Skip to content

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

New pull request New
29 Open 2,574 Closed
29 Open 2,574 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Try pti 0.12.3 for BMG 580
#4263 opened May 21, 2025 by anmyachev • Draft
[TEST] Fix typo in tensor-pointer-load-block-2d.mlir
#4256 opened May 21, 2025 by whitneywhtsang Loading…
@whitneywhtsang
[DRAFT] Fix CUTLASS benchmark when running on BMG
#4255 opened May 21, 2025 by jle-quel • Draft
1
@jle-quel
Customizable Triton XPU reference for E2E accuracy workload
#4229 opened May 16, 2025 by anmyachev • Draft
Merge OpenAI Triton commit 4b9efc5
#4225 opened May 15, 2025 by whitneywhtsang Loading…
@whitneywhtsang
Update PyTorch inductor test suite
#4224 opened May 15, 2025 by anmyachev • Draft
Do not use extractvalue if the inserted value is directly reachable
#4212 opened May 15, 2025 by AndreyPavlenko • Draft
5
Introduce Subgroup 2D Block Encoding
#4193 opened May 14, 2025 by alexbaden Loading…
1
15
Use inline VISA to optimize horizontal batched subgroup reduce
#4171 opened May 12, 2025 by chengjunlu • Draft
8
Enable SLPVectorization
#4166 opened May 11, 2025 by anmyachev • Draft
[tensor-descriptor]: Extend support when tensor descriptor created in control flow
#4152 opened May 9, 2025 by etiotto Loading…
1
@etiotto
16
Handle op with multi results case in changeAndPropagateLayout
#4146 opened May 9, 2025 by ESI-SYD Loading…
1
4
Prevent 2d block loads with dimensions larger than the tensor block size
#4088 opened May 3, 2025 by alexbaden • Draft
1
13
Use strides when computing load shuffle vector indices in the loadVals map
#4065 opened Apr 30, 2025 by alexbaden • Draft
1
[DEBUG] Disable loop versioning to check compile time
#4043 opened Apr 28, 2025 by anmyachev • Draft
2
New pass Reduce variable liveness
#3965 opened Apr 18, 2025 by mfrancepillois • Draft
1
45
Allow noinline functions to be called with correct argument types. dependencies: compiler
#3963 opened Apr 17, 2025 by etiotto • Draft
@etiotto
11
[CI][benchmarks] Option to run all liger tests
#3917 opened Apr 14, 2025 by Egor-Krivov • Draft
5
[CI] Initial integrate sglang kernels to triton-test
#3901 opened Apr 11, 2025 by LiyangLingIntel • Draft
1
@LiyangLingIntel
2
Add functional test for 2D block lowering of tensor pointers.
#3876 opened Apr 9, 2025 by chengjunlu • Draft
8
[benchmarks][ci] Initial integration of sglang kernels to benchmarks dependencies: igc
#3796 opened Mar 31, 2025 by LiyangLingIntel • Draft
3
@LiyangLingIntel
23
Change 2D block prefetch from OCL to SPV
#3767 opened Mar 26, 2025 by whitneywhtsang • Draft
1
@whitneywhtsang
Change 2D block store from OCL to SPV
#3766 opened Mar 26, 2025 by whitneywhtsang • Draft
1
@whitneywhtsang
2
[SGLANG] add sglang block fp8 gemm kernels into benchmark
#3645 opened Mar 12, 2025 by airMeng Loading…
5 of 7 tasks
1
10
[Performance] Add the support of tensor of pointer in the prefetching and loop pipelining
#3634 opened Mar 10, 2025 by chengjunlu • Draft
2
@whitneywhtsang @etiotto @chengjunlu
22
ProTip! Exclude everything labeled bug with -label:bug.