mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-19 19:12:56 +02:00
786d7d887a
As explained in r204976, because of how the allocation of VSX registers interacts with the call-lowering code, we sometimes end up generating self VSX copies. Specifically, things like this: %VSL2<def> = COPY %F2, %VSL2<imp-use,kill> (where %F2 is really a sub-register of %VSL2, and so this copy is a nop) This adds a small cleanup pass to remove these prior to post-RA scheduling. llvm-svn: 204980
28 lines
923 B
LLVM
28 lines
923 B
LLVM
; RUN: llc -mcpu=pwr7 -mattr=+vsx < %s | FileCheck %s
|
|
target datalayout = "E-m:e-i64:64-n32:64"
|
|
target triple = "powerpc64-unknown-linux-gnu"
|
|
|
|
define double @takFP(double %x, double %y, double %z) #0 {
|
|
entry:
|
|
br i1 undef, label %if.then, label %return
|
|
|
|
if.then: ; preds = %if.then, %entry
|
|
%x.tr16 = phi double [ %call, %if.then ], [ %x, %entry ]
|
|
%call = tail call double @takFP(double undef, double undef, double undef)
|
|
%call4 = tail call double @takFP(double undef, double %x.tr16, double undef)
|
|
%cmp = fcmp olt double undef, %call
|
|
br i1 %cmp, label %if.then, label %return
|
|
|
|
return: ; preds = %if.then, %entry
|
|
%z.tr.lcssa = phi double [ %z, %entry ], [ %call4, %if.then ]
|
|
ret double %z.tr.lcssa
|
|
|
|
; CHECK: @takFP
|
|
; CHECK-NOT: xxlor 0, 0, 0
|
|
; CHECK: blr
|
|
}
|
|
|
|
attributes #0 = { nounwind readnone }
|
|
attributes #1 = { nounwind }
|
|
|