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

source/libvpx/vp8/common/arm/neon/sixtappredict16x16_neon.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.
 ;
 
 
     EXPORT  |vp8_sixtap_predict16x16_neon|
     ARM
     REQUIRE8
     PRESERVE8
 
     AREA ||.text||, CODE, READONLY, ALIGN=2
+
+filter16_coeff
+    DCD     0,  0,  128,    0,   0,  0,   0,  0
+    DCD     0, -6,  123,   12,  -1,  0,   0,  0
+    DCD     2, -11, 108,   36,  -8,  1,   0,  0
+    DCD     0, -9,   93,   50,  -6,  0,   0,  0
+    DCD     3, -16,  77,   77, -16,  3,   0,  0
+    DCD     0, -6,   50,   93,  -9,  0,   0,  0
+    DCD     1, -8,   36,  108, -11,  2,   0,  0
+    DCD     0, -1,   12,  123,  -6,   0,  0,  0
+
 ; r0    unsigned char  *src_ptr,
 ; r1    int  src_pixels_per_line,
 ; r2    int  xoffset,
 ; r3    int  yoffset,
 ; r4    unsigned char *dst_ptr,
 ; stack(r5) int  dst_pitch
 
 ;Note: To take advantage of 8-bit mulplication instruction in NEON. First apply abs() to
 ; filter coeffs to make them u8. Then, use vmlsl for negtive coeffs. After multiplication,
 ; the result can be negtive. So, I treat the result as s16. But, since it is also possible
 ; that the result can be a large positive number (> 2^15-1), which could be confused as a
 ; negtive number. To avoid that error, apply filter coeffs in the order of 0, 1, 4 ,5 ,2,
 ; which ensures that the result stays in s16 range. Finally, saturated add the result by
 ; applying 3rd filter coeff. Same applys to other filter functions.
 
 |vp8_sixtap_predict16x16_neon| PROC
     push            {r4-r5, lr}
 
-    ldr             r12, _filter16_coeff_
+    adr             r12, filter16_coeff
     ldr             r4, [sp, #12]           ;load parameters from stack
     ldr             r5, [sp, #16]           ;load parameters from stack
 
     cmp             r2, #0                  ;skip first_pass filter if xoffset=0
     beq             secondpass_filter16x16_only
 
     add             r2, r12, r2, lsl #5     ;calculate filter location
 
     cmp             r3, #0                  ;skip second_pass filter if yoffset=0
 
@@ skipping 422 matching lines @@
     sub             r4, r4, r5, lsl #4
     add             r4, r4, #8
 
     bne filt_blk2d_spo16x16_outloop_neon
 
     pop             {r4-r5,pc}
 
     ENDP
 
 ;-----------------
-
-_filter16_coeff_
-    DCD     filter16_coeff
-filter16_coeff
-    DCD     0,  0,  128,    0,   0,  0,   0,  0
-    DCD     0, -6,  123,   12,  -1,  0,   0,  0
-    DCD     2, -11, 108,   36,  -8,  1,   0,  0
-    DCD     0, -9,   93,   50,  -6,  0,   0,  0
-    DCD     3, -16,  77,   77, -16,  3,   0,  0
-    DCD     0, -6,   50,   93,  -9,  0,   0,  0
-    DCD     1, -8,   36,  108, -11,  2,   0,  0
-    DCD     0, -1,   12,  123,  -6,   0,  0,  0
-
     END