Search results

IInconsistent Error Handling in <code>torch.fused_moving_avg_obs_fake_quant</code> Between CPU and GPU Implementations

Status: Open.

#153310 In pytorch/pytorch;

· SilentTester73 opened on May 10, 2025

reshape_view_helper too complex to be used for fake tensor tracing but not proxy tracing !

module: autogradRelated to torch.autograd, and the autograd engine in generalRelated to torch.autograd, and the autograd engine in general

Status: Open.

#153303 In pytorch/pytorch;

· laithsakka opened on May 10, 2025

[FSDP2] avoid GPU OOM for reshard_after_forward=int with shared post_forward_mesh

Status: Open.

#153302 In pytorch/pytorch;

· weifengpy opened on May 10, 2025

Accuracy issue in torch inductor

oncall: pt2

Status: Open.

#153299 In pytorch/pytorch;

· naveenkumarmarri opened on May 10, 2025

Cleanup autotune_fallback_to_aten post-deprecation

actionable

good first issue

module: inductor

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

Task

#153298 In pytorch/pytorch;

· henrylhtsang opened on May 10, 2025

[c10d] ProcessGroupNCCL cuda streams got merged in nightly

oncall: distributedAdd this issue/PR to distributed oncall triage queueAdd this issue/PR to distributed oncall triage queue

Status: Open.

#153296 In pytorch/pytorch;

· weifengpy opened on May 9, 2025

DISABLED test_parity__foreach_add_fastpath_inplace_cuda_float32 (__main__.TestForeachCUDA)

module: flaky-testsProblem is a flaky test in CIProblem is a flaky test in CI

module: unknownWe do not know who is responsible for this feature, bug, or test case.We do not know who is responsible for this feature, bug, or test case.

skippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.

Status: Open.

#153284 In pytorch/pytorch;

· pytorch-bot opened on May 9, 2025

<code>torch.nn.functional.rrelu</code> crashes on CPU with <code>training=True</code> when <code>lower</code> or <code>upper</code> is set to <code>inf</code>

Status: Open.

#153281 In pytorch/pytorch;

· jiren-the-gray opened on May 9, 2025

torch.nn.functional.conv_transpose2d produces inconsistent output on CPU and CUDA

Status: Open.

#153276 In pytorch/pytorch;

· jiren-the-gray opened on May 9, 2025

Auto format lint not making suggestions

module: devxRelated to PyTorch contribution experience (HUD, pytorchbot)Related to PyTorch contribution experience (HUD, pytorchbot)

Status: Open.

#153273 In pytorch/pytorch;

· clee2000 opened on May 9, 2025

DISABLED test_ddp_apply_optim_in_backward (__main__.TestDistBackendWithSpawn)

module: rocmAMD GPU support for PytorchAMD GPU support for Pytorch

skippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.

Status: Open.

#153266 In pytorch/pytorch;

· jithunnair-amd opened on May 9, 2025

[RFC][API-Unstable] Support 3rd party SYCL kernels with CPP Extension API

Status: Open.

#153265 In pytorch/pytorch;

· dvrogozh opened on May 9, 2025