[PATCH] D68082: [LV] Emitting SCEV checks with OptForSize

[Sjoerd Meijer via Phabricator via llvm-commits]

SjoerdMeijer updated this revision to Diff 223465.
SjoerdMeijer added a comment.

Comments addressed:

- cleaned up the test case a bit.
- couldn't reuse an existing run-line, I guess because of -mcpu=skx, but a separate run line seems fine to me.

> The other suggested fix in LoopAccessInfo::collectStridedAccess() indeed deserves a separate patch

Thanks for that suggestions, and I will address this separately. I have unfinished business in the vectorizer, and will add this to me my list of things to do next.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D68082/new/

https://reviews.llvm.org/D68082

Files:
  llvm/lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
  llvm/test/Transforms/LoopVectorize/X86/optsize.ll


Index: llvm/test/Transforms/LoopVectorize/X86/optsize.ll
===================================================================
--- llvm/test/Transforms/LoopVectorize/X86/optsize.ll
+++ llvm/test/Transforms/LoopVectorize/X86/optsize.ll
@@ -4,6 +4,7 @@
 ; attributes. This is a target-dependent version of the test.
 ; RUN: opt < %s -loop-vectorize -force-vector-width=64 -S -mtriple=x86_64-unknown-linux -mcpu=skx | FileCheck %s
 ; RUN: opt < %s -loop-vectorize -S -mtriple=x86_64-unknown-linux -mcpu=skx | FileCheck %s --check-prefix AUTOVF
+; RUN: opt < %s -loop-vectorize -S | FileCheck %s --check-prefix=NO-SCEV-PREDS
 
 target datalayout = "E-m:e-p:32:32-i64:32-f64:32:64-a:0:32-n32-S128"
 
@@ -196,3 +197,42 @@
 while.cond.loopexit:
   ret i32 0
 }
+
+; PR43371: don't run into an assert due to emitting SCEV runtime checks
+; with OptForSize.
+;
+ at cm_array = external global [2592 x i16], align 1
+
+define void @pr43371() optsize {
+;
+; NO-SCEV-PREDS-LABEL: @pr43371
+;
+; We do not want to generate SCEV predicates when optimising for size, because
+; that will lead to extra code generation such as the SCEV overflow runtime
+; checks. Not generating SCEV predicates can still result in vectorisation as
+; the non-consecutive loads/stores can be scalarized:
+;
+; NO-SCEV-PREDS: vector.body:
+; NO-SCEV-PREDS: store i16 0, i16* %{{.*}}, align 1
+; NO-SCEV-PREDS: store i16 0, i16* %{{.*}}, align 1
+; NO-SCEV-PREDS: br i1 {{.*}}, label %vector.body
+;
+entry:
+  br label %header
+
+header:
+  br label %for.body29
+
+for.cond.cleanup28:
+  unreachable
+
+for.body29:
+  %i24.0170 = phi i16 [ 0, %header ], [ %inc37, %for.body29]
+  %add33 = add i16 undef, %i24.0170
+  %idxprom34 = zext i16 %add33 to i32
+  %arrayidx35 = getelementptr [2592 x i16], [2592 x i16] * @cm_array, i32 0, i32 %idxprom34
+  store i16 0, i16 * %arrayidx35, align 1
+  %inc37 = add i16 %i24.0170, 1
+  %cmp26 = icmp ult i16 %inc37, 756
+  br i1 %cmp26, label %for.body29, label %for.cond.cleanup28
+}
Index: llvm/lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
===================================================================
--- llvm/lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
+++ llvm/lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
@@ -409,7 +409,8 @@
   const ValueToValueMap &Strides =
       getSymbolicStrides() ? *getSymbolicStrides() : ValueToValueMap();
 
-  int Stride = getPtrStride(PSE, Ptr, TheLoop, Strides, true, false);
+  bool CanAddPredicate = !TheLoop->getHeader()->getParent()->hasOptSize();
+  int Stride = getPtrStride(PSE, Ptr, TheLoop, Strides, CanAddPredicate, false);
   if (Stride == 1 || Stride == -1)
     return Stride;
   return 0;


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D68082.223465.patch
Type: text/x-patch
Size: 2678 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191007/b0b7f4c0/attachment.bin>