Skip to content

Issues: intel/intel-xpu-backend-for-triton

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

136 Open 999 Closed
136 Open 999 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
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Merge OpenAI Triton till Dec 13rd enhancement New feature or request upstream: rebase PR to be up-streamed
#2879 opened Nov 29, 2024 by whitneywhtsang
@whitneywhtsang
[UT] Fix test_gather tests: ut
#2878 opened Nov 29, 2024 by whitneywhtsang
[GEMM] 2048X2048X2048 has big variance. codegen: gemm performance research
#2873 opened Nov 29, 2024 by quintinwang5
@quintinwang5
Attach opencl.kernels metadata enhancement New feature or request good first issue Good for newcomers performance
#2869 opened Nov 28, 2024 by victor-eds
[Pytorch Upstream] Triton build failed with docker image pytorch/manylinux2_28-builder:cpu bug Something isn't working
#2868 opened Nov 28, 2024 by chuanqi129
1
'/debug:fastlink' and '/INCREMENTAL' aren't recognized in Windows build ci code quality windows
#2855 opened Nov 27, 2024 by anmyachev
Drop -tritonintelgpu-optimize-elementwise-locality pass code quality
#2850 opened Nov 27, 2024 by victor-eds
1
@victor-eds
SPIRVRunner: Investigate and handle multi-kernel execution (benchmark- gemm_streamk_benchmark.py) enhancement New feature or request
#2831 opened Nov 26, 2024 by kballeda
@kballeda
1
Failed core tests on A50 bug Something isn't working
#2830 opened Nov 26, 2024 by pbchekin
2
Reduce changes in common files for windows support enhancement New feature or request
#2824 opened Nov 26, 2024 by whitneywhtsang
@anmyachev
2
Remove getElemsPerThreadForOperands from MmaEncodingTrait enhancement New feature or request
#2823 opened Nov 26, 2024 by whitneywhtsang
@chengjunlu
Flaky Triton build bug Something isn't working
#2817 opened Nov 25, 2024 by pbchekin
2
Reland upstream commit 340cbc6 upstream: rebase PR to be up-streamed
#2811 opened Nov 23, 2024 by whitneywhtsang
@anmyachev
Use llvm.func's reqd_work_group_size to specify static local size enhancement New feature or request performance
#2798 opened Nov 22, 2024 by victor-eds
Fine tune sub-group transpose bank conflict prevention for PVC codegen: attention performance
#2797 opened Nov 22, 2024 by victor-eds
@victor-eds
[WIN] Parallel tests execution fails because of locked files bug Something isn't working windows
#2777 opened Nov 20, 2024 by gshimansky 5.0 [DGPU] Initial enabling
1
@gshimansky
2
Consider reusing PyTorch solution for ext_oneapi_get_default_context enhancement New feature or request
#2757 opened Nov 19, 2024 by anmyachev 0.3 [Triton] Language and Runtime
Enable all tests from test_debug.py on XPU bug Something isn't working tests: ut
#2755 opened Nov 19, 2024 by anmyachev 1.0 [UT and Tutorials][Triton 3.2] Pass rate
1
Enable -tritonintelgpu-optimize-reduction-locality by default codegen: attention enhancement New feature or request performance
#2748 opened Nov 19, 2024 by victor-eds 4.0 [Performance] Core
1
@victor-eds
Enable test_side_effectful_reduction on XPU enhancement New feature or request tests: ut
#2736 opened Nov 18, 2024 by anmyachev 1.0 [UT and Tutorials][Triton 3.0] Pass rate
1
@anmyachev
Changes in LLVM SimplifyCFG result in the elimination of mask load paths when the mask is the divisor in a division instruction bug Something isn't working upstream: triton
#2726 opened Nov 16, 2024 by alexbaden 0.3 [Triton] Language and Runtime
1
@alexbaden
[Performance] Enable the SLPVectorizer to improve flash attention performance performance
#2715 opened Nov 15, 2024 by chengjunlu 4.0 [Performance] Core
1
@chengjunlu
[CI] Use DLE for performance runners ci dependencies: ipex enhancement New feature or request performance profiling
#2701 opened Nov 13, 2024 by pbchekin 0.1.1 [Triton / Pytorch] Switch to Pytorch from IPEX
@pbchekin
[GEMM] Performance degradation out of box bug Something isn't working codegen: gemm performance
#2693 opened Nov 12, 2024 by whitneywhtsang 4.0 [Performance] Core
@ESI-SYD
2
Implement support for TritonGPU::UpcastMXFPOp for Intel XPU BE codegen: mlir enhancement New feature or request upstream: triton
#2678 opened Nov 11, 2024 by etiotto 0.3 [Triton] Language and Runtime
1
@LiyangLingIntel
1
ProTip! Mix and match filters to narrow down what you’re looking for.