[llvm] 4454157 - [LV] Move test for incomplete partial reduction chains to separate file.

Florian Hahn via llvm-commits llvm-commits at lists.llvm.org
Sun Oct 19 14:27:19 PDT 2025


Author: Florian Hahn
Date: 2025-10-19T22:23:53+01:00
New Revision: 445415709e93a375d4447688bc354eb8a1a3bac8

URL: https://github.com/llvm/llvm-project/commit/445415709e93a375d4447688bc354eb8a1a3bac8
DIFF: https://github.com/llvm/llvm-project/commit/445415709e93a375d4447688bc354eb8a1a3bac8.diff

LOG: [LV] Move test for incomplete partial reduction chains to separate file.

Move test to new file, to prepare for adding similar tests in
https://github.com/llvm/llvm-project/pull/162822.

Added: 
    llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-incomplete-chains.ll

Modified: 
    llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-chained.ll

Removed: 
    


################################################################################
diff  --git a/llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-chained.ll b/llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-chained.ll
index 5ae08393a1804..3dfa6df3313a5 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-chained.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-chained.ll
@@ -1361,132 +1361,6 @@ for.body:                                         ; preds = %for.body.preheader,
   br i1 %exitcond.not, label %for.cond.cleanup, label %for.body, !loop !1
 }
 
-define i32 @red_extended_add_incomplete_chain(ptr %start, ptr %end, i32 %offset) {
-; CHECK-NEON-LABEL: define i32 @red_extended_add_incomplete_chain(
-; CHECK-NEON-SAME: ptr [[START:%.*]], ptr [[END:%.*]], i32 [[OFFSET:%.*]]) #[[ATTR1:[0-9]+]] {
-; CHECK-NEON-NEXT:  entry:
-; CHECK-NEON-NEXT:    [[START2:%.*]] = ptrtoint ptr [[START]] to i64
-; CHECK-NEON-NEXT:    [[END1:%.*]] = ptrtoint ptr [[END]] to i64
-; CHECK-NEON-NEXT:    [[TMP0:%.*]] = add i64 [[END1]], 1
-; CHECK-NEON-NEXT:    [[TMP1:%.*]] = sub i64 [[TMP0]], [[START2]]
-; CHECK-NEON-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP1]], 16
-; CHECK-NEON-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
-; CHECK-NEON:       vector.ph:
-; CHECK-NEON-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP1]], 16
-; CHECK-NEON-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP1]], [[N_MOD_VF]]
-; CHECK-NEON-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[START]], i64 [[N_VEC]]
-; CHECK-NEON-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i32> poison, i32 [[OFFSET]], i64 0
-; CHECK-NEON-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i32> [[BROADCAST_SPLATINSERT]], <16 x i32> poison, <16 x i32> zeroinitializer
-; CHECK-NEON-NEXT:    br label [[VECTOR_BODY:%.*]]
-; CHECK-NEON:       vector.body:
-; CHECK-NEON-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEON-NEXT:    [[VEC_PHI:%.*]] = phi <16 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP4:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEON-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[START]], i64 [[INDEX]]
-; CHECK-NEON-NEXT:    [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[NEXT_GEP]], align 1
-; CHECK-NEON-NEXT:    [[TMP3:%.*]] = zext <16 x i8> [[WIDE_LOAD]] to <16 x i32>
-; CHECK-NEON-NEXT:    [[PARTIAL_REDUCE:%.*]] = add <16 x i32> [[VEC_PHI]], [[TMP3]]
-; CHECK-NEON-NEXT:    [[TMP4]] = add <16 x i32> [[PARTIAL_REDUCE]], [[BROADCAST_SPLAT]]
-; CHECK-NEON-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
-; CHECK-NEON-NEXT:    [[TMP5:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEON-NEXT:    br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
-; CHECK-NEON:       middle.block:
-; CHECK-NEON-NEXT:    [[TMP6:%.*]] = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> [[TMP4]])
-; CHECK-NEON-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP1]], [[N_VEC]]
-; CHECK-NEON-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
-; CHECK-NEON:       scalar.ph:
-;
-; CHECK-SVE-LABEL: define i32 @red_extended_add_incomplete_chain(
-; CHECK-SVE-SAME: ptr [[START:%.*]], ptr [[END:%.*]], i32 [[OFFSET:%.*]]) #[[ATTR1:[0-9]+]] {
-; CHECK-SVE-NEXT:  entry:
-; CHECK-SVE-NEXT:    [[START2:%.*]] = ptrtoint ptr [[START]] to i64
-; CHECK-SVE-NEXT:    [[END1:%.*]] = ptrtoint ptr [[END]] to i64
-; CHECK-SVE-NEXT:    [[TMP0:%.*]] = add i64 [[END1]], 1
-; CHECK-SVE-NEXT:    [[TMP1:%.*]] = sub i64 [[TMP0]], [[START2]]
-; CHECK-SVE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-SVE-NEXT:    [[TMP3:%.*]] = shl nuw i64 [[TMP2]], 2
-; CHECK-SVE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP1]], [[TMP3]]
-; CHECK-SVE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
-; CHECK-SVE:       vector.ph:
-; CHECK-SVE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-SVE-NEXT:    [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 4
-; CHECK-SVE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP1]], [[TMP5]]
-; CHECK-SVE-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP1]], [[N_MOD_VF]]
-; CHECK-SVE-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[START]], i64 [[N_VEC]]
-; CHECK-SVE-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[OFFSET]], i64 0
-; CHECK-SVE-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-; CHECK-SVE-NEXT:    br label [[VECTOR_BODY:%.*]]
-; CHECK-SVE:       vector.body:
-; CHECK-SVE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-SVE-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP9:%.*]], [[VECTOR_BODY]] ]
-; CHECK-SVE-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[START]], i64 [[INDEX]]
-; CHECK-SVE-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i8>, ptr [[NEXT_GEP]], align 1
-; CHECK-SVE-NEXT:    [[TMP7:%.*]] = zext <vscale x 4 x i8> [[WIDE_LOAD]] to <vscale x 4 x i32>
-; CHECK-SVE-NEXT:    [[TMP8:%.*]] = add <vscale x 4 x i32> [[VEC_PHI]], [[TMP7]]
-; CHECK-SVE-NEXT:    [[TMP9]] = add <vscale x 4 x i32> [[TMP8]], [[BROADCAST_SPLAT]]
-; CHECK-SVE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
-; CHECK-SVE-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-SVE-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
-; CHECK-SVE:       middle.block:
-; CHECK-SVE-NEXT:    [[TMP11:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[TMP9]])
-; CHECK-SVE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP1]], [[N_VEC]]
-; CHECK-SVE-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
-; CHECK-SVE:       scalar.ph:
-;
-; CHECK-SVE-MAXBW-LABEL: define i32 @red_extended_add_incomplete_chain(
-; CHECK-SVE-MAXBW-SAME: ptr [[START:%.*]], ptr [[END:%.*]], i32 [[OFFSET:%.*]]) #[[ATTR1:[0-9]+]] {
-; CHECK-SVE-MAXBW-NEXT:  entry:
-; CHECK-SVE-MAXBW-NEXT:    [[START2:%.*]] = ptrtoint ptr [[START]] to i64
-; CHECK-SVE-MAXBW-NEXT:    [[END1:%.*]] = ptrtoint ptr [[END]] to i64
-; CHECK-SVE-MAXBW-NEXT:    [[TMP0:%.*]] = add i64 [[END1]], 1
-; CHECK-SVE-MAXBW-NEXT:    [[TMP1:%.*]] = sub i64 [[TMP0]], [[START2]]
-; CHECK-SVE-MAXBW-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-SVE-MAXBW-NEXT:    [[TMP3:%.*]] = shl nuw i64 [[TMP2]], 3
-; CHECK-SVE-MAXBW-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP1]], [[TMP3]]
-; CHECK-SVE-MAXBW-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
-; CHECK-SVE-MAXBW:       vector.ph:
-; CHECK-SVE-MAXBW-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-SVE-MAXBW-NEXT:    [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 8
-; CHECK-SVE-MAXBW-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP1]], [[TMP5]]
-; CHECK-SVE-MAXBW-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP1]], [[N_MOD_VF]]
-; CHECK-SVE-MAXBW-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[START]], i64 [[N_VEC]]
-; CHECK-SVE-MAXBW-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 8 x i32> poison, i32 [[OFFSET]], i64 0
-; CHECK-SVE-MAXBW-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 8 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
-; CHECK-SVE-MAXBW-NEXT:    br label [[VECTOR_BODY:%.*]]
-; CHECK-SVE-MAXBW:       vector.body:
-; CHECK-SVE-MAXBW-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-SVE-MAXBW-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 8 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP8:%.*]], [[VECTOR_BODY]] ]
-; CHECK-SVE-MAXBW-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[START]], i64 [[INDEX]]
-; CHECK-SVE-MAXBW-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 8 x i8>, ptr [[NEXT_GEP]], align 1
-; CHECK-SVE-MAXBW-NEXT:    [[TMP7:%.*]] = zext <vscale x 8 x i8> [[WIDE_LOAD]] to <vscale x 8 x i32>
-; CHECK-SVE-MAXBW-NEXT:    [[PARTIAL_REDUCE:%.*]] = add <vscale x 8 x i32> [[VEC_PHI]], [[TMP7]]
-; CHECK-SVE-MAXBW-NEXT:    [[TMP8]] = add <vscale x 8 x i32> [[PARTIAL_REDUCE]], [[BROADCAST_SPLAT]]
-; CHECK-SVE-MAXBW-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
-; CHECK-SVE-MAXBW-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-SVE-MAXBW-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
-; CHECK-SVE-MAXBW:       middle.block:
-; CHECK-SVE-MAXBW-NEXT:    [[TMP10:%.*]] = call i32 @llvm.vector.reduce.add.nxv8i32(<vscale x 8 x i32> [[TMP8]])
-; CHECK-SVE-MAXBW-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP1]], [[N_VEC]]
-; CHECK-SVE-MAXBW-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
-; CHECK-SVE-MAXBW:       scalar.ph:
-;
-entry:
-  br label %loop
-
-loop:
-  %ptr.iv = phi ptr [ %start, %entry ], [ %gep.iv.next, %loop ]
-  %red = phi i32 [ 0, %entry ], [ %red.next, %loop ]
-  %l = load i8, ptr %ptr.iv, align 1
-  %l.ext = zext i8 %l to i32
-  %add = add i32 %red, %l.ext
-  %red.next = add i32 %add, %offset
-  %gep.iv.next = getelementptr i8, ptr %ptr.iv, i64 1
-  %ec = icmp eq ptr %ptr.iv, %end
-  br i1 %ec, label %exit, label %loop
-
-exit:
-  ret i32 %red.next
-}
-
 attributes #0 = { vscale_range(1,16) }
 
 

diff  --git a/llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-incomplete-chains.ll b/llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-incomplete-chains.ll
new file mode 100644
index 0000000000000..d80178fde45d9
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/partial-reduce-incomplete-chains.ll
@@ -0,0 +1,72 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 6
+; RUN: opt --mattr=+neon,+dotprod -passes=loop-vectorize -force-vector-interleave=1 -enable-epilogue-vectorization=false -S %s | FileCheck %s --check-prefixes=CHECK-NEON
+
+target triple = "arm64-apple-macosx"
+
+define i32 @red_extended_add_incomplete_chain(ptr %start, ptr %end, i32 %offset) {
+; CHECK-NEON-LABEL: define i32 @red_extended_add_incomplete_chain(
+; CHECK-NEON-SAME: ptr [[START:%.*]], ptr [[END:%.*]], i32 [[OFFSET:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEON-NEXT:  [[ENTRY:.*]]:
+; CHECK-NEON-NEXT:    [[START2:%.*]] = ptrtoint ptr [[START]] to i64
+; CHECK-NEON-NEXT:    [[END1:%.*]] = ptrtoint ptr [[END]] to i64
+; CHECK-NEON-NEXT:    [[TMP0:%.*]] = add i64 [[END1]], 1
+; CHECK-NEON-NEXT:    [[TMP1:%.*]] = sub i64 [[TMP0]], [[START2]]
+; CHECK-NEON-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP1]], 16
+; CHECK-NEON-NEXT:    br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
+; CHECK-NEON:       [[VECTOR_PH]]:
+; CHECK-NEON-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP1]], 16
+; CHECK-NEON-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP1]], [[N_MOD_VF]]
+; CHECK-NEON-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[START]], i64 [[N_VEC]]
+; CHECK-NEON-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i32> poison, i32 [[OFFSET]], i64 0
+; CHECK-NEON-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i32> [[BROADCAST_SPLATINSERT]], <16 x i32> poison, <16 x i32> zeroinitializer
+; CHECK-NEON-NEXT:    br label %[[VECTOR_BODY:.*]]
+; CHECK-NEON:       [[VECTOR_BODY]]:
+; CHECK-NEON-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEON-NEXT:    [[VEC_PHI:%.*]] = phi <16 x i32> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP5:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEON-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[START]], i64 [[INDEX]]
+; CHECK-NEON-NEXT:    [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[NEXT_GEP]], align 1
+; CHECK-NEON-NEXT:    [[TMP3:%.*]] = zext <16 x i8> [[WIDE_LOAD]] to <16 x i32>
+; CHECK-NEON-NEXT:    [[TMP4:%.*]] = add <16 x i32> [[VEC_PHI]], [[TMP3]]
+; CHECK-NEON-NEXT:    [[TMP5]] = add <16 x i32> [[TMP4]], [[BROADCAST_SPLAT]]
+; CHECK-NEON-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
+; CHECK-NEON-NEXT:    [[TMP6:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEON-NEXT:    br i1 [[TMP6]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK-NEON:       [[MIDDLE_BLOCK]]:
+; CHECK-NEON-NEXT:    [[TMP7:%.*]] = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> [[TMP5]])
+; CHECK-NEON-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP1]], [[N_VEC]]
+; CHECK-NEON-NEXT:    br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_PH]]
+; CHECK-NEON:       [[SCALAR_PH]]:
+; CHECK-NEON-NEXT:    [[BC_RESUME_VAL:%.*]] = phi ptr [ [[TMP2]], %[[MIDDLE_BLOCK]] ], [ [[START]], %[[ENTRY]] ]
+; CHECK-NEON-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i32 [ [[TMP7]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
+; CHECK-NEON-NEXT:    br label %[[LOOP:.*]]
+; CHECK-NEON:       [[LOOP]]:
+; CHECK-NEON-NEXT:    [[PTR_IV:%.*]] = phi ptr [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[GEP_IV_NEXT:%.*]], %[[LOOP]] ]
+; CHECK-NEON-NEXT:    [[RED:%.*]] = phi i32 [ [[BC_MERGE_RDX]], %[[SCALAR_PH]] ], [ [[RED_NEXT:%.*]], %[[LOOP]] ]
+; CHECK-NEON-NEXT:    [[L:%.*]] = load i8, ptr [[PTR_IV]], align 1
+; CHECK-NEON-NEXT:    [[L_EXT:%.*]] = zext i8 [[L]] to i32
+; CHECK-NEON-NEXT:    [[ADD:%.*]] = add i32 [[RED]], [[L_EXT]]
+; CHECK-NEON-NEXT:    [[RED_NEXT]] = add i32 [[ADD]], [[OFFSET]]
+; CHECK-NEON-NEXT:    [[GEP_IV_NEXT]] = getelementptr i8, ptr [[PTR_IV]], i64 1
+; CHECK-NEON-NEXT:    [[EC:%.*]] = icmp eq ptr [[PTR_IV]], [[END]]
+; CHECK-NEON-NEXT:    br i1 [[EC]], label %[[EXIT]], label %[[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK-NEON:       [[EXIT]]:
+; CHECK-NEON-NEXT:    [[RED_NEXT_LCSSA:%.*]] = phi i32 [ [[RED_NEXT]], %[[LOOP]] ], [ [[TMP7]], %[[MIDDLE_BLOCK]] ]
+; CHECK-NEON-NEXT:    ret i32 [[RED_NEXT_LCSSA]]
+;
+entry:
+  br label %loop
+
+loop:
+  %ptr.iv = phi ptr [ %start, %entry ], [ %gep.iv.next, %loop ]
+  %red = phi i32 [ 0, %entry ], [ %red.next, %loop ]
+  %l = load i8, ptr %ptr.iv, align 1
+  %l.ext = zext i8 %l to i32
+  %add = add i32 %red, %l.ext
+  %red.next = add i32 %add, %offset
+  %gep.iv.next = getelementptr i8, ptr %ptr.iv, i64 1
+  %ec = icmp eq ptr %ptr.iv, %end
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret i32 %red.next
+}


        


More information about the llvm-commits mailing list