[AMDGPU] Only run AMDGPUPrintfRuntimeBindingPass at non-prelink phase
#125162
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The pass is registered in a pipeline that will run at none LTO, LTO prelink, as well as LTO postlink phase. It uses the _current_ number of metadata operands as unique id for printf handling. However, if two TUs both have `printf`, their unique id will be starting with 1. After they are linked together, the id is no longer unique anymore. To resolve this issue, the pass should not run at LTO pre link stage.
This stack of pull requests is managed by Graphite. Learn more about stacking. |
|
@llvm/pr-subscribers-backend-amdgpu Author: Shilei Tian (shiltian) ChangesThe pass is registered in a pipeline that will run at none LTO, LTO prelink, as Full diff: https://github.com/llvm/llvm-project/pull/125162.diff 2 Files Affected:
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
index 96062b30fc0127a..3f90e05ce83babb 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
@@ -791,7 +791,8 @@ void AMDGPUTargetMachine::registerPassBuilderCallbacks(PassBuilder &PB) {
PB.registerPipelineEarlySimplificationEPCallback(
[](ModulePassManager &PM, OptimizationLevel Level,
ThinOrFullLTOPhase Phase) {
- PM.addPass(AMDGPUPrintfRuntimeBindingPass());
+ if (!isLTOPreLink(Phase))
+ PM.addPass(AMDGPUPrintfRuntimeBindingPass());
if (Level == OptimizationLevel::O0)
return;
diff --git a/llvm/test/CodeGen/AMDGPU/print-pipeline-passes.ll b/llvm/test/CodeGen/AMDGPU/print-pipeline-passes.ll
index c68143f44866f34..b1fc76f457ece26 100644
--- a/llvm/test/CodeGen/AMDGPU/print-pipeline-passes.ll
+++ b/llvm/test/CodeGen/AMDGPU/print-pipeline-passes.ll
@@ -14,6 +14,7 @@
; PRE-NOT: internalize
; PRE-NOT: amdgpu-attributor
+; PRE-NOT: printfToRuntime
define amdgpu_kernel void @kernel() {
entry:
|
|
bump |
Icohedron
pushed a commit
to Icohedron/llvm-project
that referenced
this pull request
Feb 11, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The pass is registered in a pipeline that will run at none LTO, LTO prelink, as
well as LTO postlink phase. It uses the current number of metadata operands as
unique id for printf handling. However, if two TUs both have
printf, theirunique id will be starting with 1. After they are linked together, the id is no
longer unique anymore. To resolve this issue, the pass should not run at LTO pre
link stage.