intel/llvm
1
0
Fork 0
mirror of https://github.com/intel/llvm.git synced 2026-02-06 23:31:50 +08:00
Code Issues Packages Projects Releases Wiki Activity

[BasicAA] Remove unnecessary known size requirement

Browse Source 
The size requirement on V2 was present because it was not clear
whether an unknown size would allow an access before the start of
V2, which could then overlap. This is clarified since D91649: In
this part of BasicAA, all accesses can occur only after the base
pointer, even if they have unknown size.

This makes the positive and negative offset cases symmetric.

Differential Revision: https://reviews.llvm.org/D91482
This commit is contained in:
Nikita Popov
2020-11-14 18:01:14 +01:00
parent 6ee22ca6ce
commit 1dea8ed8b7
2 changed files with 11 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
llvm/lib/Analysis/BasicAliasAnalysis.cpp
View File

@@ -1336,9 +1336,7 @@ AliasResult BasicAAResult::aliasGEP(
// ---------------->|
// |-->V1Size |-------> V2Size
// GEP1 V2
// We need to know that V2Size is not unknown, otherwise we might have
// stripped a gep with negative index ('gep <ptr>, -1, ...).
if (V1Size.hasValue() && V2Size.hasValue()) {
if (V1Size.hasValue()) {
if ((-DecompGEP1.Offset).ult(V1Size.getValue()))
return PartialAlias;
return NoAlias;

10
llvm/test/Analysis/BasicAA/negoffset.ll
View File

@@ -117,3 +117,13 @@ entry:
%2 = load i32, i32* %1
ret i32 %2
}
; CHECK-LABEL: Function: one_size_unknown:
; CHECK: NoModRef: Ptr: i8* %p.minus1 <-> call void @llvm.memset.p0i8.i32(i8* %p, i8 0, i32 %size, i1 false)
define void @one_size_unknown(i8* %p, i32 %size) {
%p.minus1 = getelementptr inbounds i8, i8* %p, i32 -1
call void @llvm.memset.p0i8.i32(i8* %p, i8 0, i32 %size, i1 false)
ret void
}
declare void @llvm.memset.p0i8.i32(i8*, i8, i32, i1)