Update libvpx snapshot to v0.9.7-p1 (Cayuga).

source/libvpx/vp8/common/x86/postproc_sse2.asm

 ;
 ;  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
 ;
 ;  Use of this source code is governed by a BSD-style license
 ;  that can be found in the LICENSE file in the root of the source
 ;  tree. An additional intellectual property rights grant can be found
 ;  in the file PATENTS.  All contributing project authors may
 ;  be found in the AUTHORS file in the root of the source tree.
 ;
 
 
 %include "vpx_ports/x86_abi_support.asm"
 
 ;void vp8_post_proc_down_and_across_xmm
 ;(
 ;    unsigned char *src_ptr,
 ;    unsigned char *dst_ptr,
 ;    int src_pixels_per_line,
 ;    int dst_pixels_per_line,
 ;    int rows,
 ;    int cols,
 ;    int flimit
 ;)
 global sym(vp8_post_proc_down_and_across_xmm)
 sym(vp8_post_proc_down_and_across_xmm):
     push        rbp
     mov         rbp, rsp
     SHADOW_ARGS_TO_STACK 7
-    SAVE_XMM
+    SAVE_XMM 7
     GET_GOT     rbx
     push        rsi
     push        rdi
     ; end prolog
 
 %if ABI_IS_32BIT=1 && CONFIG_PIC=1
     ALIGN_STACK 16, rax
     ; move the global rd onto the stack, since we don't have enough registers
     ; to do PIC addressing
     movdqa      xmm0, [GLOBAL(rd42)]
@@ skipping 209 matching lines @@
 
 
 ;void vp8_mbpost_proc_down_xmm(unsigned char *dst,
 ;                            int pitch, int rows, int cols,int flimit)
 extern sym(vp8_rv)
 global sym(vp8_mbpost_proc_down_xmm)
 sym(vp8_mbpost_proc_down_xmm):
     push        rbp
     mov         rbp, rsp
     SHADOW_ARGS_TO_STACK 5
-    SAVE_XMM
+    SAVE_XMM 7
     GET_GOT     rbx
     push        rsi
     push        rdi
     ; end prolog
 
     ALIGN_STACK 16, rax
     sub         rsp, 128+16
 
     ; unsigned char d[16][8] at [rsp]
     ; create flimit2 at [rsp+128]
@@ skipping 179 matching lines @@
 %undef flimit4
 
 
 ;void vp8_mbpost_proc_across_ip_xmm(unsigned char *src,
 ;                                int pitch, int rows, int cols,int flimit)
 global sym(vp8_mbpost_proc_across_ip_xmm)
 sym(vp8_mbpost_proc_across_ip_xmm):
     push        rbp
     mov         rbp, rsp
     SHADOW_ARGS_TO_STACK 5
-    SAVE_XMM
+    SAVE_XMM 7
     GET_GOT     rbx
     push        rsi
     push        rdi
     ; end prolog
 
     ALIGN_STACK 16, rax
     sub         rsp, 16
 
     ; create flimit4 at [rsp]
     mov         eax, dword ptr arg(4) ;flimit
@@ skipping 216 matching lines @@
     pop         rbp
     ret
 
 
 SECTION_RODATA
 align 16
 rd42:
     times 8 dw 0x04
 four8s:
     times 4 dd 8