From 759cfef3ac4c07dba1ece0bbc1207e099348816d Mon Sep 17 00:00:00 2001
From: Alan Modra <amodra@gmail.com>
Date: Sat, 17 Aug 2013 18:47:22 +0930
Subject: PowerPC LE memcpy
 http://sourceware.org/ml/libc-alpha/2013-08/msg00103.html

LIttle-endian support for memcpy.  I spent some time cleaning up the
64-bit power7 memcpy, in order to avoid the extra alignment traps
power7 takes for little-endian.  It probably would have been better
to copy the linux kernel version of memcpy.

	* sysdeps/powerpc/powerpc32/power4/memcpy.S: Add little endian support.
	* sysdeps/powerpc/powerpc32/power6/memcpy.S: Likewise.
	* sysdeps/powerpc/powerpc32/power7/memcpy.S: Likewise.
	* sysdeps/powerpc/powerpc32/power7/mempcpy.S: Likewise.
	* sysdeps/powerpc/powerpc64/memcpy.S: Likewise.
	* sysdeps/powerpc/powerpc64/power4/memcpy.S: Likewise.
	* sysdeps/powerpc/powerpc64/power6/memcpy.S: Likewise.
	* sysdeps/powerpc/powerpc64/power7/memcpy.S: Likewise.
	* sysdeps/powerpc/powerpc64/power7/mempcpy.S: Likewise.  Make better
	use of regs.  Use power7 mtocrf.  Tidy function tails.
---
 sysdeps/powerpc/powerpc64/memcpy.S | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

(limited to 'sysdeps/powerpc/powerpc64/memcpy.S')

diff --git a/sysdeps/powerpc/powerpc64/memcpy.S b/sysdeps/powerpc/powerpc64/memcpy.S
index b8c4cc8b10..5fc7401c99 100644
--- a/sysdeps/powerpc/powerpc64/memcpy.S
+++ b/sysdeps/powerpc/powerpc64/memcpy.S
@@ -212,15 +212,28 @@ EALIGN (memcpy, 5, 0)
     blt   cr6,5f
     srdi  7,6,16
     bgt	  cr6,3f
+#ifdef __LITTLE_ENDIAN__
+    sth   7,0(3)
+#else
     sth   6,0(3)
+#endif
     b     7f
     .align  4
 3:
+#ifdef __LITTLE_ENDIAN__
+    rotlwi 6,6,24
+    stb   6,0(3)
+    sth   7,1(3)
+#else
     stb   7,0(3)
     sth   6,1(3)
+#endif
     b     7f
     .align  4
 5:
+#ifdef __LITTLE_ENDIAN__
+    rotlwi 6,6,8
+#endif
     stb   6,0(3)
 7:
     cmpldi	cr1,10,16
@@ -328,7 +341,11 @@ EALIGN (memcpy, 5, 0)
     ld    7,8(5)
     subfic  9,10,64
     beq   2f
+#ifdef __LITTLE_ENDIAN__
+    srd   0,6,10
+#else
     sld   0,6,10
+#endif
     cmpldi  11,1
     mr    6,7
     addi  4,4,-8
@@ -336,15 +353,25 @@ EALIGN (memcpy, 5, 0)
     b     1f
 2:  addi  5,5,8
     .align  4
+#ifdef __LITTLE_ENDIAN__
+0:  srd   0,6,10
+    sld   8,7,9
+#else
 0:  sld   0,6,10
     srd   8,7,9
+#endif
     cmpldi  11,2
     ld    6,8(5)
     or    0,0,8
     addi  11,11,-2
     std   0,0(4)
+#ifdef __LITTLE_ENDIAN__
+    srd   0,7,10
+1:  sld   8,6,9
+#else
     sld   0,7,10
 1:  srd   8,6,9
+#endif
     or    0,0,8
     beq   8f
     ld    7,16(5)
-- 
cgit 1.4.1