Ifu 2022 07 28 #12
Merged
Ifu 2022 07 28 #12
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ytes aligned (pytorch#1212) Summary: Pull Request resolved: pytorch#1212 As title Reviewed By: divchenko Differential Revision: D37951520 fbshipit-source-id: a2c2ab57bb13ed750e986b0326566a8f0a8ea3ae
Summary: This would help us benchmark EmbeddingCollection in torchrec. Thank you for your time in reviewing this PR :) Pull Request resolved: pytorch#1194 Reviewed By: jianyuh Differential Revision: D37845758 Pulled By: colin2328 fbshipit-source-id: a65ef76b195ca2cfd56b69d39e2d59ae930edfae
Summary: Pull Request resolved: pytorch#1211 As title. Reviewed By: jianyuh Differential Revision: D37877585 fbshipit-source-id: 8063c05c5ed03b3c9c76c3a20bfb41db332a6d19
…ise_add_jagged_output() (pytorch#1213) Summary: Pull Request resolved: pytorch#1213 Reviewed By: mjanderson09 Differential Revision: D37446339 fbshipit-source-id: 88486749716d87fe618ba1f2366fb5dbe3b6691f
Reviewed By: zsol Differential Revision: D38069933 fbshipit-source-id: 006a059e4728595df0a69df2890b38d4331b05e6
Summary: Pull Request resolved: pytorch#1214 as title. Reviewed By: jiaqizhai, brad-mengchi, mjanderson09 Differential Revision: D37978119 fbshipit-source-id: c2c004dfb2e1483f6fbf6a415c9dd58d95599cb0
Summary: Pull Request resolved: pytorch#1204 This will be better shared between Trec and HPC. - We will refactor https://www.internalfb.com/code/fbsource/[history]/fbcode/caffe2/torch/fb/hpc/quantized_comms_lib.py to extract the common components in FBGEMM - It's open source so TorchRec can call it from FBGEMM. - Reuse the quantize utils functions and dedup the code. This part of code is landable. Reviewed By: YLGH Differential Revision: D37799807 fbshipit-source-id: ced3c98efd096985db02c287449fa48939fd3da3
Summary: Pull Request resolved: pytorch#1219 This provides a generic codec interface that can be used for quantized comms. Reviewed By: colin2328, jianyuh Differential Revision: D38125284 fbshipit-source-id: 12a3a1e05e2893e1931e4782c47a75aab9840e52
Summary: Pull Request resolved: pytorch#1220 This patch fixes a failure of `build_amd_gpu` on GitHub CI. ## Background We start to encounter the following error on `build_amd_gpu` recently (around July 11th). ``` Building initial module for 5.15.0-1014-azure Error! Bad return status for module build on kernel: 5.15.0-1014-azure (x86_64) Consult /var/lib/dkms/amdgpu/5.13.20.22.10-1401700/build/make.log for more information. dpkg: error processing package amdgpu-dkms (--configure): installed amdgpu-dkms package post-installation script subprocess returned error exit status 10 ``` ## Solution I did not perform any deep investigation, but basically, we don't need dkms for this test. This patch disables it. Reviewed By: sryap Differential Revision: D38133302 fbshipit-source-id: ade20dcec658ac186d8f8ad1b713d339e504ba6f
Summary: Pull Request resolved: pytorch#1215 This is a follow-up on D37951520 (pytorch@5a15342) - Minor clean-up and refactoring for non-contiguous grad output: avoid do .contiguous and clone twice when alignment is not multiple of 16. - Add more comments. - Add unit test coverage TODO: add the 16 alignment unit test coverage. Reviewed By: brad-mengchi Differential Revision: D37988742 fbshipit-source-id: ce8b7f9658c9e6229d67b420212bb3d55daea8c8
Summary: Pull Request resolved: pytorch#1223 useful if some kernel warmup is desired Reviewed By: sryap Differential Revision: D38141801 fbshipit-source-id: b7efbad77040fa7a5216a14114a975932163ed14
Summary: Pull Request resolved: pytorch#1218 Add support for Int types in jagged_index_select and update the test to validate Int types Reviewed By: jianyuh Differential Revision: D38105237 fbshipit-source-id: c9392da8378759bda384abd768a44f0aef5ad1b1
Summary: Pull Request resolved: pytorch#1202 Update OSS asmjit version to d3fbf7c9bc7c1d1365a94a45614b91c5a3706b81, sync with the third-party/asmjit. Add a -Wno-unused-variable compilation flag to enable Clang 14. Reviewed By: brad-mengchi Differential Revision: D37763232 fbshipit-source-id: 650ee2ba8601dc0922640507f28691a87d275db3
…antizedSBFloatToFloatOrHalfAvx2" (pytorch#1222) Summary: Pull Request resolved: pytorch#1222 Reviewed By: Liang-Dong Differential Revision: D38155592 fbshipit-source-id: 025fc4445029a104fe38e76ad317a2e0dfb6ede7
liligwu
pushed a commit
that referenced
this pull request
Feb 8, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No conflict. All tests pass.
20220728_IFU_tests.log.txt