Search results

Torch 2.6.0 rc is broken on Amazon linux 2023

high priority

module: binariesAnything related to official binaries that we release to usersAnything related to official binaries that we release to users

triage review

Status: Closed (completed).

#144433 In pytorch/pytorch;

· by atalman was closed on Jan 9, 2025

· 

2.6.0

RuntimeError when running backward on MPS: "view size is not compatible" with self-attention block

high priority

module: convolutionProblems related to convolutions (THNN, THCUNN, CuDNN)Problems related to convolutions (THNN, THCUNN, CuDNN)

module: mpsRelated to Apple Metal Performance Shaders frameworkRelated to Apple Metal Performance Shaders framework

module: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn't

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: Closed (completed).

#142344 In pytorch/pytorch;

· by NripeshN was closed on Dec 13, 2024

· 

2.6.0

MacOS wheels are no longer built with OpenMP runtime

high priority

module: binariesAnything related to official binaries that we release to usersAnything related to official binaries that we release to users

module: macosMac OS related issuesMac OS related issues

module: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn't

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: Closed (completed).

#142266 In pytorch/pytorch;

· by malfet was closed on Dec 17, 2024

· 

2.6.0

Output size of the matrix multiplication is larger than currently supported by the MPS backend: 72250,72250, needs to be less than 2**32 elements

high priority

module: mpsRelated to Apple Metal Performance Shaders frameworkRelated to Apple Metal Performance Shaders framework

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: Closed (completed).

#141909 In pytorch/pytorch;

· by gazonk was closed on Dec 17, 2024

· 

2.6.0

MPS Regression when rendering LTXVideo (after pytorch2.4.1)

high priority

module: correctness (silent)issue that returns an incorrect result silentlyissue that returns an incorrect result silently

module: mpsRelated to Apple Metal Performance Shaders frameworkRelated to Apple Metal Performance Shaders framework

module: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn't

triage review

Status: Closed (completed).

#141471 In pytorch/pytorch;

· by cchance27 was closed on Dec 1, 2024

· 

2.6.0

Potential BC regression in nightly: <code>torch.compile(..., fullgraph=True)</code> does not work with <code>torch.device</code> TorchFunctionMode

high priority

module: dynamo

oncall: pt2

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: Closed (completed).

#141046 In pytorch/pytorch;

· by kiukchung was closed on Dec 11, 2024

· 

2.6.0

PyTorch &gt;= 2.5.0 revive #111469 (python3.10/site-packages/torch/lib/../../nvidia/cusparse/lib/libcusparse.so.12: undefined symbol: __nvJitLinkComplete_12_4, version libnvJitLink.so.12)

high priority

module: binariesAnything related to official binaries that we release to usersAnything related to official binaries that we release to users

module: cudaRelated to torch.cuda, and CUDA support in generalRelated to torch.cuda, and CUDA support in general

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: Closed (completed).

#140797 In pytorch/pytorch;

· by mbertrait was closed on Jan 14, 2025

· 

2.6.0

[Performance] [CuDNN-Attention] CuDNN backend should return the output in the same stride order as input Query

high priority

module: cudnnRelated to torch.backends.cudnn, and CuDNN supportRelated to torch.backends.cudnn, and CuDNN support

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: Closed (completed).

#138340 In pytorch/pytorch;

· by drisspg was closed on Nov 16, 2024

· 

2.6.0

[binary] cuSPARSELt increases manywheel binary size signifantly

high priority

module: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn't

oncall: relengIn support of CI and Release EngineeringIn support of CI and Release Engineering

topic: binaries

triage review

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: Closed (completed).

#137374 In pytorch/pytorch;

· by atalman was closed on Nov 20, 2024

· 

2.6.0

torch.compile errors when tracing numpy.random.uniform with numpy2

high priority

module: numpyRelated to numpy support, and also numpy compatibility of our operatorsRelated to numpy support, and also numpy compatibility of our operators

oncall: pt2

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: Closed (completed).

#136559 In pytorch/pytorch;

· by kiukchung was closed on Nov 1, 2024

· 

2.6.0

<code>import torch</code> fails on stock ubuntu if one uses XPU nightly binaries

high priority

module: crashProblem manifests as a hard crash, as opposed to a RuntimeErrorProblem manifests as a hard crash, as opposed to a RuntimeError

module: xpuIntel XPU related issuesIntel XPU related issues

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: Closed (completed).

#135867 In pytorch/pytorch;

· by malfet was closed on Nov 29, 2024

· 

2.6.0

On Kaggle : libcusparse.so.12: undefined symbol: __nvJitLinkComplete_12_4, version libnvJitLink.so.12

high priority

module: binariesAnything related to official binaries that we release to usersAnything related to official binaries that we release to users

module: cudaRelated to torch.cuda, and CUDA support in generalRelated to torch.cuda, and CUDA support in general

module: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn't

module: sparseRelated to torch.sparseRelated to torch.sparse

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: Closed (completed).

#134929 In pytorch/pytorch;

· by FurkanGozukara was closed on Feb 3, 2025

· 

2.6.0