Skip to content

Pull requests: intel/torch-xpu-ops

New pull request New
163 Open 1,840 Closed
163 Open 1,840 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Add AOT compilation effectiveness check script disable_all Disable all ci test jobs for the PR, just keep basic lint check
#3694 opened May 18, 2026 by chuanqi129 Contributor Loading…
8
Migrate fmod implementation to sycl and improve precision handling
#3693 opened May 18, 2026 by orrangetabby17 Loading…
7
Fix FakeTensorDispatchCache.test_cache_default_device default-device cleanup leak
#3691 opened May 15, 2026 by pponikox Contributor Loading…
1
@pponikox
1
Fix DecorateInfo test-name matching for XPU overrides via register_test disable_accelerate Disable accelerate test job in PR CI testing disable_distributed Disable distributed UT test jobs for the PR disable_e2e Disable all e2e test jobs for the PR disable_win Disable Windows CI test jobs for the PR
#3690 opened May 15, 2026 by gplutop7 Contributor Loading…
12
Align test_streams_xpu.py with upstream version disable_all Disable all ci test jobs for the PR, just keep basic lint check disable_distributed Disable distributed UT test jobs for the PR disable_e2e Disable all e2e test jobs for the PR
#3687 opened May 15, 2026 by kdrozd-dev Contributor • Draft
1
Split subgroup topk kernel per-K for parallel AOT compilation
#3683 opened May 15, 2026 by jianyizh Contributor Loading…
16
test: add XPU regression test for flash attention dynamic shape generalization (#3007)
#3679 opened May 15, 2026 by Copilot AI • Draft
7 tasks done
1
@Stonepia
5
Purge usage of pytorch named tensors
#3678 opened May 14, 2026 by jbschlosser Loading…
1
Replace Euclid's algorithm with optimized std::gcd
#3676 opened May 14, 2026 by Silv3S Contributor Loading…
3
Fix uninitialized index use in pool-related operators. disable_distributed Disable distributed UT test jobs for the PR disable_e2e Disable all e2e test jobs for the PR
#3675 opened May 14, 2026 by BBBela Contributor Loading…
1
@BBBela
3
inf is returned by nn.TransformerEncoderLayer disable_all Disable all ci test jobs for the PR, just keep basic lint check
#3674 opened May 14, 2026 by Stonepia Contributor • Draft
1
2
Fix debugmode warning string parity in test_debugmode_warns_outside_ci disable_distributed Disable distributed UT test jobs for the PR disable_e2e Disable all e2e test jobs for the PR
#3673 opened May 14, 2026 by pponikox Contributor Loading…
1
@pponikox
4
Fix test ordering flakiness in TestExpandedWeightModuleXPU
#3672 opened May 14, 2026 by Copilot AI • Draft
1
@Stonepia
1
Functionality issues in TestCommon.test_out. disable_all Disable all ci test jobs for the PR, just keep basic lint check
#3670 opened May 14, 2026 by Stonepia Contributor • Draft
1
2
fix: use relaxed cast for Half/BFloat16/Float8 in fill_kernel to match CPU semantics
#3669 opened May 14, 2026 by Copilot AI • Draft
5 tasks done
1
@Stonepia
3
Fix XPU FFT float16 support by promoting at dispatch entry points in SpectralOps.cpp disable_all Disable all ci test jobs for the PR, just keep basic lint check
#3668 opened May 14, 2026 by Copilot AI • Draft
3 tasks done
1
@Stonepia
4
test: add repro for test_as_subclass TypeError regression (issue #2518)
#3667 opened May 14, 2026 by Copilot AI • Draft
4 tasks done
1
@Stonepia
3
Fix test_transformerencoderlayer XPU failures: add dtypesIfXPU and remove XPU from fast_path_device disable_all Disable all ci test jobs for the PR, just keep basic lint check
#3666 opened May 14, 2026 by Copilot AI • Draft
1
@Stonepia
Fix INTERNAL ASSERT in BatchNorm stats kernel when out= tensor is non-contiguous
#3665 opened May 14, 2026 by Copilot AI • Draft
8 tasks done
1
@Stonepia
6
Add runtime env to use copy engine on PVC
#3660 opened May 13, 2026 by frost-intel Contributor Loading…
2
Fix test_scaled_mm_deepseek_error_messages for XPU disable_distributed Disable distributed UT test jobs for the PR disable_e2e Disable all e2e test jobs for the PR disable_win Disable Windows CI test jobs for the PR
#3656 opened May 13, 2026 by pbielak Contributor Loading…
4
[SYCLTLA] Support more headdim for FlashAttention fwd/bwd
#3652 opened May 13, 2026 by LuFinch Contributor Loading…
1
9
[sparse][xpu] Fix test_gradcheck_mm error message regex for XPU and remove spurious skipTest
#3651 opened May 12, 2026 by jafraustro Contributor Loading…
1
1
Save stderr in separate file and show in console disable_accelerate Disable accelerate test job in PR CI testing disable_distributed Disable distributed UT test jobs for the PR disable_e2e Disable all e2e test jobs for the PR disable_transformers Disable transformers UT test in PR CI disable_win Disable Windows CI test jobs for the PR disable_windows_ut
#3648 opened May 12, 2026 by jmamzax Contributor • Draft
Fix FakeTensor stride for XPU fft c2r and max_unpool2d
#3646 opened May 12, 2026 by PatrykWilczewski Contributor • Draft
1
3
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.