, R4 ; /* 0x00000004150a7825 */ /* 0x000fc600078e0204 */ /*0230*/ STS [R18], R23 ; /* 0x0000001712007388 */ /* 0x004fe80000000800 */ /*0240*/ STS [R18+0x10], R25 ; /* 0x0000101912007388 */ /* 0x008fe80000000800 */ /*0250*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0260*/ LDS R8, [R13.X4+0x10] ; /* 0x000010000d087984 */ /* 0x000fe80000004800 */ /*0270*/ LDS R26, [R13.X4+0x18] ; /* 0x000018000d1a7984 */ /* 0x000fe80000004800 */ /*0280*/ LDS.64 R6, [R0.X8] ; /* 0x0000000000067984 */ /* 0x000e280000008a00 */ /*0290*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*02a0*/ LDG.E R29, [R2.64+0x8] ; /* 0x00000804021d7981 */ /* 0x000ea8000c1e1900 */ /*02b0*/ LDG.E R28, [R10.64] ; /* 0x000000040a1c7981 */ /* 0x000ee2000c1e1900 */ /*02c0*/ FFMA R6, R8, R6, R9 ; /* 0x0000000608067223 */ /* 0x001fc40000000009 */ /*02d0*/ IMAD.WIDE R8, R21, 0x4, R10 ; /* 0x0000000415087825 */ /* 0x000fe200078e020a */ /*02e0*/ STS [R18], R29 ; /* 0x0000001d12007388 */ /* 0x004fe80000000800 */ /*02f0*/ STS [R18+0x10], R28 ; /* 0x0000101c12007388 */ /* 0x008fe80000000800 */ /*0300*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0310*/ LDS R23, [R13.X4+0x10] ; /* 0x000010000d177984 */ /* 0x000fe80000004800 */ /*0320*/ LDS R24, [R13.X4+0x18] ; /* 0x000018000d187984 */ /* 0x000fe80000004800 */ /*0330*/ LDS.64 R4, [R0.X8] ; /* 0x0000000000047984 */ /* 0x000e280000008a00 */ /*0340*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0350*/ LDG.E R25, [R2.64+0x10] ; /* 0x0000100402197981 */ /* 0x000ea8000c1e1900 */ /*0360*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x000ee2000c1e1900 */ /*0370*/ FFMA R27, R26, R7, R6 ; /* 0x000000071a1b7223 */ /* 0x000fc60000000006 */ /*0380*/ STS [R18], R25 ; /* 0x0000001912007388 */ /* 0x004fe80000000800 */ /*0390*/ STS [R18+0x10], R11 ; /* 0x0000100b12007388 */ /* 0x0083e80000000800 */ /*03a0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fe20000010000 */ /*03b0*/ IMAD.WIDE R10, R21, 0x4, R8 ; /* 0x00000004150a7825 */ /* 0x002fca00078e0208 */ /*03c0*/ LDS R25, [R13.X4+0x10] ; /* 0x000010000d197984 */ /* 0x000fe80000004800 */ /*03d0*/ LDS R26, [R13.X4+0x18] ; /* 0x000018000d1a7984 */ /* 0x000fe80000004800 */ /*03e0*/ LDS.64 R6, [R0.X8] ; /* 0x0000000000067984 */ /* 0x000e680000008a00 */ /*03f0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0400*/ LDG.E R10, [R10.64] ; /* 0x000000040a0a7981 */ /* 0x000ea8000c1e1900 */ /*0410*/ LDG.E R29, [R2.64+0x18] ; /* 0x00001804021d7981 */ /* 0x000722000c1e1900 */ /*0420*/ FFMA R27, R23, R4, R27 ; /* 0x00000004171b7223 */ /* 0x001fe2000000001b */ /*0430*/ IADD3 R19, R19, 0x4, RZ ; /* 0x0000000413137810 */ /* 0x000fc40007ffe0ff */ /*0440*/ LEA R15, R14, R15, 0x3 ; /* 0x0000000f0e0f7211 */ /* 0x000fe200078e18ff */ /*0450*/ FFMA R5, R24, R5, R27 ; /* 0x0000000518057223 */ /* 0x000fc8000000001b */ /*0460*/ FFMA R5, R25, R6, R5 ; /* 0x0000000619057223 */ /* 0x002fe20000000005 */ /*0470*/ IADD3 R2, P2, R2, 0x20, RZ ; /* 0x0000002002027810 */ /* 0x008fe20007f5e0ff */ /*0480*/ IMAD.IADD R6, R22, 0x1, R19 ; /* 0x0000000116067824 */ /* 0x000fe400078e0213 */ /*0490*/ FFMA R5, R26, R7, R5 ; /* 0x000000071a057223 */ /* 0x000fe20000000005 */ /*04a0*/ IADD3.X R3, RZ, R3, RZ, P2, !PT ; /* 0x00000003ff037210 */ /* 0x000fe400017fe4ff */ /*04b0*/ ISETP.NE.AND P1, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f25270 */ /*04c0*/ STS [R18+0x10], R10 ; /* 0x0000100a12007388 */ /* 0x004fe80000000800 */ /*04d0*/ STS [R18], R29 ; /* 0x0000001d12007388 */ /* 0x010fe80000000800 */ /*04e0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*04f0*/ LDS R23, [R13.X4+0x10] ; /* 0x000010000d177984 */ /* 0x000fe80000004800 */ /*0500*/ LDS.64 R8, [R0.X8] ; /* 0x0000000000087984 */ /* 0x000e280000008a00 */ /*0510*/ LDS R4, [R13.X4+0x18] ; /* 0x000018000d047984 */ /* 0x000e620000004800 */ /*0520*/ FFMA R5, R23, R8, R5 ; /* 0x0000000817057223 */ /* 0x001fc80000000005 */ /*0530*/ FFMA R9, R4, R9, R5 ; /* 0x0000000904097223 */ /* 0x002fe20000000005 */ /*0540*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0550*/ @P1 BRA 0x1f0 ; /* 0xfffffc9000001947 */ /* 0x000fea000383ffff */ /*0560*/ @!P0 BRA 0x730 ; /* 0x000001c000008947 */ /* 0x000fea0003800000 */ /*0570*/ IMAD R6, R16, c[0x0][0x178], R13 ; /* 0x00005e0010067a24 */ /* 0x000fe200078e020d */ /*0580*/ MOV R4, 0x4 ; /* 0x0000000400047802 */ /* 0x000fe20000000f00 */ /*0590*/ IMAD R2, R19, 0x2, R0 ; /* 0x0000000213027824 */ /* 0x000fc600078e0200 */ /*05a0*/ LEA R3, R19, R6, 0x1 ; /* 0x0000000613037211 */ /* 0x000fe200078e08ff */ /*05b0*/ IMAD R5, R2, c[0x0][0x178], R13 ; /* 0x00005e0002057a24 */ /* 0x000fc800078e020d */ /*05c0*/ IMAD.WIDE R2, R3, R4, c[0x0][0x160] ; /* 0x0000580003027625 */ /* 0x000fc800078e0204 */ /*05d0*/ IMAD R5, R12, 0x2, R5 ; /* 0x000000020c057824 */ /* 0x002fe200078e0205 */ /*05e0*/ MOV R6, R2 ; /* 0x0000000200067202 */ /* 0x000fe40000000f00 */ /*05f0*/ MOV R15, R3 ; /* 0x00000003000f7202 */ /* 0x000fe20000000f00 */ /*0600*/ IMAD.WIDE R2, R5, R4, c[0x0][0x168] ; /* 0x00005a0005027625 */ /* 0x000fc800078e0204 */ /*0610*/ IMAD.MOV.U32 R7, RZ, RZ, R15 ; /* 0x000000ffff077224 */ /* 0x000fe200078e000f */ /*0620*/ LDG.E R11, [R2.64] ; /* 0x00000004020b7981 */ /* 0x0000aa000c1e1900 */ /*0630*/ LDG.E R7, [R6.64] ; /* 0x0000000406077981 */ /* 0x0002e2000c1e1900 */ /*0640*/ IADD3 R17, R17, -0x1, RZ ; /* 0xffffffff11117810 */ /* 0x000fc80007ffe0ff */ /*0650*/ ISETP.NE.AND P0, PT, R17, RZ, PT ; /* 0x000000ff1100720c */ /* 0x000fe20003f05270 */ /*0660*/ IMAD.WIDE R2, R21, 0x4, R2 ; /* 0x0000000415027825 */ /* 0x001fe200078e0202 */ /*0670*/ IADD3 R6, P1, R6, 0x8, RZ ; /* 0x0000000806067810 */ /* 0x002fc80007f3e0ff */ /*0680*/ IADD3.X R15, RZ, R15, RZ, P1, !PT ; /* 0x0000000fff0f7210 */ /* 0x000fe20000ffe4ff */ /*0690*/ STS [R18+0x10], R11 ; /* 0x0000100b12007388 */ /* 0x004fe80000000800 */ /*06a0*/ STS [R18], R7 ; /* 0x0000000712007388 */ /* 0x008fe80000000800 */ /*06b0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*06c0*/ LDS R8, [R13.X4+0x10] ; /* 0x000010000d087984 */ /* 0x000fe80000004800 */ /*06d0*/ LDS.64 R4, [R0.X8] ; /* 0x0000000000047984 */ /* 0x000e280000008a00 */ /*06e0*/ LDS R10, [R13.X4+0x18] ; /* 0x000018000d0a7984 */ /* 0x000e620000004800 */ /*06f0*/ FFMA R4, R8, R4, R9 ; /* 0x0000000408047223 */ /* 0x001fc80000000009 */ /*0700*/ FFMA R9, R10, R5, R4 ; /* 0x000000050a097223 */ /* 0x002fe20000000004 */ /*0710*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0720*/ @P0 BRA 0x610 ; /* 0xfffffee000000947 */ /* 0x000fea000383ffff */ /*0730*/ HFMA2.MMA R3, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff037435 */ /* 0x000fe200000001ff */ /*0740*/ LEA R13, R12, R13, 0x1 ; /* 0x0000000d0c0d7211 */ /* 0x006fca00078e08ff */ /*0750*/ IMAD R2, R16, c[0x0][0x178], R13 ; /* 0x00005e0010027a24 */ /* 0x000fc800078e020d */ /*0760*/ IMAD.WIDE R2, R2, R3, c[0x0][0x170] ; /* 0x00005c0002027625 */ /* 0x000fca00078e0203 */ /*0770*/ STG.E [R2.64], R9 ; /* 0x0000000902007986 */ /* 0x000fe2000c101904 */ /*0780*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0790*/ BRA 0x790; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*07a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0800*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*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 */ ..........
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <stdio.h> #include <cuda.h> /* Matrices are stored in row-major order: */ /* M(row, col) = (M.width*row +col); */ typedef struct{ /* suppose we use only square matrices */ int width; float *elements; } Matrix; /* Thread block size */ #define TILE_WIDTH 2 /***********************/ /* TODO, write KERNEL */ /***********************/ __global__ void MatMul(const float* Md, const float* Nd, float* Pd, const int Width) { __shared__ float Mds[TILE_WIDTH][TILE_WIDTH]; __shared__ float Nds[TILE_WIDTH][TILE_WIDTH]; int bx = blockIdx.x; int by = blockIdx.y; int tx = threadIdx.x; int ty = threadIdx.y; // Identify the row and column of the Pd element to work on int Row = by * TILE_WIDTH + ty; int Col = bx * TILE_WIDTH + tx; float Pvalue = 0; //Loop over the Md and Nd tiles required to compute the Pd element for (int m = 0; m < Width/TILE_WIDTH; m++) { // Coolaborative loading of Md and Nd tiles into shared memory Mds[ty][tx] = Md[Row*Width + (m*TILE_WIDTH + tx)]; Nds[ty][tx] = Nd[Col + (m*TILE_WIDTH + ty)*Width]; __syncthreads(); for (int k = 0; k < TILE_WIDTH; k++) Pvalue += Mds[ty][k] * Nds[k][tx]; __syncthreads(); } Pd[Row*Width+Col] = Pvalue; } /**/ void test(const Matrix C); /**/ int main(int argc, char* argv[]) { int i; /* init matrices */ Matrix h_A, h_B, h_C; cudaEvent_t start; cudaEvent_t stop; cudaEventCreate(&start); cudaEventCreate(&stop); /*******************/ /** READING INPUT **/ /*******************/ int size = 0; //dimension of matrices scanf("%d", &size); int full_size = sizeof(float)*size*size; h_A.width = size; h_B.width = size; h_C.width = size; /* Allocate host memory */ h_A.elements = (float*)malloc(full_size); h_B.elements = (float*)malloc(full_size); h_C.elements = (float*)malloc(full_size); /**/ for(i=0;i<size*size;++i){ scanf("%f", &h_A.elements[i]);} for(i=0;i<size*size;++i){ scanf("%f", &h_B.elements[i]);} /********************/ /** FINISHED INPUT **/ /********************/ /*************************/ /* allocate device */ /* memory for A,B,C */ /*************************/ Matrix d_A, d_B, d_C; d_A.width = size; d_B.width = size; d_C.width = size; cudaMalloc(&d_A.elements, full_size); cudaMalloc(&d_B.elements, full_size); cudaMalloc(&d_C.elements, full_size); cudaEventRecord(start,0); /***********************************/ /* copy vectors A&B to device */ /***********************************/ cudaMemcpy(d_A.elements, h_A.elements, full_size, cudaMemcpyHostToDevice); cudaMemcpy(d_B.elements, h_B.elements, full_size, cudaMemcpyHostToDevice); /*********************/ /* call kernel */ /*********************/ dim3 dimBlock(TILE_WIDTH, TILE_WIDTH); dim3 dimGrid(h_B.width/dimBlock.x, h_A.width/dimBlock.y); MatMul<<<dimGrid,dimBlock>>>(d_A.elements, d_B.elements, d_C.elements,d_A.width); /**************************/ /* copy result back */ /**************************/ cudaMemcpy(h_C.elements, d_C.elements, full_size, cudaMemcpyDeviceToHost); cudaEventRecord(stop, 0); cudaEventSynchronize(stop); float elapsedTime; cudaEventElapsedTime(&elapsedTime, start, stop); fprintf(stderr,"Elapsed time = %f (s)\n",elapsedTime); cudaEventDestroy(start); cudaEventDestroy(stop); /*******************************************/ /** Testing output, don't change anything! */ /*******************************************/ test(h_C); /* free device memory */ cudaFree(d_A.elements); cudaFree(d_B.elements); cudaFree(d_C.elements); /* free host memory */ free(h_A.elements); free(h_B.elements); free(h_C.elements); return 0; } //function to test the input, don't change anything! void test(const Matrix C) { int i,j; //int size = C.width*C.width; for(i=0;i<C.width;++i) { for(j=0;j<C.width;++j) printf("%4.1f ", C.elements[i*C.width+j]); printf("\n"); } }
.file "tmpxft_001a5ccc_00000000-6_MMult2.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2061: .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 .LFE2061: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "%4.1f " .LC1: .string "\n" .text .globl _Z4test6Matrix .type _Z4test6Matrix, @function _Z4test6Matrix: .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 subq $24, %rsp .cfi_def_cfa_offset 80 movl %edi, 4(%rsp) testl %edi, %edi jle .L3 movq %rsi, %r12 movslq %edi, %rdi leaq 0(,%rdi,4), %r15 negq %rdi leaq 0(,%rdi,4), %rax movq %rax, 8(%rsp) movq %r15, %rbp movl $0, %r14d leaq .LC0(%rip), %r13 .L5: movq 8(%rsp), %rax leaq 0(%rbp,%rax), %rbx .L6: pxor %xmm0, %xmm0 cvtss2sd (%rbx,%r12), %xmm0 movq %r13, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L6 leaq .LC1(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addl $1, %r14d addq %r15, %rbp cmpl %r14d, 4(%rsp) jne .L5 .L3: 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 .LFE2058: .size _Z4test6Matrix, .-_Z4test6Matrix .globl _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi .type _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi, @function _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi: .LFB2083: .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 .L13 .L9: movq 136(%rsp), %rax subq %fs:40, %rax jne .L14 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L13: .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 _Z6MatMulPKfS0_Pfi(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L9 .L14: call __stack_chk_fail@PLT .cfi_endproc .LFE2083: .size _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi, .-_Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi .globl _Z6MatMulPKfS0_Pfi .type _Z6MatMulPKfS0_Pfi, @function _Z6MatMulPKfS0_Pfi: .LFB2084: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2084: .size _Z6MatMulPKfS0_Pfi, .-_Z6MatMulPKfS0_Pfi .section .rodata.str1.1 .LC2: .string "%d" .LC3: .string "%f" .LC4: .string "Elapsed time = %f (s)\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 $152, %rsp .cfi_def_cfa_offset 208 movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 40(%rsp), %rdi call cudaEventCreate@PLT leaq 48(%rsp), %rdi call cudaEventCreate@PLT movl $0, 32(%rsp) leaq 32(%rsp), %rsi leaq .LC2(%rip), %rdi movl $0, %eax call __isoc23_scanf@PLT movl 32(%rsp), %r13d movl %r13d, %eax movq %rax, 24(%rsp) movl %r13d, %r12d imull %r13d, %r12d sall $2, %r12d movslq %r12d, %r12 movq %r12, %rdi call malloc@PLT movq %rax, %rbp movq %rax, 16(%rsp) movq %r12, %rdi call malloc@PLT movq %rax, 8(%rsp) movq %r12, %rdi call malloc@PLT movq %rax, %r14 movl %r13d, %eax imull %r13d, %eax testl %eax, %eax jle .L23 movl $0, %ebx leaq .LC3(%rip), %r15 .L19: movq %rbp, %rsi movq %r15, %rdi movl $0, %eax call __isoc23_scanf@PLT addl $1, %ebx movl 32(%rsp), %eax movl %eax, %edx imull %eax, %edx addq $4, %rbp cmpl %ebx, %edx jg .L19 testl %edx, %edx jle .L18 movq 8(%rsp), %rbp movl $0, %ebx leaq .LC3(%rip), %r15 .L20: movq %rbp, %rsi movq %r15, %rdi movl $0, %eax call __isoc23_scanf@PLT addl $1, %ebx movl 32(%rsp), %eax addq $4, %rbp movl %eax, %edx imull %eax, %edx cmpl %ebx, %edx jg .L20 .L18: movl %eax, 80(%rsp) movl %eax, 96(%rsp) movl %eax, 112(%rsp) leaq 88(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT leaq 104(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT leaq 120(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT movl $0, %esi movq 40(%rsp), %rdi call cudaEventRecord@PLT movl $1, %ecx movq %r12, %rdx movq 16(%rsp), %rsi movq 88(%rsp), %rdi call cudaMemcpy@PLT movl $1, %ecx movq %r12, %rdx movq 8(%rsp), %rsi movq 104(%rsp), %rdi call cudaMemcpy@PLT shrl %r13d movl %r13d, 68(%rsp) movl %r13d, 72(%rsp) movl $2, 56(%rsp) movl $2, 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 .L27 .L21: movl $2, %ecx movq %r12, %rdx movq 120(%rsp), %rsi movq %r14, %rdi call cudaMemcpy@PLT movl $0, %esi movq 48(%rsp), %rdi call cudaEventRecord@PLT movq 48(%rsp), %rdi call cudaEventSynchronize@PLT leaq 36(%rsp), %rdi movq 48(%rsp), %rdx movq 40(%rsp), %rsi call cudaEventElapsedTime@PLT pxor %xmm0, %xmm0 cvtss2sd 36(%rsp), %xmm0 leaq .LC4(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $1, %eax call __fprintf_chk@PLT movq 40(%rsp), %rdi call cudaEventDestroy@PLT movq 48(%rsp), %rdi call cudaEventDestroy@PLT movl 24(%rsp), %edi movq %r14, %rsi call _Z4test6Matrix movq 88(%rsp), %rdi call cudaFree@PLT movq 104(%rsp), %rdi call cudaFree@PLT movq 120(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call free@PLT movq 8(%rsp), %rdi call free@PLT movq %r14, %rdi call free@PLT movq 136(%rsp), %rax subq %fs:40, %rax jne .L28 movl $0, %eax addq $152, %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 .L23: .cfi_restore_state movl %r13d, %eax jmp .L18 .L27: movl 80(%rsp), %ecx movq 120(%rsp), %rdx movq 104(%rsp), %rsi movq 88(%rsp), %rdi call _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi jmp .L21 .L28: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size main, .-main .section .rodata.str1.1 .LC5: .string "_Z6MatMulPKfS0_Pfi" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2086: .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 .LC5(%rip), %rdx movq %rdx, %rcx leaq _Z6MatMulPKfS0_Pfi(%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 .LFE2086: .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> /* Matrices are stored in row-major order: */ /* M(row, col) = (M.width*row +col); */ typedef struct{ /* suppose we use only square matrices */ int width; float *elements; } Matrix; /* Thread block size */ #define TILE_WIDTH 2 /***********************/ /* TODO, write KERNEL */ /***********************/ __global__ void MatMul(const float* Md, const float* Nd, float* Pd, const int Width) { __shared__ float Mds[TILE_WIDTH][TILE_WIDTH]; __shared__ float Nds[TILE_WIDTH][TILE_WIDTH]; int bx = blockIdx.x; int by = blockIdx.y; int tx = threadIdx.x; int ty = threadIdx.y; // Identify the row and column of the Pd element to work on int Row = by * TILE_WIDTH + ty; int Col = bx * TILE_WIDTH + tx; float Pvalue = 0; //Loop over the Md and Nd tiles required to compute the Pd element for (int m = 0; m < Width/TILE_WIDTH; m++) { // Coolaborative loading of Md and Nd tiles into shared memory Mds[ty][tx] = Md[Row*Width + (m*TILE_WIDTH + tx)]; Nds[ty][tx] = Nd[Col + (m*TILE_WIDTH + ty)*Width]; __syncthreads(); for (int k = 0; k < TILE_WIDTH; k++) Pvalue += Mds[ty][k] * Nds[k][tx]; __syncthreads(); } Pd[Row*Width+Col] = Pvalue; } /**/ void test(const Matrix C); /**/ int main(int argc, char* argv[]) { int i; /* init matrices */ Matrix h_A, h_B, h_C; cudaEvent_t start; cudaEvent_t stop; cudaEventCreate(&start); cudaEventCreate(&stop); /*******************/ /** READING INPUT **/ /*******************/ int size = 0; //dimension of matrices scanf("%d", &size); int full_size = sizeof(float)*size*size; h_A.width = size; h_B.width = size; h_C.width = size; /* Allocate host memory */ h_A.elements = (float*)malloc(full_size); h_B.elements = (float*)malloc(full_size); h_C.elements = (float*)malloc(full_size); /**/ for(i=0;i<size*size;++i){ scanf("%f", &h_A.elements[i]);} for(i=0;i<size*size;++i){ scanf("%f", &h_B.elements[i]);} /********************/ /** FINISHED INPUT **/ /********************/ /*************************/ /* allocate device */ /* memory for A,B,C */ /*************************/ Matrix d_A, d_B, d_C; d_A.width = size; d_B.width = size; d_C.width = size; cudaMalloc(&d_A.elements, full_size); cudaMalloc(&d_B.elements, full_size); cudaMalloc(&d_C.elements, full_size); cudaEventRecord(start,0); /***********************************/ /* copy vectors A&B to device */ /***********************************/ cudaMemcpy(d_A.elements, h_A.elements, full_size, cudaMemcpyHostToDevice); cudaMemcpy(d_B.elements, h_B.elements, full_size, cudaMemcpyHostToDevice); /*********************/ /* call kernel */ /*********************/ dim3 dimBlock(TILE_WIDTH, TILE_WIDTH); dim3 dimGrid(h_B.width/dimBlock.x, h_A.width/dimBlock.y); MatMul<<<dimGrid,dimBlock>>>(d_A.elements, d_B.elements, d_C.elements,d_A.width); /**************************/ /* copy result back */ /**************************/ cudaMemcpy(h_C.elements, d_C.elements, full_size, cudaMemcpyDeviceToHost); cudaEventRecord(stop, 0); cudaEventSynchronize(stop); float elapsedTime; cudaEventElapsedTime(&elapsedTime, start, stop); fprintf(stderr,"Elapsed time = %f (s)\n",elapsedTime); cudaEventDestroy(start); cudaEventDestroy(stop); /*******************************************/ /** Testing output, don't change anything! */ /*******************************************/ test(h_C); /* free device memory */ cudaFree(d_A.elements); cudaFree(d_B.elements); cudaFree(d_C.elements); /* free host memory */ free(h_A.elements); free(h_B.elements); free(h_C.elements); return 0; } //function to test the input, don't change anything! void test(const Matrix C) { int i,j; //int size = C.width*C.width; for(i=0;i<C.width;++i) { for(j=0;j<C.width;++j) printf("%4.1f ", C.elements[i*C.width+j]); printf("\n"); } }
#include <stdio.h> #include <hip/hip_runtime.h> /* Matrices are stored in row-major order: */ /* M(row, col) = (M.width*row +col); */ typedef struct{ /* suppose we use only square matrices */ int width; float *elements; } Matrix; /* Thread block size */ #define TILE_WIDTH 2 /***********************/ /* TODO, write KERNEL */ /***********************/ __global__ void MatMul(const float* Md, const float* Nd, float* Pd, const int Width) { __shared__ float Mds[TILE_WIDTH][TILE_WIDTH]; __shared__ float Nds[TILE_WIDTH][TILE_WIDTH]; int bx = blockIdx.x; int by = blockIdx.y; int tx = threadIdx.x; int ty = threadIdx.y; // Identify the row and column of the Pd element to work on int Row = by * TILE_WIDTH + ty; int Col = bx * TILE_WIDTH + tx; float Pvalue = 0; //Loop over the Md and Nd tiles required to compute the Pd element for (int m = 0; m < Width/TILE_WIDTH; m++) { // Coolaborative loading of Md and Nd tiles into shared memory Mds[ty][tx] = Md[Row*Width + (m*TILE_WIDTH + tx)]; Nds[ty][tx] = Nd[Col + (m*TILE_WIDTH + ty)*Width]; __syncthreads(); for (int k = 0; k < TILE_WIDTH; k++) Pvalue += Mds[ty][k] * Nds[k][tx]; __syncthreads(); } Pd[Row*Width+Col] = Pvalue; } /**/ void test(const Matrix C); /**/ int main(int argc, char* argv[]) { int i; /* init matrices */ Matrix h_A, h_B, h_C; hipEvent_t start; hipEvent_t stop; hipEventCreate(&start); hipEventCreate(&stop); /*******************/ /** READING INPUT **/ /*******************/ int size = 0; //dimension of matrices scanf("%d", &size); int full_size = sizeof(float)*size*size; h_A.width = size; h_B.width = size; h_C.width = size; /* Allocate host memory */ h_A.elements = (float*)malloc(full_size); h_B.elements = (float*)malloc(full_size); h_C.elements = (float*)malloc(full_size); /**/ for(i=0;i<size*size;++i){ scanf("%f", &h_A.elements[i]);} for(i=0;i<size*size;++i){ scanf("%f", &h_B.elements[i]);} /********************/ /** FINISHED INPUT **/ /********************/ /*************************/ /* allocate device */ /* memory for A,B,C */ /*************************/ Matrix d_A, d_B, d_C; d_A.width = size; d_B.width = size; d_C.width = size; hipMalloc(&d_A.elements, full_size); hipMalloc(&d_B.elements, full_size); hipMalloc(&d_C.elements, full_size); hipEventRecord(start,0); /***********************************/ /* copy vectors A&B to device */ /***********************************/ hipMemcpy(d_A.elements, h_A.elements, full_size, hipMemcpyHostToDevice); hipMemcpy(d_B.elements, h_B.elements, full_size, hipMemcpyHostToDevice); /*********************/ /* call kernel */ /*********************/ dim3 dimBlock(TILE_WIDTH, TILE_WIDTH); dim3 dimGrid(h_B.width/dimBlock.x, h_A.width/dimBlock.y); MatMul<<<dimGrid,dimBlock>>>(d_A.elements, d_B.elements, d_C.elements,d_A.width); /**************************/ /* copy result back */ /**************************/ hipMemcpy(h_C.elements, d_C.elements, full_size, hipMemcpyDeviceToHost); hipEventRecord(stop, 0); hipEventSynchronize(stop); float elapsedTime; hipEventElapsedTime(&elapsedTime, start, stop); fprintf(stderr,"Elapsed time = %f (s)\n",elapsedTime); hipEventDestroy(start); hipEventDestroy(stop); /*******************************************/ /** Testing output, don't change anything! */ /*******************************************/ test(h_C); /* free device memory */ hipFree(d_A.elements); hipFree(d_B.elements); hipFree(d_C.elements); /* free host memory */ free(h_A.elements); free(h_B.elements); free(h_C.elements); return 0; } //function to test the input, don't change anything! void test(const Matrix C) { int i,j; //int size = C.width*C.width; for(i=0;i<C.width;++i) { for(j=0;j<C.width;++j) printf("%4.1f ", C.elements[i*C.width+j]); printf("\n"); } }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <stdio.h> #include <hip/hip_runtime.h> /* Matrices are stored in row-major order: */ /* M(row, col) = (M.width*row +col); */ typedef struct{ /* suppose we use only square matrices */ int width; float *elements; } Matrix; /* Thread block size */ #define TILE_WIDTH 2 /***********************/ /* TODO, write KERNEL */ /***********************/ __global__ void MatMul(const float* Md, const float* Nd, float* Pd, const int Width) { __shared__ float Mds[TILE_WIDTH][TILE_WIDTH]; __shared__ float Nds[TILE_WIDTH][TILE_WIDTH]; int bx = blockIdx.x; int by = blockIdx.y; int tx = threadIdx.x; int ty = threadIdx.y; // Identify the row and column of the Pd element to work on int Row = by * TILE_WIDTH + ty; int Col = bx * TILE_WIDTH + tx; float Pvalue = 0; //Loop over the Md and Nd tiles required to compute the Pd element for (int m = 0; m < Width/TILE_WIDTH; m++) { // Coolaborative loading of Md and Nd tiles into shared memory Mds[ty][tx] = Md[Row*Width + (m*TILE_WIDTH + tx)]; Nds[ty][tx] = Nd[Col + (m*TILE_WIDTH + ty)*Width]; __syncthreads(); for (int k = 0; k < TILE_WIDTH; k++) Pvalue += Mds[ty][k] * Nds[k][tx]; __syncthreads(); } Pd[Row*Width+Col] = Pvalue; } /**/ void test(const Matrix C); /**/ int main(int argc, char* argv[]) { int i; /* init matrices */ Matrix h_A, h_B, h_C; hipEvent_t start; hipEvent_t stop; hipEventCreate(&start); hipEventCreate(&stop); /*******************/ /** READING INPUT **/ /*******************/ int size = 0; //dimension of matrices scanf("%d", &size); int full_size = sizeof(float)*size*size; h_A.width = size; h_B.width = size; h_C.width = size; /* Allocate host memory */ h_A.elements = (float*)malloc(full_size); h_B.elements = (float*)malloc(full_size); h_C.elements = (float*)malloc(full_size); /**/ for(i=0;i<size*size;++i){ scanf("%f", &h_A.elements[i]);} for(i=0;i<size*size;++i){ scanf("%f", &h_B.elements[i]);} /********************/ /** FINISHED INPUT **/ /********************/ /*************************/ /* allocate device */ /* memory for A,B,C */ /*************************/ Matrix d_A, d_B, d_C; d_A.width = size; d_B.width = size; d_C.width = size; hipMalloc(&d_A.elements, full_size); hipMalloc(&d_B.elements, full_size); hipMalloc(&d_C.elements, full_size); hipEventRecord(start,0); /***********************************/ /* copy vectors A&B to device */ /***********************************/ hipMemcpy(d_A.elements, h_A.elements, full_size, hipMemcpyHostToDevice); hipMemcpy(d_B.elements, h_B.elements, full_size, hipMemcpyHostToDevice); /*********************/ /* call kernel */ /*********************/ dim3 dimBlock(TILE_WIDTH, TILE_WIDTH); dim3 dimGrid(h_B.width/dimBlock.x, h_A.width/dimBlock.y); MatMul<<<dimGrid,dimBlock>>>(d_A.elements, d_B.elements, d_C.elements,d_A.width); /**************************/ /* copy result back */ /**************************/ hipMemcpy(h_C.elements, d_C.elements, full_size, hipMemcpyDeviceToHost); hipEventRecord(stop, 0); hipEventSynchronize(stop); float elapsedTime; hipEventElapsedTime(&elapsedTime, start, stop); fprintf(stderr,"Elapsed time = %f (s)\n",elapsedTime); hipEventDestroy(start); hipEventDestroy(stop); /*******************************************/ /** Testing output, don't change anything! */ /*******************************************/ test(h_C); /* free device memory */ hipFree(d_A.elements); hipFree(d_B.elements); hipFree(d_C.elements); /* free host memory */ free(h_A.elements); free(h_B.elements); free(h_C.elements); return 0; } //function to test the input, don't change anything! void test(const Matrix C) { int i,j; //int size = C.width*C.width; for(i=0;i<C.width;++i) { for(j=0;j<C.width;++j) printf("%4.1f ", C.elements[i*C.width+j]); printf("\n"); } }
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z6MatMulPKfS0_Pfi .globl _Z6MatMulPKfS0_Pfi .p2align 8 .type _Z6MatMulPKfS0_Pfi,@function _Z6MatMulPKfS0_Pfi: s_load_b32 s2, s[0:1], 0x18 v_bfe_u32 v5, v0, 10, 10 v_dual_mov_b32 v2, 0 :: v_dual_and_b32 v3, 0x3ff, v0 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2) v_lshl_add_u32 v4, s15, 1, v5 v_lshl_add_u32 v0, s14, 1, v3 s_waitcnt lgkmcnt(0) s_cmp_lt_i32 s2, 2 s_cbranch_scc1 .LBB0_5 s_load_b128 s[4:7], s[0:1], 0x0 v_lshlrev_b32_e32 v8, 2, v3 v_lshlrev_b32_e32 v6, 3, v5 v_mad_u64_u32 v[1:2], null, v4, s2, v[3:4] s_lshr_b32 s3, s2, 31 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3) v_dual_mov_b32 v2, 0 :: v_dual_add_nc_u32 v7, 16, v8 v_add_nc_u32_e32 v3, v6, v8 s_add_i32 s3, s2, s3 s_mov_b32 s8, 0 s_delay_alu instid0(VALU_DEP_2) v_add_nc_u32_e32 v8, v7, v6 s_ashr_i32 s3, s3, 1 s_set_inst_prefetch_distance 0x1 .p2align 6 .LBB0_2: s_lshl_b32 s9, s8, 1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(VALU_DEP_2) v_add_nc_u32_e32 v10, s9, v5 v_add_nc_u32_e32 v9, s9, v1 s_mov_b32 s9, 0 v_mad_u64_u32 v[11:12], null, v10, s2, v[0:1] s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v10, 31, v9 v_lshlrev_b64 v[9:10], 2, v[9:10] s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_ashrrev_i32_e32 v12, 31, v11 s_waitcnt lgkmcnt(0) v_add_co_u32 v9, vcc_lo, s4, v9 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_4) v_lshlrev_b64 v[11:12], 2, v[11:12] v_add_co_ci_u32_e32 v10, vcc_lo, s5, v10, vcc_lo s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_u32 v11, vcc_lo, s6, v11 v_add_co_ci_u32_e32 v12, vcc_lo, s7, v12, vcc_lo global_load_b32 v10, v[9:10], off global_load_b32 v11, v[11:12], off v_mov_b32_e32 v9, v7 s_waitcnt vmcnt(1) ds_store_b32 v3, v10 s_waitcnt vmcnt(0) ds_store_b32 v8, v11 s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv .LBB0_3: v_add_nc_u32_e32 v10, s9, v6 s_add_i32 s9, s9, 4 ds_load_b32 v11, v9 ds_load_b32 v10, v10 v_add_nc_u32_e32 v9, 8, v9 s_cmp_lg_u32 s9, 4 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v2, v10, v11 s_cbranch_scc0 .LBB0_3 s_add_i32 s8, s8, 1 s_delay_alu instid0(SALU_CYCLE_1) s_cmp_eq_u32 s8, s3 s_barrier buffer_gl0_inv s_cbranch_scc0 .LBB0_2 .LBB0_5: s_set_inst_prefetch_distance 0x2 s_load_b64 s[0:1], s[0:1], 0x10 v_mad_u64_u32 v[5:6], null, v4, s2, v[0:1] 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], 2, 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_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 _Z6MatMulPKfS0_Pfi .amdhsa_group_segment_fixed_size 32 .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 13 .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 _Z6MatMulPKfS0_Pfi, .Lfunc_end0-_Z6MatMulPKfS0_Pfi .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: 32 .kernarg_segment_align: 8 .kernarg_segment_size: 28 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z6MatMulPKfS0_Pfi .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z6MatMulPKfS0_Pfi.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 13 .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> /* Matrices are stored in row-major order: */ /* M(row, col) = (M.width*row +col); */ typedef struct{ /* suppose we use only square matrices */ int width; float *elements; } Matrix; /* Thread block size */ #define TILE_WIDTH 2 /***********************/ /* TODO, write KERNEL */ /***********************/ __global__ void MatMul(const float* Md, const float* Nd, float* Pd, const int Width) { __shared__ float Mds[TILE_WIDTH][TILE_WIDTH]; __shared__ float Nds[TILE_WIDTH][TILE_WIDTH]; int bx = blockIdx.x; int by = blockIdx.y; int tx = threadIdx.x; int ty = threadIdx.y; // Identify the row and column of the Pd element to work on int Row = by * TILE_WIDTH + ty; int Col = bx * TILE_WIDTH + tx; float Pvalue = 0; //Loop over the Md and Nd tiles required to compute the Pd element for (int m = 0; m < Width/TILE_WIDTH; m++) { // Coolaborative loading of Md and Nd tiles into shared memory Mds[ty][tx] = Md[Row*Width + (m*TILE_WIDTH + tx)]; Nds[ty][tx] = Nd[Col + (m*TILE_WIDTH + ty)*Width]; __syncthreads(); for (int k = 0; k < TILE_WIDTH; k++) Pvalue += Mds[ty][k] * Nds[k][tx]; __syncthreads(); } Pd[Row*Width+Col] = Pvalue; } /**/ void test(const Matrix C); /**/ int main(int argc, char* argv[]) { int i; /* init matrices */ Matrix h_A, h_B, h_C; hipEvent_t start; hipEvent_t stop; hipEventCreate(&start); hipEventCreate(&stop); /*******************/ /** READING INPUT **/ /*******************/ int size = 0; //dimension of matrices scanf("%d", &size); int full_size = sizeof(float)*size*size; h_A.width = size; h_B.width = size; h_C.width = size; /* Allocate host memory */ h_A.elements = (float*)malloc(full_size); h_B.elements = (float*)malloc(full_size); h_C.elements = (float*)malloc(full_size); /**/ for(i=0;i<size*size;++i){ scanf("%f", &h_A.elements[i]);} for(i=0;i<size*size;++i){ scanf("%f", &h_B.elements[i]);} /********************/ /** FINISHED INPUT **/ /********************/ /*************************/ /* allocate device */ /* memory for A,B,C */ /*************************/ Matrix d_A, d_B, d_C; d_A.width = size; d_B.width = size; d_C.width = size; hipMalloc(&d_A.elements, full_size); hipMalloc(&d_B.elements, full_size); hipMalloc(&d_C.elements, full_size); hipEventRecord(start,0); /***********************************/ /* copy vectors A&B to device */ /***********************************/ hipMemcpy(d_A.elements, h_A.elements, full_size, hipMemcpyHostToDevice); hipMemcpy(d_B.elements, h_B.elements, full_size, hipMemcpyHostToDevice); /*********************/ /* call kernel */ /*********************/ dim3 dimBlock(TILE_WIDTH, TILE_WIDTH); dim3 dimGrid(h_B.width/dimBlock.x, h_A.width/dimBlock.y); MatMul<<<dimGrid,dimBlock>>>(d_A.elements, d_B.elements, d_C.elements,d_A.width); /**************************/ /* copy result back */ /**************************/ hipMemcpy(h_C.elements, d_C.elements, full_size, hipMemcpyDeviceToHost); hipEventRecord(stop, 0); hipEventSynchronize(stop); float elapsedTime; hipEventElapsedTime(&elapsedTime, start, stop); fprintf(stderr,"Elapsed time = %f (s)\n",elapsedTime); hipEventDestroy(start); hipEventDestroy(stop); /*******************************************/ /** Testing output, don't change anything! */ /*******************************************/ test(h_C); /* free device memory */ hipFree(d_A.elements); hipFree(d_B.elements); hipFree(d_C.elements); /* free host memory */ free(h_A.elements); free(h_B.elements); free(h_C.elements); return 0; } //function to test the input, don't change anything! void test(const Matrix C) { int i,j; //int size = C.width*C.width; for(i=0;i<C.width;++i) { for(j=0;j<C.width;++j) printf("%4.1f ", C.elements[i*C.width+j]); printf("\n"); } }
.text .file "MMult2.hip" .globl _Z21__device_stub__MatMulPKfS0_Pfi # -- Begin function _Z21__device_stub__MatMulPKfS0_Pfi .p2align 4, 0x90 .type _Z21__device_stub__MatMulPKfS0_Pfi,@function _Z21__device_stub__MatMulPKfS0_Pfi: # @_Z21__device_stub__MatMulPKfS0_Pfi .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 $_Z6MatMulPKfS0_Pfi, %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 _Z21__device_stub__MatMulPKfS0_Pfi, .Lfunc_end0-_Z21__device_stub__MatMulPKfS0_Pfi .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 leaq 32(%rsp), %rdi callq hipEventCreate leaq 16(%rsp), %rdi callq hipEventCreate movl $0, 12(%rsp) leaq 12(%rsp), %rsi movl $.L.str, %edi xorl %eax, %eax callq __isoc23_scanf movl 12(%rsp), %ebp movl %ebp, %eax imull %ebp, %eax shll $2, %eax movslq %eax, %r12 movq %r12, %rdi callq malloc movq %rax, %r14 movq %r12, %rdi callq malloc movq %rax, 24(%rsp) # 8-byte Spill movq %r12, %rdi callq malloc movq %rax, %r15 cmpl $0, 12(%rsp) je .LBB1_3 # %bb.1: # %.lr.ph.preheader movq %r14, %r13 xorl %ebx, %ebx .p2align 4, 0x90 .LBB1_2: # %.lr.ph # =>This Inner Loop Header: Depth=1 movl $.L.str.1, %edi movq %r13, %rsi xorl %eax, %eax callq __isoc23_scanf incq %rbx movl 12(%rsp), %eax imull %eax, %eax addq $4, %r13 cmpq %rax, %rbx jb .LBB1_2 .LBB1_3: # %.preheader movl 12(%rsp), %eax testl %eax, %eax je .LBB1_6 # %bb.4: # %.lr.ph40.preheader movq 24(%rsp), %r13 # 8-byte Reload xorl %ebx, %ebx .p2align 4, 0x90 .LBB1_5: # %.lr.ph40 # =>This Inner Loop Header: Depth=1 movl $.L.str.1, %edi movq %r13, %rsi xorl %eax, %eax callq __isoc23_scanf incq %rbx movl 12(%rsp), %eax movl %eax, %ecx imull %ecx, %ecx addq $4, %r13 cmpq %rcx, %rbx jb .LBB1_5 .LBB1_6: # %._crit_edge movl %eax, 48(%rsp) movl %eax, 80(%rsp) movl %eax, 64(%rsp) leaq 56(%rsp), %rdi movq %r12, %rsi callq hipMalloc leaq 88(%rsp), %rdi movq %r12, %rsi callq hipMalloc leaq 72(%rsp), %rdi movq %r12, %rsi callq hipMalloc movq 32(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 56(%rsp), %rdi movq %r14, 96(%rsp) # 8-byte Spill movq %r14, %rsi movq %r12, %rdx movl $1, %ecx callq hipMemcpy movq 88(%rsp), %rdi movq 24(%rsp), %rsi # 8-byte Reload movq %r12, %rdx movl $1, %ecx callq hipMemcpy movl %ebp, %eax shrl %eax movq %rax, %rdi shlq $32, %rdi orq %rax, %rdi movabsq $8589934594, %rdx # imm = 0x200000002 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_8 # %bb.7: movq 56(%rsp), %rax movq 88(%rsp), %rcx movq 72(%rsp), %rdx movl 48(%rsp), %esi movq %rax, 168(%rsp) movq %rcx, 160(%rsp) movq %rdx, 152(%rsp) movl %esi, 44(%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 44(%rsp), %rax movq %rax, 200(%rsp) leaq 136(%rsp), %rdi leaq 120(%rsp), %rsi leaq 112(%rsp), %rdx leaq 104(%rsp), %rcx callq __hipPopCallConfiguration movq 136(%rsp), %rsi movl 144(%rsp), %edx movq 120(%rsp), %rcx movl 128(%rsp), %r8d leaq 176(%rsp), %r9 movl $_Z6MatMulPKfS0_Pfi, %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 .LBB1_8: movq 72(%rsp), %rsi movq %r15, %rdi movq %r12, %rdx movl $2, %ecx callq hipMemcpy movq 16(%rsp), %rdi xorl %r12d, %r12d xorl %esi, %esi callq hipEventRecord movq 16(%rsp), %rdi callq hipEventSynchronize movq 32(%rsp), %rsi movq 16(%rsp), %rdx leaq 176(%rsp), %rdi callq hipEventElapsedTime movq stderr(%rip), %rdi movss 176(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.2, %esi movb $1, %al callq fprintf movq 32(%rsp), %rdi callq hipEventDestroy movq 16(%rsp), %rdi callq hipEventDestroy testl %ebp, %ebp jle .LBB1_13 # %bb.9: # %.preheader.lr.ph.i xorl %r13d, %r13d .p2align 4, 0x90 .LBB1_10: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB1_11 Depth 2 movl %r12d, %eax leaq (%r15,%rax,4), %rbx xorl %r14d, %r14d .p2align 4, 0x90 .LBB1_11: # Parent Loop BB1_10 Depth=1 # => This Inner Loop Header: Depth=2 movss (%rbx,%r14,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.3, %edi movb $1, %al callq printf incq %r14 cmpq %r14, %rbp jne .LBB1_11 # %bb.12: # %._crit_edge.i # in Loop: Header=BB1_10 Depth=1 movl $10, %edi callq putchar@PLT incq %r13 addl %ebp, %r12d cmpq %rbp, %r13 jne .LBB1_10 .LBB1_13: # %_Z4test6Matrix.exit movq 56(%rsp), %rdi callq hipFree movq 88(%rsp), %rdi callq hipFree movq 72(%rsp), %rdi callq hipFree movq 96(%rsp), %rdi # 8-byte Reload callq free movq 24(%rsp), %rdi # 8-byte Reload callq free movq %r15, %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_end1: .size main, .Lfunc_end1-main .cfi_endproc # -- End function .globl _Z4test6Matrix # -- Begin function _Z4test6Matrix .p2align 4, 0x90 .type _Z4test6Matrix,@function _Z4test6Matrix: # @_Z4test6Matrix .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 movq %rsi, (%rsp) # 8-byte Spill testl %edi, %edi jle .LBB2_5 # %bb.1: # %.preheader.lr.ph movl %edi, %ebp movl %edi, %r14d xorl %r15d, %r15d xorl %r12d, %r12d .p2align 4, 0x90 .LBB2_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB2_3 Depth 2 movl %r15d, %eax movq (%rsp), %rcx # 8-byte Reload leaq (%rcx,%rax,4), %r13 xorl %ebx, %ebx .p2align 4, 0x90 .LBB2_3: # Parent Loop BB2_2 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r13,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.3, %edi movb $1, %al callq printf incq %rbx cmpq %rbx, %r14 jne .LBB2_3 # %bb.4: # %._crit_edge # in Loop: Header=BB2_2 Depth=1 movl $10, %edi callq putchar@PLT incq %r12 addl %ebp, %r15d cmpq %r14, %r12 jne .LBB2_2 .LBB2_5: # %._crit_edge12 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_end2: .size _Z4test6Matrix, .Lfunc_end2-_Z4test6Matrix .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 .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 $_Z6MatMulPKfS0_Pfi, %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 _Z6MatMulPKfS0_Pfi,@object # @_Z6MatMulPKfS0_Pfi .section .rodata,"a",@progbits .globl _Z6MatMulPKfS0_Pfi .p2align 3, 0x0 _Z6MatMulPKfS0_Pfi: .quad _Z21__device_stub__MatMulPKfS0_Pfi .size _Z6MatMulPKfS0_Pfi, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%d" .size .L.str, 3 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "%f" .size .L.str.1, 3 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "Elapsed time = %f (s)\n" .size .L.str.2, 23 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "%4.1f " .size .L.str.3, 8 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z6MatMulPKfS0_Pfi" .size .L__unnamed_1, 19 .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__MatMulPKfS0_Pfi .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z6MatMulPKfS0_Pfi .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 : _Z6MatMulPKfS0_Pfi .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 R3, SR_CTAID.Y ; /* 0x0000000000037919 */ /* 0x000e220000002600 */ /*0020*/ MOV R14, c[0x0][0x178] ; /* 0x00005e00000e7a02 */ /* 0x000fe20000000f00 */ /*0030*/ ULDC.64 UR4, c[0x0][0x118] ; /* 0x0000460000047ab9 */ /* 0x000fe20000000a00 */ /*0040*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*0050*/ S2R R0, SR_TID.Y ; /* 0x0000000000007919 */ /* 0x000e220000002200 */ /*0060*/ ISETP.GE.AND P0, PT, R14, 0x2, PT ; /* 0x000000020e00780c */ /* 0x000fc60003f06270 */ /*0070*/ S2R R12, SR_CTAID.X ; /* 0x00000000000c7919 */ /* 0x000e680000002500 */ /*0080*/ S2R R13, SR_TID.X ; /* 0x00000000000d7919 */ /* 0x000ea20000002100 */ /*0090*/ LEA R16, R3, R0, 0x1 ; /* 0x0000000003107211 */ /* 0x001fca00078e08ff */ /*00a0*/ @!P0 BRA 0x730 ; /* 0x0000068000008947 */ /* 0x000fea0003800000 */ /*00b0*/ LEA.HI R2, R14, c[0x0][0x178], RZ, 0x1 ; /* 0x00005e000e027a11 */ /* 0x000fe200078f08ff */ /*00c0*/ HFMA2.MMA R19, -RZ, RZ, 0, 0 ; /* 0x00000000ff137435 */ /* 0x000fe200000001ff */ /*00d0*/ SHF.L.U32 R18, R0, 0x3, RZ ; /* 0x0000000300127819 */ /* 0x000fe200000006ff */ /*00e0*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*00f0*/ SHF.R.S32.HI R22, RZ, 0x1, R2 ; /* 0x00000001ff167819 */ /* 0x000fe40000011402 */ /*0100*/ SHF.L.U32 R21, R14, 0x1, RZ ; /* 0x000000010e157819 */ /* 0x000fe200000006ff */ /*0110*/ IMAD R18, R13, 0x4, R18 ; /* 0x000000040d127824 */ /* 0x004fe200078e0212 */ /*0120*/ IADD3 R2, R22.reuse, -0x1, RZ ; /* 0xffffffff16027810 */ /* 0x040fe40007ffe0ff */ /*0130*/ LOP3.LUT R17, R22, 0x3, RZ, 0xc0, !PT ; /* 0x0000000316117812 */ /* 0x000fc400078ec0ff */ /*0140*/ ISETP.GE.U32.AND P1, PT, R2, 0x3, PT ; /* 0x000000030200780c */ /* 0x000fe40003f26070 */ /*0150*/ ISETP.NE.AND P0, PT, R17, RZ, PT ; /* 0x000000ff1100720c */ /* 0x000fd60003f05270 */ /*0160*/ @!P1 BRA 0x560 ; /* 0x000003f000009947 */ /* 0x000fea0003800000 */ /*0170*/ MOV R20, 0x4 ; /* 0x0000000400147802 */ /* 0x000fe20000000f00 */ /*0180*/ IMAD R15, R0, c[0x0][0x178], R13.reuse ; /* 0x00005e00000f7a24 */ /* 0x100fe200078e020d */ /*0190*/ IADD3 R22, -R22, R17, RZ ; /* 0x0000001116167210 */ /* 0x000fe20007ffe1ff */ /*01a0*/ IMAD R3, R16, c[0x0][0x178], R13 ; /* 0x00005e0010037a24 */ /* 0x000fe200078e020d */ /*01b0*/ MOV R19, RZ ; /* 0x000000ff00137202 */ /* 0x000fe20000000f00 */ /*01c0*/ IMAD.MOV.U32 R9, RZ, RZ, RZ ; /* 0x000000ffff097224 */ /* 0x000fe200078e00ff */ /*01d0*/ LEA R15, R12, R15, 0x1 ; /* 0x0000000f0c0f7211 */ /* 0x002fe200078e08ff */ /*01e0*/ IMAD.WIDE R2, R3, R20, c[0x0][0x160] ; /* 0x0000580003027625 */ /* 0x000fc800078e0214 */ /*01f0*/ IMAD.WIDE R4, R15, R20, c[0x0][0x168] ; /* 0x00005a000f047625 */ /* 0x000fe200078e0214 */ /*0200*/ LDG.E R23, [R2.64] ; /* 0x0000000402177981 */ /* 0x000ea8000c1e1900 */ /*0210*/ LDG.E R25, [R4.64] ; /* 0x0000000404197981 */ /* 0x000ee2000c1e1900 */ /*0220*/ IMAD.WIDE R10, R21, 0x4, R4 ; /* 0x00000004150a7825 */ /* 0x000fc600078e0204 */ /*0230*/ STS [R18], R23 ; /* 0x0000001712007388 */ /* 0x004fe80000000800 */ /*0240*/ STS [R18+0x10], R25 ; /* 0x0000101912007388 */ /* 0x008fe80000000800 */ /*0250*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0260*/ LDS R8, [R13.X4+0x10] ; /* 0x000010000d087984 */ /* 0x000fe80000004800 */ /*0270*/ LDS R26, [R13.X4+0x18] ; /* 0x000018000d1a7984 */ /* 0x000fe80000004800 */ /*0280*/ LDS.64 R6, [R0.X8] ; /* 0x0000000000067984 */ /* 0x000e280000008a00 */ /*0290*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*02a0*/ LDG.E R29, [R2.64+0x8] ; /* 0x00000804021d7981 */ /* 0x000ea8000c1e1900 */ /*02b0*/ LDG.E R28, [R10.64] ; /* 0x000000040a1c7981 */ /* 0x000ee2000c1e1900 */ /*02c0*/ FFMA R6, R8, R6, R9 ; /* 0x0000000608067223 */ /* 0x001fc40000000009 */ /*02d0*/ IMAD.WIDE R8, R21, 0x4, R10 ; /* 0x0000000415087825 */ /* 0x000fe200078e020a */ /*02e0*/ STS [R18], R29 ; /* 0x0000001d12007388 */ /* 0x004fe80000000800 */ /*02f0*/ STS [R18+0x10], R28 ; /* 0x0000101c12007388 */ /* 0x008fe80000000800 */ /*0300*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0310*/ LDS R23, [R13.X4+0x10] ; /* 0x000010000d177984 */ /* 0x000fe80000004800 */ /*0320*/ LDS R24, [R13.X4+0x18] ; /* 0x000018000d187984 */ /* 0x000fe80000004800 */ /*0330*/ LDS.64 R4, [R0.X8] ; /* 0x0000000000047984 */ /* 0x000e280000008a00 */ /*0340*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0350*/ LDG.E R25, [R2.64+0x10] ; /* 0x0000100402197981 */ /* 0x000ea8000c1e1900 */ /*0360*/ LDG.E R11, [R8.64] ; /* 0x00000004080b7981 */ /* 0x000ee2000c1e1900 */ /*0370*/ FFMA R27, R26, R7, R6 ; /* 0x000000071a1b7223 */ /* 0x000fc60000000006 */ /*0380*/ STS [R18], R25 ; /* 0x0000001912007388 */ /* 0x004fe80000000800 */ /*0390*/ STS [R18+0x10], R11 ; /* 0x0000100b12007388 */ /* 0x0083e80000000800 */ /*03a0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fe20000010000 */ /*03b0*/ IMAD.WIDE R10, R21, 0x4, R8 ; /* 0x00000004150a7825 */ /* 0x002fca00078e0208 */ /*03c0*/ LDS R25, [R13.X4+0x10] ; /* 0x000010000d197984 */ /* 0x000fe80000004800 */ /*03d0*/ LDS R26, [R13.X4+0x18] ; /* 0x000018000d1a7984 */ /* 0x000fe80000004800 */ /*03e0*/ LDS.64 R6, [R0.X8] ; /* 0x0000000000067984 */ /* 0x000e680000008a00 */ /*03f0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0400*/ LDG.E R10, [R10.64] ; /* 0x000000040a0a7981 */ /* 0x000ea8000c1e1900 */ /*0410*/ LDG.E R29, [R2.64+0x18] ; /* 0x00001804021d7981 */ /* 0x000722000c1e1900 */ /*0420*/ FFMA R27, R23, R4, R27 ; /* 0x00000004171b7223 */ /* 0x001fe2000000001b */ /*0430*/ IADD3 R19, R19, 0x4, RZ ; /* 0x0000000413137810 */ /* 0x000fc40007ffe0ff */ /*0440*/ LEA R15, R14, R15, 0x3 ; /* 0x0000000f0e0f7211 */ /* 0x000fe200078e18ff */ /*0450*/ FFMA R5, R24, R5, R27 ; /* 0x0000000518057223 */ /* 0x000fc8000000001b */ /*0460*/ FFMA R5, R25, R6, R5 ; /* 0x0000000619057223 */ /* 0x002fe20000000005 */ /*0470*/ IADD3 R2, P2, R2, 0x20, RZ ; /* 0x0000002002027810 */ /* 0x008fe20007f5e0ff */ /*0480*/ IMAD.IADD R6, R22, 0x1, R19 ; /* 0x0000000116067824 */ /* 0x000fe400078e0213 */ /*0490*/ FFMA R5, R26, R7, R5 ; /* 0x000000071a057223 */ /* 0x000fe20000000005 */ /*04a0*/ IADD3.X R3, RZ, R3, RZ, P2, !PT ; /* 0x00000003ff037210 */ /* 0x000fe400017fe4ff */ /*04b0*/ ISETP.NE.AND P1, PT, R6, RZ, PT ; /* 0x000000ff0600720c */ /* 0x000fe20003f25270 */ /*04c0*/ STS [R18+0x10], R10 ; /* 0x0000100a12007388 */ /* 0x004fe80000000800 */ /*04d0*/ STS [R18], R29 ; /* 0x0000001d12007388 */ /* 0x010fe80000000800 */ /*04e0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*04f0*/ LDS R23, [R13.X4+0x10] ; /* 0x000010000d177984 */ /* 0x000fe80000004800 */ /*0500*/ LDS.64 R8, [R0.X8] ; /* 0x0000000000087984 */ /* 0x000e280000008a00 */ /*0510*/ LDS R4, [R13.X4+0x18] ; /* 0x000018000d047984 */ /* 0x000e620000004800 */ /*0520*/ FFMA R5, R23, R8, R5 ; /* 0x0000000817057223 */ /* 0x001fc80000000005 */ /*0530*/ FFMA R9, R4, R9, R5 ; /* 0x0000000904097223 */ /* 0x002fe20000000005 */ /*0540*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0550*/ @P1 BRA 0x1f0 ; /* 0xfffffc9000001947 */ /* 0x000fea000383ffff */ /*0560*/ @!P0 BRA 0x730 ; /* 0x000001c000008947 */ /* 0x000fea0003800000 */ /*0570*/ IMAD R6, R16, c[0x0][0x178], R13 ; /* 0x00005e0010067a24 */ /* 0x000fe200078e020d */ /*0580*/ MOV R4, 0x4 ; /* 0x0000000400047802 */ /* 0x000fe20000000f00 */ /*0590*/ IMAD R2, R19, 0x2, R0 ; /* 0x0000000213027824 */ /* 0x000fc600078e0200 */ /*05a0*/ LEA R3, R19, R6, 0x1 ; /* 0x0000000613037211 */ /* 0x000fe200078e08ff */ /*05b0*/ IMAD R5, R2, c[0x0][0x178], R13 ; /* 0x00005e0002057a24 */ /* 0x000fc800078e020d */ /*05c0*/ IMAD.WIDE R2, R3, R4, c[0x0][0x160] ; /* 0x0000580003027625 */ /* 0x000fc800078e0204 */ /*05d0*/ IMAD R5, R12, 0x2, R5 ; /* 0x000000020c057824 */ /* 0x002fe200078e0205 */ /*05e0*/ MOV R6, R2 ; /* 0x0000000200067202 */ /* 0x000fe40000000f00 */ /*05f0*/ MOV R15, R3 ; /* 0x00000003000f7202 */ /* 0x000fe20000000f00 */ /*0600*/ IMAD.WIDE R2, R5, R4, c[0x0][0x168] ; /* 0x00005a0005027625 */ /* 0x000fc800078e0204 */ /*0610*/ IMAD.MOV.U32 R7, RZ, RZ, R15 ; /* 0x000000ffff077224 */ /* 0x000fe200078e000f */ /*0620*/ LDG.E R11, [R2.64] ; /* 0x00000004020b7981 */ /* 0x0000aa000c1e1900 */ /*0630*/ LDG.E R7, [R6.64] ; /* 0x0000000406077981 */ /* 0x0002e2000c1e1900 */ /*0640*/ IADD3 R17, R17, -0x1, RZ ; /* 0xffffffff11117810 */ /* 0x000fc80007ffe0ff */ /*0650*/ ISETP.NE.AND P0, PT, R17, RZ, PT ; /* 0x000000ff1100720c */ /* 0x000fe20003f05270 */ /*0660*/ IMAD.WIDE R2, R21, 0x4, R2 ; /* 0x0000000415027825 */ /* 0x001fe200078e0202 */ /*0670*/ IADD3 R6, P1, R6, 0x8, RZ ; /* 0x0000000806067810 */ /* 0x002fc80007f3e0ff */ /*0680*/ IADD3.X R15, RZ, R15, RZ, P1, !PT ; /* 0x0000000fff0f7210 */ /* 0x000fe20000ffe4ff */ /*0690*/ STS [R18+0x10], R11 ; /* 0x0000100b12007388 */ /* 0x004fe80000000800 */ /*06a0*/ STS [R18], R7 ; /* 0x0000000712007388 */ /* 0x008fe80000000800 */ /*06b0*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*06c0*/ LDS R8, [R13.X4+0x10] ; /* 0x000010000d087984 */ /* 0x000fe80000004800 */ /*06d0*/ LDS.64 R4, [R0.X8] ; /* 0x0000000000047984 */ /* 0x000e280000008a00 */ /*06e0*/ LDS R10, [R13.X4+0x18] ; /* 0x000018000d0a7984 */ /* 0x000e620000004800 */ /*06f0*/ FFMA R4, R8, R4, R9 ; /* 0x0000000408047223 */ /* 0x001fc80000000009 */ /*0700*/ FFMA R9, R10, R5, R4 ; /* 0x000000050a097223 */ /* 0x002fe20000000004 */ /*0710*/ BAR.SYNC.DEFER_BLOCKING 0x0 ; /* 0x0000000000007b1d */ /* 0x000fec0000010000 */ /*0720*/ @P0 BRA 0x610 ; /* 0xfffffee000000947 */ /* 0x000fea000383ffff */ /*0730*/ HFMA2.MMA R3, -RZ, RZ, 0, 2.384185791015625e-07 ; /* 0x00000004ff037435 */ /* 0x000fe200000001ff */ /*0740*/ LEA R13, R12, R13, 0x1 ; /* 0x0000000d0c0d7211 */ /* 0x006fca00078e08ff */ /*0750*/ IMAD R2, R16, c[0x0][0x178], R13 ; /* 0x00005e0010027a24 */ /* 0x000fc800078e020d */ /*0760*/ IMAD.WIDE R2, R2, R3, c[0x0][0x170] ; /* 0x00005c0002027625 */ /* 0x000fca00078e0203 */ /*0770*/ STG.E [R2.64], R9 ; /* 0x0000000902007986 */ /* 0x000fe2000c101904 */ /*0780*/ EXIT ; /* 0x000000000000794d */ /* 0x000fea0003800000 */ /*0790*/ BRA 0x790; /* 0xfffffff000007947 */ /* 0x000fc0000383ffff */ /*07a0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07b0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07c0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07d0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07e0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*07f0*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*0800*/ NOP; /* 0x0000000000007918 */ /* 0x000fc00000000000 */ /*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 */ ..........
.text .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .protected _Z6MatMulPKfS0_Pfi .globl _Z6MatMulPKfS0_Pfi .p2align 8 .type _Z6MatMulPKfS0_Pfi,@function _Z6MatMulPKfS0_Pfi: s_load_b32 s2, s[0:1], 0x18 v_bfe_u32 v5, v0, 10, 10 v_dual_mov_b32 v2, 0 :: v_dual_and_b32 v3, 0x3ff, v0 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2) v_lshl_add_u32 v4, s15, 1, v5 v_lshl_add_u32 v0, s14, 1, v3 s_waitcnt lgkmcnt(0) s_cmp_lt_i32 s2, 2 s_cbranch_scc1 .LBB0_5 s_load_b128 s[4:7], s[0:1], 0x0 v_lshlrev_b32_e32 v8, 2, v3 v_lshlrev_b32_e32 v6, 3, v5 v_mad_u64_u32 v[1:2], null, v4, s2, v[3:4] s_lshr_b32 s3, s2, 31 s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_3) v_dual_mov_b32 v2, 0 :: v_dual_add_nc_u32 v7, 16, v8 v_add_nc_u32_e32 v3, v6, v8 s_add_i32 s3, s2, s3 s_mov_b32 s8, 0 s_delay_alu instid0(VALU_DEP_2) v_add_nc_u32_e32 v8, v7, v6 s_ashr_i32 s3, s3, 1 s_set_inst_prefetch_distance 0x1 .p2align 6 .LBB0_2: s_lshl_b32 s9, s8, 1 s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_2) | instid1(VALU_DEP_2) v_add_nc_u32_e32 v10, s9, v5 v_add_nc_u32_e32 v9, s9, v1 s_mov_b32 s9, 0 v_mad_u64_u32 v[11:12], null, v10, s2, v[0:1] s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1) v_ashrrev_i32_e32 v10, 31, v9 v_lshlrev_b64 v[9:10], 2, v[9:10] s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_2) v_ashrrev_i32_e32 v12, 31, v11 s_waitcnt lgkmcnt(0) v_add_co_u32 v9, vcc_lo, s4, v9 s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_4) v_lshlrev_b64 v[11:12], 2, v[11:12] v_add_co_ci_u32_e32 v10, vcc_lo, s5, v10, vcc_lo s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3) v_add_co_u32 v11, vcc_lo, s6, v11 v_add_co_ci_u32_e32 v12, vcc_lo, s7, v12, vcc_lo global_load_b32 v10, v[9:10], off global_load_b32 v11, v[11:12], off v_mov_b32_e32 v9, v7 s_waitcnt vmcnt(1) ds_store_b32 v3, v10 s_waitcnt vmcnt(0) ds_store_b32 v8, v11 s_waitcnt lgkmcnt(0) s_barrier buffer_gl0_inv .LBB0_3: v_add_nc_u32_e32 v10, s9, v6 s_add_i32 s9, s9, 4 ds_load_b32 v11, v9 ds_load_b32 v10, v10 v_add_nc_u32_e32 v9, 8, v9 s_cmp_lg_u32 s9, 4 s_waitcnt lgkmcnt(0) v_fmac_f32_e32 v2, v10, v11 s_cbranch_scc0 .LBB0_3 s_add_i32 s8, s8, 1 s_delay_alu instid0(SALU_CYCLE_1) s_cmp_eq_u32 s8, s3 s_barrier buffer_gl0_inv s_cbranch_scc0 .LBB0_2 .LBB0_5: s_set_inst_prefetch_distance 0x2 s_load_b64 s[0:1], s[0:1], 0x10 v_mad_u64_u32 v[5:6], null, v4, s2, v[0:1] 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], 2, 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_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 _Z6MatMulPKfS0_Pfi .amdhsa_group_segment_fixed_size 32 .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 13 .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 _Z6MatMulPKfS0_Pfi, .Lfunc_end0-_Z6MatMulPKfS0_Pfi .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: 32 .kernarg_segment_align: 8 .kernarg_segment_size: 28 .language: OpenCL C .language_version: - 2 - 0 .max_flat_workgroup_size: 1024 .name: _Z6MatMulPKfS0_Pfi .private_segment_fixed_size: 0 .sgpr_count: 18 .sgpr_spill_count: 0 .symbol: _Z6MatMulPKfS0_Pfi.kd .uniform_work_group_size: 1 .uses_dynamic_stack: false .vgpr_count: 13 .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_001a5ccc_00000000-6_MMult2.cudafe1.cpp" .text #APP #NO_APP .type _ZL26__cudaUnregisterBinaryUtilv, @function _ZL26__cudaUnregisterBinaryUtilv: .LFB2061: .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 .LFE2061: .size _ZL26__cudaUnregisterBinaryUtilv, .-_ZL26__cudaUnregisterBinaryUtilv .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "%4.1f " .LC1: .string "\n" .text .globl _Z4test6Matrix .type _Z4test6Matrix, @function _Z4test6Matrix: .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 subq $24, %rsp .cfi_def_cfa_offset 80 movl %edi, 4(%rsp) testl %edi, %edi jle .L3 movq %rsi, %r12 movslq %edi, %rdi leaq 0(,%rdi,4), %r15 negq %rdi leaq 0(,%rdi,4), %rax movq %rax, 8(%rsp) movq %r15, %rbp movl $0, %r14d leaq .LC0(%rip), %r13 .L5: movq 8(%rsp), %rax leaq 0(%rbp,%rax), %rbx .L6: pxor %xmm0, %xmm0 cvtss2sd (%rbx,%r12), %xmm0 movq %r13, %rsi movl $2, %edi movl $1, %eax call __printf_chk@PLT addq $4, %rbx cmpq %rbp, %rbx jne .L6 leaq .LC1(%rip), %rsi movl $2, %edi movl $0, %eax call __printf_chk@PLT addl $1, %r14d addq %r15, %rbp cmpl %r14d, 4(%rsp) jne .L5 .L3: 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 .LFE2058: .size _Z4test6Matrix, .-_Z4test6Matrix .globl _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi .type _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi, @function _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi: .LFB2083: .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 .L13 .L9: movq 136(%rsp), %rax subq %fs:40, %rax jne .L14 addq $152, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L13: .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 _Z6MatMulPKfS0_Pfi(%rip), %rdi call cudaLaunchKernel@PLT addq $16, %rsp .cfi_def_cfa_offset 160 jmp .L9 .L14: call __stack_chk_fail@PLT .cfi_endproc .LFE2083: .size _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi, .-_Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi .globl _Z6MatMulPKfS0_Pfi .type _Z6MatMulPKfS0_Pfi, @function _Z6MatMulPKfS0_Pfi: .LFB2084: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 call _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2084: .size _Z6MatMulPKfS0_Pfi, .-_Z6MatMulPKfS0_Pfi .section .rodata.str1.1 .LC2: .string "%d" .LC3: .string "%f" .LC4: .string "Elapsed time = %f (s)\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 $152, %rsp .cfi_def_cfa_offset 208 movq %fs:40, %rax movq %rax, 136(%rsp) xorl %eax, %eax leaq 40(%rsp), %rdi call cudaEventCreate@PLT leaq 48(%rsp), %rdi call cudaEventCreate@PLT movl $0, 32(%rsp) leaq 32(%rsp), %rsi leaq .LC2(%rip), %rdi movl $0, %eax call __isoc23_scanf@PLT movl 32(%rsp), %r13d movl %r13d, %eax movq %rax, 24(%rsp) movl %r13d, %r12d imull %r13d, %r12d sall $2, %r12d movslq %r12d, %r12 movq %r12, %rdi call malloc@PLT movq %rax, %rbp movq %rax, 16(%rsp) movq %r12, %rdi call malloc@PLT movq %rax, 8(%rsp) movq %r12, %rdi call malloc@PLT movq %rax, %r14 movl %r13d, %eax imull %r13d, %eax testl %eax, %eax jle .L23 movl $0, %ebx leaq .LC3(%rip), %r15 .L19: movq %rbp, %rsi movq %r15, %rdi movl $0, %eax call __isoc23_scanf@PLT addl $1, %ebx movl 32(%rsp), %eax movl %eax, %edx imull %eax, %edx addq $4, %rbp cmpl %ebx, %edx jg .L19 testl %edx, %edx jle .L18 movq 8(%rsp), %rbp movl $0, %ebx leaq .LC3(%rip), %r15 .L20: movq %rbp, %rsi movq %r15, %rdi movl $0, %eax call __isoc23_scanf@PLT addl $1, %ebx movl 32(%rsp), %eax addq $4, %rbp movl %eax, %edx imull %eax, %edx cmpl %ebx, %edx jg .L20 .L18: movl %eax, 80(%rsp) movl %eax, 96(%rsp) movl %eax, 112(%rsp) leaq 88(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT leaq 104(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT leaq 120(%rsp), %rdi movq %r12, %rsi call cudaMalloc@PLT movl $0, %esi movq 40(%rsp), %rdi call cudaEventRecord@PLT movl $1, %ecx movq %r12, %rdx movq 16(%rsp), %rsi movq 88(%rsp), %rdi call cudaMemcpy@PLT movl $1, %ecx movq %r12, %rdx movq 8(%rsp), %rsi movq 104(%rsp), %rdi call cudaMemcpy@PLT shrl %r13d movl %r13d, 68(%rsp) movl %r13d, 72(%rsp) movl $2, 56(%rsp) movl $2, 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 .L27 .L21: movl $2, %ecx movq %r12, %rdx movq 120(%rsp), %rsi movq %r14, %rdi call cudaMemcpy@PLT movl $0, %esi movq 48(%rsp), %rdi call cudaEventRecord@PLT movq 48(%rsp), %rdi call cudaEventSynchronize@PLT leaq 36(%rsp), %rdi movq 48(%rsp), %rdx movq 40(%rsp), %rsi call cudaEventElapsedTime@PLT pxor %xmm0, %xmm0 cvtss2sd 36(%rsp), %xmm0 leaq .LC4(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $1, %eax call __fprintf_chk@PLT movq 40(%rsp), %rdi call cudaEventDestroy@PLT movq 48(%rsp), %rdi call cudaEventDestroy@PLT movl 24(%rsp), %edi movq %r14, %rsi call _Z4test6Matrix movq 88(%rsp), %rdi call cudaFree@PLT movq 104(%rsp), %rdi call cudaFree@PLT movq 120(%rsp), %rdi call cudaFree@PLT movq 16(%rsp), %rdi call free@PLT movq 8(%rsp), %rdi call free@PLT movq %r14, %rdi call free@PLT movq 136(%rsp), %rax subq %fs:40, %rax jne .L28 movl $0, %eax addq $152, %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 .L23: .cfi_restore_state movl %r13d, %eax jmp .L18 .L27: movl 80(%rsp), %ecx movq 120(%rsp), %rdx movq 104(%rsp), %rsi movq 88(%rsp), %rdi call _Z32__device_stub__Z6MatMulPKfS0_PfiPKfS0_Pfi jmp .L21 .L28: call __stack_chk_fail@PLT .cfi_endproc .LFE2057: .size main, .-main .section .rodata.str1.1 .LC5: .string "_Z6MatMulPKfS0_Pfi" .text .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2086: .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 .LC5(%rip), %rdx movq %rdx, %rcx leaq _Z6MatMulPKfS0_Pfi(%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 .LFE2086: .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 "MMult2.hip" .globl _Z21__device_stub__MatMulPKfS0_Pfi # -- Begin function _Z21__device_stub__MatMulPKfS0_Pfi .p2align 4, 0x90 .type _Z21__device_stub__MatMulPKfS0_Pfi,@function _Z21__device_stub__MatMulPKfS0_Pfi: # @_Z21__device_stub__MatMulPKfS0_Pfi .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 $_Z6MatMulPKfS0_Pfi, %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 _Z21__device_stub__MatMulPKfS0_Pfi, .Lfunc_end0-_Z21__device_stub__MatMulPKfS0_Pfi .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 leaq 32(%rsp), %rdi callq hipEventCreate leaq 16(%rsp), %rdi callq hipEventCreate movl $0, 12(%rsp) leaq 12(%rsp), %rsi movl $.L.str, %edi xorl %eax, %eax callq __isoc23_scanf movl 12(%rsp), %ebp movl %ebp, %eax imull %ebp, %eax shll $2, %eax movslq %eax, %r12 movq %r12, %rdi callq malloc movq %rax, %r14 movq %r12, %rdi callq malloc movq %rax, 24(%rsp) # 8-byte Spill movq %r12, %rdi callq malloc movq %rax, %r15 cmpl $0, 12(%rsp) je .LBB1_3 # %bb.1: # %.lr.ph.preheader movq %r14, %r13 xorl %ebx, %ebx .p2align 4, 0x90 .LBB1_2: # %.lr.ph # =>This Inner Loop Header: Depth=1 movl $.L.str.1, %edi movq %r13, %rsi xorl %eax, %eax callq __isoc23_scanf incq %rbx movl 12(%rsp), %eax imull %eax, %eax addq $4, %r13 cmpq %rax, %rbx jb .LBB1_2 .LBB1_3: # %.preheader movl 12(%rsp), %eax testl %eax, %eax je .LBB1_6 # %bb.4: # %.lr.ph40.preheader movq 24(%rsp), %r13 # 8-byte Reload xorl %ebx, %ebx .p2align 4, 0x90 .LBB1_5: # %.lr.ph40 # =>This Inner Loop Header: Depth=1 movl $.L.str.1, %edi movq %r13, %rsi xorl %eax, %eax callq __isoc23_scanf incq %rbx movl 12(%rsp), %eax movl %eax, %ecx imull %ecx, %ecx addq $4, %r13 cmpq %rcx, %rbx jb .LBB1_5 .LBB1_6: # %._crit_edge movl %eax, 48(%rsp) movl %eax, 80(%rsp) movl %eax, 64(%rsp) leaq 56(%rsp), %rdi movq %r12, %rsi callq hipMalloc leaq 88(%rsp), %rdi movq %r12, %rsi callq hipMalloc leaq 72(%rsp), %rdi movq %r12, %rsi callq hipMalloc movq 32(%rsp), %rdi xorl %esi, %esi callq hipEventRecord movq 56(%rsp), %rdi movq %r14, 96(%rsp) # 8-byte Spill movq %r14, %rsi movq %r12, %rdx movl $1, %ecx callq hipMemcpy movq 88(%rsp), %rdi movq 24(%rsp), %rsi # 8-byte Reload movq %r12, %rdx movl $1, %ecx callq hipMemcpy movl %ebp, %eax shrl %eax movq %rax, %rdi shlq $32, %rdi orq %rax, %rdi movabsq $8589934594, %rdx # imm = 0x200000002 movl $1, %esi movl $1, %ecx xorl %r8d, %r8d xorl %r9d, %r9d callq __hipPushCallConfiguration testl %eax, %eax jne .LBB1_8 # %bb.7: movq 56(%rsp), %rax movq 88(%rsp), %rcx movq 72(%rsp), %rdx movl 48(%rsp), %esi movq %rax, 168(%rsp) movq %rcx, 160(%rsp) movq %rdx, 152(%rsp) movl %esi, 44(%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 44(%rsp), %rax movq %rax, 200(%rsp) leaq 136(%rsp), %rdi leaq 120(%rsp), %rsi leaq 112(%rsp), %rdx leaq 104(%rsp), %rcx callq __hipPopCallConfiguration movq 136(%rsp), %rsi movl 144(%rsp), %edx movq 120(%rsp), %rcx movl 128(%rsp), %r8d leaq 176(%rsp), %r9 movl $_Z6MatMulPKfS0_Pfi, %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 .LBB1_8: movq 72(%rsp), %rsi movq %r15, %rdi movq %r12, %rdx movl $2, %ecx callq hipMemcpy movq 16(%rsp), %rdi xorl %r12d, %r12d xorl %esi, %esi callq hipEventRecord movq 16(%rsp), %rdi callq hipEventSynchronize movq 32(%rsp), %rsi movq 16(%rsp), %rdx leaq 176(%rsp), %rdi callq hipEventElapsedTime movq stderr(%rip), %rdi movss 176(%rsp), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.2, %esi movb $1, %al callq fprintf movq 32(%rsp), %rdi callq hipEventDestroy movq 16(%rsp), %rdi callq hipEventDestroy testl %ebp, %ebp jle .LBB1_13 # %bb.9: # %.preheader.lr.ph.i xorl %r13d, %r13d .p2align 4, 0x90 .LBB1_10: # %.preheader.i # =>This Loop Header: Depth=1 # Child Loop BB1_11 Depth 2 movl %r12d, %eax leaq (%r15,%rax,4), %rbx xorl %r14d, %r14d .p2align 4, 0x90 .LBB1_11: # Parent Loop BB1_10 Depth=1 # => This Inner Loop Header: Depth=2 movss (%rbx,%r14,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.3, %edi movb $1, %al callq printf incq %r14 cmpq %r14, %rbp jne .LBB1_11 # %bb.12: # %._crit_edge.i # in Loop: Header=BB1_10 Depth=1 movl $10, %edi callq putchar@PLT incq %r13 addl %ebp, %r12d cmpq %rbp, %r13 jne .LBB1_10 .LBB1_13: # %_Z4test6Matrix.exit movq 56(%rsp), %rdi callq hipFree movq 88(%rsp), %rdi callq hipFree movq 72(%rsp), %rdi callq hipFree movq 96(%rsp), %rdi # 8-byte Reload callq free movq 24(%rsp), %rdi # 8-byte Reload callq free movq %r15, %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_end1: .size main, .Lfunc_end1-main .cfi_endproc # -- End function .globl _Z4test6Matrix # -- Begin function _Z4test6Matrix .p2align 4, 0x90 .type _Z4test6Matrix,@function _Z4test6Matrix: # @_Z4test6Matrix .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 movq %rsi, (%rsp) # 8-byte Spill testl %edi, %edi jle .LBB2_5 # %bb.1: # %.preheader.lr.ph movl %edi, %ebp movl %edi, %r14d xorl %r15d, %r15d xorl %r12d, %r12d .p2align 4, 0x90 .LBB2_2: # %.preheader # =>This Loop Header: Depth=1 # Child Loop BB2_3 Depth 2 movl %r15d, %eax movq (%rsp), %rcx # 8-byte Reload leaq (%rcx,%rax,4), %r13 xorl %ebx, %ebx .p2align 4, 0x90 .LBB2_3: # Parent Loop BB2_2 Depth=1 # => This Inner Loop Header: Depth=2 movss (%r13,%rbx,4), %xmm0 # xmm0 = mem[0],zero,zero,zero cvtss2sd %xmm0, %xmm0 movl $.L.str.3, %edi movb $1, %al callq printf incq %rbx cmpq %rbx, %r14 jne .LBB2_3 # %bb.4: # %._crit_edge # in Loop: Header=BB2_2 Depth=1 movl $10, %edi callq putchar@PLT incq %r12 addl %ebp, %r15d cmpq %r14, %r12 jne .LBB2_2 .LBB2_5: # %._crit_edge12 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_end2: .size _Z4test6Matrix, .Lfunc_end2-_Z4test6Matrix .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 .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 $_Z6MatMulPKfS0_Pfi, %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 _Z6MatMulPKfS0_Pfi,@object # @_Z6MatMulPKfS0_Pfi .section .rodata,"a",@progbits .globl _Z6MatMulPKfS0_Pfi .p2align 3, 0x0 _Z6MatMulPKfS0_Pfi: .quad _Z21__device_stub__MatMulPKfS0_Pfi .size _Z6MatMulPKfS0_Pfi, 8 .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "%d" .size .L.str, 3 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "%f" .size .L.str.1, 3 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "Elapsed time = %f (s)\n" .size .L.str.2, 23 .type .L.str.3,@object # @.str.3 .L.str.3: .asciz "%4.1f " .size .L.str.3, 8 .type .L__unnamed_1,@object # @0 .L__unnamed_1: .asciz "_Z6MatMulPKfS0_Pfi" .size .L__unnamed_1, 19 .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__MatMulPKfS0_Pfi .addrsig_sym __hip_module_ctor .addrsig_sym __hip_module_dtor .addrsig_sym _Z6MatMulPKfS0_Pfi .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> //void setup_cuda(int ngpus, int argc, char **argv){ //insert from Bob' Born // ; //} //void process_error( const cudaError_t &error, char *string=0, bool verbose=false ){ //insert from Bob's Born // ; //} extern "C" void rtm_gpu_init(int nt, int nz, int nx, int zrec, float * Vx0, float * Vz0, float * sigmaxx0, float * sigmazz0, float * sigmaxz0, //(nz, nx, nt) // float * Vx, float * Vz, float * sigmaxx, float * sigmazz, float * sigmaxz, //(nt, nx) float * m1_x,float * m1_z,float * aux_m2_c, float * aux_m3_c, float * aux_m2m3_c) { //set cuda devices and put all data onto gpu memory cudaError_t cuda_ret; cudaError_t err; //Set Device fprintf(stderr,"GPU init. \n"); cuda_ret = cudaSetDevice(0); if(cuda_ret != cudaSuccess){ fprintf(stderr, "Failed to set the cuda device !\n"); } else{ fprintf(stderr, "cuda device set OK\n"); } // init data //cudaMalloc(&g_,sizeof()*nx*nz*nt); }
code for sm_80
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
#include <stdio.h> //void setup_cuda(int ngpus, int argc, char **argv){ //insert from Bob' Born // ; //} //void process_error( const cudaError_t &error, char *string=0, bool verbose=false ){ //insert from Bob's Born // ; //} extern "C" void rtm_gpu_init(int nt, int nz, int nx, int zrec, float * Vx0, float * Vz0, float * sigmaxx0, float * sigmazz0, float * sigmaxz0, //(nz, nx, nt) // float * Vx, float * Vz, float * sigmaxx, float * sigmazz, float * sigmaxz, //(nt, nx) float * m1_x,float * m1_z,float * aux_m2_c, float * aux_m3_c, float * aux_m2m3_c) { //set cuda devices and put all data onto gpu memory cudaError_t cuda_ret; cudaError_t err; //Set Device fprintf(stderr,"GPU init. \n"); cuda_ret = cudaSetDevice(0); if(cuda_ret != cudaSuccess){ fprintf(stderr, "Failed to set the cuda device !\n"); } else{ fprintf(stderr, "cuda device set OK\n"); } // init data //cudaMalloc(&g_,sizeof()*nx*nz*nt); }
.file "tmpxft_00121692_00000000-6_rtmlib-modC-gpu.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 .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "GPU init. \n" .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC1: .string "Failed to set the cuda device !\n" .section .rodata.str1.1 .LC2: .string "cuda device set OK\n" .text .globl rtm_gpu_init .type rtm_gpu_init, @function rtm_gpu_init: .LFB2057: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq .LC0(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $0, %edi call cudaSetDevice@PLT testl %eax, %eax je .L4 leaq .LC1(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT .L3: addq $8, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L4: .cfi_restore_state leaq .LC2(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L3 .cfi_endproc .LFE2057: .size rtm_gpu_init, .-rtm_gpu_init .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2083: .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) call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2083: .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> //void setup_cuda(int ngpus, int argc, char **argv){ //insert from Bob' Born // ; //} //void process_error( const cudaError_t &error, char *string=0, bool verbose=false ){ //insert from Bob's Born // ; //} extern "C" void rtm_gpu_init(int nt, int nz, int nx, int zrec, float * Vx0, float * Vz0, float * sigmaxx0, float * sigmazz0, float * sigmaxz0, //(nz, nx, nt) // float * Vx, float * Vz, float * sigmaxx, float * sigmazz, float * sigmaxz, //(nt, nx) float * m1_x,float * m1_z,float * aux_m2_c, float * aux_m3_c, float * aux_m2m3_c) { //set cuda devices and put all data onto gpu memory cudaError_t cuda_ret; cudaError_t err; //Set Device fprintf(stderr,"GPU init. \n"); cuda_ret = cudaSetDevice(0); if(cuda_ret != cudaSuccess){ fprintf(stderr, "Failed to set the cuda device !\n"); } else{ fprintf(stderr, "cuda device set OK\n"); } // init data //cudaMalloc(&g_,sizeof()*nx*nz*nt); }
#include <hip/hip_runtime.h> #include <stdio.h> //void setup_cuda(int ngpus, int argc, char **argv){ //insert from Bob' Born // ; //} //void process_error( const cudaError_t &error, char *string=0, bool verbose=false ){ //insert from Bob's Born // ; //} extern "C" void rtm_gpu_init(int nt, int nz, int nx, int zrec, float * Vx0, float * Vz0, float * sigmaxx0, float * sigmazz0, float * sigmaxz0, //(nz, nx, nt) // float * Vx, float * Vz, float * sigmaxx, float * sigmazz, float * sigmaxz, //(nt, nx) float * m1_x,float * m1_z,float * aux_m2_c, float * aux_m3_c, float * aux_m2m3_c) { //set cuda devices and put all data onto gpu memory hipError_t cuda_ret; hipError_t err; //Set Device fprintf(stderr,"GPU init. \n"); cuda_ret = hipSetDevice(0); if(cuda_ret != hipSuccess){ fprintf(stderr, "Failed to set the cuda device !\n"); } else{ fprintf(stderr, "cuda device set OK\n"); } // init data //cudaMalloc(&g_,sizeof()*nx*nz*nt); }
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> #include <stdio.h> //void setup_cuda(int ngpus, int argc, char **argv){ //insert from Bob' Born // ; //} //void process_error( const cudaError_t &error, char *string=0, bool verbose=false ){ //insert from Bob's Born // ; //} extern "C" void rtm_gpu_init(int nt, int nz, int nx, int zrec, float * Vx0, float * Vz0, float * sigmaxx0, float * sigmazz0, float * sigmaxz0, //(nz, nx, nt) // float * Vx, float * Vz, float * sigmaxx, float * sigmazz, float * sigmaxz, //(nt, nx) float * m1_x,float * m1_z,float * aux_m2_c, float * aux_m3_c, float * aux_m2m3_c) { //set cuda devices and put all data onto gpu memory hipError_t cuda_ret; hipError_t err; //Set Device fprintf(stderr,"GPU init. \n"); cuda_ret = hipSetDevice(0); if(cuda_ret != hipSuccess){ fprintf(stderr, "Failed to set the cuda device !\n"); } else{ fprintf(stderr, "cuda device set OK\n"); } // init data //cudaMalloc(&g_,sizeof()*nx*nz*nt); }
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] 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> //void setup_cuda(int ngpus, int argc, char **argv){ //insert from Bob' Born // ; //} //void process_error( const cudaError_t &error, char *string=0, bool verbose=false ){ //insert from Bob's Born // ; //} extern "C" void rtm_gpu_init(int nt, int nz, int nx, int zrec, float * Vx0, float * Vz0, float * sigmaxx0, float * sigmazz0, float * sigmaxz0, //(nz, nx, nt) // float * Vx, float * Vz, float * sigmaxx, float * sigmazz, float * sigmaxz, //(nt, nx) float * m1_x,float * m1_z,float * aux_m2_c, float * aux_m3_c, float * aux_m2m3_c) { //set cuda devices and put all data onto gpu memory hipError_t cuda_ret; hipError_t err; //Set Device fprintf(stderr,"GPU init. \n"); cuda_ret = hipSetDevice(0); if(cuda_ret != hipSuccess){ fprintf(stderr, "Failed to set the cuda device !\n"); } else{ fprintf(stderr, "cuda device set OK\n"); } // init data //cudaMalloc(&g_,sizeof()*nx*nz*nt); }
.text .file "rtmlib-modC-gpu.hip" .globl rtm_gpu_init # -- Begin function rtm_gpu_init .p2align 4, 0x90 .type rtm_gpu_init,@function rtm_gpu_init: # @rtm_gpu_init .cfi_startproc # %bb.0: pushq %rax .cfi_def_cfa_offset 16 movq stderr(%rip), %rcx movl $.L.str, %edi movl $11, %esi movl $1, %edx callq fwrite@PLT xorl %edi, %edi callq hipSetDevice movq stderr(%rip), %rcx testl %eax, %eax jne .LBB0_1 # %bb.3: movl $.L.str.2, %edi movl $19, %esi jmp .LBB0_2 .LBB0_1: movl $.L.str.1, %edi movl $32, %esi .LBB0_2: movl $1, %edx popq %rax .cfi_def_cfa_offset 8 jmp fwrite@PLT # TAILCALL .Lfunc_end0: .size rtm_gpu_init, .Lfunc_end0-rtm_gpu_init .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "GPU init. \n" .size .L.str, 12 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "Failed to set the cuda device !\n" .size .L.str.1, 33 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "cuda device set OK\n" .size .L.str.2, 20 .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 __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] 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_00121692_00000000-6_rtmlib-modC-gpu.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 .section .rodata.str1.1,"aMS",@progbits,1 .LC0: .string "GPU init. \n" .section .rodata.str1.8,"aMS",@progbits,1 .align 8 .LC1: .string "Failed to set the cuda device !\n" .section .rodata.str1.1 .LC2: .string "cuda device set OK\n" .text .globl rtm_gpu_init .type rtm_gpu_init, @function rtm_gpu_init: .LFB2057: .cfi_startproc endbr64 subq $8, %rsp .cfi_def_cfa_offset 16 leaq .LC0(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT movl $0, %edi call cudaSetDevice@PLT testl %eax, %eax je .L4 leaq .LC1(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT .L3: addq $8, %rsp .cfi_remember_state .cfi_def_cfa_offset 8 ret .L4: .cfi_restore_state leaq .LC2(%rip), %rdx movl $2, %esi movq stderr(%rip), %rdi movl $0, %eax call __fprintf_chk@PLT jmp .L3 .cfi_endproc .LFE2057: .size rtm_gpu_init, .-rtm_gpu_init .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2083: .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) call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2083: .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 "rtmlib-modC-gpu.hip" .globl rtm_gpu_init # -- Begin function rtm_gpu_init .p2align 4, 0x90 .type rtm_gpu_init,@function rtm_gpu_init: # @rtm_gpu_init .cfi_startproc # %bb.0: pushq %rax .cfi_def_cfa_offset 16 movq stderr(%rip), %rcx movl $.L.str, %edi movl $11, %esi movl $1, %edx callq fwrite@PLT xorl %edi, %edi callq hipSetDevice movq stderr(%rip), %rcx testl %eax, %eax jne .LBB0_1 # %bb.3: movl $.L.str.2, %edi movl $19, %esi jmp .LBB0_2 .LBB0_1: movl $.L.str.1, %edi movl $32, %esi .LBB0_2: movl $1, %edx popq %rax .cfi_def_cfa_offset 8 jmp fwrite@PLT # TAILCALL .Lfunc_end0: .size rtm_gpu_init, .Lfunc_end0-rtm_gpu_init .cfi_endproc # -- End function .type .L.str,@object # @.str .section .rodata.str1.1,"aMS",@progbits,1 .L.str: .asciz "GPU init. \n" .size .L.str, 12 .type .L.str.1,@object # @.str.1 .L.str.1: .asciz "Failed to set the cuda device !\n" .size .L.str.1, 33 .type .L.str.2,@object # @.str.2 .L.str.2: .asciz "cuda device set OK\n" .size .L.str.2, 20 .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 __hip_cuid_
You are a helpful coding assistant specialized in converting CUDA source code to CUDA device assembly.
// have fun :-)
code for sm_80
You are a helpful coding assistant specialized in converting CUDA source code to CUDA host assembly.
// have fun :-)
.file "tmpxft_0019eaa9_00000000-6_cumulative_sum.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 .type _ZL24__sti____cudaRegisterAllv, @function _ZL24__sti____cudaRegisterAllv: .LFB2052: .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) call __cudaRegisterFatBinaryEnd@PLT leaq _ZL26__cudaUnregisterBinaryUtilv(%rip), %rdi call atexit@PLT addq $8, %rsp .cfi_def_cfa_offset 8 ret .cfi_endproc .LFE2052: .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.
// have fun :-)
#include <hip/hip_runtime.h> // have fun :-)
You are a helpful coding assistant specialized in converting HIP source code to AMD device assembly.
#include <hip/hip_runtime.h> // have fun :-)
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] 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> // have fun :-)
.text .file "cumulative_sum.hip" .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 __hip_cuid_
You are a helpful coding assistant specialized in converting from CUDA device assembly to AMD device assembly.
code for sm_80
.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_ .amdgcn_target "amdgcn-amd-amdhsa--gfx1100" .amdgpu_metadata --- amdhsa.kernels: [] amdhsa.target: amdgcn-amd-amdhsa--gfx1100 amdhsa.version: - 1 - 2 ... .end_amdgpu_metadata