Skip to content

extra patch for TensorFlow 1.12.0 to remove -B/usr/bin from linker_bin_path_flag in cuda_configure.bzl#7800

Merged
akesandgren merged 3 commits intoeasybuilders:developfrom
FokkeDijkstra:tensorflow-fosscuda-2018b-extra-patch
Apr 2, 2019
Merged

extra patch for TensorFlow 1.12.0 to remove -B/usr/bin from linker_bin_path_flag in cuda_configure.bzl#7800
akesandgren merged 3 commits intoeasybuilders:developfrom
FokkeDijkstra:tensorflow-fosscuda-2018b-extra-patch

Conversation

@FokkeDijkstra
Copy link
Copy Markdown
Contributor

@FokkeDijkstra FokkeDijkstra commented Mar 4, 2019

Added extra patch to remove -B/usr/bin from linker_bin_path_flag in
tensorflow-1.12.0/third_party/gpus/cuda_configure.bzl
This prevents the wrong linker from being used.

@FokkeDijkstra
Added extra patch to remove -B/usr/bin from linker_bin_path_flag in
9f4cec9 
tensorflow-1.12.0/third_party/gpus/cuda_configure.bzl
This prevents the wrong linker from being used.
@FokkeDijkstra
Fixed checksum for for TensorFlow-1.12.0_remove_usrbin_from_linker_bi…
9965699 
…n_path_flag.patch
@boegel
Copy link
Copy Markdown
Member

boegel commented Mar 9, 2019

@FokkeDijkstra How does this problem manifest itself, can you show an example of an error?

@boegel boegel added the bug fix label Mar 9, 2019
@boegel boegel added this to the 3.x milestone Mar 9, 2019
@boegel boegel requested a review from akesandgren March 9, 2019 21:04
@boegel boegel changed the title Extra patch to remove -B/usr/bin from linker_bin_path_flag in cuda_configure.bzl extra patch for TensorFlow 1.12.0 to remove -B/usr/bin from linker_bin_path_flag in cuda_configure.bzl Mar 9, 2019
akesandgren
akesandgren previously approved these changes Mar 9, 2019
View reviewed changes
Copy link
Copy Markdown
Contributor

@akesandgren akesandgren left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, makes sense do do this.

@boegel
Copy link
Copy Markdown
Member

boegel commented Mar 9, 2019

Patch should also be used in TensorFlow-1.12.0-fosscuda-2018b-Python-2.7.15.eb?

@akesandgren
Copy link
Copy Markdown
Contributor

akesandgren commented Mar 10, 2019

Yeah probably. I see no reason why not.

@FokkeDijkstra
Copy link
Copy Markdown
Contributor Author

FokkeDijkstra commented Mar 11, 2019

@FokkeDijkstra How does this problem manifest itself, can you show an example of an error?

I get errors like this when building TensorFlow on our CentOS 7.5 system:

  external/local_config_cuda/crosstool/clang/bin/crosstool_wrapper_driver_is_not_gcc -o bazel-out/k8-opt/bin/external/protobuf_archive/js_embed -Wl,-no-as-needed -pie -Wl,-z,relro,-z,now '-Wl,--build-id=md5' '-Wl,--hash-style=gnu' -no-canonical-prefixes -B/usr/bin -Wl,--gc-sections -Wl,@bazel-out/k8-opt/bin/external/protobuf_archive/js_embed-2.params)
/usr/bin/ld.gold: error: /software/software/GCCcore/7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/crtbeginS.o: unsupported reloc 42 against global symbol _ITM_deregisterTMCloneTable
/usr/bin/ld.gold: error: /software/software/GCCcore/7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/crtbeginS.o: unsupported reloc 42 against global symbol _ITM_registerTMCloneTable
/usr/bin/ld.gold: error: bazel-out/k8-opt/bin/external/protobuf_archive/_objs/js_embed/embed.o: unsupported reloc 42 against global symbol std::ios_base::Init::~Init()
/software/software/GCCcore/7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/crtbeginS.o(.text+0x1a): error: unsupported reloc 42
/software/software/GCCcore/7.3.0/lib/gcc/x86_64-pc-linux-gnu/7.3.0/crtbeginS.o(.text+0x6b): error: unsupported reloc 42
bazel-out/k8-opt/bin/external/protobuf_archive/_objs/js_embed/embed.o:embed.cc:function _GLOBAL__sub_I_main: error: unsupported reloc 42
collect2: error: ld returned 1 exit status

@FokkeDijkstra
Copy link
Copy Markdown
Contributor Author

FokkeDijkstra commented Mar 11, 2019

Patch should also be used in TensorFlow-1.12.0-fosscuda-2018b-Python-2.7.15.eb?

I'll add it to this PR.

@FokkeDijkstra
Added extra patch to remove -B/usr/bin from linker_bin_path_flag in
9bb98b0 
tensorflow-1.12.0/third_party/gpus/cuda_configure.bzl
This prevents the wrong linker from being used.
akesandgren
akesandgren approved these changes Apr 1, 2019
View reviewed changes
Copy link
Copy Markdown
Contributor

@akesandgren akesandgren left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@akesandgren
Copy link
Copy Markdown
Contributor

akesandgren commented Apr 2, 2019

Test report by @akesandgren
SUCCESS
Build succeeded for 6 out of 6 (2 easyconfigs in this PR)
b-an03.hpc2n.umu.se - Linux ubuntu 16.04, Intel(R) Xeon(R) CPU E5-2690 v4 @ 2.60GHz, Python 2.7.12
See https://gist.github.com/5109341fcf1dcd3119c05201948060c6 for a full test report.

@akesandgren
Copy link
Copy Markdown
Contributor

akesandgren commented Apr 2, 2019

Going in, thanks @FokkeDijkstra!

@akesandgren akesandgren merged commit 0947158 into easybuilders:develop Apr 2, 2019
@boegel boegel modified the milestones: 3.x, 3.9.0 Apr 2, 2019
@boegel
Copy link
Copy Markdown
Member

boegel commented Apr 3, 2019

Test report by @boegel
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in this PR)
nic170 - Linux centos linux 7.5.1804, Intel(R) Xeon(R) CPU E5-2683 v4 @ 2.10GHz, Python 2.7.5
See https://gist.github.com/517bac185aa9cb60c6ef0ab2c36d5531 for a full test report.

@Flamefire Flamefire mentioned this pull request May 7, 2020
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@akesandgren akesandgren akesandgren approved these changes

Assignees

No one assigned

Labels

bug fix

Projects

None yet

Milestone

3.9.0

Development

Successfully merging this pull request may close these issues.

3 participants

@FokkeDijkstra @boegel @akesandgren