My previous commit had an incomplete message, it should have been:

make the 'fp return in ST(0)' optimization smart enough to
look through token factor nodes.  THis allows us to compile 
testcases like CodeGen/X86/fp-stack-retcopy.ll into:

_carg:
	subl	$12, %esp
	call	L_foo$stub
	fstpl	(%esp)
	fldl	(%esp)
	addl	$12, %esp
	ret

instead of:

_carg:
	subl	$28, %esp
	call	L_foo$stub
	fstpl	16(%esp)
	movsd	16(%esp), %xmm0
	movsd	%xmm0, 8(%esp)
	fldl	8(%esp)
	addl	$28, %esp
	ret

Still not optimal, but much better and this is a trivial patch.  Fixing 
the rest requires invasive surgery that is is not llvm 2.2 material.

llvm-svn: 46054

1 files changed, 12 insertions, 0 deletions

diff --git a/llvm/test/CodeGen/X86/fp-stack-retcopy.ll b/llvm/test/CodeGen/X86/fp-stack-retcopy.ll
new file mode 100644
index 00000000000..997f8df20fe
--- /dev/null
+++ b/llvm/test/CodeGen/X86/fp-stack-retcopy.ll
@@ -0,0 +1,12 @@
+; This should not copy the result of foo into an xmm register.
+; RUN: llvm-as < %s | llc -march=x86 -mcpu=yonah -mtriple=i686-apple-darwin9 | not grep xmm
+; rdar://5689903
+
+declare double @foo()
+
+define double @carg({ double, double }* byval  %z) nounwind  {
+entry:
+	%tmp5 = tail call double @foo() nounwind 		; <double> [#uses=1]
+	ret double %tmp5
+}
+