Search results

[CPU] fix _weight_int8pack_mm with large output shape

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: cpuCPU specific problem (e.g., perf, algorithm)CPU specific problem (e.g., perf, algorithm)

open source

release notes: linalg_frontendrelease notes categoryrelease notes category

Status: Open (in progress).

pytorch/pytorchnumber 158341

#158341 In pytorch/pytorch;

· Xia-Weiwen opened on Jul 15, 2025

<code>torch.compile</code> fails with <code>NotImplementedError: Unsupported for now if query, key, value are the same buffer.</code> in <code>flex_attention</code>

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: cpuCPU specific problem (e.g., perf, algorithm)CPU specific problem (e.g., perf, algorithm)

module: flex attention

module: inductor

module: intelSpecific to x86 architectureSpecific to x86 architecture

module: pt2-dispatcherPT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op,PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op,

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: Open.

#157612 In pytorch/pytorch;

· raindrops-0199 opened on Jul 4, 2025

Segmentation faults with CPU FlexAttention

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: cpuCPU specific problem (e.g., perf, algorithm)CPU specific problem (e.g., perf, algorithm)

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

module: flex attention

module: higher order operatorstorch.cond and similartorch.cond and similar

module: intelSpecific to x86 architectureSpecific to x86 architecture

module: pt2-dispatcherPT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op,PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op,

oncall: cpu inductorCPU Inductor issues for Intel team to triageCPU Inductor issues for Intel team to triage

oncall: pt2

Status: Open.

#153145 In pytorch/pytorch;

· Flamefire opened on May 8, 2025

Enable explicitly vectorized <code>_weight_int8pack_mm</code> op for FP16 dtype on x86_64 CPU

ciflow/trunkTrigger trunk jobs on your pull requestTrigger trunk jobs on your pull request

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: cpuCPU specific problem (e.g., perf, algorithm)CPU specific problem (e.g., perf, algorithm)

open source

release notes: intelrelease notes categoryrelease notes category

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 (in progress).

pytorch/pytorchnumber 146777

#146777 In pytorch/pytorch;

· sanchitintel opened on Feb 9, 2025

Fix <code>USE_STATIC_MKL</code> lost functionality

ciflow/binariesTrigger all binary build and upload jobs on the PRTrigger all binary build and upload jobs on the PR

ciflow/trunkTrigger trunk jobs on your pull requestTrigger trunk jobs on your pull request

ciflow/xpuRun XPU CI tasksRun XPU CI tasks

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: mklRelated to our MKL supportRelated to our MKL support

open source

topic: not user facingtopic categorytopic category

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: Draft (not ready).

pytorch/pytorchnumber 138996

#138996 In pytorch/pytorch;

· xuhancn opened on Oct 26, 2024

Improve oneDNN memory alloction performance for pytorch Windows

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: cpuCPU specific problem (e.g., perf, algorithm)CPU specific problem (e.g., perf, algorithm)

module: performanceIssues related to performance, either of kernel code or framework glueIssues related to performance, either of kernel code or framework glue

module: windowsWindows support for PyTorchWindows support for PyTorch

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.

#125744 In pytorch/pytorch;

· xuhancn opened on May 8, 2024

[RFC] Add new CPP builder for inductor on pytorch Windows

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: inductor

module: windowsWindows support for PyTorchWindows support for PyTorch

oncall: cpu inductorCPU Inductor issues for Intel team to triageCPU Inductor issues for Intel team to triage

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: Open.

#124245 In pytorch/pytorch;

· xuhancn opened on Apr 17, 2024

[RFC] Intel GPU Upstreaming

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: inductor

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: Open.

#114723 In pytorch/pytorch;

· EikanWang opened on Nov 29, 2023

IPEX as TorchDynamo Backend Performance Dashboard

intelThis tag is for PR from IntelThis tag is for PR from Intel

oncall: cpu inductorCPU Inductor issues for Intel team to triageCPU Inductor issues for Intel team to triage

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: Open.

#101273 In pytorch/pytorch;

· DiweiSun opened on May 12, 2023

transposed 2d copy bfloat16 support

intelThis tag is for PR from IntelThis tag is for PR from Intel

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.

#97653 In pytorch/pytorch;

· zhuhaozhe opened on Mar 27, 2023

<code>torch.nn.LazyLinear</code> crash when using torch.bfloat16 dtype in pytorch 1.12.0 and 1.13.0

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: nnRelated to torch.nnRelated to torch.nn

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.

#95108 In pytorch/pytorch;

· ILoveSE opened on Feb 18, 2023

nn.Linear allocate too many space which lead to CPUAllocator "allocate memory failure" if it's BF16. good for FP32.

intelThis tag is for PR from IntelThis tag is for PR from Intel

module: cpuCPU specific problem (e.g., perf, algorithm)CPU specific problem (e.g., perf, algorithm)

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.

#88658 In pytorch/pytorch;

· sywangyi opened on Nov 8, 2022