libvpx: Pull from upstream

source/libvpx/vp9/common/arm/neon/vp9_convolve8_avg_neon.asm

 ;
 ;  Copyright (c) 2013 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 60 matching lines @@
 
     sub             r4, r3, r3, lsl #2      ; -dst_stride * 3
     add             r4, r4, #4              ; -dst_stride * 3 + 4
 
     rsb             r9, r6, r1, lsl #2      ; reset src for outer loop
     sub             r9, r9, #7
     rsb             r12, r6, r3, lsl #2     ; reset dst for outer loop
 
     mov             r10, r6                 ; w loop counter
 
-loop_horiz_v
+vp9_convolve8_avg_loop_horiz_v
     vld1.8          {d24}, [r0], r1
     vld1.8          {d25}, [r0], r1
     vld1.8          {d26}, [r0], r1
     vld1.8          {d27}, [r0], r8
 
     vtrn.16         q12, q13
     vtrn.8          d24, d25
     vtrn.8          d26, d27
 
     pld             [r0, r1, lsl #2]
 
     vmovl.u8        q8, d24
     vmovl.u8        q9, d25
     vmovl.u8        q10, d26
     vmovl.u8        q11, d27
 
     ; save a few instructions in the inner loop
     vswp            d17, d18
     vmov            d23, d21
 
     add             r0, r0, #3
 
-loop_horiz
+vp9_convolve8_avg_loop_horiz
     add             r5, r0, #64
 
     vld1.32         {d28[]}, [r0], r1
     vld1.32         {d29[]}, [r0], r1
     vld1.32         {d31[]}, [r0], r1
     vld1.32         {d30[]}, [r0], r8
 
     pld             [r5]
 
     vtrn.16         d28, d31
@@ skipping 48 matching lines @@
     vst1.u32        {d3[0]}, [r2@32], r3
     vst1.u32        {d2[1]}, [r2@32], r3
     vst1.u32        {d3[1]}, [r2@32], r4
 
     vmov            q8,  q9
     vmov            d20, d23
     vmov            q11, q12
     vmov            q9,  q13
 
     subs            r6, r6, #4              ; w -= 4
-    bgt             loop_horiz
+    bgt             vp9_convolve8_avg_loop_horiz
 
     ; outer loop
     mov             r6, r10                 ; restore w counter
     add             r0, r0, r9              ; src += src_stride * 4 - w
     add             r2, r2, r12             ; dst += dst_stride * 4 - w
     subs            r7, r7, #4              ; h -= 4
-    bgt loop_horiz_v
+    bgt vp9_convolve8_avg_loop_horiz_v
 
     pop             {r4-r10, pc}
 
     ENDP
 
 |vp9_convolve8_avg_vert_neon| PROC
     ldr             r12, [sp, #12]
     cmp             r12, #16
     bne             vp9_convolve8_avg_vert_c
 
     push            {r4-r8, lr}
 
     ; adjust for taps
     sub             r0, r0, r1
     sub             r0, r0, r1, lsl #1
 
     ldr             r4, [sp, #32]           ; filter_y
     ldr             r6, [sp, #40]           ; w
     ldr             lr, [sp, #44]           ; h
 
     vld1.s16        {q0}, [r4]              ; filter_y
 
     lsl             r1, r1, #1
     lsl             r3, r3, #1
 
-loop_vert_h
+vp9_convolve8_avg_loop_vert_h
     mov             r4, r0
     add             r7, r0, r1, asr #1
     mov             r5, r2
     add             r8, r2, r3, asr #1
     mov             r12, lr                 ; h loop counter
 
     vld1.u32        {d16[0]}, [r4], r1
     vld1.u32        {d16[1]}, [r7], r1
     vld1.u32        {d18[0]}, [r4], r1
     vld1.u32        {d18[1]}, [r7], r1
     vld1.u32        {d20[0]}, [r4], r1
     vld1.u32        {d20[1]}, [r7], r1
     vld1.u32        {d22[0]}, [r4], r1
 
     vmovl.u8        q8, d16
     vmovl.u8        q9, d18
     vmovl.u8        q10, d20
     vmovl.u8        q11, d22
 
-loop_vert
+vp9_convolve8_avg_loop_vert
     ; always process a 4x4 block at a time
     vld1.u32        {d24[0]}, [r7], r1
     vld1.u32        {d26[0]}, [r4], r1
     vld1.u32        {d26[1]}, [r7], r1
     vld1.u32        {d24[1]}, [r4], r1
 
     ; extract to s16
     vmovl.u8        q12, d24
     vmovl.u8        q13, d26
 
@@ skipping 44 matching lines @@
     vst1.u32        {d3[0]}, [r5@32], r3
     vst1.u32        {d3[1]}, [r8@32], r3
 
     vmov            q8, q10
     vmov            d18, d22
     vmov            d19, d24
     vmov            q10, q13
     vmov            d22, d25
 
     subs            r12, r12, #4            ; h -= 4
-    bgt             loop_vert
+    bgt             vp9_convolve8_avg_loop_vert
 
     ; outer loop
     add             r0, r0, #4
     add             r2, r2, #4
     subs            r6, r6, #4              ; w -= 4
-    bgt             loop_vert_h
+    bgt             vp9_convolve8_avg_loop_vert_h
 
     pop             {r4-r8, pc}
 
     ENDP
     END