[1/N]Port 3 distributed/_tools test cases to Intel GPU #159543
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/159543
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 63cf20f with merge base fc80f68 ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
|
|
||
| @skipIfTorchDynamo("https://github.com/pytorch/pytorch/issues/115653") | ||
| @unittest.skipIf(not TEST_CUDA, "CUDA not available") | ||
| @unittest.skipIf(not torch.accelerator.is_available(), "Accelerator not available") |
There was a problem hiding this comment.
@unittest.skipIf(not TEST_CUDA and not TEST_XPU, "Neither CUDA or XPU is not available")
| @@ -77,17 +77,18 @@ def _test_tracker_multi_group( | |||
| mp_policy: MixedPrecisionPolicy, | |||
| ): | |||
| debug = False | |||
| dev = torch.device(torch.cuda.current_device()) | |||
| dev = torch.device(torch.accelerator.current_device_index()) | |||
There was a problem hiding this comment.
torch.acceleartor does not apply to cpu.
There was a problem hiding this comment.
But I think the cpu has been skipped by @skip_if_lt_x_gpu(2).
|
@pytorchbot rebase |
|
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
We could enable Intel GPU with following methods and try the best to keep the original code styles: 1. use "torch.accelerator.current_accelerator()" to determine the accelerator backend 2. enabled XPU for some test path 3. skip some test cases which Intel GPU does not support
|
Successfully rebased |
70e2b22 to
63cf20f
Compare
|
@libohao1201 please help fix the lint error. |
For #114850, we will port distributed tests to Intel GPU.
We could enable Intel GPU with following methods and try the best to keep the original code styles:
cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta