-
Notifications
You must be signed in to change notification settings - Fork 39
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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
[GEMM] Improve performance of shape 16384x8192x4096 on default path
codegen: gemm
performance
#2435
opened Oct 7, 2024 by
whitneywhtsang
Replace the TritonGEN Barrier operation with something from upstream MLIR
codegen: mlir
enhancement
New feature or request
[Triton Upstream] Something isn't working
tests: e2e
AttrsDescriptor
class refactoring breaks Torch Inductor
bug
[Pytorch pin update] Something isn't working
nightly failure
failure in nightly actions
tests: torchinductor
e9a55b4
- inductor/test_triton_kernels.py::CustomOpTests::test_autotune_unbacked test failure
bug
[XeTLA] Add StreamK and SplitK implementation
codegen: gemm
enhancement
New feature or request
performance
Merge OpenAI Triton till Oct 11st
enhancement
New feature or request
upstream: rebase
PR to be up-streamed
Clean up duplication between MaterializeBlockPtr and RewriteTensorPointer
codegen: mlir
enhancement
New feature or request
Improve out-of-box performance for GEMM kernels variants
codegen: gemm
enhancement
New feature or request
performance
umbrella
Allocation analysis only consider New feature or request
NvidiaMmaEncodingAttr
codegen: mlir
enhancement
[FA2 performance] flashattention with dim=128 get ~90% of xetla
codegen: attention
enhancement
New feature or request
performance
Previous Next
ProTip!
What’s not been updated in a month: updated:<2024-09-09.