[InstCombine] Add fold for fabs(-x) -> fabs(x) #95627

VaibhavRumale · 2024-06-15T02:36:35Z

This patch folds fabs(-x) -> fabs(x)

Proofs: https://alive2.llvm.org/ce/z/gjzmgf

github-actions · 2024-06-15T02:36:53Z

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 @ followed by their GitHub username.

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.

llvmbot · 2024-06-15T02:37:20Z

@llvm/pr-subscribers-llvm-transforms

Author: Vaibhav (VaibhavRumale)

Changes

Here is my patch for #94170 for fold transformation for fabs(-x) -> fabs(x)

[InstCombine] Add tests for fold: fabs(-x) -> fabs(x)
[InstCombine] Add fold transformation: fabs(-x) -> fabs(x)

Proofs: https://alive2.llvm.org/ce/z/pGzLts

Full diff: https://github.com/llvm/llvm-project/pull/95627.diff

2 Files Affected:

(modified) llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp (+7)
(added) llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll (+138)

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
index 436cdbff75669..f91aea35d1aab 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
@@ -2511,6 +2511,13 @@ Instruction *InstCombinerImpl::visitCallInst(CallInst &CI) {
   }
   case Intrinsic::fabs: {
     Value *Cond, *TVal, *FVal;
+    Value* Arg = II->getArgOperand(0);
+    Value* X;
+    // fabs (-X) --> fabs (X)
+    if (match(Arg, m_FNeg(m_Value(X))))
+        return replaceInstUsesWith(
+            CI, Builder.CreateUnaryIntrinsic(Intrinsic::fabs, X));
+
     if (match(II->getArgOperand(0),
               m_Select(m_Value(Cond), m_Value(TVal), m_Value(FVal)))) {
       // fabs (select Cond, TrueC, FalseC) --> select Cond, AbsT, AbsF
diff --git a/llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll b/llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll
new file mode 100644
index 0000000000000..bc239f207ef42
--- /dev/null
+++ b/llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll
@@ -0,0 +1,138 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt -S -passes=instcombine %s | FileCheck %s
+
+define float @fabs_fneg_basic(float %x) {
+; CHECK-LABEL: define float @fabs_fneg_basic(
+; CHECK-SAME: float [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[FABS]]
+;
+  %neg = fneg float %x
+  %fabs = call float @llvm.fabs.f32(float %neg)
+  ret float %fabs
+}
+
+define <2 x float> @fabs_fneg_v2f32(<2 x float> %x) {
+; CHECK-LABEL: define <2 x float> @fabs_fneg_v2f32(
+; CHECK-SAME: <2 x float> [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call <2 x float> @llvm.fabs.v2f32(<2 x float> [[X]])
+; CHECK-NEXT:    ret <2 x float> [[FABS]]
+;
+  %neg = fneg <2 x float> %x
+  %fabs = call <2 x float> @llvm.fabs.v2f32(<2 x float> %neg)
+  ret <2 x float> %fabs
+}
+
+define double @fabs_fneg_f64(double %x) {
+; CHECK-LABEL: define double @fabs_fneg_f64(
+; CHECK-SAME: double [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call double @llvm.fabs.f64(double [[X]])
+; CHECK-NEXT:    ret double [[FABS]]
+;
+  %neg = fneg double %x
+  %fabs = call double @llvm.fabs.f64(double %neg)
+  ret double %fabs
+}
+
+define <4 x double> @fabs_fneg_v4f64(<4 x double> %x) {
+; CHECK-LABEL: define <4 x double> @fabs_fneg_v4f64(
+; CHECK-SAME: <4 x double> [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call <4 x double> @llvm.fabs.v4f64(<4 x double> [[X]])
+; CHECK-NEXT:    ret <4 x double> [[FABS]]
+;
+  %neg = fneg <4 x double> %x
+  %fabs = call <4 x double> @llvm.fabs.v4f64(<4 x double> %neg)
+  ret <4 x double> %fabs
+}
+
+define half @fabs_fneg_f16(half %x) {
+; CHECK-LABEL: define half @fabs_fneg_f16(
+; CHECK-SAME: half [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call half @llvm.fabs.f16(half [[X]])
+; CHECK-NEXT:    ret half [[FABS]]
+;
+  %neg = fneg half %x
+  %fabs = call half @llvm.fabs.f16(half %neg)
+  ret half %fabs
+}
+
+define float @fabs_fneg_nnan(float %x) {
+; CHECK-LABEL: define float @fabs_fneg_nnan(
+; CHECK-SAME: float [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[FABS]]
+;
+  %neg = fneg nnan float %x
+  %fabs = call nnan float @llvm.fabs.f32(float %neg)
+  ret float %fabs
+}
+
+define float @fabs_fneg_ninf(float %x) {
+; CHECK-LABEL: define float @fabs_fneg_ninf(
+; CHECK-SAME: float [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[FABS]]
+;
+  %neg = fneg ninf float %x
+  %fabs = call ninf float @llvm.fabs.f32(float %neg)
+  ret float %fabs
+}
+
+define float @fabs_fneg_no_fabs(float %x) {
+; CHECK-LABEL: define float @fabs_fneg_no_fabs(
+; CHECK-SAME: float [[X:%.*]]) {
+; CHECK-NEXT:    [[NEG:%.*]] = fneg float [[X]]
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %neg = fneg float %x
+  ret float %neg
+}
+
+define <2 x float> @fabs_fneg_splat_v2f32(<2 x float> %x) {
+; CHECK-LABEL: define <2 x float> @fabs_fneg_splat_v2f32(
+; CHECK-SAME: <2 x float> [[X:%.*]]) {
+; CHECK-NEXT:    ret <2 x float> <float 2.000000e+00, float 2.000000e+00>
+;
+  %neg = fneg <2 x float> <float -2.0, float -2.0>
+  %fabs = call <2 x float> @llvm.fabs.v2f32(<2 x float> %neg)
+  ret <2 x float> %fabs
+}
+
+
+define <2 x float> @fabs_fneg_splat_poison_v2f32(<2 x float> %x) {
+; CHECK-LABEL: define <2 x float> @fabs_fneg_splat_poison_v2f32(
+; CHECK-SAME: <2 x float> [[X:%.*]]) {
+; CHECK-NEXT:    [[FABS:%.*]] = call <2 x float> @llvm.fabs.v2f32(<2 x float> <float -2.000000e+00, float poison>)
+; CHECK-NEXT:    ret <2 x float> [[FABS]]
+;
+  %neg = fneg <2 x float> <float 2.0, float poison>
+  %fabs = call <2 x float> @llvm.fabs.v2f32(<2 x float> %neg)
+  ret <2 x float> %fabs
+}
+
+define <2 x float> @fabs_fneg_non_splat_v2f32(<2 x float> %x) {
+; CHECK-LABEL: define <2 x float> @fabs_fneg_non_splat_v2f32(
+; CHECK-SAME: <2 x float> [[X:%.*]]) {
+; CHECK-NEXT:    ret <2 x float> <float 2.000000e+00, float 3.000000e+00>
+;
+  %neg = fneg <2 x float> <float 2.0, float 3.0>
+  %fabs = call <2 x float> @llvm.fabs.v2f32(<2 x float> %neg)
+  ret <2 x float> %fabs
+}
+
+declare void @use(float)
+
+define float @fabs_fneg_multi_use(float %x) {
+; CHECK-LABEL: define float @fabs_fneg_multi_use(
+; CHECK-SAME: float [[X:%.*]]) {
+; CHECK-NEXT:    [[NEG:%.*]] = fneg float [[X]]
+; CHECK-NEXT:    call void @use(float [[NEG]])
+; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[FABS]]
+;
+  %neg = fneg float %x
+  call void @use(float %neg)
+  %fabs = call float @llvm.fabs.f32(float %neg)
+  ret float %fabs
+}
+

llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll

dtcxzyw · 2024-06-15T02:42:45Z

llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp

+    Value* Arg = II->getArgOperand(0);
+    Value* X;
+    // fabs (-X) --> fabs (X)
+    if (match(Arg, m_FNeg(m_Value(X))))


The comment #94183 (comment) is not resolved.

github-actions · 2024-06-15T04:41:53Z

✅ With the latest revision this PR passed the C/C++ code formatter.

llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll

llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp

llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll

arsenm · 2024-06-25T11:06:59Z

Reverse ping?

…with flags

dtcxzyw

LGTM. Thank you!

dtcxzyw · 2024-06-26T08:28:50Z

llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp

+    Value *Arg = II->getArgOperand(0);
+    Value *X;
+    // fabs (-X) --> fabs (X)
+    if (match(Arg, m_FNeg(m_Value(X)))) {


We can merge this fold with fabs(copysign(x, y)) -> fabs(x).

```suggestion if (match(Arg, m_CombineOr(m_FNeg(m_Value(X)), m_CopySign(m_Value(X), m_Value(Unused))))) {

VaibhavRumale · 2024-06-26T08:34:01Z

Reverse ping?

Sorry for the delay; I was busy with my graduation. All requested changes have been made. Thank you!

github-actions · 2024-06-27T18:10:50Z

@VaibhavRumale 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!

This patch folds `fabs(-x) -> fabs(x)` Closes llvm#94170 Proofs: https://alive2.llvm.org/ce/z/gjzmgf

VaibhavRumale added 2 commits June 14, 2024 18:45

[InstCombine] Add tests for fold: fabs(-x) -> fabs(x)

f310f62

[InstCombine] Add fold transformation: fabs(-x) -> fabs(x)

c8956e0

VaibhavRumale requested a review from nikic as a code owner June 15, 2024 02:36

llvmbot added the llvm:transforms label Jun 15, 2024

dtcxzyw requested review from arsenm and dtcxzyw June 15, 2024 02:39

VaibhavRumale mentioned this pull request Jun 15, 2024

[InstCombine] Add fold for fabs(-x) -> fabs(x) #94183

Closed

dtcxzyw reviewed Jun 15, 2024

View reviewed changes

[InstCombine] Use Arg variable in fabs select match

bc5a50f

arsenm reviewed Jun 15, 2024

View reviewed changes

llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll Show resolved Hide resolved

VaibhavRumale added 2 commits June 15, 2024 04:47

[InstCombine] Add tests with copysign instead of fneg

ae4ba74

[InstCombine] Ensure FMF preservation in fabs transformation

1c6e39f

dtcxzyw reviewed Jun 15, 2024

View reviewed changes

llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll Outdated Show resolved Hide resolved

arsenm reviewed Jun 17, 2024

View reviewed changes

llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp Outdated Show resolved Hide resolved

llvm/test/Transforms/InstCombine/fabs-fneg-fold.ll Outdated Show resolved Hide resolved

[InstCombine] Handle FMF in CreateUnaryIntrinsic, add copysign tests …

e9009f0

…with flags

dtcxzyw approved these changes Jun 26, 2024

View reviewed changes

arsenm approved these changes Jun 27, 2024

View reviewed changes

arsenm merged commit ea68668 into llvm:main Jun 27, 2024
7 checks passed

lravenclaw pushed a commit to lravenclaw/llvm-project that referenced this pull request Jul 3, 2024

[InstCombine] Add fold for fabs(-x) -> fabs(x) (llvm#95627)

c87554a

This patch folds `fabs(-x) -> fabs(x)` Closes llvm#94170 Proofs: https://alive2.llvm.org/ce/z/gjzmgf

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[InstCombine] Add fold for fabs(-x) -> fabs(x) #95627

[InstCombine] Add fold for fabs(-x) -> fabs(x) #95627

Uh oh!

VaibhavRumale commented Jun 15, 2024 •

edited by dtcxzyw

Loading

Uh oh!

github-actions bot commented Jun 15, 2024

Uh oh!

llvmbot commented Jun 15, 2024

Uh oh!

Uh oh!

dtcxzyw Jun 15, 2024

Uh oh!

github-actions bot commented Jun 15, 2024 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

arsenm commented Jun 25, 2024

Uh oh!

dtcxzyw left a comment

Uh oh!

dtcxzyw Jun 26, 2024

Uh oh!

VaibhavRumale commented Jun 26, 2024

Uh oh!

Uh oh!

github-actions bot commented Jun 27, 2024

Uh oh!

Uh oh!

[InstCombine] Add fold for fabs(-x) -> fabs(x) #95627

[InstCombine] Add fold for fabs(-x) -> fabs(x) #95627

Uh oh!

Conversation

VaibhavRumale commented Jun 15, 2024 • edited by dtcxzyw Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Jun 15, 2024

Uh oh!

llvmbot commented Jun 15, 2024

Uh oh!

Uh oh!

dtcxzyw Jun 15, 2024

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Jun 15, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

arsenm commented Jun 25, 2024

Uh oh!

dtcxzyw left a comment

Choose a reason for hiding this comment

Uh oh!

dtcxzyw Jun 26, 2024

Choose a reason for hiding this comment

Uh oh!

VaibhavRumale commented Jun 26, 2024

Uh oh!

Uh oh!

github-actions bot commented Jun 27, 2024

Uh oh!

Uh oh!

VaibhavRumale commented Jun 15, 2024 •

edited by dtcxzyw

Loading

github-actions bot commented Jun 15, 2024 •

edited

Loading