[PATCH] D150991: [LV] Use Legal::isUniform to detect uniform pointers.

[Florian Hahn via Phabricator via llvm-commits]

fhahn updated this revision to Diff 526573.
fhahn added a comment.

Updated to use alternative fix, thanks!


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D150991/new/

https://reviews.llvm.org/D150991

Files:
  llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
  llvm/test/Transforms/LoopVectorize/AArch64/pr60831-sve-inv-store-crash.ll
  llvm/test/Transforms/LoopVectorize/AArch64/sve-inv-store.ll
  llvm/test/Transforms/LoopVectorize/skeleton-lcssa-crash.ll


Index: llvm/test/Transforms/LoopVectorize/skeleton-lcssa-crash.ll
===================================================================
--- llvm/test/Transforms/LoopVectorize/skeleton-lcssa-crash.ll
+++ llvm/test/Transforms/LoopVectorize/skeleton-lcssa-crash.ll
@@ -50,13 +50,11 @@
 ; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i16, ptr [[TMP5]], i32 0
 ; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x i16>, ptr [[TMP6]], align 2, !alias.scope !0
 ; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i16, ptr [[L_2]], i64 0
-; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x i16> [[WIDE_LOAD]], i32 0
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x i16> [[WIDE_LOAD]], i32 1
 ; CHECK-NEXT:    store i16 [[TMP8]], ptr [[TMP7]], align 2, !alias.scope !3, !noalias !0
-; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x i16> [[WIDE_LOAD]], i32 1
-; CHECK-NEXT:    store i16 [[TMP9]], ptr [[TMP7]], align 2, !alias.scope !3, !noalias !0
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
-; CHECK-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT_LOOPEXIT:%.*]], label [[SCALAR_PH]]
Index: llvm/test/Transforms/LoopVectorize/AArch64/sve-inv-store.ll
===================================================================
--- llvm/test/Transforms/LoopVectorize/AArch64/sve-inv-store.ll
+++ llvm/test/Transforms/LoopVectorize/AArch64/sve-inv-store.ll
@@ -166,19 +166,16 @@
 ; CHECK-NEXT:    [[TMP10:%.*]] = add i64 [[INDEX]], 2
 ; CHECK-NEXT:    [[TMP11:%.*]] = add i64 [[INDEX]], 3
 ; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr i8, ptr [[STORE_DEST:%.*]], i64 0
-; CHECK-NEXT:    [[DOTSPLATINSERT1:%.*]] = insertelement <vscale x 4 x ptr> poison, ptr [[TMP12]], i64 0
-; CHECK-NEXT:    [[DOTSPLAT2:%.*]] = shufflevector <vscale x 4 x ptr> [[DOTSPLATINSERT1]], <vscale x 4 x ptr> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[TMP13:%.*]] = call i32 @llvm.vscale.i32()
 ; CHECK-NEXT:    [[TMP14:%.*]] = mul i32 [[TMP13]], 4
 ; CHECK-NEXT:    [[TMP15:%.*]] = sub i32 [[TMP14]], 1
 ; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <vscale x 4 x i64> [[TMP7]], i32 [[TMP15]]
-; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <vscale x 4 x ptr> [[DOTSPLAT2]], i32 0
-; CHECK-NEXT:    store i64 [[TMP16]], ptr [[TMP17]], align 1
-; CHECK-NEXT:    [[TMP18:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT:    [[TMP19:%.*]] = mul i64 [[TMP18]], 4
-; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP19]]
-; CHECK-NEXT:    [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT:    br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
+; CHECK-NEXT:    store i64 [[TMP16]], ptr [[TMP12]], align 1
+; CHECK-NEXT:    [[TMP17:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP18:%.*]] = mul i64 [[TMP17]], 4
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP18]]
+; CHECK-NEXT:    [[TMP19:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 100, [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
Index: llvm/test/Transforms/LoopVectorize/AArch64/pr60831-sve-inv-store-crash.ll
===================================================================
--- llvm/test/Transforms/LoopVectorize/AArch64/pr60831-sve-inv-store-crash.ll
+++ llvm/test/Transforms/LoopVectorize/AArch64/pr60831-sve-inv-store-crash.ll
@@ -3,9 +3,6 @@
 
 target triple = "aarch64-unknown-linux-gnu"
 
-; REQUIRES: asserts
-; XFAIL: *
-
 ; Test cases for PR60831.
 
 define void @test_invar_gep(ptr %dst) #0 {
Index: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
===================================================================
--- llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -4703,7 +4703,8 @@
   auto isVectorizedMemAccessUse = [&](Instruction *I, Value *Ptr) -> bool {
     if (isa<StoreInst>(I) && I->getOperand(0) == Ptr)
       return false;
-    return getLoadStorePointerOperand(I) == Ptr && isUniformDecision(I, VF);
+    return getLoadStorePointerOperand(I) == Ptr &&
+           (isUniformDecision(I, VF) || Legal->isUniform(Ptr));
   };
 
   // Holds a list of values which are known to have at least one uniform use.
@@ -4749,10 +4750,8 @@
       if (isUniformMemOpUse(&I))
         addToWorklistIfAllowed(&I);
 
-      if (isVectorizedMemAccessUse(&I, Ptr)) {
-        assert(isUniformDecision(&I, VF) && "consistency check");
+      if (isVectorizedMemAccessUse(&I, Ptr))
         HasUniformUse.insert(Ptr);
-      }
     }
 
   // Add to the worklist any operands which have *only* uniform (e.g. lane 0


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D150991.526573.patch
Type: text/x-patch
Size: 5238 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230530/6bae6984/attachment.bin>