[LV] Remove assert that VF cannot be scalable in setCostBasedWideningDecision.

[sdesmalen]

Summary

Scalarization for scalable vectors is not (yet) supported, so the
LV discards a VF when scalarization is chosen as the widening
decision. It should therefore not assert that the VF is not scalable
when it computes the decision to scalarize.

The code can get here when both the interleave-cost, gather/scatter cost
and scalarization-cost are all illegal. This may e.g. happen for SVE
when the VF=1, to avoid generating <vscale x 1 x eltty> types that
the code-generator cannot yet handle.

[sdesmalen]

Removed loop attribute from the test, because that actually caused the test to not actually test the code-change.

That's because it would always choose a VF=vscale x 4, and would therefore never try to compute a widening decision for VF=vscale x 1, which causes this issue.

[david-arm]

LGTM!

llvm/test/Transforms/LoopVectorize/AArch64/sve-gather-scatter.ll

132

nit: Can you use %[[IDX]] here instead of %index?

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

for (Instruction &I : *BB) {
       if (InterleaveCost <= GatherScatterCost &&
           InterleaveCost < ScalarizationCost) {
         Decision = CM_Interleave;
         Cost = InterleaveCost;
       } else if (GatherScatterCost < ScalarizationCost) {
         Decision = CM_GatherScatter;
         Cost = GatherScatterCost;
       } else {
-        assert(!VF.isScalable() &&
-               "We cannot yet scalarise for scalable vectors");
         Decision = CM_Scalarize;
         Cost = ScalarizationCost;
       }
       // If the instructions belongs to an interleave group, the whole group
       // receives the same decision. The whole group receives the cost, but
       // the cost will actually be assigned to one instruction.
       if (auto Group = getInterleavedAccessGroup(&I))
         setWideningDecision(Group, VF, Decision, Cost);

llvm/test/Transforms/LoopVectorize/AArch64/sve-gather-scatter.ll

-; RUN: opt -loop-vectorize -dce -instcombine -mtriple aarch64-linux-gnu -mattr=+sve -S %s -scalable-vectorization=on -o - | FileCheck %s
+; RUN: opt -loop-vectorize -dce -instcombine -mtriple aarch64-linux-gnu -mattr=+sve -S %s -scalable-vectorization=preferred -force-target-instruction-cost=1 -o - | FileCheck %s
 
 define void @gather_nxv4i32_ind64(float* noalias nocapture readonly %a, i64* noalias nocapture readonly %b, float* noalias nocapture %c, i64 %n) {
 ; CHECK-LABEL: @gather_nxv4i32_ind64
 ; CHECK: vector.body:
 ; CHECK:   %[[IND:.*]] = load <vscale x 4 x i64>, <vscale x 4 x i64>*
 ; CHECK:   %[[PTRS:.*]] = getelementptr inbounds float, float* %a, <vscale x 4 x i64> %[[IND]]
 ; CHECK:   %[[GLOAD:.*]] = call <vscale x 4 x float> @llvm.masked.gather.nxv4f32.nxv4p0f32(<vscale x 4 x float*> %[[PTRS]]
 ; CHECK:   store <vscale x 4 x float> %[[GLOAD]], <vscale x 4 x float>*
for.inc:                                          ; preds = %for.body, %if.then
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond.not = icmp eq i64 %indvars.iv.next, %n
   br i1 %exitcond.not, label %for.cond.cleanup, label %for.body, !llvm.loop !0
 
 for.cond.cleanup:                                 ; preds = %for.inc, %entry
   ret void
 }
 
+
+
+define void @gather_nxv4i32_ind64_stride2(float* noalias nocapture readonly %a, float* noalias nocapture readonly %b, i64 %n) {
+; CHECK-LABEL: @gather_nxv4i32_ind64_stride2
+; CHECK: vector.body:
+; CHECK:      %[[IDX:.*]] = phi i64 [ 0, %vector.ph ], [ %{{.*}}, %vector.body ]
+; CHECK-DAG:  %[[STEP:.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+; CHECK-DAG:  %[[IDXSPLATINS:.*]] = insertelement <vscale x 4 x i64> poison, i64 %[[IDX]], i32 0
+; CHECK-DAG:  %[[IDXSPLAT:.*]] = shufflevector <vscale x 4 x i64> %[[IDXSPLATINS]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
+; CHECK:      %[[ADD:.*]] = add <vscale x 4 x i64> %[[IDXSPLAT]], %[[STEP]]
+; CHECK:      %[[MUL:.*]] = shl <vscale x 4 x i64> %[[ADD]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> undef, i64 1, i32 0), <vscale x 4 x i64> undef, <vscale x 4 x i32> zeroinitializer)
+; CHECK:      %[[PTRS:.*]] = getelementptr inbounds float, float* %b, <vscale x 4 x i64> %[[MUL]]
+; CHECK:      call <vscale x 4 x float> @llvm.masked.gather.nxv4f32.nxv4p0f32(<vscale x 4 x float*> %[[PTRS]]
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %indvars.iv.stride2 = mul i64 %indvars.iv, 2
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv.stride2
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %a, i64 %indvars.iv
+  store float %0, float* %arrayidx2, align 4
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %exitcond.not = icmp eq i64 %indvars.iv.next, %n
+  br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
+
+for.cond.cleanup:                                 ; preds = %for.cond.cleanup.loopexit, %entry
+  ret void
+}
+
 !0 = distinct !{!0, !1, !2, !3, !4, !5}
 !1 = !{!"llvm.loop.mustprogress"}
 !2 = !{!"llvm.loop.vectorize.width", i32 4}
 !3 = !{!"llvm.loop.vectorize.scalable.enable", i1 true}
 !4 = !{!"llvm.loop.interleave.count", i32 1}
 !5 = !{!"llvm.loop.vectorize.enable", i1 true}