[xpu][feature] Supports woq_int8 inductor pattern on Intel GPU#163615
[xpu][feature] Supports woq_int8 inductor pattern on Intel GPU#163615xiaowangintel wants to merge 3 commits intopytorch:mainfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/163615
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 4c3d0c5 with merge base 0449e64 ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
| def matcher_check_fn(): | ||
| self.assertEqual( | ||
| counters["inductor"]["woq_matcher_count"], 0 if TEST_ACL else 1 | ||
| devices = ["cpu", "xpu"] |
There was a problem hiding this comment.
Hi, this will fail on device without xpu.
I suggest you move this case to TestPatternMatcherGeneric, which is already designed to be reused by both cpu and xpu.
There was a problem hiding this comment.
Gets it, I have modified.
8f0bc1a to
f14a13f
Compare
|
To add the ciflow label This helps ensure we don't trigger CI on this PR until it is actually authorized to do so. Please ping one of the reviewers if you do not have access to approve and run workflows. |
f14a13f to
3d30a78
Compare
|
@pytorchbot rebase -b main |
|
@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here |
|
Successfully rebased |
3d30a78 to
e2fc389
Compare
|
@pytorchbot rebase -b main |
|
@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here |
|
Successfully rebased |
e2fc389 to
70ce956
Compare
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: 2 jobs have failed, first few of them are: linux-aarch64 / linux-jammy-aarch64-py3.10 / test (default, 3, 3, linux.arm64.m7g.4xlarge), linux-aarch64 / linux-jammy-aarch64-py3.10 / test (default, 3, 3, linux.arm64.m8g.4xlarge) Details for Dev Infra teamRaised by workflow job |
9e5b591 to
957b2d1
Compare
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: 1 jobs have failed, first few of them are: linux-aarch64 / linux-jammy-aarch64-py3.10 / test (default, 3, 3, linux.arm64.m8g.4xlarge) Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Summary: Supports woq_int8 inductor pattern on Intel GPU. When using torch.compile, woq_int8 will be lowering to _weight_int8pack_mm instead of being falled back mul().sum(). The Intel GPU backend of _weight_int8pack_mm was supported in #160938. Pull Request resolved: #163615 Approved by: https://github.com/etaf, https://github.com/EikanWang, https://github.com/desertfire, https://github.com/jansel
…ch#163615) Summary: Supports woq_int8 inductor pattern on Intel GPU. When using torch.compile, woq_int8 will be lowering to _weight_int8pack_mm instead of being falled back mul().sum(). The Intel GPU backend of _weight_int8pack_mm was supported in pytorch#160938. Pull Request resolved: pytorch#163615 Approved by: https://github.com/etaf, https://github.com/EikanWang, https://github.com/desertfire, https://github.com/jansel
Summary:
Supports woq_int8 inductor pattern on Intel GPU. When using torch.compile, woq_int8 will be lowering to _weight_int8pack_mm instead of being falled back mul().sum(). The Intel GPU backend of _weight_int8pack_mm was supported in #160938.
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10 @voznesenskym @penguinwu @EikanWang @Guobing-Chen @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben @jataylo @chenyang78