maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

Fixed a bug in store sinking. 

The problem was in store-sink barrier check.

Store sink barrier should be checked for ModRef (read-write) mode.

http://llvm.org/bugs/show_bug.cgi?id=22613

llvm-svn: 229495
This commit is contained in:
Elena Demikhovsky 2015-02-17 13:10:05 +00:00
parent cd5a3673c3
commit ef035bb974
2 changed files with 120 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
llvm/lib/Transforms/Scalar/MergedLoadStoreMotion.cpp
View File

@ -403,7 +403,7 @@ bool MergedLoadStoreMotion::isStoreSinkBarrierInRange(const Instruction& Start,
const Instruction& End,
AliasAnalysis::Location
Loc) {
return AA->canInstructionRangeModRef(Start, End, Loc, AliasAnalysis::Ref);
return AA->canInstructionRangeModRef(Start, End, Loc, AliasAnalysis::ModRef);
}
///
@ -414,6 +414,7 @@ bool MergedLoadStoreMotion::isStoreSinkBarrierInRange(const Instruction& Start,
StoreInst *MergedLoadStoreMotion::canSinkFromBlock(BasicBlock *BB1,
StoreInst *Store0) {
DEBUG(dbgs() << "can Sink? : "; Store0->dump(); dbgs() << "\n");
BasicBlock *BB0 = Store0->getParent();
for (BasicBlock::reverse_iterator RBI = BB1->rbegin(), RBE = BB1->rend();
RBI != RBE; ++RBI) {
Instruction *Inst = &*RBI;
@ -422,13 +423,14 @@ StoreInst *MergedLoadStoreMotion::canSinkFromBlock(BasicBlock *BB1,
continue;
StoreInst *Store1 = cast<StoreInst>(Inst);
BasicBlock *BB0 = Store0->getParent();
AliasAnalysis::Location Loc0 = AA->getLocation(Store0);
AliasAnalysis::Location Loc1 = AA->getLocation(Store1);
if (AA->isMustAlias(Loc0, Loc1) && Store0->isSameOperationAs(Store1) &&
!isStoreSinkBarrierInRange(*Store1, BB1->back(), Loc1) &&
!isStoreSinkBarrierInRange(*Store0, BB0->back(), Loc0)) {
!isStoreSinkBarrierInRange(*(std::next(BasicBlock::iterator(Store1))),
BB1->back(), Loc1) &&
!isStoreSinkBarrierInRange(*(std::next(BasicBlock::iterator(Store0))),
BB0->back(), Loc0)) {
return Store1;
}
}

114
llvm/test/Transforms/InstMerge/st_sink_bugfix_22613.ll Normal file
View File

@ -0,0 +1,114 @@
; ModuleID = 'bug.c'
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
; RUN: opt -O2 -S < %s | FileCheck %s
; CHECK_LABEL: main
; CHECK: if.end
; CHECK: store
; CHECK: memset
; CHECK: if.then
; CHECK: store
; CHECK: memset
@d = common global i32 0, align 4
@b = common global i32 0, align 4
@f = common global [1 x [3 x i8]] zeroinitializer, align 1
@e = common global i32 0, align 4
@c = common global i32 0, align 4
@a = common global i32 0, align 4
; Function Attrs: nounwind uwtable
define void @fn1() #0 {
entry:
store i32 0, i32* @d, align 4
br label %for.cond
for.cond: ; preds = %for.inc8, %entry
%0 = load i32* @d, align 4
%cmp = icmp slt i32 %0, 2
br i1 %cmp, label %for.body, label %for.end10
for.body: ; preds = %for.cond
%1 = load i32* @d, align 4
%idxprom = sext i32 %1 to i64
%2 = load i32* @b, align 4
%idxprom1 = sext i32 %2 to i64
%arrayidx = getelementptr inbounds [1 x [3 x i8]]* @f, i32 0, i64 %idxprom1
%arrayidx2 = getelementptr inbounds [3 x i8]* %arrayidx, i32 0, i64 %idxprom
store i8 0, i8* %arrayidx2, align 1
store i32 0, i32* @e, align 4
br label %for.cond3
for.cond3: ; preds = %for.inc, %for.body
%3 = load i32* @e, align 4
%cmp4 = icmp slt i32 %3, 3
br i1 %cmp4, label %for.body5, label %for.end
for.body5: ; preds = %for.cond3
%4 = load i32* @c, align 4
%tobool = icmp ne i32 %4, 0
br i1 %tobool, label %if.then, label %if.end
if.then: ; preds = %for.body5
%5 = load i32* @a, align 4
%dec = add nsw i32 %5, -1
store i32 %dec, i32* @a, align 4
br label %if.end
if.end: ; preds = %if.then, %for.body5
%6 = load i32* @e, align 4
%idxprom6 = sext i32 %6 to i64
%arrayidx7 = getelementptr inbounds [3 x i8]* getelementptr inbounds ([1 x [3 x i8]]* @f, i32 0, i64 0), i32 0, i64 %idxprom6
store i8 1, i8* %arrayidx7, align 1
br label %for.inc
for.inc: ; preds = %if.end
%7 = load i32* @e, align 4
%inc = add nsw i32 %7, 1
store i32 %inc, i32* @e, align 4
br label %for.cond3
for.end: ; preds = %for.cond3
br label %for.inc8
for.inc8: ; preds = %for.end
%8 = load i32* @d, align 4
%inc9 = add nsw i32 %8, 1
store i32 %inc9, i32* @d, align 4
br label %for.cond
for.end10: ; preds = %for.cond
ret void
}
; Function Attrs: nounwind uwtable
define i32 @main() #0 {
entry:
%retval = alloca i32, align 4
store i32 0, i32* %retval
call void @fn1()
%0 = load i8* getelementptr inbounds ([1 x [3 x i8]]* @f, i32 0, i64 0, i64 1), align 1
%conv = sext i8 %0 to i32
%cmp = icmp ne i32 %conv, 1
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
call void @abort() #2
unreachable
if.end: ; preds = %entry
ret i32 0
}
; Function Attrs: noreturn nounwind
declare void @abort() #1
attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #1 = { noreturn nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #2 = { noreturn nounwind }
!llvm.ident = !{!0}
!0 = !{!"clang version 3.7.0 (trunk 229288) (llvm/trunk 229286:229290M)"}