From 21e3dfbc86955cf46a362e8ed36b5b73b42961c9 Mon Sep 17 00:00:00 2001
From: Dan Gohman <gohman@apple.com>
Date: Mon, 13 Apr 2009 16:09:41 +0000
Subject: Implement x86 h-register extract support.  - Add patterns for
 h-register extract, which avoids a shift and mask,    and in some cases a
 temporary register.  - Add address-mode matching for turning
 (X>>(8-n))&(255<<n), where    n is a valid address-mode scale value, into an
 h-register extract    and a scaled-offset address.  - Replace X86's
 MOV32to32_ and related instructions with the new    target-independent
 COPY_TO_SUBREG instruction.

On x86-64 there are complicated constraints on h registers, and
CodeGen doesn't currently provide a high-level way to express all of them,
so they are handled with a bunch of special code. This code currently only
supports extracts where the result is used by a zero-extend or a store,
though these are fairly common.

These transformations are not always beneficial; since there are only
4 h registers, they sometimes require extra move instructions, and
this sometimes increases register pressure because it can force out
values that would otherwise be in one of those registers. However,
this appears to be relatively uncommon.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@68962 91177308-0d34-0410-b5e6-96231b3b80d8
---
 test/CodeGen/X86/h-register-store.ll | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)
 create mode 100644 test/CodeGen/X86/h-register-store.ll

(limited to 'test/CodeGen/X86/h-register-store.ll')

diff --git a/test/CodeGen/X86/h-register-store.ll b/test/CodeGen/X86/h-register-store.ll
new file mode 100644
index 0000000..e867242
--- /dev/null
+++ b/test/CodeGen/X86/h-register-store.ll
@@ -0,0 +1,27 @@
+; RUN: llvm-as < %s | llc -march=x86-64 > %t
+; RUN: grep mov %t | count 6
+; RUN: grep {movb	%ah, (%rsi)} %t | count 3
+; RUN: llvm-as < %s | llc -march=x86 > %t
+; RUN: grep mov %t | count 3
+; RUN: grep {movb	%ah, (%e} %t | count 3
+
+; Use h-register extract and store.
+
+define void @foo16(i16 inreg %p, i8* inreg %z) nounwind {
+  %q = lshr i16 %p, 8
+  %t = trunc i16 %q to i8
+  store i8 %t, i8* %z
+  ret void
+}
+define void @foo32(i32 inreg %p, i8* inreg %z) nounwind {
+  %q = lshr i32 %p, 8
+  %t = trunc i32 %q to i8
+  store i8 %t, i8* %z
+  ret void
+}
+define void @foo64(i64 inreg %p, i8* inreg %z) nounwind {
+  %q = lshr i64 %p, 8
+  %t = trunc i64 %q to i8
+  store i8 %t, i8* %z
+  ret void
+}
-- 
cgit v1.1