[PowerPC] Avoid VSX FMA mutate when killed product reg = addend reg
With VSX enabled, test/CodeGen/PowerPC/recipest.ll exposes a bug in
the FMA mutation pass. If we have a situation where a killed product
register is the same register as the FMA target, such as:
%vreg5<def,tied1> = XSNMSUBADP %vreg5<tied0>, %vreg11, %vreg5,
%RM<imp-use>; VSFRC:%vreg5 F8RC:%vreg11
then the substitution makes no sense. We end up getting a crash when
we try to extend the interval associated with the killed product
register, as there is already a live range for %vreg5 there. This
patch just disables the mutation under those circumstances.
Since recipest.ll generates different code with VMX enabled, I've
modified that test to use -mattr=-vsx. I've borrowed the code from
that test that exposed the bug and placed it in fma-mutate.ll, where
it tests several mutation opportunities including the "bad" one.
llvm-svn: 220290
diff --git a/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp b/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp
index ff2633e..daf8790 100644
--- a/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp
+++ b/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp
@@ -1754,6 +1754,12 @@
unsigned OldFMAReg = MI->getOperand(0).getReg();
+ // The transformation doesn't work well with things like:
+ // %vreg5 = A-form-op %vreg5, %vreg11, %vreg5;
+ // so leave such things alone.
+ if (OldFMAReg == KilledProdReg)
+ continue;
+
assert(OldFMAReg == AddendMI->getOperand(0).getReg() &&
"Addend copy not tied to old FMA output!");