Skip to content
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

Merge OpenAI Triton commit 14800bf #1174

Merged
merged 5 commits into from
May 22, 2024
Merged

Conversation

whitneywhtsang
Copy link
Contributor

This PR change the Triton base from 5e6952d to 14800bf (May 20).
Pass rate: 98.64%

Please do not squash and merge this PR.

MaskRay and others added 4 commits May 20, 2024 09:59
Follow-up to #3933.

This might be copied from llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp.
`setUseAssemblerInfoForParsing` was only used for -mattr=dumpcode and
should not be needed here.

The llvm-project plan is to continue improving the integrated assembler
and remove `setUseAssemblerInfoForParsing`.

While here, drop the byte-order mark.
This PR adds the functionality to profile MLIR/LLVM passes.

A user can set `MLIR_ENABLE_TIMING` or `LLVM_ENABLE_TIMING` env.
variable to track the elapsed time of each MLIR or LLVM pass.
Like num_stages and num_warps, this tunes the generated kernel,
controlling the maximum number of registers that one thread can consume.
Reduce usages of `check_cuda_or_hip`, so that more tests can be run for
downstream backends by default.

Signed-off-by: Whitney Tsang <whitney.tsang@intel.com>
@whitneywhtsang whitneywhtsang self-assigned this May 22, 2024
@whitneywhtsang whitneywhtsang force-pushed the whitneywhtsang/merge branch from 090e8a0 to 323baad Compare May 22, 2024 18:20
@whitneywhtsang whitneywhtsang marked this pull request as ready for review May 22, 2024 18:47
@whitneywhtsang whitneywhtsang merged commit 323baad into llvm-target May 22, 2024
2 checks passed
@vlad-penkin vlad-penkin linked an issue Jun 22, 2024 that may be closed by this pull request
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Merge OpenAI Triton till May 24th
5 participants