[CVP] Add samesign flag to icmp
#115642
Merged
[CVP] Add samesign flag to icmp
#115642
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
|
Thank you for submitting a Pull Request (PR) to the LLVM Project! This PR will be automatically labeled and the relevant teams will be notified. If you wish to, you can add reviewers by using the "Reviewers" section on this page. If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers. If you have further questions, they may be answered by the LLVM GitHub User Guide. You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums. |
|
@llvm/pr-subscribers-clang @llvm/pr-subscribers-llvm-transforms Author: Han Qi (rhanqtl) ChangesPatch is 36.07 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/115642.diff 15 Files Affected:
diff --git a/llvm/lib/Transforms/Scalar/CorrelatedValuePropagation.cpp b/llvm/lib/Transforms/Scalar/CorrelatedValuePropagation.cpp
index 7f838340410b51..4b4364ee1e88ef 100644
--- a/llvm/lib/Transforms/Scalar/CorrelatedValuePropagation.cpp
+++ b/llvm/lib/Transforms/Scalar/CorrelatedValuePropagation.cpp
@@ -38,6 +38,7 @@
#include "llvm/IR/Type.h"
#include "llvm/IR/Value.h"
#include "llvm/Support/Casting.h"
+#include "llvm/Support/KnownBits.h"
#include "llvm/Transforms/Utils/Local.h"
#include <cassert>
#include <optional>
@@ -288,24 +289,37 @@ static bool processICmp(ICmpInst *Cmp, LazyValueInfo *LVI) {
if (!Cmp->getOperand(0)->getType()->isIntOrIntVectorTy())
return false;
- if (!Cmp->isSigned())
+ if (!(Cmp->isSigned() || (Cmp->isUnsigned() && !Cmp->hasSameSign())))
return false;
- ICmpInst::Predicate UnsignedPred =
- ConstantRange::getEquivalentPredWithFlippedSignedness(
- Cmp->getPredicate(),
- LVI->getConstantRangeAtUse(Cmp->getOperandUse(0),
- /*UndefAllowed*/ true),
- LVI->getConstantRangeAtUse(Cmp->getOperandUse(1),
- /*UndefAllowed*/ true));
+ bool Changed = false;
- if (UnsignedPred == ICmpInst::Predicate::BAD_ICMP_PREDICATE)
- return false;
+ ConstantRange CR1 = LVI->getConstantRangeAtUse(Cmp->getOperandUse(0),
+ /*UndefAllowed*/ true),
+ CR2 = LVI->getConstantRangeAtUse(Cmp->getOperandUse(1),
+ /*UndefAllowed*/ true);
- ++NumSICmps;
- Cmp->setPredicate(UnsignedPred);
+ if (Cmp->isSigned()) {
+ ICmpInst::Predicate UnsignedPred =
+ ConstantRange::getEquivalentPredWithFlippedSignedness(
+ Cmp->getPredicate(), CR1, CR2);
- return true;
+ if (UnsignedPred == ICmpInst::Predicate::BAD_ICMP_PREDICATE)
+ return false;
+
+ ++NumSICmps;
+ Cmp->setPredicate(UnsignedPred);
+ Changed = true;
+ }
+
+ auto Op0Known = CR1.toKnownBits(), Op1Known = CR2.toKnownBits();
+ if ((Op0Known.Zero.isNegative() && Op1Known.Zero.isNegative()) ||
+ (Op0Known.One.isNegative() && Op1Known.One.isNegative())) {
+ Cmp->setSameSign();
+ Changed = true;
+ }
+
+ return Changed;
}
/// See if LazyValueInfo's ability to exploit edge conditions or range
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/ashr.ll b/llvm/test/Transforms/CorrelatedValuePropagation/ashr.ll
index f719effac113e9..dcc90d548d698e 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/ashr.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/ashr.ll
@@ -133,10 +133,10 @@ define void @test5(i32 %n) {
; CHECK-NEXT: br i1 [[CMP]], label %[[LOOP:.*]], label %[[EXIT:.*]]
; CHECK: [[LOOP]]:
; CHECK-NEXT: [[A:%.*]] = phi i32 [ [[N]], %[[ENTRY]] ], [ [[SHR:%.*]], %[[LOOP]] ]
-; CHECK-NEXT: [[COND:%.*]] = icmp ugt i32 [[A]], 4
+; CHECK-NEXT: [[COND:%.*]] = icmp samesign ugt i32 [[A]], 4
; CHECK-NEXT: call void @llvm.assume(i1 [[COND]])
; CHECK-NEXT: [[SHR]] = lshr i32 [[A]], 1
-; CHECK-NEXT: [[LOOPCOND:%.*]] = icmp ugt i32 [[SHR]], 8
+; CHECK-NEXT: [[LOOPCOND:%.*]] = icmp samesign ugt i32 [[SHR]], 8
; CHECK-NEXT: br i1 [[LOOPCOND]], label %[[LOOP]], label %[[EXIT]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/basic.ll b/llvm/test/Transforms/CorrelatedValuePropagation/basic.ll
index 3c3b1d4bef45bb..3c82edf27aef97 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/basic.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/basic.ll
@@ -579,7 +579,7 @@ define i1 @umin(i32 %a, i32 %b) {
; CHECK-NEXT: [[CMP2:%.*]] = icmp ult i32 [[B]], 20
; CHECK-NEXT: br i1 [[CMP2]], label [[B_GUARD:%.*]], label [[OUT]]
; CHECK: b_guard:
-; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp ult i32 [[A]], [[B]]
+; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp samesign ult i32 [[A]], [[B]]
; CHECK-NEXT: [[MIN:%.*]] = select i1 [[SEL_CMP]], i32 [[A]], i32 [[B]]
; CHECK-NEXT: ret i1 false
; CHECK: out:
@@ -612,7 +612,7 @@ define i1 @smin(i32 %a, i32 %b) {
; CHECK-NEXT: [[CMP2:%.*]] = icmp ult i32 [[B]], 20
; CHECK-NEXT: br i1 [[CMP2]], label [[B_GUARD:%.*]], label [[OUT]]
; CHECK: b_guard:
-; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp ule i32 [[A]], [[B]]
+; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp samesign ule i32 [[A]], [[B]]
; CHECK-NEXT: [[MIN:%.*]] = select i1 [[SEL_CMP]], i32 [[A]], i32 [[B]]
; CHECK-NEXT: ret i1 false
; CHECK: out:
@@ -645,7 +645,7 @@ define i1 @smax(i32 %a, i32 %b) {
; CHECK-NEXT: [[CMP2:%.*]] = icmp sgt i32 [[B]], 20
; CHECK-NEXT: br i1 [[CMP2]], label [[B_GUARD:%.*]], label [[OUT]]
; CHECK: b_guard:
-; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp uge i32 [[A]], [[B]]
+; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp samesign uge i32 [[A]], [[B]]
; CHECK-NEXT: [[MAX:%.*]] = select i1 [[SEL_CMP]], i32 [[A]], i32 [[B]]
; CHECK-NEXT: ret i1 false
; CHECK: out:
@@ -678,7 +678,7 @@ define i1 @umax(i32 %a, i32 %b) {
; CHECK-NEXT: [[CMP2:%.*]] = icmp sgt i32 [[B]], 20
; CHECK-NEXT: br i1 [[CMP2]], label [[B_GUARD:%.*]], label [[OUT]]
; CHECK: b_guard:
-; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp uge i32 [[A]], [[B]]
+; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp samesign uge i32 [[A]], [[B]]
; CHECK-NEXT: [[MAX:%.*]] = select i1 [[SEL_CMP]], i32 [[A]], i32 [[B]]
; CHECK-NEXT: ret i1 false
; CHECK: out:
@@ -824,7 +824,7 @@ define i1 @clamp_low3(i32 noundef %a) {
; CHECK-NEXT: [[CMP:%.*]] = icmp sge i32 [[A]], 5
; CHECK-NEXT: br i1 [[CMP]], label [[A_GUARD:%.*]], label [[OUT:%.*]]
; CHECK: a_guard:
-; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp ugt i32 [[A]], 5
+; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp samesign ugt i32 [[A]], 5
; CHECK-NEXT: [[ADD:%.*]] = add nsw i32 [[A]], -1
; CHECK-NEXT: [[SEL:%.*]] = select i1 [[SEL_CMP]], i32 [[ADD]], i32 5
; CHECK-NEXT: ret i1 false
@@ -852,7 +852,7 @@ define i1 @clamp_low4(i32 noundef %a) {
; CHECK-NEXT: [[CMP:%.*]] = icmp sge i32 [[A]], 5
; CHECK-NEXT: br i1 [[CMP]], label [[A_GUARD:%.*]], label [[OUT:%.*]]
; CHECK: a_guard:
-; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp ule i32 [[A]], 5
+; CHECK-NEXT: [[SEL_CMP:%.*]] = icmp samesign ule i32 [[A]], 5
; CHECK-NEXT: [[ADD:%.*]] = add nsw i32 [[A]], -1
; CHECK-NEXT: [[SEL:%.*]] = select i1 [[SEL_CMP]], i32 5, i32 [[ADD]]
; CHECK-NEXT: ret i1 false
@@ -1085,10 +1085,10 @@ define void @abs1(i32 %a, ptr %p) {
; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[A]], 0
; CHECK-NEXT: [[ABS:%.*]] = select i1 [[CMP]], i32 [[SUB]], i32 [[A]]
; CHECK-NEXT: store i1 true, ptr [[P]], align 1
-; CHECK-NEXT: [[C2:%.*]] = icmp ult i32 [[ABS]], 19
+; CHECK-NEXT: [[C2:%.*]] = icmp samesign ult i32 [[ABS]], 19
; CHECK-NEXT: store i1 [[C2]], ptr [[P]], align 1
; CHECK-NEXT: store i1 true, ptr [[P]], align 1
-; CHECK-NEXT: [[C4:%.*]] = icmp uge i32 [[ABS]], 1
+; CHECK-NEXT: [[C4:%.*]] = icmp samesign uge i32 [[ABS]], 1
; CHECK-NEXT: store i1 [[C4]], ptr [[P]], align 1
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
@@ -1131,10 +1131,10 @@ define void @abs2(i32 %a, ptr %p) {
; CHECK-NEXT: [[CMP:%.*]] = icmp sge i32 [[A]], 0
; CHECK-NEXT: [[ABS:%.*]] = select i1 [[CMP]], i32 [[A]], i32 [[SUB]]
; CHECK-NEXT: store i1 true, ptr [[P]], align 1
-; CHECK-NEXT: [[C2:%.*]] = icmp ult i32 [[ABS]], 19
+; CHECK-NEXT: [[C2:%.*]] = icmp samesign ult i32 [[ABS]], 19
; CHECK-NEXT: store i1 [[C2]], ptr [[P]], align 1
; CHECK-NEXT: store i1 true, ptr [[P]], align 1
-; CHECK-NEXT: [[C4:%.*]] = icmp uge i32 [[ABS]], 1
+; CHECK-NEXT: [[C4:%.*]] = icmp samesign uge i32 [[ABS]], 1
; CHECK-NEXT: store i1 [[C4]], ptr [[P]], align 1
; CHECK-NEXT: br label [[EXIT]]
; CHECK: exit:
@@ -1934,7 +1934,7 @@ define void @select_assume(i32 %a, i32 %b, i1 %c, ptr %p) {
; CHECK-NEXT: [[C2:%.*]] = icmp ult i32 [[B]], 20
; CHECK-NEXT: call void @llvm.assume(i1 [[C2]])
; CHECK-NEXT: [[S:%.*]] = select i1 [[C]], i32 [[A]], i32 [[B]]
-; CHECK-NEXT: [[C3:%.*]] = icmp ult i32 [[S]], 19
+; CHECK-NEXT: [[C3:%.*]] = icmp samesign ult i32 [[S]], 19
; CHECK-NEXT: store i1 [[C3]], ptr [[P]], align 1
; CHECK-NEXT: store i1 true, ptr [[P]], align 1
; CHECK-NEXT: ret void
@@ -1957,10 +1957,10 @@ define void @xor(i8 %a, ptr %p) {
; CHECK-NEXT: [[A_MASK:%.*]] = and i8 [[A]], 15
; CHECK-NEXT: [[XOR:%.*]] = xor i8 [[A_MASK]], -86
; CHECK-NEXT: store i1 true, ptr [[P]], align 1
-; CHECK-NEXT: [[C2:%.*]] = icmp ugt i8 [[XOR]], -96
+; CHECK-NEXT: [[C2:%.*]] = icmp samesign ugt i8 [[XOR]], -96
; CHECK-NEXT: store i1 [[C2]], ptr [[P]], align 1
; CHECK-NEXT: store i1 true, ptr [[P]], align 1
-; CHECK-NEXT: [[C4:%.*]] = icmp ult i8 [[XOR]], -81
+; CHECK-NEXT: [[C4:%.*]] = icmp samesign ult i8 [[XOR]], -81
; CHECK-NEXT: store i1 [[C4]], ptr [[P]], align 1
; CHECK-NEXT: ret void
;
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/cond-using-block-value.ll b/llvm/test/Transforms/CorrelatedValuePropagation/cond-using-block-value.ll
index a7a1803bccc263..89ce59b96a5788 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/cond-using-block-value.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/cond-using-block-value.ll
@@ -13,7 +13,7 @@ define void @test_icmp_from_implied_cond(i32 %a, i32 %b) {
; CHECK-NEXT: br i1 [[COND]], label [[L2:%.*]], label [[END]]
; CHECK: l2:
; CHECK-NEXT: call void @use(i1 true)
-; CHECK-NEXT: [[B_CMP2:%.*]] = icmp ult i32 [[B]], 31
+; CHECK-NEXT: [[B_CMP2:%.*]] = icmp samesign ult i32 [[B]], 31
; CHECK-NEXT: call void @use(i1 [[B_CMP2]])
; CHECK-NEXT: ret void
; CHECK: end:
@@ -74,7 +74,7 @@ define void @test_icmp_from_implied_range(i16 %x, i32 %b) {
; CHECK-NEXT: br i1 [[COND]], label [[L1:%.*]], label [[END:%.*]]
; CHECK: l1:
; CHECK-NEXT: call void @use(i1 true)
-; CHECK-NEXT: [[B_CMP2:%.*]] = icmp ult i32 [[B]], 65534
+; CHECK-NEXT: [[B_CMP2:%.*]] = icmp samesign ult i32 [[B]], 65534
; CHECK-NEXT: call void @use(i1 [[B_CMP2]])
; CHECK-NEXT: ret void
; CHECK: end:
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/deopt.ll b/llvm/test/Transforms/CorrelatedValuePropagation/deopt.ll
index 96cc6c17503123..03815828ba6d50 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/deopt.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/deopt.ll
@@ -97,7 +97,7 @@ define void @test3(i1 %c, i1 %c2) {
; CHECK-LABEL: @test3(
; CHECK-NEXT: [[SEL:%.*]] = select i1 [[C:%.*]], i64 0, i64 1
; CHECK-NEXT: [[SEL2:%.*]] = select i1 [[C2:%.*]], i64 [[SEL]], i64 2
-; CHECK-NEXT: [[CMP:%.*]] = icmp ugt i64 [[SEL2]], 1
+; CHECK-NEXT: [[CMP:%.*]] = icmp samesign ugt i64 [[SEL2]], 1
; CHECK-NEXT: br i1 [[CMP]], label [[TAKEN:%.*]], label [[UNTAKEN:%.*]]
; CHECK: taken:
; CHECK-NEXT: call void @use() [ "deopt"(i64 2) ]
@@ -122,7 +122,7 @@ define void @test4(i1 %c, i1 %c2) {
; CHECK-NEXT: [[SEL2:%.*]] = select i1 [[C2:%.*]], i64 0, i64 1
; CHECK-NEXT: [[ADD1:%.*]] = add nuw nsw i64 0, [[SEL]]
; CHECK-NEXT: [[ADD2:%.*]] = add nuw nsw i64 [[ADD1]], [[SEL2]]
-; CHECK-NEXT: [[CMP:%.*]] = icmp ugt i64 [[ADD2]], 1
+; CHECK-NEXT: [[CMP:%.*]] = icmp samesign ugt i64 [[ADD2]], 1
; CHECK-NEXT: br i1 [[CMP]], label [[TAKEN:%.*]], label [[UNTAKEN:%.*]]
; CHECK: taken:
; CHECK-NEXT: call void @use() [ "deopt"(i64 2) ]
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/icmp.ll b/llvm/test/Transforms/CorrelatedValuePropagation/icmp.ll
index 57c9f8926b524f..72f09a949a060d 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/icmp.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/icmp.ll
@@ -128,7 +128,7 @@ define i1 @test4(i32 %x, i32 %y) #0 {
; CHECK-NEXT: br i1 [[CMP2]], label [[CONT2:%.*]], label [[OUT]]
; CHECK: cont2:
; CHECK-NEXT: [[ADD:%.*]] = add nuw nsw i32 [[X]], [[Y]]
-; CHECK-NEXT: [[CMP3:%.*]] = icmp ult i32 [[ADD]], 15
+; CHECK-NEXT: [[CMP3:%.*]] = icmp samesign ult i32 [[ADD]], 15
; CHECK-NEXT: br label [[OUT]]
; CHECK: out:
; CHECK-NEXT: [[RET:%.*]] = phi i1 [ true, [[ENTRY:%.*]] ], [ true, [[CONT1]] ], [ [[CMP3]], [[CONT2]] ]
@@ -198,7 +198,7 @@ define i1 @test6(i32 %x, i32 %y) #0 {
; CHECK-NEXT: br i1 [[CMP2]], label [[CONT2:%.*]], label [[OUT]]
; CHECK: cont2:
; CHECK-NEXT: [[SHIFTED:%.*]] = shl nuw nsw i32 [[X]], [[Y]]
-; CHECK-NEXT: [[CMP3:%.*]] = icmp ult i32 [[SHIFTED]], 65536
+; CHECK-NEXT: [[CMP3:%.*]] = icmp samesign ult i32 [[SHIFTED]], 65536
; CHECK-NEXT: br label [[OUT]]
; CHECK: out:
; CHECK-NEXT: [[RET:%.*]] = phi i1 [ true, [[ENTRY:%.*]] ], [ true, [[CONT1]] ], [ [[CMP3]], [[CONT2]] ]
@@ -1265,7 +1265,7 @@ define void @ashr_sgt(i8 %x) {
; CHECK-NEXT: br i1 [[C]], label [[IF:%.*]], label [[ELSE:%.*]]
; CHECK: if:
; CHECK-NEXT: call void @check1(i1 true)
-; CHECK-NEXT: [[C3:%.*]] = icmp ugt i8 [[X]], 8
+; CHECK-NEXT: [[C3:%.*]] = icmp samesign ugt i8 [[X]], 8
; CHECK-NEXT: call void @check1(i1 [[C3]])
; CHECK-NEXT: ret void
; CHECK: else:
@@ -1291,7 +1291,7 @@ define void @ashr_sge(i8 %x) {
; CHECK-NEXT: br i1 [[C]], label [[IF:%.*]], label [[ELSE:%.*]]
; CHECK: if:
; CHECK-NEXT: call void @check1(i1 true)
-; CHECK-NEXT: [[C3:%.*]] = icmp uge i8 [[X]], 5
+; CHECK-NEXT: [[C3:%.*]] = icmp samesign uge i8 [[X]], 5
; CHECK-NEXT: call void @check1(i1 [[C3]])
; CHECK-NEXT: ret void
; CHECK: else:
@@ -1374,7 +1374,7 @@ define i1 @pr69928(i64 noundef %arg, i64 noundef %arg1) {
; CHECK-NEXT: entry:
; CHECK-NEXT: [[CMP1:%.*]] = icmp ult i64 [[ARG:%.*]], 64424509440
; CHECK-NEXT: [[AND:%.*]] = and i64 [[ARG1:%.*]], 4294967295
-; CHECK-NEXT: [[CMP2:%.*]] = icmp ult i64 [[ARG]], [[AND]]
+; CHECK-NEXT: [[CMP2:%.*]] = icmp samesign ult i64 [[ARG]], [[AND]]
; CHECK-NEXT: [[SELECT:%.*]] = select i1 [[CMP1]], i1 [[CMP2]], i1 false
; CHECK-NEXT: ret i1 [[SELECT]]
;
@@ -1390,7 +1390,7 @@ define i1 @test_select_flip(i64 noundef %arg) {
; CHECK-LABEL: @test_select_flip(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[CMP1:%.*]] = icmp ult i64 [[ARG:%.*]], 1000
-; CHECK-NEXT: [[CMP2:%.*]] = icmp ult i64 [[ARG]], 100
+; CHECK-NEXT: [[CMP2:%.*]] = icmp samesign ult i64 [[ARG]], 100
; CHECK-NEXT: [[SELECT:%.*]] = select i1 [[CMP1]], i1 [[CMP2]], i1 false
; CHECK-NEXT: ret i1 [[SELECT]]
;
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/minmaxabs.ll b/llvm/test/Transforms/CorrelatedValuePropagation/minmaxabs.ll
index a13ec50bd053a4..49a17fd4160679 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/minmaxabs.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/minmaxabs.ll
@@ -12,7 +12,7 @@ define void @test_umin(i32 %x) {
; CHECK-LABEL: @test_umin(
; CHECK-NEXT: [[M:%.*]] = call i32 @llvm.umin.i32(i32 [[X:%.*]], i32 10)
; CHECK-NEXT: call void @use(i1 true)
-; CHECK-NEXT: [[C2:%.*]] = icmp ult i32 [[M]], 10
+; CHECK-NEXT: [[C2:%.*]] = icmp samesign ult i32 [[M]], 10
; CHECK-NEXT: call void @use(i1 [[C2]])
; CHECK-NEXT: ret void
;
@@ -60,7 +60,7 @@ define void @test_smax(i32 %x) {
; CHECK-LABEL: @test_smax(
; CHECK-NEXT: [[M:%.*]] = call i32 @llvm.smax.i32(i32 [[X:%.*]], i32 10)
; CHECK-NEXT: call void @use(i1 true)
-; CHECK-NEXT: [[C2:%.*]] = icmp ugt i32 [[M]], 10
+; CHECK-NEXT: [[C2:%.*]] = icmp samesign ugt i32 [[M]], 10
; CHECK-NEXT: call void @use(i1 [[C2]])
; CHECK-NEXT: ret void
;
@@ -77,7 +77,7 @@ define void @test_abs1(ptr %p) {
; CHECK-NEXT: [[X:%.*]] = load i32, ptr [[P:%.*]], align 4, !range [[RNG0:![0-9]+]]
; CHECK-NEXT: [[A:%.*]] = call i32 @llvm.abs.i32(i32 [[X]], i1 true)
; CHECK-NEXT: call void @use(i1 true)
-; CHECK-NEXT: [[C2:%.*]] = icmp ult i32 [[A]], 15
+; CHECK-NEXT: [[C2:%.*]] = icmp samesign ult i32 [[A]], 15
; CHECK-NEXT: call void @use(i1 [[C2]])
; CHECK-NEXT: ret void
;
@@ -110,7 +110,7 @@ define void @test_abs3(i32 %x) {
; CHECK-LABEL: @test_abs3(
; CHECK-NEXT: [[A:%.*]] = call i32 @llvm.abs.i32(i32 [[X:%.*]], i1 true)
; CHECK-NEXT: call void @use(i1 true)
-; CHECK-NEXT: [[C2:%.*]] = icmp ugt i32 [[A]], 0
+; CHECK-NEXT: [[C2:%.*]] = icmp samesign ugt i32 [[A]], 0
; CHECK-NEXT: call void @use(i1 [[C2]])
; CHECK-NEXT: ret void
;
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/overflow_predicate.ll b/llvm/test/Transforms/CorrelatedValuePropagation/overflow_predicate.ll
index 75f66c5a89e7e2..8591ab73ebfeb8 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/overflow_predicate.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/overflow_predicate.ll
@@ -49,7 +49,7 @@ define i1 @uadd_ov_true(i8 %x, ptr %px, ptr %pc) {
; CHECK-NEXT: [[OV:%.*]] = extractvalue { i8, i1 } [[VAL_OV]], 1
; CHECK-NEXT: br i1 [[OV]], label [[OVERFLOW:%.*]], label [[TRAP:%.*]]
; CHECK: overflow:
-; CHECK-NEXT: [[C1:%.*]] = icmp ugt i8 [[X]], -100
+; CHECK-NEXT: [[C1:%.*]] = icmp samesign ugt i8 [[X]], -100
; CHECK-NEXT: store i1 [[C1]], ptr [[PC:%.*]], align 1
; CHECK-NEXT: ret i1 true
; CHECK: trap:
@@ -113,7 +113,7 @@ define i1 @sadd_ov_true(i8 %x, ptr %px, ptr %pc) {
; CHECK-NEXT: [[OV:%.*]] = extractvalue { i8, i1 } [[VAL_OV]], 1
; CHECK-NEXT: br i1 [[OV]], label [[OVERFLOW:%.*]], label [[TRAP:%.*]]
; CHECK: overflow:
-; CHECK-NEXT: [[C1:%.*]] = icmp ugt i8 [[X]], 28
+; CHECK-NEXT: [[C1:%.*]] = icmp samesign ugt i8 [[X]], 28
; CHECK-NEXT: store i1 [[C1]], ptr [[PC:%.*]], align 1
; CHECK-NEXT: ret i1 true
; CHECK: trap:
@@ -177,7 +177,7 @@ define i1 @usub_ov_true(i8 %x, ptr %px, ptr %pc) {
; CHECK-NEXT: [[OV:%.*]] = extractvalue { i8, i1 } [[VAL_OV]], 1
; CHECK-NEXT: br i1 [[OV]], label [[OVERFLOW:%.*]], label [[TRAP:%.*]]
; CHECK: overflow:
-; CHECK-NEXT: [[C1:%.*]] = icmp ult i8 [[X]], 99
+; CHECK-NEXT: [[C1:%.*]] = icmp samesign ult i8 [[X]], 99
; CHECK-NEXT: store i1 [[C1]], ptr [[PC:%.*]], align 1
; CHECK-NEXT: ret i1 true
; CHECK: trap:
@@ -241,7 +241,7 @@ define i1 @ssub_ov_true(i8 %x, ptr %px, ptr %pc) {
; CHECK-NEXT: [[OV:%.*]] = extractvalue { i8, i1 } [[VAL_OV]], 1
; CHECK-NEXT: br i1 [[OV]], label [[OVERFLOW:%.*]], label [[TRAP:%.*]]
; CHECK: overflow:
-; CHECK-NEXT: [[C1:%.*]] = icmp ult i8 [[X]], -29
+; CHECK-NEXT: [[C1:%.*]] = icmp samesign ult i8 [[X]], -29
; CHECK-NEXT: store i1 [[C1]], ptr [[PC:%.*]], align 1
; CHECK-NEXT: ret i1 true
; CHECK: trap:
@@ -273,7 +273,7 @@ define i1 @umul_ov_false(i8 %x, ptr %px, ptr %pc) {
; CHECK-NEXT: [[OV:%.*]] = extractvalue { i8, i1 } [[VAL_OV]], 1
; CHECK-NEXT: br i1 [[OV]], label [[TRAP:%.*]], label [[NO_OVERFLOW:%.*]]
; CHECK: no_overflow:
-; CHECK-NEXT: [[C1:%.*]] = icmp ugt i8 [[X]], 24
+; CHECK-NEXT: [[C1:%.*]] = icmp samesign ugt i8 [[X]], 24
; CHECK-NEXT: store i1 [[C1]], ptr [[PC:%.*]], align 1
; CHECK-NEXT: ret i1 false
; CHECK: trap:
diff --git a/llvm/test/Transforms/CorrelatedValuePropagation/range.ll b/llvm/test/Transforms/CorrelatedValuePropagation/range.ll
index ce1b591218d1b1..03d71fa9b52773 100644
--- a/llvm/test/Transforms/CorrelatedValuePropagation/range.ll
+++ b/llvm/test/Transforms/CorrelatedValuePropagation/range.ll
@@ -64,7 +64,7 @@ define i32 @test3(i32 %c) nounwind {
; CHECK: if.then:
; CHECK-NEXT: ret i32 1
; CHECK: if.end:
-; CHECK-NEXT: [[CMP1:%.*]] = icmp ult i32 [[C]], 3
+; CHECK-NEXT: [[CMP1:%.*]] = icmp samesign ult i32 [[C]], 3
; CHECK-NEXT: br i1 [[CMP1]], label [[IF_THEN2:%.*]], label [[IF_END8:%.*]]
; CHECK: if.then2:
; CHECK-NEXT: br i1 true, label [[IF_THEN4:%.*]], label [[IF_END6:%.*]]
@@ -989,11 +989,11 @@ define i1 @ctlz_nofold(i16 %x) {
; CHECK-NEXT: br i1 [[CMP]], label [[IF:%.*]], label [[ELSE:%.*]]
; CHECK: if:
; CHECK-NEXT: [[CTLZ:%.*]] = call i16 @llvm.ctlz.i16(i16 [[X]], i1 false)
-; CHECK-NEXT: [[RES:%.*]] = icmp uge i16 [[CTLZ]], 9
+; CHECK-NEXT: [[RES:%.*]] = icmp samesign uge i16 [[CTLZ]], 9
; CHECK-NEXT: ret i1 [[RES]]
; CHECK: else:
; CHECK-NEXT: [[CTLZ2:%.*]] = call i16 @llvm.ctl...
[truncated]
|
dtcxzyw
reviewed
Nov 10, 2024
dtcxzyw
reviewed
Nov 10, 2024
dtcxzyw
reviewed
Nov 10, 2024
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
nikic
reviewed
Nov 10, 2024
|
It looks like you also need to update a clang test: |
antoniofrighetto
approved these changes
Nov 11, 2024
|
@rhanqtl Congratulations on having your first Pull Request (PR) merged into the LLVM Project! Your changes will be combined with recent changes from other authors, then tested by our build bots. If there is a problem with a build, you may receive a report in an email or a comment on this PR. Please check whether problems have been caused by your change specifically, as the builds can include changes from many authors. It is not uncommon for your change to be included in a build that fails due to someone else's changes, or infrastructure issues. How to do this, and the rest of the post-merge process, is covered in detail here. If your change does cause a problem, it may be reverted, or you can revert it yourself. This is a normal part of LLVM development. You can fix your changes and open a new PR to merge them again. If you don't get any reports, no action is required from you. Your changes are working as expected, well done! |
antoniofrighetto
changed the title
samesign flag to icmp
Groverkss
pushed a commit
to iree-org/llvm-project
that referenced
this pull request
Nov 15, 2024
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.
Closes #114820