[llvm] [VPlan] Preserve IR flags when widening casts (PR #115373)

[Florian Hahn via llvm-commits]

================
@@ -0,0 +1,78 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes="default<O3>,loop-vectorize" -force-vector-interleave=1 -force-vector-width=4 -S < %s | FileCheck %s
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+define dso_local void @uitofp_preserve_nneg(ptr nocapture noundef writeonly %result, i32 noundef %size, float noundef %y, float noundef %delta) {
+; CHECK-LABEL: @uitofp_preserve_nneg(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[CMP4:%.*]] = icmp sgt i32 [[SIZE:%.*]], 0
+; CHECK-NEXT:    br i1 [[CMP4]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]]
+; CHECK:       for.body.preheader:
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext nneg i32 [[SIZE]] to i64
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[SIZE]], 4
+; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[FOR_BODY_PREHEADER4:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[N_VEC:%.*]] = and i64 [[WIDE_TRIP_COUNT]], 2147483644
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[DELTA:%.*]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <4 x float> poison, float [[Y:%.*]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT3:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT2]], <4 x float> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = uitofp nneg <4 x i32> [[VEC_IND]] to <4 x float>
+; CHECK-NEXT:    [[TMP1:%.*]] = tail call <4 x float> @llvm.fmuladd.v4f32(<4 x float> [[BROADCAST_SPLAT]], <4 x float> [[TMP0]], <4 x float> [[BROADCAST_SPLAT3]])
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds float, ptr [[RESULT:%.*]], i64 [[INDEX]]
+; CHECK-NEXT:    store <4 x float> [[TMP1]], ptr [[TMP2]], align 4
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], splat (i32 4)
+; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP3]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[N_VEC]], [[WIDE_TRIP_COUNT]]
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[FOR_END]], label [[FOR_BODY_PREHEADER4]]
+; CHECK:       for.body.preheader4:
+; CHECK-NEXT:    [[INDVARS_IV_PH:%.*]] = phi i64 [ 0, [[FOR_BODY_PREHEADER]] ], [ [[N_VEC]], [[MIDDLE_BLOCK]] ]
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ [[INDVARS_IV_PH]], [[FOR_BODY_PREHEADER4]] ]
+; CHECK-NEXT:    [[TMP4:%.*]] = trunc nuw nsw i64 [[INDVARS_IV]] to i32
+; CHECK-NEXT:    [[CONV:%.*]] = uitofp nneg i32 [[TMP4]] to float
+; CHECK-NEXT:    [[TMP5:%.*]] = tail call float @llvm.fmuladd.f32(float [[DELTA]], float [[CONV]], float [[Y]])
+; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds float, ptr [[RESULT]], i64 [[INDVARS_IV]]
+; CHECK-NEXT:    store float [[TMP5]], ptr [[ARRAYIDX]], align 4
+; CHECK-NEXT:    [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
+; CHECK-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], [[WIDE_TRIP_COUNT]]
+; CHECK-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_END_LOOPEXIT:%.*]], label [[FOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       for.end.loopexit:
+; CHECK-NEXT:    br label [[FOR_END]]
+; CHECK:       for.end:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %cmp4 = icmp sgt i32 %size, 0
+  br i1 %cmp4, label %for.body.preheader, label %for.end
----------------
fhahn wrote:

this shouldn't be needed

https://github.com/llvm/llvm-project/pull/115373