Skip to content

[LV][EVL] Remove metadata on EVL vectorized loops #155760

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 2 commits into from
Sep 22, 2025
Merged

[LV][EVL] Remove metadata on EVL vectorized loops #155760

merged 2 commits into from
Sep 22, 2025

Conversation

@arcbbb
Copy link
Contributor

@arcbbb arcbbb commented Aug 28, 2025 •
edited
Loading

This patch removes the metadata emission for EVL‑vectorized loops,
since there is no current in-tree consumer:

  1. after VPlan performs canonical IV replacement [VPlan] Convert EVL loops to variable-length stepping after dissolution #147222 and
  2. RISCV dropped EVLIndVarSimplifyPass [RISCV][EVL] Drop EVLIndVarSimplifyPass from the pipeline #151483, which was the only user of this metadata.

@arcbbb arcbbb requested review from fhahn, lukel97 and mshockwave August 28, 2025 06:22
@llvmbot
Copy link
Member

llvmbot commented Aug 28, 2025 •
edited
Loading

@llvm/pr-subscribers-vectorizers

@llvm/pr-subscribers-backend-risc-v

Author: Shih-Po Hung (arcbbb)

Changes

Following #151296 and #151483, introduce a switch to enable/disable metadata emission for EVL‑vectorized loops,
disabled by default.

Patch is 110.14 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/155760.diff

14 Files Affected:

  • (modified) llvm/lib/Transforms/Vectorize/LoopVectorize.cpp (+5-1)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll (+1)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll (+4-5)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll (+72-73)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll (+36-37)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-cast-intrinsics.ll (+72-73)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-cond-reduction.ll (+32-34)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-div.ll (+16-17)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-fixed-order-recurrence.ll (+20-21)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-intermediate-store.ll (+14-16)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-known-no-overflow.ll (+12-13)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/truncate-to-minimal-bitwidth-cost.ll (+7-8)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/truncate-to-minimal-bitwidth-evl-crash.ll (+4-5)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/type-info-cache-evl-crash.ll (+4-5)
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 9667b506e594f..604a3e1a85bae 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -402,6 +402,10 @@ static cl::opt<bool> EnableEarlyExitVectorization(
     cl::desc(
         "Enable vectorization of early exit loops with uncountable exits."));
 
+static cl::opt<bool> EmitMetadataForEVLLoops(
+    "emit-metadata-for-evl-loops", cl::init(false), cl::Hidden,
+    cl::desc("Emit metadata for EVL-vectorized loops."));
+
 // Likelyhood of bypassing the vectorized loop because there are zero trips left
 // after prolog. See `emitIterationCountCheck`.
 static constexpr uint32_t MinItersBypassWeights[] = {1, 127};
@@ -7479,7 +7483,7 @@ DenseMap<const SCEV *, Value *> LoopVectorizationPlanner::executePlan(
               return VI->getOpcode() == VPInstruction::ExplicitVectorLength;
             return false;
           });
-      if (IsEVLVectorized) {
+      if (EmitMetadataForEVLLoops && IsEVLVectorized) {
         LLVMContext &Context = L->getHeader()->getContext();
         MDNode *LoopID = L->getLoopID();
         auto *IsEVLVectorizedMD = MDNode::get(
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll b/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll
index 00521b1fd8546..a108424a8221c 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll
@@ -1,6 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ; RUN: opt -passes=loop-vectorize \
 ; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
+; RUN: -emit-metadata-for-evl-loops \
 ; RUN: -mtriple=riscv64 -mattr=+v -S < %s | FileCheck %s
 
 ; Make sure we do not vectorize a loop with a widened int induction.
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll b/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll
index d41d47a6448be..0a324cd61ce00 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll
@@ -74,7 +74,7 @@ define void @test(ptr %p, i64 %a, i8 %b) {
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i32 [[SHL_I32]] to i8
 ; CHECK-NEXT:    store i8 [[TRUNC]], ptr [[P]], align 1
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[IV]], 8
-; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_COND1]], label [[EXIT1]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_COND1]], label [[EXIT1]], !llvm.loop [[LOOP3:![0-9]+]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret void
 ;
@@ -106,9 +106,8 @@ exit:                                             ; preds = %for.body
   ret void
 }
 ;.
-; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]], [[META3:![0-9]+]]}
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
 ; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
-; CHECK: [[META2]] = !{!"llvm.loop.isvectorized.tailfoldingstyle", !"evl"}
-; CHECK: [[META3]] = !{!"llvm.loop.unroll.runtime.disable"}
-; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META3]], [[META1]]}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
 ;.
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll
index 38e78321a4f36..48a33c4be2c99 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll
@@ -59,7 +59,7 @@ define void @test_and(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP4:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -170,7 +170,7 @@ define void @test_or(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -185,7 +185,7 @@ define void @test_or(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP6:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -296,7 +296,7 @@ define void @test_xor(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -311,7 +311,7 @@ define void @test_xor(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP8:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -422,7 +422,7 @@ define void @test_shl(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -437,7 +437,7 @@ define void @test_shl(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP10:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -548,7 +548,7 @@ define void @test_lshr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -563,7 +563,7 @@ define void @test_lshr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP12:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -674,7 +674,7 @@ define void @test_ashr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -689,7 +689,7 @@ define void @test_ashr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP14:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP13:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -800,7 +800,7 @@ define void @test_add(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP15:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -815,7 +815,7 @@ define void @test_add(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP16:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP15:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -926,7 +926,7 @@ define void @test_sub(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP17:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP16:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -941,7 +941,7 @@ define void @test_sub(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP18:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP17:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1052,7 +1052,7 @@ define void @test_mul(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP19:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP18:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1067,7 +1067,7 @@ define void @test_mul(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP20:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP19:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1178,7 +1178,7 @@ define void @test_sdiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP21:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1193,7 +1193,7 @@ define void @test_sdiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP22:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP21:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1304,7 +1304,7 @@ define void @test_udiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP23:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1319,7 +1319,7 @@ define void @test_udiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP24:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP23:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1430,7 +1430,7 @@ define void @test_srem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP25:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP24:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1445,7 +1445,7 @@ define void @test_srem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP26:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP25:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1556,7 +1556,7 @@ define void @test_urem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP27:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1571,7 +1571,7 @@ define void @test_urem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP28:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP27:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1685,7 +1685,7 @@ define void @test_fadd(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP19]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP11]], [[TMP19]]
 ; IF-EVL-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP16]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP29:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP16]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP28:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1700,7 +1700,7 @@ define void @test_fadd(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store float [[TMP]], ptr [[ARRAYIDX1]], align 4
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-...
[truncated]

@llvmbot
Copy link
Member

llvmbot commented Aug 28, 2025

@llvm/pr-subscribers-llvm-transforms

Author: Shih-Po Hung (arcbbb)

Changes

Following #151296 and #151483, introduce a switch to enable/disable metadata emission for EVL‑vectorized loops,
disabled by default.

Patch is 110.14 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/155760.diff

14 Files Affected:

  • (modified) llvm/lib/Transforms/Vectorize/LoopVectorize.cpp (+5-1)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll (+1)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll (+4-5)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll (+72-73)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-call-intrinsics.ll (+36-37)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-cast-intrinsics.ll (+72-73)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-cond-reduction.ll (+32-34)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-div.ll (+16-17)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-fixed-order-recurrence.ll (+20-21)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-intermediate-store.ll (+14-16)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-known-no-overflow.ll (+12-13)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/truncate-to-minimal-bitwidth-cost.ll (+7-8)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/truncate-to-minimal-bitwidth-evl-crash.ll (+4-5)
  • (modified) llvm/test/Transforms/LoopVectorize/RISCV/type-info-cache-evl-crash.ll (+4-5)
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 9667b506e594f..604a3e1a85bae 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -402,6 +402,10 @@ static cl::opt<bool> EnableEarlyExitVectorization(
     cl::desc(
         "Enable vectorization of early exit loops with uncountable exits."));
 
+static cl::opt<bool> EmitMetadataForEVLLoops(
+    "emit-metadata-for-evl-loops", cl::init(false), cl::Hidden,
+    cl::desc("Emit metadata for EVL-vectorized loops."));
+
 // Likelyhood of bypassing the vectorized loop because there are zero trips left
 // after prolog. See `emitIterationCountCheck`.
 static constexpr uint32_t MinItersBypassWeights[] = {1, 127};
@@ -7479,7 +7483,7 @@ DenseMap<const SCEV *, Value *> LoopVectorizationPlanner::executePlan(
               return VI->getOpcode() == VPInstruction::ExplicitVectorLength;
             return false;
           });
-      if (IsEVLVectorized) {
+      if (EmitMetadataForEVLLoops && IsEVLVectorized) {
         LLVMContext &Context = L->getHeader()->getContext();
         MDNode *LoopID = L->getLoopID();
         auto *IsEVLVectorizedMD = MDNode::get(
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll b/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll
index 00521b1fd8546..a108424a8221c 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/evl-compatible-loops.ll
@@ -1,6 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ; RUN: opt -passes=loop-vectorize \
 ; RUN: -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
+; RUN: -emit-metadata-for-evl-loops \
 ; RUN: -mtriple=riscv64 -mattr=+v -S < %s | FileCheck %s
 
 ; Make sure we do not vectorize a loop with a widened int induction.
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll b/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll
index d41d47a6448be..0a324cd61ce00 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/pr88802.ll
@@ -74,7 +74,7 @@ define void @test(ptr %p, i64 %a, i8 %b) {
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i32 [[SHL_I32]] to i8
 ; CHECK-NEXT:    store i8 [[TRUNC]], ptr [[P]], align 1
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[IV]], 8
-; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_COND1]], label [[EXIT1]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_COND1]], label [[EXIT1]], !llvm.loop [[LOOP3:![0-9]+]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret void
 ;
@@ -106,9 +106,8 @@ exit:                                             ; preds = %for.body
   ret void
 }
 ;.
-; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]], [[META3:![0-9]+]]}
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
 ; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
-; CHECK: [[META2]] = !{!"llvm.loop.isvectorized.tailfoldingstyle", !"evl"}
-; CHECK: [[META3]] = !{!"llvm.loop.unroll.runtime.disable"}
-; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META3]], [[META1]]}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
 ;.
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll
index 38e78321a4f36..48a33c4be2c99 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-bin-unary-ops-args.ll
@@ -59,7 +59,7 @@ define void @test_and(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP4:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -170,7 +170,7 @@ define void @test_or(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -185,7 +185,7 @@ define void @test_or(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP6:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -296,7 +296,7 @@ define void @test_xor(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -311,7 +311,7 @@ define void @test_xor(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP8:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -422,7 +422,7 @@ define void @test_shl(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -437,7 +437,7 @@ define void @test_shl(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP10:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -548,7 +548,7 @@ define void @test_lshr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -563,7 +563,7 @@ define void @test_lshr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP12:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -674,7 +674,7 @@ define void @test_ashr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -689,7 +689,7 @@ define void @test_ashr(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP14:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP13:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -800,7 +800,7 @@ define void @test_add(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP15:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -815,7 +815,7 @@ define void @test_add(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP16:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP15:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -926,7 +926,7 @@ define void @test_sub(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP17:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP16:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -941,7 +941,7 @@ define void @test_sub(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP18:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP17:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1052,7 +1052,7 @@ define void @test_mul(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP19:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP18:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1067,7 +1067,7 @@ define void @test_mul(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP20:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP19:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1178,7 +1178,7 @@ define void @test_sdiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP21:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1193,7 +1193,7 @@ define void @test_sdiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP22:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP21:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1304,7 +1304,7 @@ define void @test_udiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP23:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1319,7 +1319,7 @@ define void @test_udiv(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP24:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP23:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1430,7 +1430,7 @@ define void @test_srem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP25:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP24:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1445,7 +1445,7 @@ define void @test_srem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP26:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP25:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1556,7 +1556,7 @@ define void @test_urem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP18]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP10]], [[TMP18]]
 ; IF-EVL-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP27:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1571,7 +1571,7 @@ define void @test_urem(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store i8 [[TMP]], ptr [[ARRAYIDX1]], align 1
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP28:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[DOTNOT]], label %[[FINISH_LOOPEXIT]], label %[[LOOP]], !llvm.loop [[LOOP27:![0-9]+]]
 ; IF-EVL:       [[FINISH_LOOPEXIT]]:
 ; IF-EVL-NEXT:    ret void
 ;
@@ -1685,7 +1685,7 @@ define void @test_fadd(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add nuw i64 [[TMP19]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[TMP11]], [[TMP19]]
 ; IF-EVL-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], 100
-; IF-EVL-NEXT:    br i1 [[TMP16]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP29:![0-9]+]]
+; IF-EVL-NEXT:    br i1 [[TMP16]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP28:![0-9]+]]
 ; IF-EVL:       [[MIDDLE_BLOCK]]:
 ; IF-EVL-NEXT:    br label %[[FINISH_LOOPEXIT:.*]]
 ; IF-EVL:       [[SCALAR_PH]]:
@@ -1700,7 +1700,7 @@ define void @test_fadd(ptr nocapture %a, ptr nocapture readonly %b) {
 ; IF-EVL-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[LEN]]
 ; IF-EVL-NEXT:    store float [[TMP]], ptr [[ARRAYIDX1]], align 4
 ; IF-EVL-NEXT:    [[DOTNOT:%.*]] = icmp eq i64 [[DEC]], 100
-; IF-...
[truncated]

@arcbbb arcbbb force-pushed the disable-evl-metadata branch from 9f9d38a to 45d2ffb Compare August 28, 2025 07:02
fhahn
fhahn reviewed Aug 28, 2025
View reviewed changes
Copy link
Contributor

@fhahn fhahn left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you add a rational for needing an option for this and why the default should be false?

My understanding is that the metadata is used somewhere and there should be no harm from adding it for all EVL vectorized loops.

@arcbbb
Copy link
Contributor Author

arcbbb commented Sep 1, 2025

Could you add a rational for needing an option for this and why the default should be false?

My understanding is that the metadata is used somewhere and there should be no harm from adding it for all EVL vectorized loops.

Thanks. I’ve updated the PR description to include the rationale.

@lukel97
Copy link
Contributor

lukel97 commented Sep 1, 2025

I'm not sure if there's much point in adding a cl::opt. Since nothing uses the metadata after #151483, and there's no flag to run EVLIndVarSimplifyPass anyway. So I don't think anyone is going to end up using it anyway.

Should we just go ahead and remove the metadata entirely?

fhahn
fhahn reviewed Sep 1, 2025
View reviewed changes
Copy link
Contributor

@fhahn fhahn left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If it is not used in-tree, we should remove it instead of keeping dead code around

@lukel97
Copy link
Contributor

lukel97 commented Sep 16, 2025

Reverse ping

@arcbbb arcbbb force-pushed the disable-evl-metadata branch from 45d2ffb to 71f7787 Compare September 22, 2025 09:37
@arcbbb arcbbb changed the title [LV][EVL] Gate EVL loop metadata emission behind a cl::opt [LV][EVL] Remove metadata on EVL vectorized loops Sep 22, 2025
@arcbbb
Copy link
Contributor Author

arcbbb commented Sep 22, 2025

Updated as suggested! Thanks!

lukel97
lukel97 approved these changes Sep 22, 2025
View reviewed changes
Copy link
Contributor

@lukel97 lukel97 left a comment •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM thanks!

fhahn
fhahn approved these changes Sep 22, 2025
View reviewed changes
Copy link
Contributor

@fhahn fhahn left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thanks. Very nice to see it’s no longer needed

@arcbbb arcbbb merged commit 0d22f83 into llvm:main Sep 22, 2025
9 checks passed
@arcbbb arcbbb deleted the disable-evl-metadata branch September 22, 2025 23:39
@arcbbb arcbbb mentioned this pull request Sep 24, 2025
Closed
lukel97 added a commit to lukel97/llvm-project that referenced this pull request Sep 24, 2025
@lukel97
[LV] Remove EVLIndVarSimplify pass
d03e907 
Initially this was needed to replace the fixed-step canonical IV with the variable-step EVL IV, but this was eventually superseded by the loop vectorizer doing this transform itself in llvm#147222. The pass was then removed from the RISC-V pipeline in llvm#151483 and the loop vectorizer stopped emitting the metadata used by the pass in llvm#155760, so now there's no users of it.
@lukel97 lukel97 mentioned this pull request Sep 24, 2025
Merged
lukel97 added a commit that referenced this pull request Sep 25, 2025
@lukel97
[LV] Remove EVLIndVarSimplify pass (#160454)
aa6a33a 
Initially this was needed to replace the fixed-step canonical IV with
the variable-step EVL IV, but this was eventually superseded by the loop
vectorizer doing this transform itself in #147222. The pass was then
removed from the RISC-V pipeline in #151483 and the loop vectorizer
stopped emitting the metadata used by the pass in #155760, so now
there's no users of it.
mahesh-attarde pushed a commit to mahesh-attarde/llvm-project that referenced this pull request Oct 3, 2025
@lukel97 @mahesh-attarde
[LV] Remove EVLIndVarSimplify pass (llvm#160454)
5181347 
Initially this was needed to replace the fixed-step canonical IV with
the variable-step EVL IV, but this was eventually superseded by the loop
vectorizer doing this transform itself in llvm#147222. The pass was then
removed from the RISC-V pipeline in llvm#151483 and the loop vectorizer
stopped emitting the metadata used by the pass in llvm#155760, so now
there's no users of it.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@fhahn fhahn fhahn approved these changes

@lukel97 lukel97 lukel97 approved these changes

@mshockwave mshockwave Awaiting requested review from mshockwave

Assignees

No one assigned

Labels

backend:RISC-V llvm:transforms vectorizers

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@arcbbb @llvmbot @lukel97 @fhahn