Skip to content

Pull requests: intel/intel-xpu-backend-for-triton

New pull request New
43 Open 4,627 Closed
43 Open 4,627 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

[LoopDistribute] Add loop distribution pass for multi-dot loops
#7356 opened Jul 4, 2026 by whitneywhtsang Contributor • Draft
[LLVMIR] Fix IGC segfault on sub-byte int bitcast
#7352 opened Jul 3, 2026 by dev-tomek Contributor • Draft
1
Merge OpenAI Triton commit 326ee7f
#7351 opened Jul 3, 2026 by wdziurdz Contributor Loading…
[BLOCKIO] Support VNNI transform 2D block IO when sub-group-size=32.
#7350 opened Jul 3, 2026 by chengjunlu Contributor • Draft
7
Use sub-group-size=32 for fused swiglu gemm kernel which has better performance.
#7349 opened Jul 3, 2026 by chengjunlu Contributor • Draft
2
[vLLM] [spec_decode] Update spec_decode skiplist
#7348 opened Jul 2, 2026 by sarah12121212 Contributor Loading…
2
[vLLM] Add kda tests
#7346 opened Jul 2, 2026 by quinnlp Contributor Loading…
1
[vLLM] Add tests to fill coverage gap
#7344 opened Jul 2, 2026 by quinnlp Contributor Loading…
fix: include libsycl_dir in compile_module_from_src cache key to prevent cross-environment collision
#7338 opened Jul 2, 2026 by ZhaoqiongZ Loading…
1
[vLLM] Increase TRITON_ATTN test coverage for priority kernels
#7331 opened Jun 30, 2026 by quinnlp Contributor • Draft
[vLLM] Enable tensor-descriptor Q load for non-power-of-2 GQA (Qwen2-7B)
#7328 opened Jun 30, 2026 by gurwinderintel Contributor Loading…
Add pass to convert tt.descriptor gather/scatter to ttig ops.
#7326 opened Jun 30, 2026 by chengjunlu Contributor Loading…
9
Bump actions/cache from 5 to 6 dependencies Pull requests that update a dependency file github_actions Pull requests that update GitHub Actions code
#7323 opened Jun 29, 2026 by dependabot Bot Loading…
[Analysis] Add reusable RegisterPressureAnalysis
#7320 opened Jun 29, 2026 by etiotto Contributor Loading…
@etiotto
4
[benchmarks][flash-attention] Fix backward autotuning config constraint
#7317 opened Jun 29, 2026 by gurwinderintel Contributor Loading…
1
[Benchmarks] Integrate vLLM benchmarks with the triton_kernels_benchmark CLI run-benchmarks
#7313 opened Jun 29, 2026 by januszjah Contributor Loading…
1
3
Fix fresh_triton_cache fixture to isolate on-disk cache dir
#7312 opened Jun 28, 2026 by januszjah Contributor Loading…
3 tasks
3
[XPU] Support more SYCL capable compilers building kernel launchers
#7311 opened Jun 26, 2026 by dvrogozh Contributor Loading…
3
Search tensor descriptors even if defining op is tt::CallOp
#7307 opened Jun 25, 2026 by anmyachev Contributor • Draft
[Tests] add unittests to check SYCL arch detection logic in arch_parser.c
#7293 opened Jun 24, 2026 by aliharbaji Contributor Loading…
5 of 7 tasks
1
7
[Compiler] Add StageLargeFMADotsViaSLM: SLM-staged K-tiling on non-DPAS hw
#7291 opened Jun 24, 2026 by gurwinderintel Contributor Loading…
1
4
[vLLM] Add infrastructure to record kernel to test map
#7288 opened Jun 23, 2026 by quinnlp Contributor Loading…
1
[RemoveMasks] Fix unsound IV-equivalent iter_arg classification
#7286 opened Jun 23, 2026 by wdziurdz Contributor • Draft
[Compiler] Add CooperativeFMADots: K-tile FMA dots before LLVM unrolling
#7282 opened Jun 22, 2026 by gurwinderintel Contributor • Draft
3 of 6 tasks
2
Update PyTorch pin run-benchmarks
#7279 opened Jun 22, 2026 by wdziurdz Contributor Loading…
7
ProTip! Filter pull requests by the default branch with base:main.