Skip to content

[oneDNN] Update of onednn to 2.2 #31067

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

Merged
merged 1 commit into from
Feb 23, 2021
Merged

[oneDNN] Update of onednn to 2.2 #31067

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 1 addition & 1 deletion cmake/external/mkldnn.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ SET(MKLDNN_SOURCE_DIR ${THIRD_PARTY_PATH}/mkldnn/src/extern_mkldnn)
SET(MKLDNN_INSTALL_DIR ${THIRD_PARTY_PATH}/install/mkldnn)
SET(MKLDNN_INC_DIR "${MKLDNN_INSTALL_DIR}/include" CACHE PATH "mkldnn include directory." FORCE)
SET(MKLDNN_REPOSITORY ${GIT_URL}/oneapi-src/oneDNN.git)
SET(MKLDNN_TAG a18f78f1f058437e9efee403655d671633360f98)
SET(MKLDNN_TAG 3d53cd3f17ce7ca365c980f0e1e50359751ca038)

# Introduce variables:
# * CMAKE_INSTALL_LIBDIR
Expand Down
9 changes: 7 additions & 2 deletions paddle/fluid/operators/mkldnn/conv_mkldnn_op.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@

#include "paddle/fluid/framework/data_layout_transform.h"
#include "paddle/fluid/operators/conv_op.h"
#include "paddle/fluid/platform/cpu_info.h"
#include "paddle/fluid/platform/mkldnn_reuse.h"

namespace paddle {
Expand Down Expand Up @@ -976,8 +977,12 @@ class ConvMKLDNNGradOpKernel : public paddle::framework::OpKernel<T> {
* ('any') which lets a primitive (conv backward in this case) choose
* the memory format preferred for best performance
*/

auto chosen_memory_format = MKLDNNMemoryFormat::any;
// TODO: NHWC is preferred starting from oneDNN 2.1 . Any may crash
auto chosen_memory_format =
platform::MayIUse(platform::cpu_isa_t::avx512_core) &&
is_conv3d == false
? MKLDNNMemoryFormat::nhwc
: MKLDNNMemoryFormat::any;
weights_format = MKLDNNMemoryFormat::any;

auto src_md = platform::MKLDNNMemDesc(
Expand Down