RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-02-08 05:16:53 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[CVP] Fix assertion failure on vector with.overflow

Browse Source 
Noticed on D62703. LVI only handles plain integers, not vectors of
integers. This was previously not an issue, because vector support
for with.overflow is only a relatively recent addition.

llvm-svn: 362261
This commit is contained in:
Nikita Popov 2019-05-31 20:42:07 +00:00
parent bfe0c48b1d
commit d1d1320557
2 changed files with 13 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
lib/Transforms/Scalar/CorrelatedValuePropagation.cpp
View File

@ -434,7 +434,7 @@ static bool processCallSite(CallSite CS, LazyValueInfo *LVI) {
unsigned ArgNo = 0;
if (auto *WO = dyn_cast<WithOverflowInst>(CS.getInstruction())) {
if (willNotOverflow(WO, LVI)) {
if (WO->getLHS()->getType()->isIntegerTy() && willNotOverflow(WO, LVI)) {
processOverflowIntrinsic(WO);
return true;
}

12
test/Transforms/CorrelatedValuePropagation/overflows.ll
View File

@ -21,6 +21,8 @@ declare { i32, i1 } @llvm.umul.with.overflow.i32(i32, i32)
declare { i8, i1 } @llvm.umul.with.overflow.i8(i8, i8)
declare { <2 x i32>, <2 x i1> } @llvm.uadd.with.overflow.v2i32(<2 x i32>, <2 x i32>)
declare i8 @llvm.uadd.sat.i8(i8, i8)
declare i8 @llvm.sadd.sat.i8(i8, i8)
declare i8 @llvm.usub.sat.i8(i8, i8)
@ -731,6 +733,16 @@ define { i8, i1 } @signed_mul_constant_folding() {
ret { i8, i1 } %mul
}
define { <2 x i32>, <2 x i1> } @uaddo_vec(<2 x i32> %a) {
; CHECK-LABEL: @uaddo_vec(
; CHECK-NEXT: [[ADD:%.*]] = call { <2 x i32>, <2 x i1> } @llvm.uadd.with.overflow.v2i32(<2 x i32> [[A:%.*]], <2 x i32> <i32 1, i32 1>)
; CHECK-NEXT: ret { <2 x i32>, <2 x i1> } [[ADD]]
;
%add = call { <2 x i32>, <2 x i1> } @llvm.uadd.with.overflow.v2i32(<2 x i32> %a, <2 x i32> <i32 1, i32 1>)
ret { <2 x i32>, <2 x i1> } %add
}
define i8 @uadd_sat_no_overflow(i8 %x) {
; CHECK-LABEL: @uadd_sat_no_overflow(
; CHECK-NEXT: [[CMP:%.*]] = icmp ugt i8 [[X:%.*]], 100