Make sure FindLAPACK finds the same BLAS library

[isuruf]

BLAS library is found by cmake/Dependencies.cmake and then
LAPACK library is found by FindLAPACK.cmake which in turn calls
FindBLAS.cmake. This means that we are searching for BLAS twice
and they might be different things. By setting a few variables,
this can be avoided.

cc @seemethere

[facebook-github-bot]

🔗 Helpful links

• 🧪 See artifacts and rendered test results at hud.pytorch.org/pr/49647
• 📄 Preview docs built from this PR

💊 CI failures summary and remediations

As of commit 57250c8 (more details on the Dr. CI page):

---

• 2/2 failures introduced in this PR

---

🕵️ 2 new failures recognized by patterns

The following CI failures do not appear to be due to upstream breakages:

pytorch_xla_linux_bionic_py3_6_clang9_test (1/2)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Jul 28 07:48:33 2021-07-28 07:48:33.426355: E t...w/compiler/xla/service/slow_operation_alarm.cc:55]

Jul 28 07:36:45   test_AvgPool3d_backward_after_cat_dim1_device_xla (__main__.TestNNDeviceTypeXLA) ... skip (0.002s)
Jul 28 07:36:45   test_BatchNorm_empty_xla (__main__.TestNNDeviceTypeXLA) ... ok (0.142s)
Jul 28 07:36:45   test_Bilinear_empty_xla (__main__.TestNNDeviceTypeXLA) ... skip (0.002s)
Jul 28 07:36:45   test_CTCLoss_cudnn_xla (__main__.TestNNDeviceTypeXLA) ... skip (0.002s)
Jul 28 07:36:46   test_CTCLoss_empty_target_xla (__main__.TestNNDeviceTypeXLA) ... ok (0.842s)
Jul 28 07:38:56   test_Conv2d_backward_depthwise_xla_float64 (__main__.TestNNDeviceTypeXLA) ... 2021-07-28 07:38:56.357716: E tensorflow/compiler/xla/service/slow_operation_alarm.cc:55] 
Jul 28 07:38:56 ********************************
Jul 28 07:38:56 Very slow compile?  If you want to file a bug, run with envvar XLA_FLAGS=--xla_dump_to=/tmp/foo and attach the results.
Jul 28 07:38:56 Compiling module SyncTensorsGraph.30789
Jul 28 07:38:56 ********************************
Jul 28 07:48:33 2021-07-28 07:48:33.426355: E tensorflow/compiler/xla/service/slow_operation_alarm.cc:55] 
Jul 28 07:48:33 ********************************
Jul 28 07:48:33 Very slow compile?  If you want to file a bug, run with envvar XLA_FLAGS=--xla_dump_to=/tmp/foo and attach the results.
Jul 28 07:48:33 Compiling module SyncTensorsGraph.35441
Jul 28 07:48:33 ********************************


Too long with no output (exceeded 1h30m0s): context deadline exceeded

Linux CI (pytorch-linux-bionic-py3.8-gcc9-coverage) / render_test_results (default) (2/2)

Step: "Chown workspace" (full log | diagnosis details | 🔁 rerun)

2021-07-31T19:20:59.9842153Z docker: Error resp...pine/manifests/latest": no basic auth credentials.

2021-07-31T19:20:59.9049547Z env:
2021-07-31T19:20:59.9050222Z   BUILD_ENVIRONMENT: pytorch-linux-bionic-py3.8-gcc9-coverage
2021-07-31T19:20:59.9051306Z   DOCKER_IMAGE_BASE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-bionic-py3.8-gcc9
2021-07-31T19:20:59.9052302Z   SCCACHE_BUCKET: ossci-compiler-cache-circleci-v2
2021-07-31T19:20:59.9052899Z   TORCH_CUDA_ARCH_LIST: 5.2
2021-07-31T19:20:59.9053385Z   IN_CI: 1
2021-07-31T19:20:59.9053861Z   CUSTOM_TEST_ARTIFACT_BUILD_DIR: build/custom_test_artifacts
2021-07-31T19:20:59.9054566Z   ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine
2021-07-31T19:20:59.9055140Z ##[endgroup]
2021-07-31T19:20:59.9403631Z Unable to find image '308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine:latest' locally
2021-07-31T19:20:59.9842153Z docker: Error response from daemon: Head "https://308535385114.dkr.ecr.us-east-1.amazonaws.com/v2/tool/alpine/manifests/latest": no basic auth credentials.
2021-07-31T19:20:59.9843421Z See 'docker run --help'.
2021-07-31T19:20:59.9869272Z ##[error]Process completed with exit code 125.
2021-07-31T19:21:00.0007460Z Cleaning up orphan processes

---

1 job timed out:

• pytorch_xla_linux_bionic_py3_6_clang9_test

---

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

[codecov]

Codecov Report

Merging #49647 (d42edce) into master (c0deb23) will decrease coverage by 0.00%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master   #49647      +/-   ##
==========================================
- Coverage   41.55%   41.54%   -0.01%     
==========================================
  Files         503      503              
  Lines       68006    68006              
==========================================
- Hits        28258    28255       -3     
- Misses      39748    39751       +3     

[isuruf]

It would be good to merge this for the conda-forge build.

cc @rgommers

[rgommers]

Thanks @isuruf, this looks good to me.

@malfet would you be able to land this?

[malfet]

LGTM, although it misses similar checks for Atlas and Eigen.

[facebook-github-bot]

@malfet has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

[isuruf]

LGTM, although it misses similar checks for Atlas and Eigen.

I added support for atlas. Eigen is not supported by cmake/Modules/FindBLAS.cmake

[rgommers]

This seems ready to land. The three CI failures are all unrelated (1 a connectivity issue, 2 timeouts)

[facebook-github-bot]

@malfet has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

[facebook-github-bot]

@malfet merged this pull request in b58e04f.