marcoabreu commented on a change in pull request #10731: Fix a bug in getting MKLDNN memory
URL: https://github.com/apache/incubator-mxnet/pull/10731#discussion_r184919852
##########
File path: ci/docker/runtime_functions.sh
##########
@@ -313,16 +313,22 @@ build_ubuntu_amalgamation_min() {
build_ubuntu_gpu_cmake_mkldnn() {
set -ex
cd /work/build
+ # We need to use generic archtecture. Otherwise, MKLDNN compiled in one
+ # CPU architecture (e.g., C5) can't run on another architecture (e.g., g3).
cmake \
-DUSE_CUDA=1 \
-DUSE_CUDNN=1 \
-DUSE_MKLML_MKL=1 \
-DUSE_MKLDNN=1 \
-DCMAKE_BUILD_TYPE=Release \
+ -DARCH_OPT_FLAGS="-mtune=generic" \
-G Ninja \
/work/mxnet
ninja -v
+ # libmkldnn.so.0 is a link file. We need an actual binary file named libmkldnn.so.0.
+ cp 3rdparty/mkldnn/src/libmkldnn.so.0 3rdparty/mkldnn/src/libmkldnn.so.0.tmp
+ mv 3rdparty/mkldnn/src/libmkldnn.so.0.tmp 3rdparty/mkldnn/src/libmkldnn.so.0
Review comment:
This is a no-op. Please elaborate
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
|