libvpx: Pull from upstream

source/libvpx/vp9/common/arm/neon/vp9_iht8x8_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 558 matching lines @@
 ; This function will only handle tx_type of 1,2,3.
 |vp9_iht8x8_64_add_neon| PROC
 
     ; load the inputs into d16-d19
     vld1.s16        {q8,q9}, [r0]!
     vld1.s16        {q10,q11}, [r0]!
     vld1.s16        {q12,q13}, [r0]!
     vld1.s16        {q14,q15}, [r0]!
 
     push            {r0-r10}
+    vpush           {d8-d15}
 
     ; transpose the input data
     TRANSPOSE8X8
 
     ; decide the type of transform
     cmp         r3, #2
     beq         idct_iadst
     cmp         r3, #3
     beq         iadst_iadst
 
@@ skipping 40 matching lines @@
     ; first transform rows
     IADST8X8_1D
 
     ; transpose the matrix
     TRANSPOSE8X8
 
     ; then transform columns
     IADST8X8_1D
 
 end_vp9_iht8x8_64_add_neon
+    vpop           {d8-d15}
     pop            {r0-r10}
 
     ; ROUND_POWER_OF_TWO(temp_out[j], 5)
     vrshr.s16       q8, q8, #5
     vrshr.s16       q9, q9, #5
     vrshr.s16       q10, q10, #5
     vrshr.s16       q11, q11, #5
     vrshr.s16       q12, q12, #5
     vrshr.s16       q13, q13, #5
     vrshr.s16       q14, q14, #5
@@ skipping 38 matching lines @@
     vst1.64         {d2}, [r0], r2
     vst1.64         {d3}, [r0], r2
     vst1.64         {d4}, [r0], r2
     vst1.64         {d5}, [r0], r2
     vst1.64         {d6}, [r0], r2
     vst1.64         {d7}, [r0], r2
     bx          lr
     ENDP  ; |vp9_iht8x8_64_add_neon|
 
     END