file stringlengths 6 242 | name stringlengths 1 11 | kernel stringclasses 4
values | input stringlengths 531 14.6k | output stringlengths 176 17.2k |
|---|---|---|---|---|
merged_1 | 67 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<35>;
.reg .b32 %r<30>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 125 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 176 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<17>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 201 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<18>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = (in... |
merged_1 | 12 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<11>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 110 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<28>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 232 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<21>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 88 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<19>;
.reg .b32 %r<24>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 180 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 127 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 81 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<35>;
.reg .b32 %r<24>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 202 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<21>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 126 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<30>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 238 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<28>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 172 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<24>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 70 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<5>;
.reg .b32 %r<24>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_sh... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 36 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .b16 %rs<7>;
.reg .f32 %f<19>;
.reg .b32 %r<32>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Gro... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 128 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<17>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 34 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<19>;
.reg .b32 %r<24>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 239 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 72 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<35>;
.reg .b32 %r<18>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = (in... |
merged_1 | 124 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<24>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 269 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 1, 1, 1
{
.reg .f32 %f<4>;
.reg .b32 %r<8>;
.reg .b64 %rd<11>;
ld.param.u64 %rd1, [_Z6Group7PfS_S__param_0];
ld.param.u64 %rd2, [_Z6Group7PfS_S__param_... | __global__ void __launch_bounds__(1) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
int index0 = (int)blockIdx.x;
int index1 = (((int)blockIdx.x & 32767) >> 6) * 1536 + ((int)blockIdx.x >> 15) * 64 + ((int)blockIdx.x & 63);
output0[index0] = input1[index1] / input0[0... |
merged_1 | 234 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 266 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 2, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<5>;
.reg .b32 %r<16>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_shar... | __global__ void __launch_bounds__(2) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = (int)... |
merged_1 | 186 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 230 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 8 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<11>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 231 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<17>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 76 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .b16 %rs<6>;
.reg .f32 %f<5>;
.reg .b32 %r<32>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Grou... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 0 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<11>;
.reg .b32 %r<27>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 23 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<19>;
.reg .b32 %r<18>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = (in... |
merged_1 | 129 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<30>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 42 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<19>;
.reg .b32 %r<24>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 92 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<35>;
.reg .b32 %r<17>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 249 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 8, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<5>;
.reg .b32 %r<28>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_shar... | __global__ void __launch_bounds__(8) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((int... |
merged_1 | 75 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<35>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 123 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<26>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 37 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .b16 %rs<5>;
.reg .f32 %f<7>;
.reg .b32 %r<30>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Grou... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 30 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .b16 %rs<6>;
.reg .f32 %f<19>;
.reg .b32 %r<32>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Gro... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 99 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 64, 1, 1
{
.reg .pred %p<2>;
.reg .b16 %rs<6>;
.reg .f32 %f<5>;
.reg .b32 %r<32>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group... | __global__ void __launch_bounds__(64) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((in... |
merged_1 | 122 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .b16 %rs<4>;
.reg .f32 %f<67>;
.reg .b32 %r<32>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Gro... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 62 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<35>;
.reg .b32 %r<28>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 233 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 119 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<18>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = (in... |
merged_1 | 198 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<11>;
.reg .b32 %r<12>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 111 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<67>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 4 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<11>;
.reg .b32 %r<16>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = (in... |
merged_1 | 89 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<19>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_s... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 185 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 51 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .b16 %rs<6>;
.reg .f32 %f<5>;
.reg .b32 %r<32>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Grou... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 174 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<131>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 236 | null | .visible .entry _Z6Group7PfS_S_(
.param .u64 _Z6Group7PfS_S__param_0,
.param .u64 _Z6Group7PfS_S__param_1,
.param .u64 _Z6Group7PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<259>;
.reg .b32 %r<25>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ6Group7PfS_S_E13input0_... | __global__ void __launch_bounds__(128) Group7(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
__shared__ float input0_shared[8];
if ((int)threadIdx.x < 1) {
int index0 = (int)threadIdx.x;
input0_shared[index0] = input0[index0];
}
__syncthreads();
int index1 = ((i... |
merged_1 | 32 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<91>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[2];
__shared__ float mediate1_shared[4096];
__shared__ float input1_shared[256];
output0_local[0] = 0.000000e+00f;
output0_local[1] =... |
merged_1 | 24 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<85>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[4];
__shared__ float mediate1_shared[8192];
__shared__ float input1_shared[4096];
output0_local[0] = 0.000000e+00f;
output0_local[1] ... |
merged_1 | 26 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<93>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[4];
__shared__ float mediate1_shared[6144];
__shared__ float input1_shared[1536];
output0_local[0] = 0.000000e+00f;
output0_local[1] ... |
merged_1 | 52 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<5>;
.reg .f32 %f<55>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float mediate1_shared[1024];
__shared__ float input1_shared[256];
float red_buf0[1];
normal_reduce_temp0[0] = 0.... |
merged_1 | 1 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<211>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[32];
__shared__ float mediate1_shared[8192];
__shared__ float input1_shared[512];
output0_local[0] = 0.000000e+00f;
output0_local[16]... |
merged_1 | 11 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<129>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[16];
__shared__ float mediate1_shared[512];
__shared__ float input1_shared[4096];
output0_local[0] = 0.000000e+00f;
output0_local[4] ... |
merged_1 | 31 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<97>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[8];
__shared__ float mediate1_shared[2048];
__shared__ float input1_shared[2048];
output0_local[0] = 0.000000e+00f;
output0_local[2] ... |
merged_1 | 139 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<4>;
.reg .f32 %f<102>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[1];
__shared__ float mediate1_shared[10240];
__shared__ float input1_shared[160];
output0_local[0] = 0.000000e+00f;
for (int K_outer ... |
merged_1 | 17 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<205>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[32];
__shared__ float mediate1_shared[512];
__shared__ float input1_shared[8192];
output0_local[0] = 0.000000e+00f;
output0_local[4] ... |
merged_1 | 65 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<6>;
.reg .f32 %f<33>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float mediate1_shared[512];
__shared__ float input1_shared[512];
float red_buf0[1];
normal_reduce_temp0[0] = 0.0... |
merged_1 | 15 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<129>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[16];
__shared__ float mediate1_shared[4096];
__shared__ float input1_shared[2048];
output0_local[0] = 0.000000e+00f;
output0_local[4]... |
merged_1 | 6 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<205>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[32];
__shared__ float mediate1_shared[4096];
__shared__ float input1_shared[1024];
output0_local[0] = 0.000000e+00f;
output0_local[8]... |
merged_1 | 114 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<5>;
.reg .f32 %f<170>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[2];
__shared__ float mediate1_shared[10240];
__shared__ float input1_shared[320];
output0_local[0] = 0.000000e+00f;
output0_local[1] ... |
merged_1 | 12 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<129>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[16];
__shared__ float mediate1_shared[2048];
__shared__ float input1_shared[4096];
output0_local[0] = 0.000000e+00f;
output0_local[4]... |
merged_1 | 13 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<205>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[32];
__shared__ float mediate1_shared[1024];
__shared__ float input1_shared[4096];
output0_local[0] = 0.000000e+00f;
output0_local[4]... |
merged_1 | 8 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<133>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[16];
__shared__ float mediate1_shared[4096];
__shared__ float input1_shared[512];
output0_local[0] = 0.000000e+00f;
output0_local[8] ... |
merged_1 | 28 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<77>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[4];
__shared__ float mediate1_shared[2048];
__shared__ float input1_shared[4096];
output0_local[0] = 0.000000e+00f;
output0_local[2] ... |
merged_1 | 142 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<26>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float mediate1_shared[256];
__shared__ float input1_shared[256];
__shared__ float red_buf0[128];
normal_reduce_t... |
merged_1 | 14 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<205>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[32];
__shared__ float mediate1_shared[2048];
__shared__ float input1_shared[2048];
output0_local[0] = 0.000000e+00f;
output0_local[4]... |
merged_1 | 91 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<52>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[1];
__shared__ float mediate1_shared[512];
__shared__ float input1_shared[4096];
output0_local[0] = 0.000000e+00f;
*(float4*)(mediate... |
merged_1 | 10 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<109>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[8];
__shared__ float mediate1_shared[4096];
__shared__ float input1_shared[256];
output0_local[0] = 0.000000e+00f;
output0_local[1] =... |
merged_1 | 120 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<168>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[2];
__shared__ float mediate1_shared[32];
__shared__ float input1_shared[8192];
output0_local[0] = 0.000000e+00f;
output0_local[1] = ... |
merged_1 | 49 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<172>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[2];
__shared__ float mediate1_shared[5120];
__shared__ float input1_shared[640];
output0_local[0] = 0.000000e+00f;
output0_local[1] =... |
merged_1 | 59 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<6>;
.reg .f32 %f<33>;
.reg .b32 ... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float mediate1_shared[512];
__shared__ float input1_shared[256];
float red_buf0[1];
normal_reduce_temp0[0] = 0.0... |
merged_1 | 23 | null | .visible .entry _Z30Reshape_Reshape_DotSplitK_4278PfS_S_(
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_0,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_1,
.param .u64 _Z30Reshape_Reshape_DotSplitK_4278PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<110>;
.reg .b32... | __global__ void __launch_bounds__(128) Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float output0_local[8];
__shared__ float mediate1_shared[8192];
__shared__ float input1_shared[128];
output0_local[0] = 0.000000e+00f;
output0_local[1] =... |
merged_1 | 0 | null | .visible .entry _Z7Not_564PaS_(
.param .u64 _Z7Not_564PaS__param_0,
.param .u64 _Z7Not_564PaS__param_1
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .b16 %rs<7>;
.reg .b32 %r<5>;
.reg .b64 %rd<8>;
ld.param.u64 %rd1, [_Z7Not_564PaS__param_0];
ld.param.u64 %rd2, [_Z7Not_564PaS__param_1];
cvta.to.global.u... | __global__ void __launch_bounds__(128) Not_564(int8_t* __restrict__ input0, int8_t* __restrict__ output0) {
int index0 = (int)blockIdx.x * 256 + (int)threadIdx.x;
output0[index0] = signed char((int)input0[index0] == 0);
output0[index0 + 128] = signed char((int)input0[index0 + 128] == 0);
}
|
merged_1 | 3 | null | .visible .entry _Z17SoftmaxBasic_3920PfS_(
.param .u64 _Z17SoftmaxBasic_3920PfS__param_0,
.param .u64 _Z17SoftmaxBasic_3920PfS__param_1
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<36>;
.reg .b32 %r<41>;
.reg .b64 %rd<9>;
// demoted variable
.shared .align 4 .b8 _ZZ17SoftmaxBasic_3920PfS_E13input0_... | __global__ void __launch_bounds__(128) SoftmaxBasic_3920(float* __restrict__ input0, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float input0_shared[1024];
__shared__ float red_buf0[128];
normal_reduce_temp0[0] = 0.000000e+00f;
for (int ax0_ax1_fused_ax2_fused_outer_outer_outer = 0... |
merged_1 | 4 | null | .visible .entry _Z17SoftmaxBasic_3920PfS_(
.param .u64 _Z17SoftmaxBasic_3920PfS__param_0,
.param .u64 _Z17SoftmaxBasic_3920PfS__param_1
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<36>;
.reg .b32 %r<47>;
.reg .b64 %rd<9>;
// demoted variable
.shared .align 4 .b8 _ZZ17SoftmaxBasic_3920PfS_E13input0_... | __global__ void __launch_bounds__(128) SoftmaxBasic_3920(float* __restrict__ input0, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float input0_shared[1024];
__shared__ float red_buf0[128];
normal_reduce_temp0[0] = 0.000000e+00f;
for (int ax0_ax1_fused_ax2_fused_outer_outer_outer = 0... |
merged_1 | 5 | null | .visible .entry _Z17SoftmaxBasic_3920PfS_(
.param .u64 _Z17SoftmaxBasic_3920PfS__param_0,
.param .u64 _Z17SoftmaxBasic_3920PfS__param_1
)
.maxntid 128, 1, 1
{
.reg .pred %p<6>;
.reg .f32 %f<73>;
.reg .b32 %r<101>;
.reg .b64 %rd<9>;
// demoted variable
.shared .align 4 .b8 _ZZ17SoftmaxBasic_3920PfS_E13input0... | __global__ void __launch_bounds__(128) SoftmaxBasic_3920(float* __restrict__ input0, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float input0_shared[4096];
float red_buf0[1];
normal_reduce_temp0[0] = 0.000000e+00f;
for (int ax0_ax1_fused_ax2_fused_outer_outer_outer = 0; ax0_ax1_fus... |
merged_1 | 0 | null | .visible .entry _Z17SoftmaxBasic_3920PfS_(
.param .u64 _Z17SoftmaxBasic_3920PfS__param_0,
.param .u64 _Z17SoftmaxBasic_3920PfS__param_1
)
.maxntid 128, 1, 1
{
.reg .pred %p<7>;
.reg .f32 %f<43>;
.reg .b32 %r<81>;
.reg .b64 %rd<9>;
// demoted variable
.shared .align 4 .b8 _ZZ17SoftmaxBasic_3920PfS_E13input0_... | __global__ void __launch_bounds__(128) SoftmaxBasic_3920(float* __restrict__ input0, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float input0_shared[2048];
float red_buf0[1];
normal_reduce_temp0[0] = 0.000000e+00f;
for (int ax0_ax1_fused_ax2_fused_outer_outer_outer = 0; ax0_ax1_fus... |
merged_1 | 10 | null | .visible .entry _Z17SoftmaxBasic_3920PfS_(
.param .u64 _Z17SoftmaxBasic_3920PfS__param_0,
.param .u64 _Z17SoftmaxBasic_3920PfS__param_1
)
.maxntid 128, 1, 1
{
.reg .pred %p<5>;
.reg .f32 %f<72>;
.reg .b32 %r<115>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ17SoftmaxBasic_3920PfS_E13input... | __global__ void __launch_bounds__(128) SoftmaxBasic_3920(float* __restrict__ input0, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float input0_shared[4096];
float red_buf0[1];
normal_reduce_temp0[0] = 0.000000e+00f;
for (int N2_outer = 0; N2_outer < 4; ++N2_outer) {
__syncthread... |
merged_1 | 22 | null | .visible .entry _Z17SoftmaxBasic_3920PfS_(
.param .u64 _Z17SoftmaxBasic_3920PfS__param_0,
.param .u64 _Z17SoftmaxBasic_3920PfS__param_1
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<68>;
.reg .b32 %r<85>;
.reg .b64 %rd<13>;
// demoted variable
.shared .align 4 .b8 _ZZ17SoftmaxBasic_3920PfS_E13input0... | __global__ void __launch_bounds__(128) SoftmaxBasic_3920(float* __restrict__ input0, float* __restrict__ output0) {
float output0_local[1];
__shared__ float input0_shared[4096];
output0_local[0] = 0.000000e+00f;
for (int N2_outer = 0; N2_outer < 16; ++N2_outer) {
__syncthreads();
for (int ax0_ax1_fused_... |
merged_1 | 1 | null | .visible .entry _Z7Group13PfS_S_S_(
.param .u64 _Z7Group13PfS_S_S__param_0,
.param .u64 _Z7Group13PfS_S_S__param_1,
.param .u64 _Z7Group13PfS_S_S__param_2,
.param .u64 _Z7Group13PfS_S_S__param_3
)
.maxntid 128, 1, 1
{
.reg .pred %p<5>;
.reg .f32 %f<57>;
.reg .b32 %r<239>;
.reg .b64 %rd<17>;
// demoted vari... | __device__ void Group13_0_Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0, char* shared) {
float output0_local[2];
float* mediate1_shared = (float*)(shared+0);
float* input1_shared = (float*)(shared+32768);
output0_local[0] = 0.000000e+00f;
ou... |
merged_1 | 0 | null | .visible .entry _Z7Group13PfS_S_S_(
.param .u64 _Z7Group13PfS_S_S__param_0,
.param .u64 _Z7Group13PfS_S_S__param_1,
.param .u64 _Z7Group13PfS_S_S__param_2,
.param .u64 _Z7Group13PfS_S_S__param_3
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<88>;
.reg .b32 %r<257>;
.reg .b64 %rd<17>;
// demoted vari... | __device__ void Group13_0_Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0, char* shared) {
float output0_local[4];
float* mediate1_shared = (float*)(shared+0);
float* input1_shared = (float*)(shared+32768);
output0_local[0] = 0.000000e+00f;
ou... |
merged_1 | 19 | null | .visible .entry _Z7Group13PfS_S_S_(
.param .u64 _Z7Group13PfS_S_S__param_0,
.param .u64 _Z7Group13PfS_S_S__param_1,
.param .u64 _Z7Group13PfS_S_S__param_2,
.param .u64 _Z7Group13PfS_S_S__param_3
)
.maxntid 64, 1, 1
{
.reg .pred %p<10>;
.reg .f32 %f<71>;
.reg .b32 %r<197>;
.reg .b64 %rd<23>;
// demoted vari... | __device__ void Group13_0_Reshape_Reshape_DotSplitK_4278(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0, char* shared) {
float output0_local[1];
float* mediate1_shared = (float*)(shared+0);
float* input1_shared = (float*)(shared+28672);
output0_local[0] = 0.000000e+00f;
fo... |
merged_1 | 22 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .f32 %f<149>;
.reg .b32 %r<148>;
.reg .b64 %rd<13>;
// demoted variable... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[16];
__shared__ float input1_shared[4096];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[8] = 0.000000e+00f;
mediate1_local[1] = 0.00000... |
merged_1 | 13 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .f32 %f<21>;
.reg .b32 %r<33>;
.reg .b64 %rd<13>;
// demoted variable
... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[2];
__shared__ float input1_shared[1024];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[1] = 0.000000e+00f;
for (int ax0_ax1_fused_ax2_f... |
merged_1 | 25 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<12>;
.reg .b32 %r<47>;
.reg .b64 %rd<13>;
/... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[1];
__shared__ float input1_shared[512];
mediate1_local[0] = 0.000000e+00f;
*(float4*)(input1_shared + (int)threadIdx.x * 4) = *(float4*)(inpu... |
merged_1 | 32 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<195>;
.reg .b32 %r<96>;
.reg .b64 %rd<17>;
... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[32];
__shared__ float input1_shared[4096];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[1] = 0.000000e+00f;
mediate1_local[2] = 0.00000... |
merged_1 | 34 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<195>;
.reg .b32 %r<110>;
.reg .b64 %rd<17>;
... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[32];
__shared__ float input1_shared[4096];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[1] = 0.000000e+00f;
mediate1_local[2] = 0.00000... |
merged_1 | 35 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<201>;
.reg .b32 %r<85>;
.reg .b64 %rd<17>;
... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[32];
__shared__ float input1_shared[4096];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[8] = 0.000000e+00f;
mediate1_local[16] = 0.0000... |
merged_1 | 0 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<39>;
.reg .b32 %r<60>;
.reg .b64 %rd<13>;
/... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[4];
__shared__ float input1_shared[2048];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[1] = 0.000000e+00f;
mediate1_local[2] = 0.000000... |
merged_1 | 54 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<147>;
.reg .b32 %r<157>;
.reg .b64 %rd<13>;
... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[16];
__shared__ float input1_shared[4096];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[1] = 0.000000e+00f;
mediate1_local[2] = 0.00000... |
merged_1 | 36 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<147>;
.reg .b32 %r<160>;
.reg .b64 %rd<13>;
... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[16];
__shared__ float input1_shared[4096];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[1] = 0.000000e+00f;
mediate1_local[2] = 0.00000... |
merged_1 | 70 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<2>;
.reg .f32 %f<16>;
.reg .b32 %r<31>;
.reg .b64 %rd<13>;
/... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[1];
__shared__ float input1_shared[512];
__shared__ float input0_shared[8];
mediate1_local[0] = 0.000000e+00f;
int4 __1 = make_int4(((int)th... |
merged_1 | 33 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<195>;
.reg .b32 %r<110>;
.reg .b64 %rd<17>;
... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float mediate1_local[32];
__shared__ float input1_shared[4096];
mediate1_local[0] = 0.000000e+00f;
mediate1_local[1] = 0.000000e+00f;
mediate1_local[2] = 0.00000... |
merged_1 | 40 | null | .visible .entry _Z22Broadcast_Add_Sum_4516PfS_S_(
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_0,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_1,
.param .u64 _Z22Broadcast_Add_Sum_4516PfS_S__param_2
)
.maxntid 128, 1, 1
{
.reg .pred %p<3>;
.reg .f32 %f<10>;
.reg .b32 %r<39>;
.reg .b64 %rd<13>;
/... | __global__ void __launch_bounds__(128) Broadcast_Add_Sum_4516(float* __restrict__ input0, float* __restrict__ input1, float* __restrict__ output0) {
float normal_reduce_temp0[1];
__shared__ float input1_shared[256];
float red_buf0[1];
normal_reduce_temp0[0] = 0.000000e+00f;
*(float2*)(input1_shared + ((int)th... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.