Skip to content

[StructurizeCFG] Fix a crash caused by not updating Predicates properly #124051

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.

Sign up for GitHub

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

Closed
shiltian wants to merge 1 commit into from
Closed

[StructurizeCFG] Fix a crash caused by not updating Predicates properly #124051

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions llvm/lib/Transforms/Scalar/StructurizeCFG.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -971,6 +971,10 @@ void StructurizeCFG::changeExit(RegionNode *Node, BasicBlock *NewExit,
SubRegion->replaceExit(NewExit);
} else {
BasicBlock *BB = Node->getNodeAs<BasicBlock>();
for (BasicBlock *Succ : successors(BB)) {
if (Succ != NewExit)
Predicates[Succ].erase(BB);
}
Comment on lines +974 to +977
Copy link
Contributor

@arsenm arsenm Jan 23, 2025

Choose a reason for hiding this comment

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

Probably should move this into killTerminator where the actual successor removal occurs

Copy link
Contributor Author

@shiltian shiltian Jan 23, 2025

Choose a reason for hiding this comment

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

does it mean that we indeed need to update Predicates here? I thought the no-update was on purpose.

Copy link
Contributor

@arsenm arsenm Jan 23, 2025

Choose a reason for hiding this comment

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

What makes you think it's on purpose?

The predicates collects branch conditions used in the CFG, so it makes sense to me that you would need to adjust those when the CFG changes

Copy link
Contributor Author

@shiltian shiltian Jan 23, 2025

Choose a reason for hiding this comment

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

Hmm, alright. It has been sitting there for 10+ years so my first reaction is, probably it was designed in that way...

killTerminator(BB);
BranchInst *Br = BranchInst::Create(NewExit, BB);
Br->setDebugLoc(TermDL[BB]);
Expand Down
26 changes: 26 additions & 0 deletions llvm/test/Transforms/StructurizeCFG/simple-structurizecfg-crash.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -S -passes=structurizecfg %s -o - | FileCheck %s

define void @foo() {
; CHECK-LABEL: define void @foo() {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: br label %[[COND_FALSE:.*]]
; CHECK: [[COND_TRUE:.*]]:
; CHECK-NEXT: br label %[[COND_END:.*]]
; CHECK: [[COND_FALSE]]:
; CHECK-NEXT: br i1 false, label %[[COND_TRUE]], label %[[COND_END]]
; CHECK: [[COND_END]]:
; CHECK-NEXT: ret void
;
entry:
br i1 false, label %cond.true, label %cond.false

cond.true: ; preds = %entry
br label %cond.end

cond.false: ; preds = %entry
br label %cond.end

cond.end: ; preds = %cond.false, %cond.true
ret void
}
Loading