-
Notifications
You must be signed in to change notification settings - Fork 21.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add warpSize to Device properties #128449
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/128449
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (4 Unrelated Failures)As of commit 1e68689 with merge base c12a4f2 (): FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
@pytorchbot rebase |
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
Successfully rebased |
36071ef
to
5b77544
Compare
@pytorchbot merge |
Merge failedReason: This PR needs a If not, please add the To add a label, you can comment to pytorchbot, for example For more information, see 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 |
The merge job was canceled or timed out. This most often happen if two merge requests were issued for the same PR, or if merge job was waiting for more than 6 hours for tests to finish. In later case, please do not hesitate to reissue the merge command |
@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 |
Adding warp_size to CudaDeviceProperties. >>> import torch >>> prop = torch.cuda.get_device_properties(torch.cuda.current_device()) >>> prop.warp_size 64 >>> @jeffdaily @pruthvistony @jithunnair-amd @ROCmSupport Co-authored-by: Jithun Nair <[email protected]> Pull Request resolved: pytorch#128449 Approved by: https://github.com/eqy, https://github.com/jataylo, https://github.com/jithunnair-amd, https://github.com/malfet
…9663) As of ROCm 6.1 [hipDeviceProp_t::regsPerMultiprocessor](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/structhip_device_prop__t.html#a7390d5b180d63978c81aa971060270b4) is now available allowing us to enable this attribute on ROCm. ``` >>> torch.cuda.get_device_properties(0) _CudaDeviceProperties(name='AMD Instinct MI250X/MI250', major=9, minor=0, gcnArchName='gfx90a:sramecc+:xnack-', total_memory=65520MB, multi_processor_count=104) >>> torch.cuda.get_device_properties(0).regs_per_multiprocessor 65536 ``` With https://github.com/triton-lang/triton/pull/3962we can extract n_regs and n_spells from a triton binary with AMD backend allowing us to enable inductor's dynamic_rblock_scaling on ROCm initially implemented in #115094 Leaving this in draft until following PRs have landed: - #129361 to bump the triton commit pin - #128449 to allow us to grab warp_size from device properties instead of hard coding 64 on ROCm. Pull Request resolved: #129663 Approved by: https://github.com/jansel, https://github.com/shunting314
…orch#129663) As of ROCm 6.1 [hipDeviceProp_t::regsPerMultiprocessor](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/structhip_device_prop__t.html#a7390d5b180d63978c81aa971060270b4) is now available allowing us to enable this attribute on ROCm. ``` >>> torch.cuda.get_device_properties(0) _CudaDeviceProperties(name='AMD Instinct MI250X/MI250', major=9, minor=0, gcnArchName='gfx90a:sramecc+:xnack-', total_memory=65520MB, multi_processor_count=104) >>> torch.cuda.get_device_properties(0).regs_per_multiprocessor 65536 ``` With https://github.com/triton-lang/triton/pull/3962we can extract n_regs and n_spells from a triton binary with AMD backend allowing us to enable inductor's dynamic_rblock_scaling on ROCm initially implemented in pytorch#115094 Leaving this in draft until following PRs have landed: - pytorch#129361 to bump the triton commit pin - pytorch#128449 to allow us to grab warp_size from device properties instead of hard coding 64 on ROCm. Pull Request resolved: pytorch#129663 Approved by: https://github.com/jansel, https://github.com/shunting314
…orch#129663) As of ROCm 6.1 [hipDeviceProp_t::regsPerMultiprocessor](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/structhip_device_prop__t.html#a7390d5b180d63978c81aa971060270b4) is now available allowing us to enable this attribute on ROCm. ``` >>> torch.cuda.get_device_properties(0) _CudaDeviceProperties(name='AMD Instinct MI250X/MI250', major=9, minor=0, gcnArchName='gfx90a:sramecc+:xnack-', total_memory=65520MB, multi_processor_count=104) >>> torch.cuda.get_device_properties(0).regs_per_multiprocessor 65536 ``` With https://github.com/triton-lang/triton/pull/3962we can extract n_regs and n_spells from a triton binary with AMD backend allowing us to enable inductor's dynamic_rblock_scaling on ROCm initially implemented in pytorch#115094 Leaving this in draft until following PRs have landed: - pytorch#129361 to bump the triton commit pin - pytorch#128449 to allow us to grab warp_size from device properties instead of hard coding 64 on ROCm. Pull Request resolved: pytorch#129663 Approved by: https://github.com/jansel, https://github.com/shunting314
Adding warp_size to CudaDeviceProperties.
@jeffdaily @pruthvistony @jithunnair-amd @ROCmSupport
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang