[PATCH] D128033: [LoopVectorize] Fix createInductionResumeValues

[Serguei Katkov via Phabricator via llvm-commits]

skatkov marked 3 inline comments as done.
skatkov added inline comments.


================
Comment at: llvm/test/Transforms/LoopVectorize/create-induction-resume.ll:5
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128-ni:1-p2:32:8:8:32-ni:2"
+target triple = "x86_64-unknown-linux-gnu"
+define void @hoge(i32 %arg) {
----------------
fhahn wrote:
> Does the test depend on x86? If so, it needs to be moved to the X86 subdirectory, otherwise it may fail on targets that don't build the backend. 
> 
> If `-force-vector-with=x` and `-force-vector-interleave=x` are sufficient, it would be good to remove the triple.
The test crashes without triple and any additional flags.


================
Comment at: llvm/test/Transforms/LoopVectorize/create-induction-resume.ll:6
+target triple = "x86_64-unknown-linux-gnu"
+define void @hoge(i32 %arg) {
+; CHECK-LABEL: @hoge(
----------------
fhahn wrote:
> It might be good to include a reference to the bug here
I did not file a bug.


================
Comment at: llvm/test/Transforms/LoopVectorize/create-induction-resume.ll:89
+
+bb8:                                              ; preds = %bb19, %bb6
+  switch i32 undef, label %bb19 [
----------------
david-arm wrote:
> Is it possible to simplify this test a little? It's quite difficult to follow what's going on here. It looks like the actual loop only requires bb8, bb9, bb10, bb11, bb18 and bb19.
It seems the first loop is required to force SSA updater.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D128033/new/

https://reviews.llvm.org/D128033