intel intel-xpu-backend-for-triton issues

intel / intel-xpu-backend-for-triton

OpenAI Triton backend for Intel® GPUs

MIT License

98 stars 28 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

[INTERPRETER][test_tl_range][test_dot_max_num_imprecise_acc] RuntimeError: "addmm_impl_cpu_" not implemented for 'Half'

#1541 AshburnLee opened 7 hours ago
1
add support for attention in match-target-size

#1540 Dewei-Wang-sh opened 8 hours ago
0
Integrate flash attention XeTLA kernel into triton repo

#1539 ESI-SYD opened 8 hours ago
0
[compile-triton.sh] Fail on executing compile-triton.sh

#1538 AshburnLee opened 12 hours ago
0
[Large 2D load for GEMM - #6] Support repCluster in emitOffset/emitIndices of dot layout.

#1537 chengjunlu opened 13 hours ago
0
Add workflow input 'device'

#1536 pbchekin opened 19 hours ago
0
[test_core.py/test_reduce] Fix failure when `op == 'xor_sum'`

#1535 whitneywhtsang opened 21 hours ago
0
[test_core.py::test_store_constant_default_dtype] Fix interpreter UT failure on XPU

#1534 whitneywhtsang opened 21 hours ago
0
Merge OpenAI Triton commit `8e96b71`

#1533 whitneywhtsang closed 21 hours ago
0
[GEN] Fix incorrect `sub_group_reduce` lowering

#1532 whitneywhtsang opened 1 day ago
0
Fix workflows nightly-wheels and coverity

#1531 pbchekin closed 1 day ago
1
[Security][Binaries in the repo] Remove spirv-dis from source control

#1530 vlad-penkin opened 2 days ago
1
[Release] Support build with GCC 9

#1529 vlad-penkin closed 59 minutes ago
0
Build with gcc9

#1528 alexbaden closed 59 minutes ago
5
Remove spirv-dis from source control

#1527 pbchekin opened 2 days ago
0
[REFACTOR] Create facilities to simplify setting function and parameter attribute

#1526 vlad-penkin opened 2 days ago
0
Create facilities to simplify setting function and parameter attributes

#1525 etiotto opened 2 days ago
0
Remove dependency to `libGenISAIntrinsics.a` (#1509)

#1524 whitneywhtsang closed 2 days ago
1
[Release branch][Cherry pick] [GEN] Replace GenISA usages with OCL for sub_group_reduce

#1523 vlad-penkin closed 2 days ago
0
[Releases] Pin numpy<2.0 to make pytorch working in the release branch

#1522 vlad-penkin closed 2 days ago
0
Pin numpy<2.0 to make pytorch working (#1388)

#1521 whitneywhtsang closed 2 days ago
2
[GEN] Replace GenISA usages with OCL for `sub_group_reduce` (#1277)

#1520 whitneywhtsang closed 2 days ago
4
[PyTorch] Update PyTorch and IPEX commit pin to PyTorch 2.3

#1519 LiyangLingIntel closed 2 days ago
2
Update IPEX pin and add cl extension feature query

#1518 LiyangLingIntel closed 3 days ago
2
[UT] rm pytest.skip add within our team and use skiplist instead

#1517 AshburnLee opened 3 days ago
0
[Large 2D load for GEMM - #5] Use the maximum 2D load capacity as possible to load the dot operands.

#1516 chengjunlu opened 3 days ago
0
[Large 2D load for GEMM - #4] Support repetition cluster in 2D store.

#1515 chengjunlu opened 3 days ago
0
[Large 2D load for GEMM - #3] Support the repCluster field in tt.dot operation lowering.

#1514 chengjunlu opened 3 days ago
0
[Large 2D load for GEMM - #2] Support the repCluster field in convert DPAS layout from/to other layouts

#1513 chengjunlu opened 3 days ago
0
[Large 2D load for GEMM - #1] Support repCluster field in converting shared layout to dot layout.

#1512 chengjunlu opened 3 days ago
0
Add workaround to avoid the long time in jitting the SYCL kernel on ATSM.

#1511 chengjunlu closed 3 days ago
0
Cache Nvidia binaries

#1510 pbchekin closed 3 days ago
0
Remove dependency to `libGenISAIntrinsics.a`

#1509 whitneywhtsang closed 3 days ago
0
Fix coverity reported issues

#1508 etiotto closed 3 days ago
0
Pass information from `ext_oneapi_supports_cl_extension` to Triton

#1507 whitneywhtsang closed 3 days ago
0
[TritonGEN]: Add operation for subgroup_scan_[ex|in]clusive

#1506 etiotto closed 2 days ago
0
[PyTorch Upstream] Triton xpu manylinux build failed in Pytorch XPU CD enabling

#1505 chuanqi129 opened 3 days ago
14
[NFC]: Address post-review comments for PR #1492

#1504 etiotto closed 3 days ago
0
[RAISE-BP] Add support for tt.broadcast increasing tensor rank

#1503 mfrancepillois opened 4 days ago
2
Fix compilation errors with clang 17.0.6

#1502 whitneywhtsang closed 3 days ago
0
Compilation errors with clang 17.0.6

#1501 pbchekin closed 3 days ago
0
Merge OpenAI Triton commit `948a3e8`

#1500 whitneywhtsang closed 4 days ago
0
Enable fp8 2d block read with fp16 DPAS format

#1499 hwnam831 opened 4 days ago
0
[TEST] Run `test_fp8_dot_acc` with subgroup size 16

#1498 whitneywhtsang closed 4 days ago
0
Do not build LLVM in CI by default

#1497 pbchekin closed 4 days ago
0
03-tutorial fails with upstream pytorch

#1496 ZzEeKkAa opened 4 days ago
0
Reland [TUTORIAL] persistent softmax kernel

#1495 victor-eds opened 5 days ago
3
[Productize GEMM #0] add batch support for gemm

#1494 vlad-penkin closed 5 days ago
1
Add batch support for gemm

#1493 LiyangLingIntel closed 3 days ago
1
[Large 2D load for GEMM - #0] Add repCluster field to the DPAS layout.

#1492 chengjunlu closed 4 days ago
2