[WebAssembly] Fold any/alltrue (setcc x, 0, eq/ne) to [not] any/alltrue x #144741

badumbatish · 2025-06-18T16:08:30Z

Fixes #50142, a miss of further vectorization, where we can only achieve zext (xor (any_true), -1).

Now in test case simd-setcc-reductions, it's converted to all_true.

Also fixes #145177, which is

all_true (setcc x, 0, eq) -> not any_true
any_true (setcc x, 0, ne) -> any_true
all_true (setcc x, 0, ne) -> all_true

github-actions · 2025-06-18T16:08:49Z

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 · 2025-06-18T16:09:21Z

@llvm/pr-subscribers-backend-webassembly

Author: jjasmine (badumbatish)

Changes

Fix a miss of further vectorization introduced in #50142 , where we can only achieve zext (xor (any_true), -1).

Now in test case issue50142, it's converted to all_true.

Full diff: https://github.com/llvm/llvm-project/pull/144741.diff

2 Files Affected:

(modified) llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp (+31)
(added) llvm/test/CodeGen/WebAssembly/issue50142.ll (+20)

diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
index 3cd923c0ba058..84a18f74867a5 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
@@ -19,6 +19,7 @@
 #include "WebAssemblyTargetMachine.h"
 #include "WebAssemblyUtilities.h"
 #include "llvm/CodeGen/CallingConvLower.h"
+#include "llvm/CodeGen/ISDOpcodes.h"
 #include "llvm/CodeGen/MachineFrameInfo.h"
 #include "llvm/CodeGen/MachineInstrBuilder.h"
 #include "llvm/CodeGen/MachineJumpTableInfo.h"
@@ -36,6 +37,7 @@
 #include "llvm/Support/KnownBits.h"
 #include "llvm/Support/MathExtras.h"
 #include "llvm/Target/TargetOptions.h"
+#include <iostream>
 using namespace llvm;
 
 #define DEBUG_TYPE "wasm-lower"
@@ -3248,6 +3250,35 @@ static SDValue performSETCCCombine(SDNode *N,
   ISD::CondCode Cond = cast<CondCodeSDNode>(N->getOperand(2))->get();
   SDLoc DL(N);
   EVT VT = N->getValueType(0);
+  //  N           LHS     LhsL        LhsLL    LhsLR   InnerCond RHS Cond
+  // setcc (iN (bitcast (setcc vNi1 (vNiY X), <vNiY 0>, eq)),     0, eq
+  // => all_true (vNi1 X)
+  if (DCI.isBeforeLegalize() && VT.isScalarInteger() && (Cond == ISD::SETEQ) &&
+      (isNullConstant(RHS)) && LHS->getOpcode() == ISD::BITCAST) {
+    SDValue LhsL = LHS.getOperand(0);
+    EVT LhsLType = LhsL.getValueType();
+    ISD::CondCode InnerCond = cast<CondCodeSDNode>(LhsL->getOperand(2))->get();
+    if (LhsL.getOpcode() == ISD::SETCC && InnerCond == ISD::SETEQ) {
+      SDValue LhsLL = LhsL.getOperand(0); // vNi1 X
+      SDValue LhsLR = LhsL.getOperand(1); // 0
+      unsigned NumElts = LhsLType.getVectorNumElements();
+      bool Vectorizable =
+          NumElts == 2 || NumElts == 4 || NumElts == 8 || NumElts == 16;
+      EVT Width = MVT::getIntegerVT(128 / NumElts);
+      // EVT LhsLLType = LhsLL.getValueType();
+
+      if (Vectorizable && LhsLR.getOpcode() == ISD::BUILD_VECTOR &&
+          LhsLType.isFixedLengthVector()) {
+        return DAG.getZExtOrTrunc(
+            DAG.getNode(
+                ISD::INTRINSIC_WO_CHAIN, DL, MVT::i32,
+                {DAG.getConstant(Intrinsic::wasm_alltrue, DL, MVT::i32),
+                 DAG.getSExtOrTrunc(LhsLL, DL,
+                                    LhsLType.changeVectorElementType(Width))}),
+            DL, MVT::i1);
+      }
+    }
+  }
 
   // setcc (iN (bitcast (vNi1 X))), 0, ne
   //   ==> any_true (vNi1 X)
diff --git a/llvm/test/CodeGen/WebAssembly/issue50142.ll b/llvm/test/CodeGen/WebAssembly/issue50142.ll
new file mode 100644
index 0000000000000..24ba941e76ee2
--- /dev/null
+++ b/llvm/test/CodeGen/WebAssembly/issue50142.ll
@@ -0,0 +1,20 @@
+; RUN: llc < %s -mtriple=wasm32-- -mattr=+simd128 | FileCheck --check-prefix=CHECK %s
+
+; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(argmem: read)
+define hidden range(i32 0, 2) i32 @all_true(ptr noundef readonly captures(none) %a) local_unnamed_addr #0 {
+; CHECK-LABEL: all_true:
+; CHECK:         .functype all_true (i32) -> (i32)
+; CHECK: local.get 0
+; CHECK-NEXT: v128.load 0:p2align=0
+; CHECK-NEXT: i8x16.all_true
+; CHECK-NEXT:    # fallthrough-return
+; CHECK-NEXT: end_function 
+entry:
+  %0 = load <16 x i8>, ptr %a, align 1
+  %.fr = freeze <16 x i8> %0
+  %1 = icmp eq <16 x i8> %.fr, zeroinitializer
+  %2 = bitcast <16 x i1> %1 to i16
+  %3 = icmp eq i16 %2, 0
+  %conv3 = zext i1 %3 to i32
+  ret i32 %conv3
+}

badumbatish · 2025-06-18T17:26:30Z

@tlively hi, I saw you're one of the issue filers for #50142, I'm not sure who's the code reviewers for the webassembly components to review my code. Can i get some pointers on who to tag?

dschuff · 2025-06-18T19:02:05Z

@tlively is probably good (unless he wants to delegate that off), also @sparker-arm has been working on wasm SIMD recently.

lukel97

Can you precommit the test in this PR? I.e. rebase this so there's one commit adding the test without the changes, and another adding the changes to WebAssemblyISelLowering.cpp and the changes to the test case. That way you can see the diff in the PR

llvm/test/CodeGen/WebAssembly/simd-setcc-reductions.ll

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp

badumbatish · 2025-06-19T23:32:46Z

i updated the code, in the case of 4xi16, the fold doesn't work since i was targeting setcc

define i32 @all_true_4_i16(<4 x i16> %v) {
; CHECK-LABEL: all_true_4_i16:
; CHECK:         .functype all_true_4_i16 (v128) -> (i32)
; CHECK-NEXT:  # %bb.0:
; CHECK-NEXT:    v128.const $push0=, 0, 0, 0, 0, 0, 0, 0, 0
; CHECK-NEXT:    i16x8.eq $push1=, $0, $pop0
; CHECK-NEXT:    i32x4.extend_low_i16x8_s $push2=, $pop1
; CHECK-NEXT:    v128.any_true $push3=, $pop2
; CHECK-NEXT:    i32.const $push4=, -1
; CHECK-NEXT:    i32.xor $push5=, $pop3, $pop4
; CHECK-NEXT:    i32.const $push6=, 1
; CHECK-NEXT:    i32.and $push7=, $pop5, $pop6
; CHECK-NEXT:    return $pop7
  %1 = icmp eq <4 x i16> %v, zeroinitializer
  %2 = bitcast <4 x i1> %1 to i4
  %3 = icmp eq i4 %2, 0
  %conv3 = zext i1 %3 to i32
  ret i32 %conv3
}

ok i just check the specs, haha there isn't any 4 x i16 in web assembly, will remove

lukel97

Thanks, this new approach looks good!

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp

lukel97

LGTM, thanks! Although we should also wait for a WebAssembly maintainer to approve this too (this isn't my neck of the woods in LLVM!)

Can you also update your PR title to not reference the issue, but instead something like "[WebAssembly] Combine any_true (setcc x, 0, eq) to not all_true"?

We should also follow up on this to also handle:

all_true (setcc x, 0, eq) -> not any_true
any_true (setcc x, 0, ne) -> any_true
all_true (setcc x, 0, ne) -> all_true

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp

badumbatish · 2025-06-23T17:31:29Z

really sorry about the extra ping @tlively , can I please get the PR either approved or reviewed?

I think I can reuse some of the code for issue #145177. I just want to make sure the code structure is good before I base the new PR ontop of this PR

badumbatish · 2025-06-26T19:55:34Z

I've refactored, and added support and test case for three more patterns. This will fix #145177

all_true (setcc x, 0, eq) -> not any_true
any_true (setcc x, 0, ne) -> any_true
all_true (setcc x, 0, ne) -> all_true

lukel97

Thanks for handling the new test cases! You should add "Fixes #50142, fixes #145177" to your PR description so it will automatically close those issues when this gets merged

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp

llvm/test/CodeGen/WebAssembly/simd-setcc-reductions.ll

This introduces the fold (any_true (setcc <X> 0, eq)) to (not (all_true)), allowing potential extra fold of (not (not ...)) Introduces test simd-setcc-reductions and readjusts simd-vecreduce-bool

Use SDPatternMatching and remove truncation. Also added 4xi64 case to reflect that.

all_true (setcc x, 0, eq) -> not any_true any_true (setcc x, 0, ne) -> any_true all_true (setcc x, 0, ne) -> all_true

lukel97

LGTM with the new changes, thanks! Gentle ping to the other WebAssembly maintainers for another review

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp

dschuff

Otherwise LGTM

clang-tools-extra/test/clang-doc/json/class.cpp

…e50142

dschuff · 2025-07-01T22:14:44Z

It looks like this one is ready to go too. is the commit message up-to-date?

badumbatish · 2025-07-01T22:16:11Z

Otherwise LGTM

just changed, tysm for merging

dschuff · 2025-07-01T22:19:36Z

just changed, tysm for merging

No problem, well done!

I'm just going to wait for the CI tests to finish, then I'll merge.

github-actions · 2025-07-01T22:27:57Z

@badumbatish 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!

lukel97 · 2025-07-01T22:52:57Z

Thanks for the review and the commit @dschuff, it's much appreciated!

llvm-ci · 2025-07-02T00:14:07Z

LLVM Buildbot has detected a new failure on builder clang-arm64-windows-msvc running on linaro-armv8-windows-msvc-04 while building llvm at step 6 "test-build-unified-tree-check-all".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/161/builds/6806

Here is the relevant piece of the build log for the reference

Step 6 (test-build-unified-tree-check-all) failure: 1200 seconds without output running [b'ninja', b'check-all'], attempting to kill
...
UNSUPPORTED: lld :: wasm/wrap.s (84340 of 84350)
UNSUPPORTED: mlgo-utils :: corpus/extract_ir_script.test (84341 of 84350)
UNSUPPORTED: mlgo-utils :: corpus/extract_ir_test.py (84342 of 84350)
UNSUPPORTED: mlgo-utils :: corpus/make_corpus_script.test (84343 of 84350)
UNSUPPORTED: mlgo-utils :: pytype.test (84344 of 84350)
PASS: lit :: xunit-output-report-failures-only.py (84345 of 84350)
PASS: lit :: selecting.py (84346 of 84350)
PASS: lit :: xfail-cl.py (84347 of 84350)
PASS: lit :: shtest-shell.py (84348 of 84350)
PASS: lit :: shtest-define.py (84349 of 84350)
command timed out: 1200 seconds without output running [b'ninja', b'check-all'], attempting to kill
program finished with exit code 1
elapsedTime=3858.992169

badumbatish · 2025-07-02T00:51:02Z

ruh roh

i digged into the build log and couldn't find any failures, just that its timing out on task 84350???

PASS: lit :: shtest-shell.py (84348 of 84350)
PASS: lit :: shtest-define.py (84349 of 84350)
....
command timed out: 1200 seconds without output running [b'ninja', b'check-all'], attempting to kill
program finished with exit code 1

how can i make sure this is not some kind of performance regression on my end?

lukel97 · 2025-07-02T01:19:47Z

ruh roh

i digged into the build log and couldn't find any failures, just that its timing out on task 84350???
PASS: lit :: shtest-shell.py (84348 of 84350)

PASS: lit :: shtest-define.py (84349 of 84350)

....

command timed out: 1200 seconds without output running [b'ninja', b'check-all'], attempting to kill

program finished with exit code 1
how can i make sure this is not some kind of performance regression on my end?

I wouldn't worry about that buildbot failure, it's likely just a flakey test/bot. There's quite a lot of false alarms as you'll soon come to see :)

llvmbot added the backend:WebAssembly label Jun 18, 2025

badumbatish force-pushed the issue50142 branch 2 times, most recently from 4d84fc3 to 79479c6 Compare June 18, 2025 17:04

badumbatish changed the title ~~Fix 50142~~ [WebAssembly] Fix missed optimization in 50142 Jun 18, 2025

badumbatish force-pushed the issue50142 branch from 79479c6 to 48595b3 Compare June 18, 2025 20:47

lukel97 reviewed Jun 18, 2025

View reviewed changes

llvm/test/CodeGen/WebAssembly/simd-setcc-reductions.ll Outdated Show resolved Hide resolved

llvm/test/CodeGen/WebAssembly/simd-setcc-reductions.ll Outdated Show resolved Hide resolved

llvm/test/CodeGen/WebAssembly/simd-setcc-reductions.ll Show resolved Hide resolved

badumbatish force-pushed the issue50142 branch from 48595b3 to 9d38a54 Compare June 18, 2025 23:42

lukel97 reviewed Jun 19, 2025

View reviewed changes

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

badumbatish force-pushed the issue50142 branch from 9d38a54 to 185ae46 Compare June 19, 2025 23:28

badumbatish force-pushed the issue50142 branch from 185ae46 to 80e558c Compare June 19, 2025 23:38

lukel97 reviewed Jun 20, 2025

View reviewed changes

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

lukel97 requested a review from tlively June 20, 2025 22:44

badumbatish force-pushed the issue50142 branch from a1977d2 to d74dfac Compare June 20, 2025 23:06

lukel97 approved these changes Jun 20, 2025

View reviewed changes

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

badumbatish force-pushed the issue50142 branch from d74dfac to b9667b1 Compare June 21, 2025 16:41

badumbatish changed the title ~~[WebAssembly] Fix missed optimization in 50142~~ [WebAssembly] Combine any_true (setcc x, 0, eq) to not all_true Jun 21, 2025

badumbatish mentioned this pull request Jun 21, 2025

[WebAssembly] Missed optimization in any_true/all_true #145177

Closed

badumbatish force-pushed the issue50142 branch from b9667b1 to 0a086da Compare June 26, 2025 19:54

badumbatish changed the title ~~[WebAssembly] Combine any_true (setcc x, 0, eq) to not all_true~~ [WebAssembly] Fold any/alltrue (setcc x, 0, eq/ne) Jun 27, 2025

badumbatish changed the title ~~[WebAssembly] Fold any/alltrue (setcc x, 0, eq/ne)~~ [WebAssembly] Fold any/alltrue (setcc x, 0, eq/ne) to [not] any/alltrue x Jun 27, 2025

lukel97 reviewed Jun 30, 2025

View reviewed changes

badumbatish force-pushed the issue50142 branch from 0a086da to 6cbded9 Compare June 30, 2025 18:36

badumbatish and others added 5 commits June 30, 2025 11:50

Precommit missed optimization test for llvm#50142

8d963b7

Fix issue 50142 by adding AnyTrueCombine

4415235

This introduces the fold (any_true (setcc <X> 0, eq)) to (not (all_true)), allowing potential extra fold of (not (not ...)) Introduces test simd-setcc-reductions and readjusts simd-vecreduce-bool

Use SDPatternMatching and remove truncate...

dd1bc67

Use SDPatternMatching and remove truncation. Also added 4xi64 case to reflect that.

Precommit test to add 3 more any/all true patterns

c389843

[WebAssembly] Add 3 more optimization for any/all

3feb8d0

all_true (setcc x, 0, eq) -> not any_true any_true (setcc x, 0, ne) -> any_true all_true (setcc x, 0, ne) -> all_true

lukel97 approved these changes Jun 30, 2025

View reviewed changes

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp Outdated Show resolved Hide resolved

lukel97 requested a review from sparker-arm June 30, 2025 21:22

[WebAssembly] Fix nit in PR 144741

8cafed8

dschuff approved these changes Jul 1, 2025

View reviewed changes

clang-tools-extra/test/clang-doc/json/class.cpp Outdated Show resolved Hide resolved

Merge branch 'main' of https://github.com/llvm/llvm-project into issu…

4236431

…e50142

badumbatish force-pushed the issue50142 branch from ecf1cdf to 4236431 Compare July 1, 2025 21:17

dschuff merged commit e9c9f8f into llvm:main Jul 1, 2025
7 checks passed

[WebAssembly] Fold any/alltrue (setcc x, 0, eq/ne) to [not] any/alltrue x #144741

[WebAssembly] Fold any/alltrue (setcc x, 0, eq/ne) to [not] any/alltrue x #144741

Conversation

badumbatish commented Jun 18, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Jun 18, 2025

Uh oh!

llvmbot commented Jun 18, 2025

Uh oh!

badumbatish commented Jun 18, 2025

Uh oh!

dschuff commented Jun 18, 2025

Uh oh!

lukel97 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

badumbatish commented Jun 19, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lukel97 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

lukel97 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

badumbatish commented Jun 23, 2025

Uh oh!

badumbatish commented Jun 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lukel97 left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

lukel97 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

dschuff left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

dschuff commented Jul 1, 2025

Uh oh!

badumbatish commented Jul 1, 2025

Uh oh!

dschuff commented Jul 1, 2025

Uh oh!

Uh oh!

github-actions bot commented Jul 1, 2025

Uh oh!

lukel97 commented Jul 1, 2025

Uh oh!

llvm-ci commented Jul 2, 2025

Uh oh!

badumbatish commented Jul 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lukel97 commented Jul 2, 2025

Uh oh!

Uh oh!

badumbatish commented Jun 18, 2025 •

edited

Loading

badumbatish commented Jun 19, 2025 •

edited

Loading

badumbatish commented Jun 26, 2025 •

edited

Loading

lukel97 left a comment •

edited

Loading

badumbatish commented Jul 2, 2025 •

edited

Loading