[SCCP] Add support for trunc nuw range. #152990
Merged
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
|
@llvm/pr-subscribers-llvm-ir @llvm/pr-subscribers-llvm-transforms Author: Andreas Jonson (andjo403) Changesproof: https://alive2.llvm.org/ce/z/_7PVxq part of changes needed for move of #151961 to SCCP Full diff: https://github.com/llvm/llvm-project/pull/152990.diff 5 Files Affected:
diff --git a/llvm/include/llvm/IR/ConstantRange.h b/llvm/include/llvm/IR/ConstantRange.h
index 9a6a9db65688a..4b2fda364fdf4 100644
--- a/llvm/include/llvm/IR/ConstantRange.h
+++ b/llvm/include/llvm/IR/ConstantRange.h
@@ -380,8 +380,9 @@ class [[nodiscard]] ConstantRange {
/// Return a new range in the specified integer type, which must be
/// strictly smaller than the current type. The returned range will
/// correspond to the possible range of values if the source range had been
- /// truncated to the specified type.
- LLVM_ABI ConstantRange truncate(uint32_t BitWidth) const;
+ /// truncated to the specified type with wrap type \p NoWrapKind.
+ LLVM_ABI ConstantRange truncate(uint32_t BitWidth,
+ unsigned NoWrapKind = 0) const;
/// Make this range have the bit width given by \p BitWidth. The
/// value is zero extended, truncated, or left alone to make it that width.
diff --git a/llvm/lib/IR/ConstantRange.cpp b/llvm/lib/IR/ConstantRange.cpp
index 2fcdbcc6a3db2..b1ea4bbd5ae6b 100644
--- a/llvm/lib/IR/ConstantRange.cpp
+++ b/llvm/lib/IR/ConstantRange.cpp
@@ -872,7 +872,8 @@ ConstantRange ConstantRange::signExtend(uint32_t DstTySize) const {
return ConstantRange(Lower.sext(DstTySize), Upper.sext(DstTySize));
}
-ConstantRange ConstantRange::truncate(uint32_t DstTySize) const {
+ConstantRange ConstantRange::truncate(uint32_t DstTySize,
+ unsigned NoWrapKind) const {
assert(getBitWidth() > DstTySize && "Not a value truncation");
if (isEmptySet())
return getEmpty(DstTySize);
@@ -886,22 +887,36 @@ ConstantRange ConstantRange::truncate(uint32_t DstTySize) const {
// We use the non-wrapped set code to analyze the [Lower, MaxValue) part, and
// then we do the union with [MaxValue, Upper)
if (isUpperWrapped()) {
- // If Upper is greater than or equal to MaxValue(DstTy), it covers the whole
- // truncated range.
- if (Upper.getActiveBits() > DstTySize || Upper.countr_one() == DstTySize)
+ // If Upper is greater than MaxValue(DstTy), it covers the whole truncated
+ // range.
+ if (Upper.getActiveBits() > DstTySize)
return getFull(DstTySize);
- Union = ConstantRange(APInt::getMaxValue(DstTySize),Upper.trunc(DstTySize));
- UpperDiv.setAllBits();
-
- // Union covers the MaxValue case, so return if the remaining range is just
- // MaxValue(DstTy).
- if (LowerDiv == UpperDiv)
- return Union;
+ // For nuw the two parts is: [0, Upper) \/ [Lower, MaxValue(DstTy) + 1]
+ if (NoWrapKind & TruncInst::NoUnsignedWrap) {
+ Union = ConstantRange(APInt::getZero(DstTySize), Upper.trunc(DstTySize));
+ UpperDiv = APInt::getOneBitSet(getBitWidth(), DstTySize);
+ } else {
+ // If Upper is equal to MaxValue(DstTy), it covers the whole truncated
+ // range.
+ if (Upper.countr_one() == DstTySize)
+ return getFull(DstTySize);
+ Union =
+ ConstantRange(APInt::getMaxValue(DstTySize), Upper.trunc(DstTySize));
+ UpperDiv.setAllBits();
+ // Union covers the MaxValue case, so return if the remaining range is
+ // just MaxValue(DstTy).
+ if (LowerDiv == UpperDiv)
+ return Union;
+ }
}
// Chop off the most significant bits that are past the destination bitwidth.
if (LowerDiv.getActiveBits() > DstTySize) {
+ // For trunc nuw if LowerDiv is greater than MaxValue(DstTy), the range is
+ // outside the whole truncated range.
+ if (NoWrapKind & TruncInst::NoUnsignedWrap)
+ return Union;
// Mask to just the signficant bits and subtract from LowerDiv/UpperDiv.
APInt Adjust = LowerDiv & APInt::getBitsSetFrom(getBitWidth(), DstTySize);
LowerDiv -= Adjust;
@@ -913,6 +928,10 @@ ConstantRange ConstantRange::truncate(uint32_t DstTySize) const {
return ConstantRange(LowerDiv.trunc(DstTySize),
UpperDiv.trunc(DstTySize)).unionWith(Union);
+ if (!LowerDiv.isZero() && NoWrapKind & TruncInst::NoUnsignedWrap)
+ return ConstantRange(LowerDiv.trunc(DstTySize), APInt::getZero(DstTySize))
+ .unionWith(Union);
+
// The truncated value wraps around. Check if we can do better than fullset.
if (UpperDivWidth == DstTySize + 1) {
// Clear the MSB so that UpperDiv wraps around.
diff --git a/llvm/lib/Transforms/Utils/SCCPSolver.cpp b/llvm/lib/Transforms/Utils/SCCPSolver.cpp
index b78c7022b9be0..a03194ded5746 100644
--- a/llvm/lib/Transforms/Utils/SCCPSolver.cpp
+++ b/llvm/lib/Transforms/Utils/SCCPSolver.cpp
@@ -1433,8 +1433,12 @@ void SCCPInstVisitor::visitCastInst(CastInst &I) {
OpSt.asConstantRange(I.getSrcTy(), /*UndefAllowed=*/false);
Type *DestTy = I.getDestTy();
- ConstantRange Res =
- OpRange.castOp(I.getOpcode(), DestTy->getScalarSizeInBits());
+ ConstantRange Res = ConstantRange::getEmpty(DestTy->getScalarSizeInBits());
+ if (auto *Trunc = dyn_cast<TruncInst>(&I))
+ Res = OpRange.truncate(DestTy->getScalarSizeInBits(),
+ Trunc->getNoWrapKind());
+ else
+ Res = OpRange.castOp(I.getOpcode(), DestTy->getScalarSizeInBits());
mergeInValue(LV, &I, ValueLatticeElement::getRange(Res));
} else
markOverdefined(&I);
diff --git a/llvm/test/Transforms/SCCP/conditions-ranges.ll b/llvm/test/Transforms/SCCP/conditions-ranges.ll
index bb3764160f724..c2270287c7170 100644
--- a/llvm/test/Transforms/SCCP/conditions-ranges.ll
+++ b/llvm/test/Transforms/SCCP/conditions-ranges.ll
@@ -1436,6 +1436,47 @@ if.end:
ret i32 0
}
+define void @trunc_nuw_i1_dominating_icmp_ne_0(i8 %x) {
+; CHECK-LABEL: @trunc_nuw_i1_dominating_icmp_ne_0(
+; CHECK-NEXT: [[ICMP:%.*]] = icmp ne i8 [[X:%.*]], 0
+; CHECK-NEXT: br i1 [[ICMP]], label [[BB1:%.*]], label [[BB2:%.*]]
+; CHECK: bb1:
+; CHECK-NEXT: call void @use(i1 true)
+; CHECK-NEXT: ret void
+; CHECK: bb2:
+; CHECK-NEXT: ret void
+;
+ %icmp = icmp ne i8 %x, 0
+ br i1 %icmp, label %bb1, label %bb2
+bb1:
+ %c1 = trunc nuw i8 %x to i1
+ call void @use(i1 %c1)
+ ret void
+bb2:
+ ret void
+}
+
+define void @neg_trunc_i1_dominating_icmp_ne_0(i8 %x) {
+; CHECK-LABEL: @neg_trunc_i1_dominating_icmp_ne_0(
+; CHECK-NEXT: [[ICMP:%.*]] = icmp ne i8 [[X:%.*]], 0
+; CHECK-NEXT: br i1 [[ICMP]], label [[BB1:%.*]], label [[BB2:%.*]]
+; CHECK: bb1:
+; CHECK-NEXT: [[C1:%.*]] = trunc i8 [[X]] to i1
+; CHECK-NEXT: call void @use(i1 [[C1]])
+; CHECK-NEXT: ret void
+; CHECK: bb2:
+; CHECK-NEXT: ret void
+;
+ %icmp = icmp ne i8 %x, 0
+ br i1 %icmp, label %bb1, label %bb2
+bb1:
+ %c1 = trunc i8 %x to i1
+ call void @use(i1 %c1)
+ ret void
+bb2:
+ ret void
+}
+
define i1 @ptr_icmp_data_layout() {
; CHECK-LABEL: @ptr_icmp_data_layout(
; CHECK-NEXT: ret i1 false
diff --git a/llvm/unittests/IR/ConstantRangeTest.cpp b/llvm/unittests/IR/ConstantRangeTest.cpp
index bcb5d498c8cb9..6bb6903bae703 100644
--- a/llvm/unittests/IR/ConstantRangeTest.cpp
+++ b/llvm/unittests/IR/ConstantRangeTest.cpp
@@ -451,6 +451,45 @@ TEST_F(ConstantRangeTest, Trunc) {
EXPECT_EQ(SevenOne.truncate(2), ConstantRange(APInt(2, 3), APInt(2, 1)));
}
+TEST_F(ConstantRangeTest, TruncNuw) {
+ auto Range = [](unsigned NumBits, unsigned Lower, unsigned Upper) {
+ return ConstantRange(APInt(NumBits, Lower), APInt(NumBits, Upper));
+ };
+ // trunc([0, 4), 3->2) = full
+ EXPECT_TRUE(
+ Range(3, 0, 4).truncate(2, TruncInst::NoUnsignedWrap).isFullSet());
+ // trunc([0, 3), 3->2) = [0, 3)
+ EXPECT_EQ(Range(3, 0, 3).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 0, 3));
+ // trunc([1, 3), 3->2) = [1, 3)
+ EXPECT_EQ(Range(3, 1, 3).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 1, 3));
+ // trunc([1, 5), 3->2) = [1, 0)
+ EXPECT_EQ(Range(3, 1, 5).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 1, 0));
+ // trunc([4, 7), 3->2) = empty
+ EXPECT_TRUE(
+ Range(3, 4, 7).truncate(2, TruncInst::NoUnsignedWrap).isEmptySet());
+ // trunc([4, 0), 3->2) = empty
+ EXPECT_TRUE(
+ Range(3, 4, 0).truncate(2, TruncInst::NoUnsignedWrap).isEmptySet());
+ // trunc([4, 1), 3->2) = [0, 1)
+ EXPECT_EQ(Range(3, 4, 1).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 0, 1));
+ // trunc([3, 1), 3->2) = [3, 1)
+ EXPECT_EQ(Range(3, 3, 1).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 3, 1));
+ // trunc([3, 0), 3->2) = [3, 0)
+ EXPECT_EQ(Range(3, 3, 0).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 3, 0));
+ // trunc([1, 0), 2->1) = [1, 0)
+ EXPECT_EQ(Range(2, 1, 0).truncate(1, TruncInst::NoUnsignedWrap),
+ Range(1, 1, 0));
+ // trunc([2, 1), 2->1) = [0, 1)
+ EXPECT_EQ(Range(2, 2, 1).truncate(1, TruncInst::NoUnsignedWrap),
+ Range(1, 0, 1));
+}
+
TEST_F(ConstantRangeTest, ZExt) {
ConstantRange ZFull = Full.zeroExtend(20);
ConstantRange ZEmpty = Empty.zeroExtend(20);
|
|
@llvm/pr-subscribers-function-specialization Author: Andreas Jonson (andjo403) Changesproof: https://alive2.llvm.org/ce/z/_7PVxq part of changes needed for move of #151961 to SCCP Full diff: https://github.com/llvm/llvm-project/pull/152990.diff 5 Files Affected:
diff --git a/llvm/include/llvm/IR/ConstantRange.h b/llvm/include/llvm/IR/ConstantRange.h
index 9a6a9db65688a..4b2fda364fdf4 100644
--- a/llvm/include/llvm/IR/ConstantRange.h
+++ b/llvm/include/llvm/IR/ConstantRange.h
@@ -380,8 +380,9 @@ class [[nodiscard]] ConstantRange {
/// Return a new range in the specified integer type, which must be
/// strictly smaller than the current type. The returned range will
/// correspond to the possible range of values if the source range had been
- /// truncated to the specified type.
- LLVM_ABI ConstantRange truncate(uint32_t BitWidth) const;
+ /// truncated to the specified type with wrap type \p NoWrapKind.
+ LLVM_ABI ConstantRange truncate(uint32_t BitWidth,
+ unsigned NoWrapKind = 0) const;
/// Make this range have the bit width given by \p BitWidth. The
/// value is zero extended, truncated, or left alone to make it that width.
diff --git a/llvm/lib/IR/ConstantRange.cpp b/llvm/lib/IR/ConstantRange.cpp
index 2fcdbcc6a3db2..b1ea4bbd5ae6b 100644
--- a/llvm/lib/IR/ConstantRange.cpp
+++ b/llvm/lib/IR/ConstantRange.cpp
@@ -872,7 +872,8 @@ ConstantRange ConstantRange::signExtend(uint32_t DstTySize) const {
return ConstantRange(Lower.sext(DstTySize), Upper.sext(DstTySize));
}
-ConstantRange ConstantRange::truncate(uint32_t DstTySize) const {
+ConstantRange ConstantRange::truncate(uint32_t DstTySize,
+ unsigned NoWrapKind) const {
assert(getBitWidth() > DstTySize && "Not a value truncation");
if (isEmptySet())
return getEmpty(DstTySize);
@@ -886,22 +887,36 @@ ConstantRange ConstantRange::truncate(uint32_t DstTySize) const {
// We use the non-wrapped set code to analyze the [Lower, MaxValue) part, and
// then we do the union with [MaxValue, Upper)
if (isUpperWrapped()) {
- // If Upper is greater than or equal to MaxValue(DstTy), it covers the whole
- // truncated range.
- if (Upper.getActiveBits() > DstTySize || Upper.countr_one() == DstTySize)
+ // If Upper is greater than MaxValue(DstTy), it covers the whole truncated
+ // range.
+ if (Upper.getActiveBits() > DstTySize)
return getFull(DstTySize);
- Union = ConstantRange(APInt::getMaxValue(DstTySize),Upper.trunc(DstTySize));
- UpperDiv.setAllBits();
-
- // Union covers the MaxValue case, so return if the remaining range is just
- // MaxValue(DstTy).
- if (LowerDiv == UpperDiv)
- return Union;
+ // For nuw the two parts is: [0, Upper) \/ [Lower, MaxValue(DstTy) + 1]
+ if (NoWrapKind & TruncInst::NoUnsignedWrap) {
+ Union = ConstantRange(APInt::getZero(DstTySize), Upper.trunc(DstTySize));
+ UpperDiv = APInt::getOneBitSet(getBitWidth(), DstTySize);
+ } else {
+ // If Upper is equal to MaxValue(DstTy), it covers the whole truncated
+ // range.
+ if (Upper.countr_one() == DstTySize)
+ return getFull(DstTySize);
+ Union =
+ ConstantRange(APInt::getMaxValue(DstTySize), Upper.trunc(DstTySize));
+ UpperDiv.setAllBits();
+ // Union covers the MaxValue case, so return if the remaining range is
+ // just MaxValue(DstTy).
+ if (LowerDiv == UpperDiv)
+ return Union;
+ }
}
// Chop off the most significant bits that are past the destination bitwidth.
if (LowerDiv.getActiveBits() > DstTySize) {
+ // For trunc nuw if LowerDiv is greater than MaxValue(DstTy), the range is
+ // outside the whole truncated range.
+ if (NoWrapKind & TruncInst::NoUnsignedWrap)
+ return Union;
// Mask to just the signficant bits and subtract from LowerDiv/UpperDiv.
APInt Adjust = LowerDiv & APInt::getBitsSetFrom(getBitWidth(), DstTySize);
LowerDiv -= Adjust;
@@ -913,6 +928,10 @@ ConstantRange ConstantRange::truncate(uint32_t DstTySize) const {
return ConstantRange(LowerDiv.trunc(DstTySize),
UpperDiv.trunc(DstTySize)).unionWith(Union);
+ if (!LowerDiv.isZero() && NoWrapKind & TruncInst::NoUnsignedWrap)
+ return ConstantRange(LowerDiv.trunc(DstTySize), APInt::getZero(DstTySize))
+ .unionWith(Union);
+
// The truncated value wraps around. Check if we can do better than fullset.
if (UpperDivWidth == DstTySize + 1) {
// Clear the MSB so that UpperDiv wraps around.
diff --git a/llvm/lib/Transforms/Utils/SCCPSolver.cpp b/llvm/lib/Transforms/Utils/SCCPSolver.cpp
index b78c7022b9be0..a03194ded5746 100644
--- a/llvm/lib/Transforms/Utils/SCCPSolver.cpp
+++ b/llvm/lib/Transforms/Utils/SCCPSolver.cpp
@@ -1433,8 +1433,12 @@ void SCCPInstVisitor::visitCastInst(CastInst &I) {
OpSt.asConstantRange(I.getSrcTy(), /*UndefAllowed=*/false);
Type *DestTy = I.getDestTy();
- ConstantRange Res =
- OpRange.castOp(I.getOpcode(), DestTy->getScalarSizeInBits());
+ ConstantRange Res = ConstantRange::getEmpty(DestTy->getScalarSizeInBits());
+ if (auto *Trunc = dyn_cast<TruncInst>(&I))
+ Res = OpRange.truncate(DestTy->getScalarSizeInBits(),
+ Trunc->getNoWrapKind());
+ else
+ Res = OpRange.castOp(I.getOpcode(), DestTy->getScalarSizeInBits());
mergeInValue(LV, &I, ValueLatticeElement::getRange(Res));
} else
markOverdefined(&I);
diff --git a/llvm/test/Transforms/SCCP/conditions-ranges.ll b/llvm/test/Transforms/SCCP/conditions-ranges.ll
index bb3764160f724..c2270287c7170 100644
--- a/llvm/test/Transforms/SCCP/conditions-ranges.ll
+++ b/llvm/test/Transforms/SCCP/conditions-ranges.ll
@@ -1436,6 +1436,47 @@ if.end:
ret i32 0
}
+define void @trunc_nuw_i1_dominating_icmp_ne_0(i8 %x) {
+; CHECK-LABEL: @trunc_nuw_i1_dominating_icmp_ne_0(
+; CHECK-NEXT: [[ICMP:%.*]] = icmp ne i8 [[X:%.*]], 0
+; CHECK-NEXT: br i1 [[ICMP]], label [[BB1:%.*]], label [[BB2:%.*]]
+; CHECK: bb1:
+; CHECK-NEXT: call void @use(i1 true)
+; CHECK-NEXT: ret void
+; CHECK: bb2:
+; CHECK-NEXT: ret void
+;
+ %icmp = icmp ne i8 %x, 0
+ br i1 %icmp, label %bb1, label %bb2
+bb1:
+ %c1 = trunc nuw i8 %x to i1
+ call void @use(i1 %c1)
+ ret void
+bb2:
+ ret void
+}
+
+define void @neg_trunc_i1_dominating_icmp_ne_0(i8 %x) {
+; CHECK-LABEL: @neg_trunc_i1_dominating_icmp_ne_0(
+; CHECK-NEXT: [[ICMP:%.*]] = icmp ne i8 [[X:%.*]], 0
+; CHECK-NEXT: br i1 [[ICMP]], label [[BB1:%.*]], label [[BB2:%.*]]
+; CHECK: bb1:
+; CHECK-NEXT: [[C1:%.*]] = trunc i8 [[X]] to i1
+; CHECK-NEXT: call void @use(i1 [[C1]])
+; CHECK-NEXT: ret void
+; CHECK: bb2:
+; CHECK-NEXT: ret void
+;
+ %icmp = icmp ne i8 %x, 0
+ br i1 %icmp, label %bb1, label %bb2
+bb1:
+ %c1 = trunc i8 %x to i1
+ call void @use(i1 %c1)
+ ret void
+bb2:
+ ret void
+}
+
define i1 @ptr_icmp_data_layout() {
; CHECK-LABEL: @ptr_icmp_data_layout(
; CHECK-NEXT: ret i1 false
diff --git a/llvm/unittests/IR/ConstantRangeTest.cpp b/llvm/unittests/IR/ConstantRangeTest.cpp
index bcb5d498c8cb9..6bb6903bae703 100644
--- a/llvm/unittests/IR/ConstantRangeTest.cpp
+++ b/llvm/unittests/IR/ConstantRangeTest.cpp
@@ -451,6 +451,45 @@ TEST_F(ConstantRangeTest, Trunc) {
EXPECT_EQ(SevenOne.truncate(2), ConstantRange(APInt(2, 3), APInt(2, 1)));
}
+TEST_F(ConstantRangeTest, TruncNuw) {
+ auto Range = [](unsigned NumBits, unsigned Lower, unsigned Upper) {
+ return ConstantRange(APInt(NumBits, Lower), APInt(NumBits, Upper));
+ };
+ // trunc([0, 4), 3->2) = full
+ EXPECT_TRUE(
+ Range(3, 0, 4).truncate(2, TruncInst::NoUnsignedWrap).isFullSet());
+ // trunc([0, 3), 3->2) = [0, 3)
+ EXPECT_EQ(Range(3, 0, 3).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 0, 3));
+ // trunc([1, 3), 3->2) = [1, 3)
+ EXPECT_EQ(Range(3, 1, 3).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 1, 3));
+ // trunc([1, 5), 3->2) = [1, 0)
+ EXPECT_EQ(Range(3, 1, 5).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 1, 0));
+ // trunc([4, 7), 3->2) = empty
+ EXPECT_TRUE(
+ Range(3, 4, 7).truncate(2, TruncInst::NoUnsignedWrap).isEmptySet());
+ // trunc([4, 0), 3->2) = empty
+ EXPECT_TRUE(
+ Range(3, 4, 0).truncate(2, TruncInst::NoUnsignedWrap).isEmptySet());
+ // trunc([4, 1), 3->2) = [0, 1)
+ EXPECT_EQ(Range(3, 4, 1).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 0, 1));
+ // trunc([3, 1), 3->2) = [3, 1)
+ EXPECT_EQ(Range(3, 3, 1).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 3, 1));
+ // trunc([3, 0), 3->2) = [3, 0)
+ EXPECT_EQ(Range(3, 3, 0).truncate(2, TruncInst::NoUnsignedWrap),
+ Range(2, 3, 0));
+ // trunc([1, 0), 2->1) = [1, 0)
+ EXPECT_EQ(Range(2, 1, 0).truncate(1, TruncInst::NoUnsignedWrap),
+ Range(1, 1, 0));
+ // trunc([2, 1), 2->1) = [0, 1)
+ EXPECT_EQ(Range(2, 2, 1).truncate(1, TruncInst::NoUnsignedWrap),
+ Range(1, 0, 1));
+}
+
TEST_F(ConstantRangeTest, ZExt) {
ConstantRange ZFull = Full.zeroExtend(20);
ConstantRange ZEmpty = Empty.zeroExtend(20);
|
|
Together with the changes from #152988 llvm-opt-benchmark almost have the same improvements as #151961 but eg. bench/ruff-rs/optimized/8dg5gv1ul0w7vccunqd3ii3jp.ll from dtcxzyw/llvm-opt-benchmark@02089b6#diff-14dbc7fa1ad32f106f10c98325db165a9e400decefabbe288b5bd15a4ef70d7c is not not handled hmm seems like I hade updated lvi also when I run locally so hade more files updated so more that is not handled with this PR. |
nikic
reviewed
Aug 11, 2025
andjo403
force-pushed
the
truncNuwConstantRange
branch
from
403ff12 to
5535520
Compare
|
rebased on topp of #152988 |
nikic
reviewed
Aug 11, 2025
nikic
reviewed
Aug 11, 2025
andjo403
added a commit
that referenced
this pull request
Aug 12, 2025
andjo403
force-pushed
the
truncNuwConstantRange
branch
from
0f8fb96 to
d5cd785
Compare
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/55/builds/15537 Here is the relevant piece of the build log for the reference |
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.
proof: https://alive2.llvm.org/ce/z/_7PVxq
part of changes needed for move of #151961 to SCCP