libvpx: Pull from upstream

source/libvpx/vp9/common/arm/neon/vp9_reconintra_neon_asm.asm

 ;
 ;  Copyright (c) 2014 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 280 matching lines @@
 ;                                const uint8_t *above,
 ;                                const uint8_t *left)
 ; r0  uint8_t *dst
 ; r1  ptrdiff_t y_stride
 ; r2  const uint8_t *above
 ; r3  const uint8_t *left
 
 |vp9_tm_predictor_4x4_neon| PROC
     ; Load ytop_left = above[-1];
     sub                 r12, r2, #1
-    ldrb                r12, [r12]
-    vdup.u8             d0, r12
+    vld1.u8             {d0[]}, [r12]
 
     ; Load above 4 pixels
     vld1.32             {d2[0]}, [r2]
 
     ; Compute above - ytop_left
     vsubl.u8            q3, d2, d0
 
     ; Load left row by row and compute left + (above - ytop_left)
     ; 1st row and 2nd row
-    ldrb                r12, [r3], #1
-    ldrb                r2, [r3], #1
-    vdup.u16            q1, r12
-    vdup.u16            q2, r2
+    vld1.u8             {d2[]}, [r3]!
+    vld1.u8             {d4[]}, [r3]!
+    vmovl.u8            q1, d2
+    vmovl.u8            q2, d4
     vadd.s16            q1, q1, q3
     vadd.s16            q2, q2, q3
     vqmovun.s16         d0, q1
     vqmovun.s16         d1, q2
     vst1.32             {d0[0]}, [r0], r1
     vst1.32             {d1[0]}, [r0], r1
 
     ; 3rd row and 4th row
-    ldrb                r12, [r3], #1
-    ldrb                r2, [r3], #1
-    vdup.u16            q1, r12
-    vdup.u16            q2, r2
+    vld1.u8             {d2[]}, [r3]!
+    vld1.u8             {d4[]}, [r3]
+    vmovl.u8            q1, d2
+    vmovl.u8            q2, d4
     vadd.s16            q1, q1, q3
     vadd.s16            q2, q2, q3
     vqmovun.s16         d0, q1
     vqmovun.s16         d1, q2
     vst1.32             {d0[0]}, [r0], r1
     vst1.32             {d1[0]}, [r0], r1
     bx                  lr
     ENDP                ; |vp9_tm_predictor_4x4_neon|
 
 ;void vp9_tm_predictor_8x8_neon (uint8_t *dst, ptrdiff_t y_stride,
@@ skipping 284 matching lines @@
     vmovl.u8            q3, d0
     vst1.64             {d24-d27}, [r0], r1
 
     subs                r2, r2, #1
     bgt                 loop_32x32_neon
 
     bx                  lr
     ENDP                ; |vp9_tm_predictor_32x32_neon|
 
     END