[X86] X86ISelDAGToDAG - don't let ADD/SUB(X,1) -> SUB/ADD(X,-1) constant fold

[RKSimon]

Extension to #168726 - ensure we peek through bitcasts to look for constants (as constant folding will)

DAG should have constant folded this, but we're still fighting the lack of proper topological sorting.

Fixes #169205

[llvmbot]

@llvm/pr-subscribers-backend-x86

Author: Simon Pilgrim (RKSimon)

Changes

Extension to #168726 - ensure we peek through bitcasts to look for constants (as constant folding will)

DAG should have constant folded this, but we're still fighting the lack of proper topological sorting.

Fixes #169205

---

Full diff: https://github.com/llvm/llvm-project/pull/169217.diff

2 Files Affected:

• (modified) llvm/lib/Target/X86/X86ISelDAGToDAG.cpp (+2-1)
• (added) llvm/test/CodeGen/X86/pr169205.ll (+23)

diff --git a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
index 2b5a63aa66926..e7903a72d85bb 100644
--- a/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
+++ b/llvm/lib/Target/X86/X86ISelDAGToDAG.cpp
@@ -1004,7 +1004,8 @@ void X86DAGToDAGISel::PreprocessISelDAG() {
     if ((N->getOpcode() == ISD::ADD || N->getOpcode() == ISD::SUB) &&
         N->getSimpleValueType(0).isVector() && !mayPreventLoadFold()) {
       APInt SplatVal;
-      if (!ISD::isBuildVectorOfConstantSDNodes(N->getOperand(0).getNode()) &&
+      if (!ISD::isBuildVectorOfConstantSDNodes(
+              peekThroughBitcasts(N->getOperand(0)).getNode()) &&
           X86::isConstantSplat(N->getOperand(1), SplatVal) &&
           SplatVal.isOne()) {
         SDLoc DL(N);
diff --git a/llvm/test/CodeGen/X86/pr169205.ll b/llvm/test/CodeGen/X86/pr169205.ll
new file mode 100644
index 0000000000000..1416102d6ba77
--- /dev/null
+++ b/llvm/test/CodeGen/X86/pr169205.ll
@@ -0,0 +1,23 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-- -mcpu=x86-64    | FileCheck %s --check-prefixes=SSE
+; RUN: llc < %s -mtriple=x86_64-- -mcpu=x86-64-v2 | FileCheck %s --check-prefixes=SSE
+; RUN: llc < %s -mtriple=x86_64-- -mcpu=x86-64-v3 | FileCheck %s --check-prefixes=AVX
+; RUN: llc < %s -mtriple=x86_64-- -mcpu=x86-64-v4 | FileCheck %s --check-prefixes=AVX
+
+define <4 x i16> @PR169205() {
+; SSE-LABEL: PR169205:
+; SSE:       # %bb.0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [1,1,1,1,u,u,u,u]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: PR169205:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
+; AVX-NEXT:    vpaddw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX-NEXT:    retq
+  %avg = tail call <16 x i8> @llvm.x86.sse2.pavg.b(<16 x i8> zeroinitializer, <16 x i8> zeroinitializer)
+  %shuffle24 = shufflevector <16 x i8> %avg, <16 x i8> zeroinitializer, <4 x i32> <i32 2, i32 4, i32 9, i32 9>
+  %conv25 = zext <4 x i8> %shuffle24 to <4 x i16>
+  %not.neg = add <4 x i16> %conv25, splat (i16 1)
+  ret <4 x i16> %not.neg
+}

[llvm-ci]

LLVM Buildbot has detected a new failure on builder llvm-clang-x86_64-expensive-checks-ubuntu running on as-builder-4 while building llvm at step 7 "test-check-all".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/187/builds/13883

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

Step 7 (test-check-all) failure: Test just built components: check-all completed (failure)
******************** TEST 'LLVM :: tools/llvm-exegesis/RISCV/rvv/filter.test' FAILED ********************
Exit Code: 2

Command Output (stdout):
--
# RUN: at line 1
/home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llvm-exegesis -mtriple=riscv64 -mcpu=sifive-x280 -benchmark-phase=assemble-measured-code --mode=inverse_throughput --opcode-name=PseudoVNCLIPU_WX_M1_MASK     --riscv-filter-config='vtype = {VXRM: rod, AVL: VLMAX, SEW: e(8|16), Policy: ta/mu}' --max-configs-per-opcode=1000 --min-instructions=10 | /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/tools/llvm-exegesis/RISCV/rvv/filter.test
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/llvm-exegesis -mtriple=riscv64 -mcpu=sifive-x280 -benchmark-phase=assemble-measured-code --mode=inverse_throughput --opcode-name=PseudoVNCLIPU_WX_M1_MASK '--riscv-filter-config=vtype = {VXRM: rod, AVL: VLMAX, SEW: e(8|16), Policy: ta/mu}' --max-configs-per-opcode=1000 --min-instructions=10
# .---command stderr------------
# | PseudoVNCLIPU_WX_M1_MASK: Failed to produce any snippet via: instruction has tied variables, avoiding Read-After-Write issue, picking random def and use registers not aliasing each other, for uses, one unique register for each position
# `-----------------------------
# executed command: /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/tools/llvm-exegesis/RISCV/rvv/filter.test
# .---command stderr------------
# | FileCheck error: '<stdin>' is empty.
# | FileCheck command line:  /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/build/bin/FileCheck /home/buildbot/worker/as-builder-4/ramdisk/expensive-checks/llvm-project/llvm/test/tools/llvm-exegesis/RISCV/rvv/filter.test
# `-----------------------------
# error: command failed with exit status: 2

--

********************

[llvm-ci]

LLVM Buildbot has detected a new failure on builder llvm-nvptx-nvidia-win running on as-builder-8 while building llvm at step 7 "test-build-unified-tree-check-llvm".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/54/builds/14951

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

Step 7 (test-build-unified-tree-check-llvm) failure: test (failure)
******************** TEST 'LLVM-Unit :: Support/./SupportTests.exe/79/105' FAILED ********************
Script(shard):
--
GTEST_OUTPUT=json:C:\buildbot\as-builder-8\llvm-nvptx-nvidia-win\build\unittests\Support\.\SupportTests.exe-LLVM-Unit-24684-79-105.json GTEST_SHUFFLE=0 GTEST_TOTAL_SHARDS=105 GTEST_SHARD_INDEX=79 C:\buildbot\as-builder-8\llvm-nvptx-nvidia-win\build\unittests\Support\.\SupportTests.exe
--


Note: This is test shard 80 of 105.

[==========] Running 16 tests from 16 test suites.

[----------] Global test environment set-up.

[----------] 1 test from BinaryStreamTest

[ RUN      ] BinaryStreamTest.DropOperations

[       OK ] BinaryStreamTest.DropOperations (0 ms)

[----------] 1 test from BinaryStreamTest (0 ms total)



[----------] 1 test from CommandLineTest

[ RUN      ] CommandLineTest.TokenizeAndMarkEOLs

[       OK ] CommandLineTest.TokenizeAndMarkEOLs (0 ms)

[----------] 1 test from CommandLineTest (0 ms total)



[----------] 1 test from DataExtractorTest

[ RUN      ] DataExtractorTest.LEB128_error

[       OK ] DataExtractorTest.LEB128_error (0 ms)

[----------] 1 test from DataExtractorTest (0 ms total)



[----------] 1 test from Error

[ RUN      ] Error.ForwardToExpected

[       OK ] Error.ForwardToExpected (0 ms)

[----------] 1 test from Error (0 ms total)
...