libvpx: Pull from upstream

source/libvpx/vpx_dsp/arm/vpx_convolve8_avg_neon_asm.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.
 ;
 
 
     ; These functions are only valid when:
     ; x_step_q4 == 16
     ; w%4 == 0
     ; h%4 == 0
     ; taps == 8
     ; VP9_FILTER_WEIGHT == 128
     ; VP9_FILTER_SHIFT == 7
 
     EXPORT  |vpx_convolve8_avg_horiz_neon|
     EXPORT  |vpx_convolve8_avg_vert_neon|
-    IMPORT  |vpx_convolve8_avg_horiz_c|
-    IMPORT  |vpx_convolve8_avg_vert_c|
     ARM
     REQUIRE8
     PRESERVE8
 
     AREA ||.text||, CODE, READONLY, ALIGN=2
 
     ; Multiply and accumulate by q0
     MACRO
     MULTIPLY_BY_Q0 $dst, $src0, $src1, $src2, $src3, $src4, $src5, $src6, $src7
     vmull.s16 $dst, $src0, d0[0]
@@ skipping 11 matching lines @@
 ; r2    uint8_t *dst
 ; r3    int dst_stride
 ; sp[]const int16_t *filter_x
 ; sp[]int x_step_q4
 ; sp[]const int16_t *filter_y ; unused
 ; sp[]int y_step_q4           ; unused
 ; sp[]int w
 ; sp[]int h
 
 |vpx_convolve8_avg_horiz_neon| PROC
-    ldr             r12, [sp, #4]           ; x_step_q4
-    cmp             r12, #16
-    bne             vpx_convolve8_avg_horiz_c
-
     push            {r4-r10, lr}
 
     sub             r0, r0, #3              ; adjust for taps
 
     ldr             r5, [sp, #32]           ; filter_x
     ldr             r6, [sp, #48]           ; w
     ldr             r7, [sp, #52]           ; h
 
     vld1.s16        {q0}, [r5]              ; filter_x
 
@@ skipping 108 matching lines @@
     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 vpx_convolve8_avg_loop_horiz_v
 
     pop             {r4-r10, pc}
 
     ENDP
 
 |vpx_convolve8_avg_vert_neon| PROC
-    ldr             r12, [sp, #12]
-    cmp             r12, #16
-    bne             vpx_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
 
@@ skipping 92 matching lines @@
     ; outer loop
     add             r0, r0, #4
     add             r2, r2, #4
     subs            r6, r6, #4              ; w -= 4
     bgt             vpx_convolve8_avg_loop_vert_h
 
     pop             {r4-r8, pc}
 
     ENDP
     END