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

Unify Intel LLVM optimizations under lib/Target/LLVMIR #2900

Merged
merged 1 commit into from
Dec 6, 2024

Conversation

alexbaden
Copy link
Contributor

Depends on #2775

I agree it would be nice to match upstream directory structure for our LLVM IR post processing passes - I've unified them here on top of the branch from 2775.

bin/CMakeLists.txt Outdated Show resolved Hide resolved
@alexbaden alexbaden force-pushed the alex/unify_intel_llvmir_processing branch from 951335f to ae8e6f8 Compare December 3, 2024 03:06
@alexbaden
Copy link
Contributor Author

Rebased on #2775 - with this change all the Intel LLVM passes are in third_party/intel/lib/Target/LLVMIR (https://github.com/intel/intel-xpu-backend-for-triton/tree/alex/unify_intel_llvmir_processing/third_party/intel/lib/Target/LLVMIR)

@@ -1,7 +1,7 @@
/// Trimmed down clone of llvm opt to be able to test triton custom llvm ir
/// passes.
#include "lib/Target/LLVMIR/LLVMPasses.h"
#include "third_party/intel/lib/LLVMIR/LLVMPasses.h"
#include "third_party/intel/lib/Target/LLVMIR/LLVMPasses.h"
Copy link
Contributor

Choose a reason for hiding this comment

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

I wonder if we really need to add this to this (common) makefile, it wasn't required for the PostProcessing library.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

PostProcessing does not test passes via lit test, but the FreezeMaskedLayout pass (which is now part of the same library) does.

Copy link
Contributor

Choose a reason for hiding this comment

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

I see

third_party/intel/CMakeLists.txt Show resolved Hide resolved
@alexbaden alexbaden marked this pull request as ready for review December 3, 2024 14:22
@alexbaden alexbaden force-pushed the alex/unify_intel_llvmir_processing branch 2 times, most recently from caf261e to 18a084a Compare December 4, 2024 14:42
@@ -1,7 +1,7 @@
/// Trimmed down clone of llvm opt to be able to test triton custom llvm ir
/// passes.
#include "lib/Target/LLVMIR/LLVMPasses.h"
#include "third_party/intel/lib/LLVMIR/LLVMPasses.h"
#include "third_party/intel/lib/Target/LLVMIR/LLVMPasses.h"
Copy link
Contributor

Choose a reason for hiding this comment

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

I see

@alexbaden alexbaden force-pushed the alex/unify_intel_llvmir_processing branch from 18a084a to 139df71 Compare December 6, 2024 02:43
@alexbaden alexbaden enabled auto-merge (squash) December 6, 2024 02:46
@alexbaden alexbaden merged commit daa8c28 into main Dec 6, 2024
5 checks passed
@alexbaden alexbaden deleted the alex/unify_intel_llvmir_processing branch December 6, 2024 03:29
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.

3 participants