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

source/libvpx/vp8/encoder/x86/dct_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.
 ;
 
@@ skipping 15 matching lines @@
     mov         rsi, arg(0)
     mov         rdi, arg(1)
 
     movsxd      rax, dword ptr arg(2)
     lea         rcx, [rsi + rax*2]
 %else
   %ifidn __OUTPUT_FORMAT__,x64
     %define     input       rcx
     %define     output      rdx
     %define     pitch       r8
+    SAVE_XMM 7, u
   %else
     %define     input       rdi
     %define     output      rsi
     %define     pitch       rdx
   %endif
 %endif
 %endmacro
 
 %macro STACK_FRAME_DESTROY 0
   %define     input
   %define     output
   %define     pitch
 
 %if ABI_IS_32BIT
     pop         rdi
     pop         rsi
     RESTORE_GOT
     pop         rbp
 %else
   %ifidn __OUTPUT_FORMAT__,x64
+    RESTORE_XMM
   %endif
 %endif
     ret
 %endmacro
 
 ;void vp8_short_fdct4x4_sse2(short *input, short *output, int pitch)
 global sym(vp8_short_fdct4x4_sse2)
 sym(vp8_short_fdct4x4_sse2):
 
     STACK_FRAME_CREATE
@@ skipping 355 matching lines @@
     times 4 dd 14500
 align 16
 _7500:
     times 4 dd 7500
 align 16
 _12000:
     times 4 dd 12000
 align 16
 _51000:
     times 4 dd 51000