[PowerPC] Avoid VSX FMA mutate when killed product reg = addend reg

[wschmidt]

Summary

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.

[hfinkel]

Oops. LGTM, thanks!

[wschmidt]

r220290. Thanks for the quick review!

lib/Target/PowerPC/PPCInstrInfo.cpp

bool processBlock(MachineBasicBlock &MBB) {
         bool OtherProdRegKill  = MI->getOperand(OtherProdOp).isKill();
 
         bool AddRegUndef = AddendMI->getOperand(1).isUndef();
         bool KilledProdRegUndef = MI->getOperand(KilledProdOp).isUndef();
         bool OtherProdRegUndef  = MI->getOperand(OtherProdOp).isUndef();
 
         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!");
 
         DEBUG(dbgs() << "VSX FMA Mutation:\n    " << *MI;);
 
         MI->getOperand(0).setReg(KilledProdReg);
         MI->getOperand(1).setReg(KilledProdReg);
         MI->getOperand(3).setReg(AddReg);

test/CodeGen/PowerPC/fma-mutate.ll

+; Test several VSX FMA mutation opportunities.  The first one isn't a
+; reasonable transformation because the killed product register is the
+; same as the FMA target register.  The second one is legal.  The third
+; one doesn't fit the feeding-copy pattern.
+
+; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -enable-unsafe-fp-math -mattr=+vsx | FileCheck %s
+target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+declare double @llvm.sqrt.f64(double)
+
+define double @foo3(double %a) nounwind {
+  %r = call double @llvm.sqrt.f64(double %a)
+  ret double %r
+
+; CHECK: @foo3
+; CHECK: xsnmsubadp [[REG:[0-9]+]], {{[0-9]+}}, [[REG]]
+; CHECK: xsmaddmdp
+; CHECK: xsmaddadp
+}
+

test/CodeGen/PowerPC/recipest.ll

-; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -enable-unsafe-fp-math | FileCheck %s
-; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck -check-prefix=CHECK-SAFE %s
+; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -enable-unsafe-fp-math -mattr=-vsx | FileCheck %s
+; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=-vsx | FileCheck -check-prefix=CHECK-SAFE %s
 target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64"
 target triple = "powerpc64-unknown-linux-gnu"
 
 declare double @llvm.sqrt.f64(double)
 declare float @llvm.sqrt.f32(float)
 declare <4 x float> @llvm.sqrt.v4f32(<4 x float>)
 
 define double @foo(double %a, double %b) nounwind {