Skip to content

Tags: AlbertDachiChen/FBGEMM

Tags

v0.6.0

Toggle v0.6.0's commit message
v0.6.0-release

v0.6.0-rc3

Toggle v0.6.0-rc3's commit message
v0.6.0-release

v0.6.0-rc2

Toggle v0.6.0-rc2's commit message
Clean up redundant AWS credentials (pytorch#2230)

Summary:
After pytorch/test-infra#4839, the credentials are not needed anymore.

Pull Request resolved: pytorch#2230

Reviewed By: spcyppt

Differential Revision: D52297582

Pulled By: huydhn

fbshipit-source-id: 3077901d9b23800c1c36d4fb28dfe0ad07955f73

ciflow/rocm/2229

Toggle ciflow/rocm/2229's commit message
[WIP] Skip JaggedTensorOpsTestt

v0.6.0-rc1

Toggle v0.6.0-rc1's commit message
Revert D51999938: Multisect successfully blamed "D51999938: Add early…

… exit in sparse_async_cumsum ops" for otest failure (pytorch#2208)

Summary:
Pull Request resolved: pytorch#2208

This diff is reverting D51999938
D51999938: Add early exit in sparse_async_cumsum ops by meremeev has been identified to be causing the following test failure:

Tests affected:
- [deeplearning/fbgemm/fbgemm_gpu:sparse_ops_test - test_schema__test_asynchronous_complete_cumsum_2d (deeplearning.fbgemm.fbgemm_gpu.test.sparse_ops_test.SparseOpsTest)](https://www.internalfb.com/intern/test/562950068047718/)

Here's the Multisect link:
https://www.internalfb.com/multisect/3757962
Here are the tasks that are relevant to this breakage:

We're generating a revert to back out the changes in this diff, please note the backout may land if someone accepts it.

If you believe this diff has been generated in error you may Commandeer and Abandon it.

Reviewed By: jasonjk-park

Differential Revision: D52099677

fbshipit-source-id: 4e36745864148e5bb337465b0a9afcfe80846389

v0.6.0-rc0

Toggle v0.6.0-rc0's commit message
Release v0.6.0-rc0

v0.5.0

Toggle v0.5.0's commit message
Release FBGEMM v0.5.0

v0.5.0-rc4

Toggle v0.5.0-rc4's commit message
Fix jagged_test_index_select_2d that hangs in OSS and revert skip tes…

…ts (pytorch#2036)

Summary:
Pull Request resolved: pytorch#2036

Before c++20, `std::atomic_flag` is initialized to an unspecified state, hence the loop `while (lock.test_and_set(std::memory_order_acquire)` is never broken and causes the test to hang in OSS. This diff properly initializes the `std::atomic_flag`.

Reviewed By: q10, sryap

Differential Revision: D49528661

fbshipit-source-id: ba2213cb9bf8c0abbd1e169db03f0e32dd2a7ebb

v0.5.0-rc3

Toggle v0.5.0-rc3's commit message
[FBGEMM][v0.5.0-rc3] Fix using package version and missing bash path

v0.5.0-rc0

Toggle v0.5.0-rc0's commit message
Skip PooledEmbeddingModulesTest until FailedHealthCheck is fixed (pyt…

…orch#1999)

Summary:
Pull Request resolved: pytorch#1999

Hypothesis version 6.83.2 onwards introduces
`HealthCheck.differing_executors` that causes tests
in`permute_pooled_embedding_test.py` to fail with error:

`The method PooledEmbeddingModulesTest.setUp was called from multiple different executors. This may lead to flaky tests and nonreproducible errors when replaying from database`.

Currently, we're using the latest version of hypothesis on CI:

https://github.com/pytorch/FBGEMM/actions/runs/6084855480/job/16515052387

Current hypothesis on FBCode is 6.70.1 which does not have
`HealthCheck.differing_executors`.

Reviewed By: shintaro-iwasaki

Differential Revision: D49020046

fbshipit-source-id: 8ab1350411260c771baf05efe607f91c12df2385