[PATCH] D91446: [AlwaysInliner] Call mergeAttributesForInlining after inlining

[Guozhi Wei via Phabricator via llvm-commits]

Carrot created this revision.
Herald added subscribers: llvm-commits, hiraditya, eraman.
Herald added a project: LLVM.
Carrot requested review of this revision.

Like inlineCallIfPossible and InlinerPass, after inlining mergeAttributesForInlining should be called to merge callee's attributes to caller. But it is not called in AlwaysInliner, causes caller's attributes inconsistent with inlined code.

Attached test case demonstrates that attribute "min-legal-vector-width"="512" is not merged into caller without this patch, and it causes failure in SelectionDAG when lowering the inlined AVX512 intrinsic.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D91446

Files:
  llvm/lib/Transforms/IPO/AlwaysInliner.cpp
  llvm/test/Transforms/Inline/always-inline-attr.ll


Index: llvm/test/Transforms/Inline/always-inline-attr.ll
===================================================================
--- /dev/null
+++ llvm/test/Transforms/Inline/always-inline-attr.ll
@@ -0,0 +1,48 @@
+; RUN: opt -passes=always-inline -S < %s | FileCheck %s
+
+
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-grtev4-linux-gnu"
+
+; After AlwaysInline the callee's attributes should be merged into caller's attibutes.
+
+; CHECK:  define dso_local <2 x i64> @foo(<8 x i64>* byval(<8 x i64>) align 64 %0) #0
+; CHECK:  attributes #0 = { uwtable mustprogress "disable-tail-calls"="false" "frame-pointer"="none" "less-precise-fpmad"="false" "min-legal-vector-width"="512"
+
+; Function Attrs: uwtable mustprogress
+define dso_local <2 x i64> @foo(<8 x i64>* byval(<8 x i64>) align 64 %0) #0 {
+entry:
+  %kBias.addr = alloca <8 x i64>, align 64
+  %indirect-arg-temp = alloca <8 x i64>, align 64
+  %kBias = load <8 x i64>, <8 x i64>* %0, align 64, !tbaa !2
+  store <8 x i64> %kBias, <8 x i64>* %kBias.addr, align 64, !tbaa !2
+  %1 = load <8 x i64>, <8 x i64>* %kBias.addr, align 64, !tbaa !2
+  store <8 x i64> %1, <8 x i64>* %indirect-arg-temp, align 64, !tbaa !2
+  %call = call <2 x i64> @bar(<8 x i64>* byval(<8 x i64>) align 64 %indirect-arg-temp)
+  ret <2 x i64> %call
+}
+
+; Function Attrs: alwaysinline nounwind uwtable mustprogress
+define internal <2 x i64> @bar(<8 x i64>* byval(<8 x i64>) align 64 %0) #1 {
+entry:
+  %__A.addr = alloca <8 x i64>, align 64
+  %__A = load <8 x i64>, <8 x i64>* %0, align 64, !tbaa !2
+  store <8 x i64> %__A, <8 x i64>* %__A.addr, align 64, !tbaa !2
+  %1 = load <8 x i64>, <8 x i64>* %__A.addr, align 64, !tbaa !2
+  %2 = bitcast <8 x i64> %1 to <16 x i32>
+  %3 = call <16 x i8> @llvm.x86.avx512.mask.pmovs.db.512(<16 x i32> %2, <16 x i8> zeroinitializer, i16 -1)
+  %4 = bitcast <16 x i8> %3 to <2 x i64>
+  ret <2 x i64> %4
+}
+
+; Function Attrs: nounwind readnone
+declare <16 x i8> @llvm.x86.avx512.mask.pmovs.db.512(<16 x i32>, <16 x i8>, i16) #2
+
+attributes #0 = { uwtable mustprogress "disable-tail-calls"="false" "frame-pointer"="none" "less-precise-fpmad"="false" "min-legal-vector-width"="128" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="true" "prefer-vector-width"="128" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+aes,+avx,+avx2,+avx512bw,+avx512dq,+avx512f,+avx512vl,+bmi2,+cx16,+cx8,+f16c,+fma,+fxsr,+mmx,+pclmul,+popcnt,+sse,+sse2,+sse3,+sse4.1,+sse4.2,+ssse3,+x87,+xsave" "tune-cpu"="generic" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #1 = { alwaysinline nounwind uwtable mustprogress "disable-tail-calls"="false" "frame-pointer"="none" "less-precise-fpmad"="false" "min-legal-vector-width"="512" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="true" "prefer-vector-width"="128" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+aes,+avx,+avx2,+avx512f,+cx16,+cx8,+f16c,+fma,+fxsr,+mmx,+pclmul,+popcnt,+sse,+sse2,+sse3,+sse4.1,+sse4.2,+ssse3,+x87,+xsave" "tune-cpu"="generic" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #2 = { nounwind readnone }
+
+
+!2 = !{!3, !3, i64 0}
+!3 = !{!"omnipotent char", !4, i64 0}
+!4 = !{!"Simple C++ TBAA"}
Index: llvm/lib/Transforms/IPO/AlwaysInliner.cpp
===================================================================
--- llvm/lib/Transforms/IPO/AlwaysInliner.cpp
+++ llvm/lib/Transforms/IPO/AlwaysInliner.cpp
@@ -78,6 +78,10 @@
             *CB, IFI, &FAM.getResult<AAManager>(F), InsertLifetime);
         assert(Res.isSuccess() && "unexpected failure to inline");
         (void)Res;
+
+        // Merge the attributes based on the inlining.
+        AttributeFuncs::mergeAttributesForInlining(*Caller, F);
+
         Changed = true;
       }
 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D91446.305214.patch
Type: text/x-patch
Size: 4006 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201113/e873146a/attachment.bin>