[X86] Use a MOVSX instruction instead of a MOVZX instruction in isel for an any_extend of the remainder from an 8-bit sdivrem.

The sdivrem will emit its own MOVSX to move %ah to the low byte of a register. By using a MOVSX for an any_extend this allows a post-isel peephole to merge them.

llvm-svn: 346581

1 files changed, 9 insertions, 0 deletions

diff --git a/llvm/lib/Target/X86/X86InstrCompiler.td b/llvm/lib/Target/X86/X86InstrCompiler.td
index 71b43a38dc2..bc08147506c 100644
--- a/llvm/lib/Target/X86/X86InstrCompiler.td
+++ b/llvm/lib/Target/X86/X86InstrCompiler.td
@@ -1320,6 +1320,15 @@ def : Pat<(i64 (anyext GR16:$src)),
 def : Pat<(i64 (anyext GR32:$src)),
           (INSERT_SUBREG (i64 (IMPLICIT_DEF)), GR32:$src, sub_32bit)>;
 
+// If this is an anyext of the remainder of an 8-bit sdivrem, use a MOVSX
+// instead of a MOVZX. The sdivrem lowering will emit emit a MOVSX to move
+// %ah to the lower byte of a register. By using a MOVSX here we allow a
+// post-isel peephole to merge the two MOVSX instructions into one.
+def anyext_sdiv : PatFrag<(ops node:$lhs), (anyext node:$lhs),[{
+  return (N->getOperand(0).getOpcode() == ISD::SDIVREM &&
+          N->getOperand(0).getResNo() == 1);
+}]>;
+def : Pat<(i32 (anyext_sdiv GR8:$src)), (MOVSX32rr8 GR8:$src)>;
 
 // Any instruction that defines a 32-bit result leaves the high half of the
 // register. Truncate can be lowered to EXTRACT_SUBREG. CopyFromReg may