VP8 for ARMv8 by using NEON intrinsics 08

Add loopfilter_neon.c
- vp8_loop_filter_horizontal_edge_y_neon
- vp8_loop_filter_horizontal_edge_uv_neon
- vp8_loop_filter_vertical_edge_y_neon
- vp8_loop_filter_vertical_edge_uv_neon

Change-Id: I50b57dedabd42d2a3c183c1738cc5346f0e71ed8
Signed-off-by: James Yu <james.yu@linaro.org>

2 files changed, 538 insertions, 409 deletions

diff --git a/vp8/common/arm/neon/loopfilter_neon.asm b/vp8/common/arm/neon/loopfilter_neon.asm
deleted file mode 100644
index c4f09c775..000000000
--- a/vp8/common/arm/neon/loopfilter_neon.asm
+++ /dev/null
@@ -1,409 +0,0 @@
-;
-;  Copyright (c) 2010 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.
-;
-
-
-    EXPORT  |vp8_loop_filter_horizontal_edge_y_neon|
-    EXPORT  |vp8_loop_filter_horizontal_edge_uv_neon|
-    EXPORT  |vp8_loop_filter_vertical_edge_y_neon|
-    EXPORT  |vp8_loop_filter_vertical_edge_uv_neon|
-    ARM
-
-    AREA ||.text||, CODE, READONLY, ALIGN=2
-
-; r0    unsigned char *src
-; r1    int pitch
-; r2    unsigned char blimit
-; r3    unsigned char limit
-; sp    unsigned char thresh,
-|vp8_loop_filter_horizontal_edge_y_neon| PROC
-    push        {lr}
-    vpush       {d8-d15}
-
-    vdup.u8     q0, r2                     ; duplicate blimit
-    vdup.u8     q1, r3                     ; duplicate limit
-    sub         r2, r0, r1, lsl #2         ; move src pointer down by 4 lines
-    ldr         r3, [sp, #68]              ; load thresh
-    add         r12, r2, r1
-    add         r1, r1, r1
-
-    vdup.u8     q2, r3                     ; duplicate thresh
-
-    vld1.u8     {q3}, [r2@128], r1              ; p3
-    vld1.u8     {q4}, [r12@128], r1             ; p2
-    vld1.u8     {q5}, [r2@128], r1              ; p1
-    vld1.u8     {q6}, [r12@128], r1             ; p0
-    vld1.u8     {q7}, [r2@128], r1              ; q0
-    vld1.u8     {q8}, [r12@128], r1             ; q1
-    vld1.u8     {q9}, [r2@128]                  ; q2
-    vld1.u8     {q10}, [r12@128]                ; q3
-
-    sub         r2, r2, r1, lsl #1
-    sub         r12, r12, r1, lsl #1
-
-    bl          vp8_loop_filter_neon
-
-    vst1.u8     {q5}, [r2@128], r1              ; store op1
-    vst1.u8     {q6}, [r12@128], r1             ; store op0
-    vst1.u8     {q7}, [r2@128], r1              ; store oq0
-    vst1.u8     {q8}, [r12@128], r1             ; store oq1
-
-    vpop        {d8-d15}
-    pop         {pc}
-    ENDP        ; |vp8_loop_filter_horizontal_edge_y_neon|
-
-
-; r0    unsigned char *u,
-; r1    int pitch,
-; r2    unsigned char blimit
-; r3    unsigned char limit
-; sp    unsigned char thresh,
-; sp+4  unsigned char *v
-|vp8_loop_filter_horizontal_edge_uv_neon| PROC
-    push        {lr}
-    vpush       {d8-d15}
-
-    vdup.u8     q0, r2                      ; duplicate blimit
-    vdup.u8     q1, r3                      ; duplicate limit
-    ldr         r12, [sp, #68]              ; load thresh
-    ldr         r2, [sp, #72]               ; load v ptr
-    vdup.u8     q2, r12                     ; duplicate thresh
-
-    sub         r3, r0, r1, lsl #2          ; move u pointer down by 4 lines
-    sub         r12, r2, r1, lsl #2         ; move v pointer down by 4 lines
-
-    vld1.u8     {d6}, [r3@64], r1              ; p3
-    vld1.u8     {d7}, [r12@64], r1             ; p3
-    vld1.u8     {d8}, [r3@64], r1              ; p2
-    vld1.u8     {d9}, [r12@64], r1             ; p2
-    vld1.u8     {d10}, [r3@64], r1             ; p1
-    vld1.u8     {d11}, [r12@64], r1            ; p1
-    vld1.u8     {d12}, [r3@64], r1             ; p0
-    vld1.u8     {d13}, [r12@64], r1            ; p0
-    vld1.u8     {d14}, [r3@64], r1             ; q0
-    vld1.u8     {d15}, [r12@64], r1            ; q0
-    vld1.u8     {d16}, [r3@64], r1             ; q1
-    vld1.u8     {d17}, [r12@64], r1            ; q1
-    vld1.u8     {d18}, [r3@64], r1             ; q2
-    vld1.u8     {d19}, [r12@64], r1            ; q2
-    vld1.u8     {d20}, [r3@64]                 ; q3
-    vld1.u8     {d21}, [r12@64]                ; q3
-
-    bl          vp8_loop_filter_neon
-
-    sub         r0, r0, r1, lsl #1
-    sub         r2, r2, r1, lsl #1
-
-    vst1.u8     {d10}, [r0@64], r1             ; store u op1
-    vst1.u8     {d11}, [r2@64], r1             ; store v op1
-    vst1.u8     {d12}, [r0@64], r1             ; store u op0
-    vst1.u8     {d13}, [r2@64], r1             ; store v op0
-    vst1.u8     {d14}, [r0@64], r1             ; store u oq0
-    vst1.u8     {d15}, [r2@64], r1             ; store v oq0
-    vst1.u8     {d16}, [r0@64]                 ; store u oq1
-    vst1.u8     {d17}, [r2@64]                 ; store v oq1
-
-    vpop        {d8-d15}
-    pop         {pc}
-    ENDP        ; |vp8_loop_filter_horizontal_edge_uv_neon|
-
-; void vp8_loop_filter_vertical_edge_y_neon(unsigned char *src, int pitch,
-;                                           const signed char *flimit,
-;                                           const signed char *limit,
-;                                           const signed char *thresh,
-;                                           int count)
-; r0    unsigned char *src
-; r1    int pitch
-; r2    unsigned char blimit
-; r3    unsigned char limit
-; sp    unsigned char thresh,
-
-|vp8_loop_filter_vertical_edge_y_neon| PROC
-    push        {lr}
-    vpush       {d8-d15}
-
-    vdup.u8     q0, r2                     ; duplicate blimit
-    vdup.u8     q1, r3                     ; duplicate limit
-    sub         r2, r0, #4                 ; src ptr down by 4 columns
-    add         r1, r1, r1
-    ldr         r3, [sp, #68]              ; load thresh
-    add         r12, r2, r1, asr #1
-
-    vld1.u8     {d6}, [r2], r1
-    vld1.u8     {d8}, [r12], r1
-    vld1.u8     {d10}, [r2], r1
-    vld1.u8     {d12}, [r12], r1
-    vld1.u8     {d14}, [r2], r1
-    vld1.u8     {d16}, [r12], r1
-    vld1.u8     {d18}, [r2], r1
-    vld1.u8     {d20}, [r12], r1
-
-    vld1.u8     {d7}, [r2], r1              ; load second 8-line src data
-    vld1.u8     {d9}, [r12], r1
-    vld1.u8     {d11}, [r2], r1
-    vld1.u8     {d13}, [r12], r1
-    vld1.u8     {d15}, [r2], r1
-    vld1.u8     {d17}, [r12], r1
-    vld1.u8     {d19}, [r2]
-    vld1.u8     {d21}, [r12]
-
-    ;transpose to 8x16 matrix
-    vtrn.32     q3, q7
-    vtrn.32     q4, q8
-    vtrn.32     q5, q9
-    vtrn.32     q6, q10
-
-    vdup.u8     q2, r3                     ; duplicate thresh
-
-    vtrn.16     q3, q5
-    vtrn.16     q4, q6
-    vtrn.16     q7, q9
-    vtrn.16     q8, q10
-
-    vtrn.8      q3, q4
-    vtrn.8      q5, q6
-    vtrn.8      q7, q8
-    vtrn.8      q9, q10
-
-    bl          vp8_loop_filter_neon
-
-    vswp        d12, d11
-    vswp        d16, d13
-
-    sub         r0, r0, #2                 ; dst ptr
-
-    vswp        d14, d12
-    vswp        d16, d15
-
-    add         r12, r0, r1, asr #1
-
-    ;store op1, op0, oq0, oq1
-    vst4.8      {d10[0], d11[0], d12[0], d13[0]}, [r0], r1
-    vst4.8      {d10[1], d11[1], d12[1], d13[1]}, [r12], r1
-    vst4.8      {d10[2], d11[2], d12[2], d13[2]}, [r0], r1
-    vst4.8      {d10[3], d11[3], d12[3], d13[3]}, [r12], r1
-    vst4.8      {d10[4], d11[4], d12[4], d13[4]}, [r0], r1
-    vst4.8      {d10[5], d11[5], d12[5], d13[5]}, [r12], r1
-    vst4.8      {d10[6], d11[6], d12[6], d13[6]}, [r0], r1
-    vst4.8      {d10[7], d11[7], d12[7], d13[7]}, [r12], r1
-
-    vst4.8      {d14[0], d15[0], d16[0], d17[0]}, [r0], r1
-    vst4.8      {d14[1], d15[1], d16[1], d17[1]}, [r12], r1
-    vst4.8      {d14[2], d15[2], d16[2], d17[2]}, [r0], r1
-    vst4.8      {d14[3], d15[3], d16[3], d17[3]}, [r12], r1
-    vst4.8      {d14[4], d15[4], d16[4], d17[4]}, [r0], r1
-    vst4.8      {d14[5], d15[5], d16[5], d17[5]}, [r12], r1
-    vst4.8      {d14[6], d15[6], d16[6], d17[6]}, [r0]
-    vst4.8      {d14[7], d15[7], d16[7], d17[7]}, [r12]
-
-    vpop        {d8-d15}
-    pop         {pc}
-    ENDP        ; |vp8_loop_filter_vertical_edge_y_neon|
-
-; void vp8_loop_filter_vertical_edge_uv_neon(unsigned char *u, int pitch
-;                                            const signed char *flimit,
-;                                            const signed char *limit,
-;                                            const signed char *thresh,
-;                                            unsigned char *v)
-; r0    unsigned char *u,
-; r1    int pitch,
-; r2    unsigned char blimit
-; r3    unsigned char limit
-; sp    unsigned char thresh,
-; sp+4  unsigned char *v
-|vp8_loop_filter_vertical_edge_uv_neon| PROC
-    push        {lr}
-    vpush       {d8-d15}
-
-    vdup.u8     q0, r2                      ; duplicate blimit
-    sub         r12, r0, #4                 ; move u pointer down by 4 columns
-    ldr         r2, [sp, #72]               ; load v ptr
-    vdup.u8     q1, r3                      ; duplicate limit
-    sub         r3, r2, #4                  ; move v pointer down by 4 columns
-
-    vld1.u8     {d6}, [r12], r1             ;load u data
-    vld1.u8     {d7}, [r3], r1              ;load v data
-    vld1.u8     {d8}, [r12], r1
-    vld1.u8     {d9}, [r3], r1
-    vld1.u8     {d10}, [r12], r1
-    vld1.u8     {d11}, [r3], r1
-    vld1.u8     {d12}, [r12], r1
-    vld1.u8     {d13}, [r3], r1
-    vld1.u8     {d14}, [r12], r1
-    vld1.u8     {d15}, [r3], r1
-    vld1.u8     {d16}, [r12], r1
-    vld1.u8     {d17}, [r3], r1
-    vld1.u8     {d18}, [r12], r1
-    vld1.u8     {d19}, [r3], r1
-    vld1.u8     {d20}, [r12]
-    vld1.u8     {d21}, [r3]
-
-    ldr        r12, [sp, #68]              ; load thresh
-
-    ;transpose to 8x16 matrix
-    vtrn.32     q3, q7
-    vtrn.32     q4, q8
-    vtrn.32     q5, q9
-    vtrn.32     q6, q10
-
-    vdup.u8     q2, r12                     ; duplicate thresh
-
-    vtrn.16     q3, q5
-    vtrn.16     q4, q6
-    vtrn.16     q7, q9
-    vtrn.16     q8, q10
-
-    vtrn.8      q3, q4
-    vtrn.8      q5, q6
-    vtrn.8      q7, q8
-    vtrn.8      q9, q10
-
-    bl          vp8_loop_filter_neon
-
-    vswp        d12, d11
-    vswp        d16, d13
-    vswp        d14, d12
-    vswp        d16, d15
-
-    sub         r0, r0, #2
-    sub         r2, r2, #2
-
-    ;store op1, op0, oq0, oq1
-    vst4.8      {d10[0], d11[0], d12[0], d13[0]}, [r0], r1
-    vst4.8      {d14[0], d15[0], d16[0], d17[0]}, [r2], r1
-    vst4.8      {d10[1], d11[1], d12[1], d13[1]}, [r0], r1
-    vst4.8      {d14[1], d15[1], d16[1], d17[1]}, [r2], r1
-    vst4.8      {d10[2], d11[2], d12[2], d13[2]}, [r0], r1
-    vst4.8      {d14[2], d15[2], d16[2], d17[2]}, [r2], r1
-    vst4.8      {d10[3], d11[3], d12[3], d13[3]}, [r0], r1
-    vst4.8      {d14[3], d15[3], d16[3], d17[3]}, [r2], r1
-    vst4.8      {d10[4], d11[4], d12[4], d13[4]}, [r0], r1
-    vst4.8      {d14[4], d15[4], d16[4], d17[4]}, [r2], r1
-    vst4.8      {d10[5], d11[5], d12[5], d13[5]}, [r0], r1
-    vst4.8      {d14[5], d15[5], d16[5], d17[5]}, [r2], r1
-    vst4.8      {d10[6], d11[6], d12[6], d13[6]}, [r0], r1
-    vst4.8      {d14[6], d15[6], d16[6], d17[6]}, [r2], r1
-    vst4.8      {d10[7], d11[7], d12[7], d13[7]}, [r0]
-    vst4.8      {d14[7], d15[7], d16[7], d17[7]}, [r2]
-
-    vpop        {d8-d15}
-    pop         {pc}
-    ENDP        ; |vp8_loop_filter_vertical_edge_uv_neon|
-
-; void vp8_loop_filter_neon();
-; This is a helper function for the loopfilters. The invidual functions do the
-; necessary load, transpose (if necessary) and store.
-
-; r0-r3 PRESERVE
-; q0    flimit
-; q1    limit
-; q2    thresh
-; q3    p3
-; q4    p2
-; q5    p1
-; q6    p0
-; q7    q0
-; q8    q1
-; q9    q2
-; q10   q3
-|vp8_loop_filter_neon| PROC
-
-    ; vp8_filter_mask
-    vabd.u8     q11, q3, q4                 ; abs(p3 - p2)
-    vabd.u8     q12, q4, q5                 ; abs(p2 - p1)
-    vabd.u8     q13, q5, q6                 ; abs(p1 - p0)
-    vabd.u8     q14, q8, q7                 ; abs(q1 - q0)
-    vabd.u8     q3, q9, q8                  ; abs(q2 - q1)
-    vabd.u8     q4, q10, q9                 ; abs(q3 - q2)
-
-    vmax.u8     q11, q11, q12
-    vmax.u8     q12, q13, q14
-    vmax.u8     q3, q3, q4
-    vmax.u8     q15, q11, q12
-
-    vabd.u8     q9, q6, q7                  ; abs(p0 - q0)
-
-    ; vp8_hevmask
-    vcgt.u8     q13, q13, q2                ; (abs(p1 - p0) > thresh)*-1
-    vcgt.u8     q14, q14, q2                ; (abs(q1 - q0) > thresh)*-1
-    vmax.u8     q15, q15, q3
-
-    vmov.u8     q10, #0x80                   ; 0x80
-
-    vabd.u8     q2, q5, q8                  ; a = abs(p1 - q1)
-    vqadd.u8    q9, q9, q9                  ; b = abs(p0 - q0) * 2
-
-    vcge.u8     q15, q1, q15
-
-    ; vp8_filter() function
-    ; convert to signed
-    veor        q7, q7, q10                 ; qs0
-    vshr.u8     q2, q2, #1                  ; a = a / 2
-    veor        q6, q6, q10                 ; ps0
-
-    veor        q5, q5, q10                 ; ps1
-    vqadd.u8    q9, q9, q2                  ; a = b + a
-
-    veor        q8, q8, q10                 ; qs1
-
-    vmov.u8     q10, #3                     ; #3
-
-    vsubl.s8    q2, d14, d12                ; ( qs0 - ps0)
-    vsubl.s8    q11, d15, d13
-
-    vcge.u8     q9, q0, q9                  ; (a > flimit * 2 + limit) * -1
-
-    vmovl.u8    q4, d20
-
-    vqsub.s8    q1, q5, q8                  ; vp8_filter = clamp(ps1-qs1)
-    vorr        q14, q13, q14               ; vp8_hevmask
-
-    vmul.i16    q2, q2, q4                  ; 3 * ( qs0 - ps0)
-    vmul.i16    q11, q11, q4
-
-    vand        q1, q1, q14                 ; vp8_filter &= hev
-    vand        q15, q15, q9                ; vp8_filter_mask
-
-    vaddw.s8    q2, q2, d2
-    vaddw.s8    q11, q11, d3
-
-    vmov.u8     q9, #4                      ; #4
-
-    ; vp8_filter = clamp(vp8_filter + 3 * ( qs0 - ps0))
-    vqmovn.s16  d2, q2
-    vqmovn.s16  d3, q11
-    vand        q1, q1, q15                 ; vp8_filter &= mask
-
-    vqadd.s8    q2, q1, q10                 ; Filter2 = clamp(vp8_filter+3)
-    vqadd.s8    q1, q1, q9                  ; Filter1 = clamp(vp8_filter+4)
-    vshr.s8     q2, q2, #3                  ; Filter2 >>= 3
-    vshr.s8     q1, q1, #3                  ; Filter1 >>= 3
-
-
-    vqadd.s8    q11, q6, q2                 ; u = clamp(ps0 + Filter2)
-    vqsub.s8    q10, q7, q1                 ; u = clamp(qs0 - Filter1)
-
-    ; outer tap adjustments: ++vp8_filter >> 1
-    vrshr.s8    q1, q1, #1
-    vbic        q1, q1, q14                 ; vp8_filter &= ~hev
-    vmov.u8     q0, #0x80                   ; 0x80
-    vqadd.s8    q13, q5, q1                 ; u = clamp(ps1 + vp8_filter)
-    vqsub.s8    q12, q8, q1                 ; u = clamp(qs1 - vp8_filter)
-
-    veor        q6, q11, q0                 ; *op0 = u^0x80
-    veor        q7, q10, q0                 ; *oq0 = u^0x80
-    veor        q5, q13, q0                 ; *op1 = u^0x80
-    veor        q8, q12, q0                 ; *oq1 = u^0x80
-
-    bx          lr
-    ENDP        ; |vp8_loop_filter_horizontal_edge_y_neon|
-
-;-----------------
-
-    END
diff --git a/vp8/common/arm/neon/loopfilter_neon.c b/vp8/common/arm/neon/loopfilter_neon.c
new file mode 100644
index 000000000..47f522bc6
--- /dev/null
+++ b/vp8/common/arm/neon/loopfilter_neon.c
@@ -0,0 +1,538 @@
+/*
+ *  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.
+ */
+
+#include <arm_neon.h>
+
+static inline void vp8_loop_filter_neon(
+        uint8x16_t qblimit,  // flimit
+        uint8x16_t qlimit,   // limit
+        uint8x16_t qthresh,  // thresh
+        uint8x16_t q3,       // p3
+        uint8x16_t q4,       // p2
+        uint8x16_t q5,       // p1
+        uint8x16_t q6,       // p0
+        uint8x16_t q7,       // q0
+        uint8x16_t q8,       // q1
+        uint8x16_t q9,       // q2
+        uint8x16_t q10,      // q3
+        uint8x16_t *q5r,     // p1
+        uint8x16_t *q6r,     // p0
+        uint8x16_t *q7r,     // q0
+        uint8x16_t *q8r) {   // q1
+    uint8x16_t q0u8, q1u8, q2u8, q11u8, q12u8, q13u8, q14u8, q15u8;
+    int16x8_t q2s16, q11s16;
+    uint16x8_t q4u16;
+    int8x16_t q1s8, q2s8, q10s8, q11s8, q12s8, q13s8;
+    int8x8_t d2s8, d3s8;
+
+    q11u8 = vabdq_u8(q3, q4);
+    q12u8 = vabdq_u8(q4, q5);
+    q13u8 = vabdq_u8(q5, q6);
+    q14u8 = vabdq_u8(q8, q7);
+    q3    = vabdq_u8(q9, q8);
+    q4    = vabdq_u8(q10, q9);
+
+    q11u8 = vmaxq_u8(q11u8, q12u8);
+    q12u8 = vmaxq_u8(q13u8, q14u8);
+    q3    = vmaxq_u8(q3, q4);
+    q15u8 = vmaxq_u8(q11u8, q12u8);
+
+    q9 = vabdq_u8(q6, q7);
+
+    // vp8_hevmask
+    q13u8 = vcgtq_u8(q13u8, qthresh);
+    q14u8 = vcgtq_u8(q14u8, qthresh);
+    q15u8 = vmaxq_u8(q15u8, q3);
+
+    q2u8 = vabdq_u8(q5, q8);
+    q9 = vqaddq_u8(q9, q9);
+
+    q15u8 = vcgeq_u8(qlimit, q15u8);
+
+    // vp8_filter() function
+    // convert to signed
+    q10 = vdupq_n_u8(0x80);
+    q8 = veorq_u8(q8, q10);
+    q7 = veorq_u8(q7, q10);
+    q6 = veorq_u8(q6, q10);
+    q5 = veorq_u8(q5, q10);
+
+    q2u8 = vshrq_n_u8(q2u8, 1);
+    q9 = vqaddq_u8(q9, q2u8);
+
+    q10 = vdupq_n_u8(3);
+
+    q2s16 = vsubl_s8(vget_low_s8(vreinterpretq_s8_u8(q7)),
+                     vget_low_s8(vreinterpretq_s8_u8(q6)));
+    q11s16 = vsubl_s8(vget_high_s8(vreinterpretq_s8_u8(q7)),
+                      vget_high_s8(vreinterpretq_s8_u8(q6)));
+
+    q9 = vcgeq_u8(qblimit, q9);
+
+    q1s8 = vqsubq_s8(vreinterpretq_s8_u8(q5),
+                    vreinterpretq_s8_u8(q8));
+
+    q14u8 = vorrq_u8(q13u8, q14u8);
+
+    q4u16 = vmovl_u8(vget_low_u8(q10));
+    q2s16 = vmulq_s16(q2s16, vreinterpretq_s16_u16(q4u16));
+    q11s16 = vmulq_s16(q11s16, vreinterpretq_s16_u16(q4u16));
+
+    q1u8 = vandq_u8(vreinterpretq_u8_s8(q1s8), q14u8);
+    q15u8 = vandq_u8(q15u8, q9);
+
+    q1s8 = vreinterpretq_s8_u8(q1u8);
+    q2s16 = vaddw_s8(q2s16, vget_low_s8(q1s8));
+    q11s16 = vaddw_s8(q11s16, vget_high_s8(q1s8));
+
+    q9 = vdupq_n_u8(4);
+    // vp8_filter = clamp(vp8_filter + 3 * ( qs0 - ps0))
+    d2s8 = vqmovn_s16(q2s16);
+    d3s8 = vqmovn_s16(q11s16);
+    q1s8 = vcombine_s8(d2s8, d3s8);
+    q1u8 = vandq_u8(vreinterpretq_u8_s8(q1s8), q15u8);
+    q1s8 = vreinterpretq_s8_u8(q1u8);
+
+    q2s8 = vqaddq_s8(q1s8, vreinterpretq_s8_u8(q10));
+    q1s8 = vqaddq_s8(q1s8, vreinterpretq_s8_u8(q9));
+    q2s8 = vshrq_n_s8(q2s8, 3);
+    q1s8 = vshrq_n_s8(q1s8, 3);
+
+    q11s8 = vqaddq_s8(vreinterpretq_s8_u8(q6), q2s8);
+    q10s8 = vqsubq_s8(vreinterpretq_s8_u8(q7), q1s8);
+
+    q1s8 = vrshrq_n_s8(q1s8, 1);
+    q1s8 = vbicq_s8(q1s8, vreinterpretq_s8_u8(q14u8));
+
+    q13s8 = vqaddq_s8(vreinterpretq_s8_u8(q5), q1s8);
+    q12s8 = vqsubq_s8(vreinterpretq_s8_u8(q8), q1s8);
+
+    q0u8 = vdupq_n_u8(0x80);
+    *q8r = veorq_u8(vreinterpretq_u8_s8(q12s8), q0u8);
+    *q7r = veorq_u8(vreinterpretq_u8_s8(q10s8), q0u8);
+    *q6r = veorq_u8(vreinterpretq_u8_s8(q11s8), q0u8);
+    *q5r = veorq_u8(vreinterpretq_u8_s8(q13s8), q0u8);
+    return;
+}
+
+void vp8_loop_filter_horizontal_edge_y_neon(
+        unsigned char *src,
+        int pitch,
+        unsigned char blimit,
+        unsigned char limit,
+        unsigned char thresh) {
+    uint8x16_t qblimit, qlimit, qthresh, q3, q4;
+    uint8x16_t q5, q6, q7, q8, q9, q10;
+
+    qblimit = vdupq_n_u8(blimit);
+    qlimit  = vdupq_n_u8(limit);
+    qthresh = vdupq_n_u8(thresh);
+    src -= (pitch << 2);
+
+    q3 = vld1q_u8(src);
+    src += pitch;
+    q4 = vld1q_u8(src);
+    src += pitch;
+    q5 = vld1q_u8(src);
+    src += pitch;
+    q6 = vld1q_u8(src);
+    src += pitch;
+    q7 = vld1q_u8(src);
+    src += pitch;
+    q8 = vld1q_u8(src);
+    src += pitch;
+    q9 = vld1q_u8(src);
+    src += pitch;
+    q10 = vld1q_u8(src);
+
+    vp8_loop_filter_neon(qblimit, qlimit, qthresh, q3, q4,
+                         q5, q6, q7, q8, q9, q10,
+                         &q5, &q6, &q7, &q8);
+
+    src -= (pitch * 5);
+    vst1q_u8(src, q5);
+    src += pitch;
+    vst1q_u8(src, q6);
+    src += pitch;
+    vst1q_u8(src, q7);
+    src += pitch;
+    vst1q_u8(src, q8);
+    return;
+}
+
+void vp8_loop_filter_horizontal_edge_uv_neon(
+        unsigned char *u,
+        int pitch,
+        unsigned char blimit,
+        unsigned char limit,
+        unsigned char thresh,
+        unsigned char *v) {
+    uint8x16_t qblimit, qlimit, qthresh, q3, q4;
+    uint8x16_t q5, q6, q7, q8, q9, q10;
+    uint8x8_t d6, d7, d8, d9, d10, d11, d12, d13, d14;
+    uint8x8_t d15, d16, d17, d18, d19, d20, d21;
+
+    qblimit = vdupq_n_u8(blimit);
+    qlimit  = vdupq_n_u8(limit);
+    qthresh = vdupq_n_u8(thresh);
+
+    u -= (pitch << 2);
+    v -= (pitch << 2);
+
+    d6  = vld1_u8(u);
+    u += pitch;
+    d7  = vld1_u8(v);
+    v += pitch;
+    d8  = vld1_u8(u);
+    u += pitch;
+    d9  = vld1_u8(v);
+    v += pitch;
+    d10 = vld1_u8(u);
+    u += pitch;
+    d11 = vld1_u8(v);
+    v += pitch;
+    d12 = vld1_u8(u);
+    u += pitch;
+    d13 = vld1_u8(v);
+    v += pitch;
+    d14 = vld1_u8(u);
+    u += pitch;
+    d15 = vld1_u8(v);
+    v += pitch;
+    d16 = vld1_u8(u);
+    u += pitch;
+    d17 = vld1_u8(v);
+    v += pitch;
+    d18 = vld1_u8(u);
+    u += pitch;
+    d19 = vld1_u8(v);
+    v += pitch;
+    d20 = vld1_u8(u);
+    d21 = vld1_u8(v);
+
+    q3 = vcombine_u8(d6, d7);
+    q4 = vcombine_u8(d8, d9);
+    q5 = vcombine_u8(d10, d11);
+    q6 = vcombine_u8(d12, d13);
+    q7 = vcombine_u8(d14, d15);
+    q8 = vcombine_u8(d16, d17);
+    q9 = vcombine_u8(d18, d19);
+    q10 = vcombine_u8(d20, d21);
+
+    vp8_loop_filter_neon(qblimit, qlimit, qthresh, q3, q4,
+                         q5, q6, q7, q8, q9, q10,
+                         &q5, &q6, &q7, &q8);
+
+    u -= (pitch * 5);
+    vst1_u8(u, vget_low_u8(q5));
+    u += pitch;
+    vst1_u8(u, vget_low_u8(q6));
+    u += pitch;
+    vst1_u8(u, vget_low_u8(q7));
+    u += pitch;
+    vst1_u8(u, vget_low_u8(q8));
+
+    v -= (pitch * 5);
+    vst1_u8(v, vget_high_u8(q5));
+    v += pitch;
+    vst1_u8(v, vget_high_u8(q6));
+    v += pitch;
+    vst1_u8(v, vget_high_u8(q7));
+    v += pitch;
+    vst1_u8(v, vget_high_u8(q8));
+    return;
+}
+
+void vp8_loop_filter_vertical_edge_y_neon(
+        unsigned char *src,
+        int pitch,
+        unsigned char blimit,
+        unsigned char limit,
+        unsigned char thresh) {
+    unsigned char *s, *d;
+    uint8x16_t qblimit, qlimit, qthresh, q3, q4;
+    uint8x16_t q5, q6, q7, q8, q9, q10;
+    uint8x8_t d6, d7, d8, d9, d10, d11, d12, d13, d14;
+    uint8x8_t d15, d16, d17, d18, d19, d20, d21;
+    uint32x4x2_t q2tmp0, q2tmp1, q2tmp2, q2tmp3;
+    uint16x8x2_t q2tmp4, q2tmp5, q2tmp6, q2tmp7;
+    uint8x16x2_t q2tmp8, q2tmp9, q2tmp10, q2tmp11;
+    uint8x8x4_t q4ResultH, q4ResultL;
+
+    qblimit = vdupq_n_u8(blimit);
+    qlimit  = vdupq_n_u8(limit);
+    qthresh = vdupq_n_u8(thresh);
+
+    s = src - 4;
+    d6  = vld1_u8(s);
+    s += pitch;
+    d8  = vld1_u8(s);
+    s += pitch;
+    d10 = vld1_u8(s);
+    s += pitch;
+    d12 = vld1_u8(s);
+    s += pitch;
+    d14 = vld1_u8(s);
+    s += pitch;
+    d16 = vld1_u8(s);
+    s += pitch;
+    d18 = vld1_u8(s);
+    s += pitch;
+    d20 = vld1_u8(s);
+    s += pitch;
+    d7  = vld1_u8(s);
+    s += pitch;
+    d9  = vld1_u8(s);
+    s += pitch;
+    d11 = vld1_u8(s);
+    s += pitch;
+    d13 = vld1_u8(s);
+    s += pitch;
+    d15 = vld1_u8(s);
+    s += pitch;
+    d17 = vld1_u8(s);
+    s += pitch;
+    d19 = vld1_u8(s);
+    s += pitch;
+    d21 = vld1_u8(s);
+
+    q3 = vcombine_u8(d6, d7);
+    q4 = vcombine_u8(d8, d9);
+    q5 = vcombine_u8(d10, d11);
+    q6 = vcombine_u8(d12, d13);
+    q7 = vcombine_u8(d14, d15);
+    q8 = vcombine_u8(d16, d17);
+    q9 = vcombine_u8(d18, d19);
+    q10 = vcombine_u8(d20, d21);
+
+    q2tmp0 = vtrnq_u32(vreinterpretq_u32_u8(q3), vreinterpretq_u32_u8(q7));
+    q2tmp1 = vtrnq_u32(vreinterpretq_u32_u8(q4), vreinterpretq_u32_u8(q8));
+    q2tmp2 = vtrnq_u32(vreinterpretq_u32_u8(q5), vreinterpretq_u32_u8(q9));
+    q2tmp3 = vtrnq_u32(vreinterpretq_u32_u8(q6), vreinterpretq_u32_u8(q10));
+
+    q2tmp4 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp0.val[0]),
+                       vreinterpretq_u16_u32(q2tmp2.val[0]));
+    q2tmp5 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp1.val[0]),
+                       vreinterpretq_u16_u32(q2tmp3.val[0]));
+    q2tmp6 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp0.val[1]),
+                       vreinterpretq_u16_u32(q2tmp2.val[1]));
+    q2tmp7 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp1.val[1]),
+                       vreinterpretq_u16_u32(q2tmp3.val[1]));
+
+    q2tmp8  = vtrnq_u8(vreinterpretq_u8_u16(q2tmp4.val[0]),
+                       vreinterpretq_u8_u16(q2tmp5.val[0]));
+    q2tmp9  = vtrnq_u8(vreinterpretq_u8_u16(q2tmp4.val[1]),
+                       vreinterpretq_u8_u16(q2tmp5.val[1]));
+    q2tmp10 = vtrnq_u8(vreinterpretq_u8_u16(q2tmp6.val[0]),
+                       vreinterpretq_u8_u16(q2tmp7.val[0]));
+    q2tmp11 = vtrnq_u8(vreinterpretq_u8_u16(q2tmp6.val[1]),
+                       vreinterpretq_u8_u16(q2tmp7.val[1]));
+
+    q3 = q2tmp8.val[0];
+    q4 = q2tmp8.val[1];
+    q5 = q2tmp9.val[0];
+    q6 = q2tmp9.val[1];
+    q7 = q2tmp10.val[0];
+    q8 = q2tmp10.val[1];
+    q9 = q2tmp11.val[0];
+    q10 = q2tmp11.val[1];
+
+    vp8_loop_filter_neon(qblimit, qlimit, qthresh, q3, q4,
+                         q5, q6, q7, q8, q9, q10,
+                         &q5, &q6, &q7, &q8);
+
+    q4ResultL.val[0] = vget_low_u8(q5);   // d10
+    q4ResultL.val[1] = vget_low_u8(q6);   // d12
+    q4ResultL.val[2] = vget_low_u8(q7);   // d14
+    q4ResultL.val[3] = vget_low_u8(q8);   // d16
+    q4ResultH.val[0] = vget_high_u8(q5);  // d11
+    q4ResultH.val[1] = vget_high_u8(q6);  // d13
+    q4ResultH.val[2] = vget_high_u8(q7);  // d15
+    q4ResultH.val[3] = vget_high_u8(q8);  // d17
+
+    d = src - 2;
+    vst4_lane_u8(d, q4ResultL, 0);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultL, 1);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultL, 2);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultL, 3);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultL, 4);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultL, 5);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultL, 6);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultL, 7);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 0);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 1);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 2);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 3);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 4);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 5);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 6);
+    d += pitch;
+    vst4_lane_u8(d, q4ResultH, 7);
+    return;
+}
+
+void vp8_loop_filter_vertical_edge_uv_neon(
+        unsigned char *u,
+        int pitch,
+        unsigned char blimit,
+        unsigned char limit,
+        unsigned char thresh,
+        unsigned char *v) {
+    unsigned char *us, *ud;
+    unsigned char *vs, *vd;
+    uint8x16_t qblimit, qlimit, qthresh, q3, q4;
+    uint8x16_t q5, q6, q7, q8, q9, q10;
+    uint8x8_t d6, d7, d8, d9, d10, d11, d12, d13, d14;
+    uint8x8_t d15, d16, d17, d18, d19, d20, d21;
+    uint32x4x2_t q2tmp0, q2tmp1, q2tmp2, q2tmp3;
+    uint16x8x2_t q2tmp4, q2tmp5, q2tmp6, q2tmp7;
+    uint8x16x2_t q2tmp8, q2tmp9, q2tmp10, q2tmp11;
+    uint8x8x4_t q4ResultH, q4ResultL;
+
+    qblimit = vdupq_n_u8(blimit);
+    qlimit  = vdupq_n_u8(limit);
+    qthresh = vdupq_n_u8(thresh);
+
+    us = u - 4;
+    d6 = vld1_u8(us);
+    us += pitch;
+    d8 = vld1_u8(us);
+    us += pitch;
+    d10 = vld1_u8(us);
+    us += pitch;
+    d12 = vld1_u8(us);
+    us += pitch;
+    d14 = vld1_u8(us);
+    us += pitch;
+    d16 = vld1_u8(us);
+    us += pitch;
+    d18 = vld1_u8(us);
+    us += pitch;
+    d20 = vld1_u8(us);
+
+    vs = v - 4;
+    d7 = vld1_u8(vs);
+    vs += pitch;
+    d9 = vld1_u8(vs);
+    vs += pitch;
+    d11 = vld1_u8(vs);
+    vs += pitch;
+    d13 = vld1_u8(vs);
+    vs += pitch;
+    d15 = vld1_u8(vs);
+    vs += pitch;
+    d17 = vld1_u8(vs);
+    vs += pitch;
+    d19 = vld1_u8(vs);
+    vs += pitch;
+    d21 = vld1_u8(vs);
+
+    q3 = vcombine_u8(d6, d7);
+    q4 = vcombine_u8(d8, d9);
+    q5 = vcombine_u8(d10, d11);
+    q6 = vcombine_u8(d12, d13);
+    q7 = vcombine_u8(d14, d15);
+    q8 = vcombine_u8(d16, d17);
+    q9 = vcombine_u8(d18, d19);
+    q10 = vcombine_u8(d20, d21);
+
+    q2tmp0 = vtrnq_u32(vreinterpretq_u32_u8(q3), vreinterpretq_u32_u8(q7));
+    q2tmp1 = vtrnq_u32(vreinterpretq_u32_u8(q4), vreinterpretq_u32_u8(q8));
+    q2tmp2 = vtrnq_u32(vreinterpretq_u32_u8(q5), vreinterpretq_u32_u8(q9));
+    q2tmp3 = vtrnq_u32(vreinterpretq_u32_u8(q6), vreinterpretq_u32_u8(q10));
+
+    q2tmp4 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp0.val[0]),
+                       vreinterpretq_u16_u32(q2tmp2.val[0]));
+    q2tmp5 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp1.val[0]),
+                       vreinterpretq_u16_u32(q2tmp3.val[0]));
+    q2tmp6 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp0.val[1]),
+                       vreinterpretq_u16_u32(q2tmp2.val[1]));
+    q2tmp7 = vtrnq_u16(vreinterpretq_u16_u32(q2tmp1.val[1]),
+                       vreinterpretq_u16_u32(q2tmp3.val[1]));
+
+    q2tmp8  = vtrnq_u8(vreinterpretq_u8_u16(q2tmp4.val[0]),
+                       vreinterpretq_u8_u16(q2tmp5.val[0]));
+    q2tmp9  = vtrnq_u8(vreinterpretq_u8_u16(q2tmp4.val[1]),
+                       vreinterpretq_u8_u16(q2tmp5.val[1]));
+    q2tmp10 = vtrnq_u8(vreinterpretq_u8_u16(q2tmp6.val[0]),
+                       vreinterpretq_u8_u16(q2tmp7.val[0]));
+    q2tmp11 = vtrnq_u8(vreinterpretq_u8_u16(q2tmp6.val[1]),
+                       vreinterpretq_u8_u16(q2tmp7.val[1]));
+
+    q3 = q2tmp8.val[0];
+    q4 = q2tmp8.val[1];
+    q5 = q2tmp9.val[0];
+    q6 = q2tmp9.val[1];
+    q7 = q2tmp10.val[0];
+    q8 = q2tmp10.val[1];
+    q9 = q2tmp11.val[0];
+    q10 = q2tmp11.val[1];
+
+    vp8_loop_filter_neon(qblimit, qlimit, qthresh, q3, q4,
+                         q5, q6, q7, q8, q9, q10,
+                         &q5, &q6, &q7, &q8);
+
+    q4ResultL.val[0] = vget_low_u8(q5);   // d10
+    q4ResultL.val[1] = vget_low_u8(q6);   // d12
+    q4ResultL.val[2] = vget_low_u8(q7);   // d14
+    q4ResultL.val[3] = vget_low_u8(q8);   // d16
+    ud = u - 2;
+    vst4_lane_u8(ud, q4ResultL, 0);
+    ud += pitch;
+    vst4_lane_u8(ud, q4ResultL, 1);
+    ud += pitch;
+    vst4_lane_u8(ud, q4ResultL, 2);
+    ud += pitch;
+    vst4_lane_u8(ud, q4ResultL, 3);
+    ud += pitch;
+    vst4_lane_u8(ud, q4ResultL, 4);
+    ud += pitch;
+    vst4_lane_u8(ud, q4ResultL, 5);
+    ud += pitch;
+    vst4_lane_u8(ud, q4ResultL, 6);
+    ud += pitch;
+    vst4_lane_u8(ud, q4ResultL, 7);
+
+    q4ResultH.val[0] = vget_high_u8(q5);  // d11
+    q4ResultH.val[1] = vget_high_u8(q6);  // d13
+    q4ResultH.val[2] = vget_high_u8(q7);  // d15
+    q4ResultH.val[3] = vget_high_u8(q8);  // d17
+    vd = v - 2;
+    vst4_lane_u8(vd, q4ResultH, 0);
+    vd += pitch;
+    vst4_lane_u8(vd, q4ResultH, 1);
+    vd += pitch;
+    vst4_lane_u8(vd, q4ResultH, 2);
+    vd += pitch;
+    vst4_lane_u8(vd, q4ResultH, 3);
+    vd += pitch;
+    vst4_lane_u8(vd, q4ResultH, 4);
+    vd += pitch;
+    vst4_lane_u8(vd, q4ResultH, 5);
+    vd += pitch;
+    vst4_lane_u8(vd, q4ResultH, 6);
+    vd += pitch;
+    vst4_lane_u8(vd, q4ResultH, 7);
+    return;
+}