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

17 Open 781 Closed

Author

Filter by author

author:

Filter by this user

Label

Filter by label

Projects

Filter by project

Milestones

Filter by milestone

Reviews

Filter by reviews

No reviews Review required Approved review Changes requested

Assignee

Filter by who’s assigned

Assigned to nobody

assignee:

Filter by this user

Sort

Sort by

Newest Oldest Most commented Least commented Recently updated Least recently updated Best match

Most reactions

👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Enable upstream pytorch (no ipex) for e2e accuracy workflow

#1279 opened Jun 7, 2024 by gshimansky Loading…

5 of 8 tasks

[TMP]: Disable layout rematerialization cache

#1275 opened Jun 6, 2024 by alexbaden Loading…

Improve robustness of MMA layout propagation to tt.store with block pointer

#1272 opened Jun 6, 2024 by mfrancepillois Loading…

1

[Draft] Enable the large 2D load for dot operands

#1263 opened Jun 6, 2024 by chengjunlu • Draft

2

Post LLVM codegen loop invariant code motion

#1261 opened Jun 5, 2024 by etiotto Loading…

1

[GEN] Extend Matrix2DBlockLoadOp OCL usages

#1259 opened Jun 5, 2024 by whitneywhtsang • Draft

1

5

Separate accumulator and result types

#1258 opened Jun 5, 2024 by jopperm Loading…

1

6

[GEN] Add cache controls decoration to 2d_block_read builtin calls

#1247 opened Jun 4, 2024 by victor-eds Loading… 4.3 [Performance] Tracking

1

7

[GEN] Change existing OCL lowering to the new interface

#1224 opened Jun 1, 2024 by whitneywhtsang Loading…

1

9

Introduce standalone runner for Triton SPIRV Kernels

#1215 opened May 30, 2024 by alexbaden Loading…

1

[UT] Automatically generate skiplist when on a new platform

#1196 opened May 28, 2024 by AshburnLee Loading…

1

6

Rewrite RewriteTensorPointer pass to support 2D block store

#1193 opened May 27, 2024 by mfrancepillois Loading…

1

15

[UT] add skiplist in pvc w/ lts for skipped cases

#1156 opened May 20, 2024 by AshburnLee Loading…

1

5

[UT] add skiplist for PVC w/ rolling driver

#1093 opened May 11, 2024 by AshburnLee Loading…

1

9

Setup micro benchmarks for non-gemm kernels

#1076 opened May 9, 2024 by LiyangLingIntel Loading…

1

5

Use 32 threads per warp for dot test on Arc

#885 opened Apr 16, 2024 by alexbaden • Draft

1

4

[GEN] Use OCL builtin for tf32 dpas

#853 opened Apr 10, 2024 by whitneywhtsang • Draft

1

4