Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

softmax fwd: force vec size to 1 when dtype is float #54183

Merged
merged 3 commits into from
May 30, 2023
Merged

softmax fwd: force vec size to 1 when dtype is float #54183

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
6ac477d
softmax fwd: force vec size to 1 when dtype is float
shaojiewang May 29, 2023
cacecf3
use 1024 as threshold to use cudnn
shaojiewang May 29, 2023
0cb4e81
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
shaojiewang May 29, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
58 changes: 36 additions & 22 deletions paddle/phi/kernels/gpudnn/softmax_gpudnn.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -798,6 +798,7 @@ void SwitchWarpSoftmaxForward(const IndexType blocks,
SOFTMAX_WARP_FORWARD_CASE(7, AccT);
SOFTMAX_WARP_FORWARD_CASE(8, AccT);
SOFTMAX_WARP_FORWARD_CASE(9, AccT);
SOFTMAX_WARP_FORWARD_CASE(10, AccT);
default:
break;
}
Expand Down Expand Up @@ -836,6 +837,7 @@ void SwitchWarpSoftmaxBackward(const int blocks,
SOFTMAX_WARP_BACKWARD_CASE(7, AccT);
SOFTMAX_WARP_BACKWARD_CASE(8, AccT);
SOFTMAX_WARP_BACKWARD_CASE(9, AccT);
SOFTMAX_WARP_BACKWARD_CASE(10, AccT);
default:
break;
}
Expand Down Expand Up @@ -1262,7 +1264,7 @@ bool UseCudnnSoftmax(const GPUContext& ctx,
#endif
}
}
constexpr int max_dim = 512;
constexpr int max_dim = 1024;
if (!cudnn_available || !last_dim ||
(softmax_dim <= max_dim && sizeof(T) <= 4) ||
softmax_dim >= MATRIX_SOFTMAX_THREAHOLD) {
Expand Down Expand Up @@ -1311,27 +1313,7 @@ void SoftmaxForwardCUDAKernelDriverImpl(const GPUContext& dev_ctx,
using T4 = typename VecT4<T>::Type;
using T2 = typename VecT2<T>::Type;

if (dim % 4 == 0) {
SwitchWarpSoftmaxForward<T, T4, IndexType, LogMode>(blocks,
threads,
dev_ctx,
out_data,
x.data<T>(),
N,
dim,
dim,
dim_log2);
} else if (dim % 2 == 0) {
SwitchWarpSoftmaxForward<T, T2, IndexType, LogMode>(blocks,
threads,
dev_ctx,
out_data,
x.data<T>(),
N,
dim,
dim,
dim_log2);
} else {
if (std::is_same<T, float>::value) {
SwitchWarpSoftmaxForward<T, T, IndexType, LogMode>(blocks,
threads,
dev_ctx,
Expand All @@ -1341,6 +1323,38 @@ void SoftmaxForwardCUDAKernelDriverImpl(const GPUContext& dev_ctx,
dim,
dim,
dim_log2);
} else {
if (dim % 4 == 0) {
SwitchWarpSoftmaxForward<T, T4, IndexType, LogMode>(blocks,
threads,
dev_ctx,
out_data,
x.data<T>(),
N,
dim,
dim,
dim_log2);
} else if (dim % 2 == 0) {
SwitchWarpSoftmaxForward<T, T2, IndexType, LogMode>(blocks,
threads,
dev_ctx,
out_data,
x.data<T>(),
N,
dim,
dim,
dim_log2);
} else {
SwitchWarpSoftmaxForward<T, T, IndexType, LogMode>(blocks,
threads,
dev_ctx,
out_data,
x.data<T>(),
N,
dim,
dim,
dim_log2);
}
}
} else {
LaunchSoftmaxForwardCudnnKernel<T>(dev_ctx, x, axis, LogMode, out);
Expand Down