libvpx: Pull from upstream

source/libvpx/vpx_dsp/arm/idct32x32_add_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 25 matching lines @@
 cospi_24_64 EQU  6270
 cospi_25_64 EQU  5520
 cospi_26_64 EQU  4756
 cospi_27_64 EQU  3981
 cospi_28_64 EQU  3196
 cospi_29_64 EQU  2404
 cospi_30_64 EQU  1606
 cospi_31_64 EQU   804
 
 
-    EXPORT  |vp9_idct32x32_1024_add_neon|
+    EXPORT  |vpx_idct32x32_1024_add_neon|
     ARM
     REQUIRE8
     PRESERVE8
 
     AREA ||.text||, CODE, READONLY, ALIGN=2
 
     AREA     Block, CODE, READONLY
 
     ; --------------------------------------------------------------------------
     ; Load from transposed_buffer
@@ skipping 224 matching lines @@
     MEND
     ; --------------------------------------------------------------------------
     ; Touches q8-q12, q15 (q13-q14 are preserved)
     ; valid output registers are anything but q8-q11
     MACRO
     DO_BUTTERFLY_STD $first_constant, $second_constant, $reg1, $reg2, $reg3, $reg4
     DO_BUTTERFLY d28, d29, d26, d27, $first_constant, $second_constant, $reg1, $reg2, $reg3, $reg4
     MEND
     ; --------------------------------------------------------------------------
 
-;void vp9_idct32x32_1024_add_neon(int16_t *input, uint8_t *dest, int dest_stride);
+;void vpx_idct32x32_1024_add_neon(int16_t *input, uint8_t *dest, int dest_stride);
 ;
 ;   r0  int16_t *input,
 ;   r1  uint8_t *dest,
 ;   r2  int dest_stride)
 ; loop counters
 ;   r4  bands loop counter
 ;   r5  pass loop counter
 ;   r8  transpose loop counter
 ; combine-add pointers
 ;   r6  dest + 31 * dest_stride, descending (30, 29, 28, ...)
 ;   r7  dest +  0 * dest_stride, ascending  (1, 2, 3, ...)
 ;   r9  dest + 15 * dest_stride, descending (14, 13, 12, ...)
 ;   r10 dest + 16 * dest_stride, ascending  (17, 18, 19, ...)
 
-|vp9_idct32x32_1024_add_neon| PROC
+|vpx_idct32x32_1024_add_neon| PROC
     ; This function does one pass of idct32x32 transform.
     ;
     ; This is done by transposing the input and then doing a 1d transform on
     ; columns. In the first pass, the transposed columns are the original
     ; rows. In the second pass, after the transposition, the colums are the
     ; original columns.
     ; The 1d transform is done by looping over bands of eight columns (the
     ; idct32_bands loop). For each band, the transform input transposition
     ; is done on demand, one band of four 8x8 matrices at a time. The four
     ; matrices are transposed by pairs (the idct32_transpose_pair loop).
@@ skipping 971 matching lines @@
 
     ; bands loop processing
     subs r4, r4, #1
     bne idct32_bands_loop
 
     ; stack operation
     add sp, sp, #512+2048+2048
     vpop {d8-d15}
     pop  {r4-r11}
     bx              lr
-    ENDP  ; |vp9_idct32x32_1024_add_neon|
+    ENDP  ; |vpx_idct32x32_1024_add_neon|
     END