Skip to content

[SYCL][NFCI] Move default optimization level configuration #7885

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

Merged
merged 2 commits into from
Jan 4, 2023

Conversation

bader
Copy link
Contributor

@bader bader commented Dec 28, 2022

Right now the front-end compiler modifies default optimization level in SYCL device compilation mode (upstream clang uses O0, whereas DPC++ uses O2). As some of CodeGen module patches landed to upstream, this difference causes conflicts and regresions when we pull changes from the upstream to the sycl branch.

Current optimization level configuration is modeled after OpenCL compiler, which is a JIT compiler, so it doesn't use clang Driver. For DPC++ compiler it's more reasonable to make this type of changes in the driver to avoid significant divergence in LIT tests.

Right now the front-end compiler modifies default optimization level in
SYCL device compilation mode (upstream clang uses O0, whereas DPC++ uses
O2). As some of CodeGen module patches landed to upstream, this
difference causes conflicts and regresions when we pull changes from the
upstream to the `sycl` branch.

Current optimization level configuration is modeled after OpenCL
compiler, which is a JIT compiler, so it doesn't use clang Driver.
For DPC++ compiler it's more reasonable to make this type of changes in
the driver to avoid significant divergence in LIT tests.
@bader bader requested review from a team as code owners December 28, 2022 23:16
@bader bader requested a review from Fznamznon December 28, 2022 23:17
@bader bader temporarily deployed to aws December 28, 2022 23:40 — with GitHub Actions Inactive
@bader bader temporarily deployed to aws December 29, 2022 00:09 — with GitHub Actions Inactive
@bader
Copy link
Contributor Author

bader commented Jan 4, 2023

@intel/dpcpp-clang-driver-reviewers, could you take a look, please?

@bader bader temporarily deployed to aws January 4, 2023 21:33 — with GitHub Actions Inactive
@bader bader requested a review from mdtoguchi January 4, 2023 21:41
Copy link
Contributor

@mdtoguchi mdtoguchi left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks - LGTM

@bader bader temporarily deployed to aws January 4, 2023 22:16 — with GitHub Actions Inactive
@bader
Copy link
Contributor Author

bader commented Jan 4, 2023

Pre-commit failure on AMDGPU is an issue in the test that will be fixed with intel/llvm-test-suite#1487.

@bader bader merged commit 102e154 into intel:sycl Jan 4, 2023
@bader bader deleted the sycl-opt-config branch January 4, 2023 22:32
whitneywhtsang added a commit to whitneywhtsang/llvm that referenced this pull request Jan 7, 2023
whitneywhtsang added a commit to whitneywhtsang/llvm that referenced this pull request Jan 13, 2023
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.

6 participants