Merge commit '53c20f17c78d1d8a0fc2505868f201e69ff59cc5'

* commit '53c20f17c78d1d8a0fc2505868f201e69ff59cc5':
  vp8: K&R formatting cosmetics

Conflicts:
	libavcodec/vp8.c
	libavcodec/vp8.h
	libavcodec/vp8data.h
	libavcodec/vp8dsp.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>

1 files changed, 307 insertions, 272 deletions

diff --git a/libavcodec/vp8dsp.c b/libavcodec/vp8dsp.c
index 6b8fb3d043..eec48cacf5 100644
--- a/libavcodec/vp8dsp.c
+++ b/libavcodec/vp8dsp.c
@@ -25,26 +25,27 @@
  * VP8 compatible video decoder
  */
 
-#include "mathops.h"
-#include "vp8dsp.h"
 #include "libavutil/common.h"
 #include "libavutil/intreadwrite.h"
 
+#include "mathops.h"
+#include "vp8dsp.h"
+
 #define MK_IDCT_DC_ADD4_C(name) \
 static void name ## _idct_dc_add4uv_c(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride)\
 {\
-    name ## _idct_dc_add_c(dst+stride*0+0, block[0], stride);\
-    name ## _idct_dc_add_c(dst+stride*0+4, block[1], stride);\
-    name ## _idct_dc_add_c(dst+stride*4+0, block[2], stride);\
-    name ## _idct_dc_add_c(dst+stride*4+4, block[3], stride);\
+    name ## _idct_dc_add_c(dst + stride * 0 + 0, block[0], stride);\
+    name ## _idct_dc_add_c(dst + stride * 0 + 4, block[1], stride);\
+    name ## _idct_dc_add_c(dst + stride * 4 + 0, block[2], stride);\
+    name ## _idct_dc_add_c(dst + stride * 4 + 4, block[3], stride);\
 }\
 \
 static void name ## _idct_dc_add4y_c(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride)\
 {\
-    name ## _idct_dc_add_c(dst+ 0, block[0], stride);\
-    name ## _idct_dc_add_c(dst+ 4, block[1], stride);\
-    name ## _idct_dc_add_c(dst+ 8, block[2], stride);\
-    name ## _idct_dc_add_c(dst+12, block[3], stride);\
+    name ## _idct_dc_add_c(dst + 0, block[0], stride);\
+    name ## _idct_dc_add_c(dst + 4, block[1], stride);\
+    name ## _idct_dc_add_c(dst + 8, block[2], stride);\
+    name ## _idct_dc_add_c(dst + 12, block[3], stride);\
 }
 
 #if CONFIG_VP7_DECODER
@@ -54,14 +55,14 @@ static void vp7_luma_dc_wht_c(int16_t block[4][4][16], int16_t dc[16])
     int16_t tmp[16];
 
     for (i = 0; i < 4; i++) {
-        a1 = (dc[i*4+0] + dc[i*4+2]) * 23170;
-        b1 = (dc[i*4+0] - dc[i*4+2]) * 23170;
-        c1 = dc[i*4+1] * 12540 - dc[i*4+3] * 30274;
-        d1 = dc[i*4+1] * 30274 + dc[i*4+3] * 12540;
-        tmp[i*4+0] = (a1 + d1) >> 14;
-        tmp[i*4+3] = (a1 - d1) >> 14;
-        tmp[i*4+1] = (b1 + c1) >> 14;
-        tmp[i*4+2] = (b1 - c1) >> 14;
+        a1 = (dc[i * 4 + 0] + dc[i * 4 + 2]) * 23170;
+        b1 = (dc[i * 4 + 0] - dc[i * 4 + 2]) * 23170;
+        c1 = dc[i * 4 + 1] * 12540 - dc[i * 4 + 3] * 30274;
+        d1 = dc[i * 4 + 1] * 30274 + dc[i * 4 + 3] * 12540;
+        tmp[i * 4 + 0] = (a1 + d1) >> 14;
+        tmp[i * 4 + 3] = (a1 - d1) >> 14;
+        tmp[i * 4 + 1] = (b1 + c1) >> 14;
+        tmp[i * 4 + 2] = (b1 - c1) >> 14;
     }
 
     for (i = 0; i < 4; i++) {
@@ -69,7 +70,7 @@ static void vp7_luma_dc_wht_c(int16_t block[4][4][16], int16_t dc[16])
         b1 = (tmp[i + 0] - tmp[i + 8]) * 23170;
         c1 = tmp[i + 4] * 12540 - tmp[i + 12] * 30274;
         d1 = tmp[i + 4] * 30274 + tmp[i + 12] * 12540;
-        AV_ZERO64(dc + i*4);
+        AV_ZERO64(dc + i * 4);
         block[0][i][0] = (a1 + d1 + 0x20000) >> 18;
         block[3][i][0] = (a1 - d1 + 0x20000) >> 18;
         block[1][i][0] = (b1 + c1 + 0x20000) >> 18;
@@ -96,15 +97,15 @@ static void vp7_idct_add_c(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
     int16_t tmp[16];
 
     for (i = 0; i < 4; i++) {
-        a1 = (block[i*4+0] + block[i*4+2]) * 23170;
-        b1 = (block[i*4+0] - block[i*4+2]) * 23170;
-        c1 = block[i*4+1] * 12540 - block[i*4+3] * 30274;
-        d1 = block[i*4+1] * 30274 + block[i*4+3] * 12540;
-        AV_ZERO64(block + i*4);
-        tmp[i*4+0] = (a1 + d1) >> 14;
-        tmp[i*4+3] = (a1 - d1) >> 14;
-        tmp[i*4+1] = (b1 + c1) >> 14;
-        tmp[i*4+2] = (b1 - c1) >> 14;
+        a1 = (block[i * 4 + 0] + block[i * 4 + 2]) * 23170;
+        b1 = (block[i * 4 + 0] - block[i * 4 + 2]) * 23170;
+        c1 = block[i * 4 + 1] * 12540 - block[i * 4 + 3] * 30274;
+        d1 = block[i * 4 + 1] * 30274 + block[i * 4 + 3] * 12540;
+        AV_ZERO64(block + i * 4);
+        tmp[i * 4 + 0] = (a1 + d1) >> 14;
+        tmp[i * 4 + 3] = (a1 - d1) >> 14;
+        tmp[i * 4 + 1] = (b1 + c1) >> 14;
+        tmp[i * 4 + 2] = (b1 - c1) >> 14;
     }
 
     for (i = 0; i < 4; i++) {
@@ -112,10 +113,10 @@ static void vp7_idct_add_c(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
         b1 = (tmp[i + 0] - tmp[i + 8]) * 23170;
         c1 = tmp[i + 4] * 12540 - tmp[i + 12] * 30274;
         d1 = tmp[i + 4] * 30274 + tmp[i + 12] * 12540;
-        dst[0*stride+i] = av_clip_uint8(dst[0*stride+i] + ((a1 + d1 + 0x20000) >> 18));
-        dst[3*stride+i] = av_clip_uint8(dst[3*stride+i] + ((a1 - d1 + 0x20000) >> 18));
-        dst[1*stride+i] = av_clip_uint8(dst[1*stride+i] + ((b1 + c1 + 0x20000) >> 18));
-        dst[2*stride+i] = av_clip_uint8(dst[2*stride+i] + ((b1 - c1 + 0x20000) >> 18));
+        dst[0 * stride + i] = av_clip_uint8(dst[0 * stride + i] + ((a1 + d1 + 0x20000) >> 18));
+        dst[3 * stride + i] = av_clip_uint8(dst[3 * stride + i] + ((a1 - d1 + 0x20000) >> 18));
+        dst[1 * stride + i] = av_clip_uint8(dst[1 * stride + i] + ((b1 + c1 + 0x20000) >> 18));
+        dst[2 * stride + i] = av_clip_uint8(dst[2 * stride + i] + ((b1 - c1 + 0x20000) >> 18));
     }
 }
 
@@ -143,23 +144,23 @@ static void vp8_luma_dc_wht_c(int16_t block[4][4][16], int16_t dc[16])
     int i, t0, t1, t2, t3;
 
     for (i = 0; i < 4; i++) {
-        t0 = dc[0*4+i] + dc[3*4+i];
-        t1 = dc[1*4+i] + dc[2*4+i];
-        t2 = dc[1*4+i] - dc[2*4+i];
-        t3 = dc[0*4+i] - dc[3*4+i];
-
-        dc[0*4+i] = t0 + t1;
-        dc[1*4+i] = t3 + t2;
-        dc[2*4+i] = t0 - t1;
-        dc[3*4+i] = t3 - t2;
+        t0 = dc[0 * 4 + i] + dc[3 * 4 + i];
+        t1 = dc[1 * 4 + i] + dc[2 * 4 + i];
+        t2 = dc[1 * 4 + i] - dc[2 * 4 + i];
+        t3 = dc[0 * 4 + i] - dc[3 * 4 + i];
+
+        dc[0 * 4 + i] = t0 + t1;
+        dc[1 * 4 + i] = t3 + t2;
+        dc[2 * 4 + i] = t0 - t1;
+        dc[3 * 4 + i] = t3 - t2;
     }
 
     for (i = 0; i < 4; i++) {
-        t0 = dc[i*4+0] + dc[i*4+3] + 3; // rounding
-        t1 = dc[i*4+1] + dc[i*4+2];
-        t2 = dc[i*4+1] - dc[i*4+2];
-        t3 = dc[i*4+0] - dc[i*4+3] + 3; // rounding
-        AV_ZERO64(dc + i*4);
+        t0 = dc[i * 4 + 0] + dc[i * 4 + 3] + 3; // rounding
+        t1 = dc[i * 4 + 1] + dc[i * 4 + 2];
+        t2 = dc[i * 4 + 1] - dc[i * 4 + 2];
+        t3 = dc[i * 4 + 0] - dc[i * 4 + 3] + 3; // rounding
+        AV_ZERO64(dc + i * 4);
 
         block[i][0][0] = (t0 + t1) >> 3;
         block[i][1][0] = (t3 + t2) >> 3;
@@ -181,8 +182,8 @@ static void vp8_luma_dc_wht_dc_c(int16_t block[4][4][16], int16_t dc[16])
     }
 }
 
-#define MUL_20091(a) ((((a)*20091) >> 16) + (a))
-#define MUL_35468(a)  (((a)*35468) >> 16)
+#define MUL_20091(a) ((((a) * 20091) >> 16) + (a))
+#define MUL_35468(a)  (((a) * 35468) >> 16)
 
 static void vp8_idct_add_c(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
 {
@@ -190,32 +191,32 @@ static void vp8_idct_add_c(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
     int16_t tmp[16];
 
     for (i = 0; i < 4; i++) {
-        t0 = block[0*4+i] + block[2*4+i];
-        t1 = block[0*4+i] - block[2*4+i];
-        t2 = MUL_35468(block[1*4+i]) - MUL_20091(block[3*4+i]);
-        t3 = MUL_20091(block[1*4+i]) + MUL_35468(block[3*4+i]);
-        block[0*4+i] = 0;
-        block[1*4+i] = 0;
-        block[2*4+i] = 0;
-        block[3*4+i] = 0;
-
-        tmp[i*4+0] = t0 + t3;
-        tmp[i*4+1] = t1 + t2;
-        tmp[i*4+2] = t1 - t2;
-        tmp[i*4+3] = t0 - t3;
+        t0 = block[0 * 4 + i] + block[2 * 4 + i];
+        t1 = block[0 * 4 + i] - block[2 * 4 + i];
+        t2 = MUL_35468(block[1 * 4 + i]) - MUL_20091(block[3 * 4 + i]);
+        t3 = MUL_20091(block[1 * 4 + i]) + MUL_35468(block[3 * 4 + i]);
+        block[0 * 4 + i] = 0;
+        block[1 * 4 + i] = 0;
+        block[2 * 4 + i] = 0;
+        block[3 * 4 + i] = 0;
+
+        tmp[i * 4 + 0] = t0 + t3;
+        tmp[i * 4 + 1] = t1 + t2;
+        tmp[i * 4 + 2] = t1 - t2;
+        tmp[i * 4 + 3] = t0 - t3;
     }
 
     for (i = 0; i < 4; i++) {
-        t0 = tmp[0*4+i] + tmp[2*4+i];
-        t1 = tmp[0*4+i] - tmp[2*4+i];
-        t2 = MUL_35468(tmp[1*4+i]) - MUL_20091(tmp[3*4+i]);
-        t3 = MUL_20091(tmp[1*4+i]) + MUL_35468(tmp[3*4+i]);
+        t0 = tmp[0 * 4 + i] + tmp[2 * 4 + i];
+        t1 = tmp[0 * 4 + i] - tmp[2 * 4 + i];
+        t2 = MUL_35468(tmp[1 * 4 + i]) - MUL_20091(tmp[3 * 4 + i]);
+        t3 = MUL_20091(tmp[1 * 4 + i]) + MUL_35468(tmp[3 * 4 + i]);
 
         dst[0] = av_clip_uint8(dst[0] + ((t0 + t3 + 4) >> 3));
         dst[1] = av_clip_uint8(dst[1] + ((t1 + t2 + 4) >> 3));
         dst[2] = av_clip_uint8(dst[2] + ((t1 - t2 + 4) >> 3));
         dst[3] = av_clip_uint8(dst[3] + ((t0 - t3 + 4) >> 3));
-        dst += stride;
+        dst   += stride;
     }
 }
 
@@ -229,7 +230,7 @@ static void vp8_idct_dc_add_c(uint8_t *dst, int16_t block[16], ptrdiff_t stride)
         dst[1] = av_clip_uint8(dst[1] + dc);
         dst[2] = av_clip_uint8(dst[2] + dc);
         dst[3] = av_clip_uint8(dst[3] + dc);
-        dst += stride;
+        dst   += stride;
     }
 }
 
@@ -237,25 +238,26 @@ MK_IDCT_DC_ADD4_C(vp8)
 #endif
 
 // because I like only having two parameters to pass functions...
-#define LOAD_PIXELS\
-    int av_unused p3 = p[-4*stride];\
-    int av_unused p2 = p[-3*stride];\
-    int av_unused p1 = p[-2*stride];\
-    int av_unused p0 = p[-1*stride];\
-    int av_unused q0 = p[ 0*stride];\
-    int av_unused q1 = p[ 1*stride];\
-    int av_unused q2 = p[ 2*stride];\
-    int av_unused q3 = p[ 3*stride];
-
-#define clip_int8(n) (cm[n+0x80]-0x80)
-
-static av_always_inline void filter_common(uint8_t *p, ptrdiff_t stride, int is4tap, int vpn)
+#define LOAD_PIXELS                                                           \
+    int av_unused p3 = p[-4 * stride];                                        \
+    int av_unused p2 = p[-3 * stride];                                        \
+    int av_unused p1 = p[-2 * stride];                                        \
+    int av_unused p0 = p[-1 * stride];                                        \
+    int av_unused q0 = p[ 0 * stride];                                        \
+    int av_unused q1 = p[ 1 * stride];                                        \
+    int av_unused q2 = p[ 2 * stride];                                        \
+    int av_unused q3 = p[ 3 * stride];
+
+#define clip_int8(n) (cm[n + 0x80] - 0x80)
+
+static av_always_inline void filter_common(uint8_t *p, ptrdiff_t stride,
+                                           int is4tap, int vpn)
 {
     LOAD_PIXELS
     int a, f1, f2;
     const uint8_t *cm = ff_crop_tab + MAX_NEG_CROP;
 
-    a = 3*(q0 - p0);
+    a = 3 * (q0 - p0);
 
     if (is4tap)
         a += clip_int8(p1 - q1);
@@ -264,23 +266,23 @@ static av_always_inline void filter_common(uint8_t *p, ptrdiff_t stride, int is4
 
     // We deviate from the spec here with c(a+3) >> 3
     // since that's what libvpx does.
-    f1 = FFMIN(a+4, 127) >> 3;
+    f1 = FFMIN(a + 4, 127) >> 3;
 
     if (vpn == 7)
         f2 = f1 - ((a & 7) == 4);
     else
-        f2 = FFMIN(a+3, 127) >> 3;
+        f2 = FFMIN(a + 3, 127) >> 3;
 
     // Despite what the spec says, we do need to clamp here to
     // be bitexact with libvpx.
-    p[-1*stride] = cm[p0 + f2];
-    p[ 0*stride] = cm[q0 - f1];
+    p[-1 * stride] = cm[p0 + f2];
+    p[ 0 * stride] = cm[q0 - f1];
 
     // only used for _inner on blocks without high edge variance
     if (!is4tap) {
-        a = (f1+1)>>1;
-        p[-2*stride] = cm[p1 + a];
-        p[ 1*stride] = cm[q1 - a];
+        a = (f1 + 1) >> 1;
+        p[-2 * stride] = cm[p1 + a];
+        p[ 1 * stride] = cm[q1 - a];
     }
 }
 
@@ -293,7 +295,7 @@ static av_always_inline int vp7_simple_limit(uint8_t *p, ptrdiff_t stride, int f
 static av_always_inline int vp8_simple_limit(uint8_t *p, ptrdiff_t stride, int flim)
 {
     LOAD_PIXELS
-    return 2*FFABS(p0-q0) + (FFABS(p1-q1) >> 1) <= flim;
+    return 2 * FFABS(p0 - q0) + (FFABS(p1 - q1) >> 1) <= flim;
 }
 
 /**
@@ -301,12 +303,17 @@ static av_always_inline int vp8_simple_limit(uint8_t *p, ptrdiff_t stride, int f
  * I - limit for interior difference
  */
 #define NORMAL_LIMIT(vpn) \
-static av_always_inline int vp ## vpn ## _normal_limit(uint8_t *p, ptrdiff_t stride, int E, int I)\
-{\
-    LOAD_PIXELS\
-    return vp ## vpn ## _simple_limit(p, stride, E)\
-        && FFABS(p3-p2) <= I && FFABS(p2-p1) <= I && FFABS(p1-p0) <= I\
-        && FFABS(q3-q2) <= I && FFABS(q2-q1) <= I && FFABS(q1-q0) <= I;\
+static av_always_inline int vp ## vpn ## _normal_limit(uint8_t *p, ptrdiff_t stride, \
+                                                       int E, int I)\
+{                                                      \
+    LOAD_PIXELS                                        \
+    return vp ## vpn ## _simple_limit(p, stride, E) && \
+           FFABS(p3 - p2) <= I &&                      \
+           FFABS(p2 - p1) <= I &&                      \
+           FFABS(p1 - p0) <= I &&                      \
+           FFABS(q3 - q2) <= I &&                      \
+           FFABS(q2 - q1) <= I &&                      \
+           FFABS(q1 - q0) <= I;                        \
 }
 
 NORMAL_LIMIT(7)
@@ -316,7 +323,7 @@ NORMAL_LIMIT(8)
 static av_always_inline int hev(uint8_t *p, ptrdiff_t stride, int thresh)
 {
     LOAD_PIXELS
-    return FFABS(p1-p0) > thresh || FFABS(q1-q0) > thresh;
+    return FFABS(p1 - p0) > thresh || FFABS(q1 - q0) > thresh;
 }
 
 static av_always_inline void filter_mbedge(uint8_t *p, ptrdiff_t stride)
@@ -326,61 +333,68 @@ static av_always_inline void filter_mbedge(uint8_t *p, ptrdiff_t stride)
 
     LOAD_PIXELS
 
-    w = clip_int8(p1-q1);
-    w = clip_int8(w + 3*(q0-p0));
+    w = clip_int8(p1 - q1);
+    w = clip_int8(w + 3 * (q0 - p0));
 
-    a0 = (27*w + 63) >> 7;
-    a1 = (18*w + 63) >> 7;
-    a2 = ( 9*w + 63) >> 7;
+    a0 = (27 * w + 63) >> 7;
+    a1 = (18 * w + 63) >> 7;
+    a2 =  (9 * w + 63) >> 7;
 
-    p[-3*stride] = cm[p2 + a2];
-    p[-2*stride] = cm[p1 + a1];
-    p[-1*stride] = cm[p0 + a0];
-    p[ 0*stride] = cm[q0 - a0];
-    p[ 1*stride] = cm[q1 - a1];
-    p[ 2*stride] = cm[q2 - a2];
+    p[-3 * stride] = cm[p2 + a2];
+    p[-2 * stride] = cm[p1 + a1];
+    p[-1 * stride] = cm[p0 + a0];
+    p[ 0 * stride] = cm[q0 - a0];
+    p[ 1 * stride] = cm[q1 - a1];
+    p[ 2 * stride] = cm[q2 - a2];
 }
 
-#define LOOP_FILTER(vpn, dir, size, stridea, strideb, maybe_inline) \
-static maybe_inline void vp ## vpn ## _ ## dir ## _loop_filter ## size ## _c(uint8_t *dst, ptrdiff_t stride,\
-                                     int flim_E, int flim_I, int hev_thresh)\
-{\
-    int i;\
-\
-    for (i = 0; i < size; i++)\
-        if (vp ## vpn ## _normal_limit(dst+i*stridea, strideb, flim_E, flim_I)) {\
-            if (hev(dst+i*stridea, strideb, hev_thresh))\
-                filter_common(dst+i*stridea, strideb, 1, vpn);\
-            else\
-                filter_mbedge(dst+i*stridea, strideb);\
-        }\
-}\
-\
-static maybe_inline void vp ## vpn ## _ ## dir ## _loop_filter ## size ## _inner_c(uint8_t *dst, ptrdiff_t stride,\
-                                      int flim_E, int flim_I, int hev_thresh)\
-{\
-    int i;\
-\
-    for (i = 0; i < size; i++)\
-        if (vp ## vpn ## _normal_limit(dst+i*stridea, strideb, flim_E, flim_I)) {\
-            int hv = hev(dst+i*stridea, strideb, hev_thresh);\
-            if (hv) \
-                filter_common(dst+i*stridea, strideb, 1, vpn);\
-            else \
-                filter_common(dst+i*stridea, strideb, 0, vpn);\
-        }\
+#define LOOP_FILTER(vpn, dir, size, stridea, strideb, maybe_inline)           \
+static maybe_inline                                                           \
+void vp ## vpn ## _ ## dir ## _loop_filter ## size ## _c(uint8_t *dst,        \
+                                               ptrdiff_t stride,              \
+                                               int flim_E, int flim_I,        \
+                                               int hev_thresh)                \
+{                                                                             \
+    int i;                                                                    \
+    for (i = 0; i < size; i++)                                                \
+        if (vp ## vpn ## _normal_limit(dst + i * stridea, strideb, flim_E, flim_I)) {       \
+            if (hev(dst + i * stridea, strideb, hev_thresh))                  \
+                filter_common(dst + i * stridea, strideb, 1, vpn);            \
+            else                                                              \
+                filter_mbedge(dst + i * stridea, strideb);                    \
+        }                                                                     \
+}                                                                             \
+                                                                              \
+static maybe_inline                                                           \
+void vp ## vpn ## _ ## dir ## _loop_filter ## size ## _inner_c(uint8_t *dst,  \
+                                                     ptrdiff_t stride,        \
+                                                     int flim_E, int flim_I,  \
+                                                     int hev_thresh)          \
+{                                                                             \
+    int i;                                                                    \
+    for (i = 0; i < size; i++)                                                \
+        if (vp ## vpn ## _normal_limit(dst + i * stridea, strideb, flim_E, flim_I)) {       \
+            int hv = hev(dst + i * stridea, strideb, hev_thresh);             \
+            if (hv)                                                           \
+                filter_common(dst + i * stridea, strideb, 1, vpn);            \
+            else                                                              \
+                filter_common(dst + i * stridea, strideb, 0, vpn);            \
+        }                                                                     \
 }
 
 #define UV_LOOP_FILTER(vpn, dir, stridea, strideb) \
 LOOP_FILTER(vpn, dir, 8, stridea, strideb, av_always_inline) \
-static void vp ## vpn ## _ ## dir ## _loop_filter8uv_c(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,\
-                                      int fE, int fI, int hev_thresh)\
+static void vp ## vpn ## _ ## dir ## _loop_filter8uv_c(uint8_t *dstU, uint8_t *dstV, \
+                                                       ptrdiff_t stride, int fE, \
+                                                       int fI, int hev_thresh)\
 {\
   vp ## vpn ## _ ## dir ## _loop_filter8_c(dstU, stride, fE, fI, hev_thresh);\
   vp ## vpn ## _ ## dir ## _loop_filter8_c(dstV, stride, fE, fI, hev_thresh);\
 }\
-static void vp ## vpn ## _ ## dir ## _loop_filter8uv_inner_c(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,\
-                                      int fE, int fI, int hev_thresh)\
+static void vp ## vpn ## _ ## dir ## _loop_filter8uv_inner_c(uint8_t *dstU, \
+                                                             uint8_t *dstV, \
+                                                             ptrdiff_t stride, int fE, \
+                                                             int fI, int hev_thresh) \
 {\
   vp ## vpn ## _ ## dir ## _loop_filter8_inner_c(dstU, stride, fE, fI, hev_thresh);\
   vp ## vpn ## _ ## dir ## _loop_filter8_inner_c(dstV, stride, fE, fI, hev_thresh);\
@@ -392,8 +406,8 @@ static void vp ## vpn ## _v_loop_filter_simple_c(uint8_t *dst, ptrdiff_t stride,
     int i;\
 \
     for (i = 0; i < 16; i++)\
-        if (vp ## vpn ## _simple_limit(dst+i, stride, flim))\
-            filter_common(dst+i, stride, 1, vpn);\
+        if (vp ## vpn ## _simple_limit(dst + i, stride, flim))\
+            filter_common(dst + i, stride, 1, vpn);\
 }\
 \
 static void vp ## vpn ## _h_loop_filter_simple_c(uint8_t *dst, ptrdiff_t stride, int flim)\
@@ -401,8 +415,8 @@ static void vp ## vpn ## _h_loop_filter_simple_c(uint8_t *dst, ptrdiff_t stride,
     int i;\
 \
     for (i = 0; i < 16; i++)\
-        if (vp ## vpn ## _simple_limit(dst+i*stride, 1, flim))\
-            filter_common(dst+i*stride, 1, 1, vpn);\
+        if (vp ## vpn ## _simple_limit(dst + i * stride, 1, flim))\
+            filter_common(dst + i * stride, 1, 1, vpn);\
 }
 
 #if CONFIG_VP7_DECODER
@@ -422,89 +436,105 @@ LOOP_FILTER_SIMPLE(8)
 #endif
 
 static const uint8_t subpel_filters[7][6] = {
-    { 0,   6, 123,  12,   1,   0 },
-    { 2,  11, 108,  36,   8,   1 },
-    { 0,   9,  93,  50,   6,   0 },
-    { 3,  16,  77,  77,  16,   3 },
-    { 0,   6,  50,  93,   9,   0 },
-    { 1,   8,  36, 108,  11,   2 },
-    { 0,   1,  12, 123,   6,   0 },
+    { 0,  6, 123,  12,  1, 0 },
+    { 2, 11, 108,  36,  8, 1 },
+    { 0,  9,  93,  50,  6, 0 },
+    { 3, 16,  77,  77, 16, 3 },
+    { 0,  6,  50,  93,  9, 0 },
+    { 1,  8,  36, 108, 11, 2 },
+    { 0,  1,  12, 123,  6, 0 },
 };
 
-#define PUT_PIXELS(WIDTH) \
-static void put_vp8_pixels ## WIDTH ##_c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int x, int y) { \
-    int i; \
-    for (i = 0; i < h; i++, dst+= dststride, src+= srcstride) { \
-        memcpy(dst, src, WIDTH); \
-    } \
+#define PUT_PIXELS(WIDTH)                                                     \
+static void put_vp8_pixels ## WIDTH ## _c(uint8_t *dst, ptrdiff_t dststride,  \
+                                          uint8_t *src, ptrdiff_t srcstride,  \
+                                          int h, int x, int y)                \
+{                                                                             \
+    int i;                                                                    \
+    for (i = 0; i < h; i++, dst += dststride, src += srcstride)               \
+        memcpy(dst, src, WIDTH);                                              \
 }
 
 PUT_PIXELS(16)
 PUT_PIXELS(8)
 PUT_PIXELS(4)
 
-#define FILTER_6TAP(src, F, stride) \
-    cm[(F[2]*src[x+0*stride] - F[1]*src[x-1*stride] + F[0]*src[x-2*stride] + \
-        F[3]*src[x+1*stride] - F[4]*src[x+2*stride] + F[5]*src[x+3*stride] + 64) >> 7]
-
-#define FILTER_4TAP(src, F, stride) \
-    cm[(F[2]*src[x+0*stride] - F[1]*src[x-1*stride] + \
-        F[3]*src[x+1*stride] - F[4]*src[x+2*stride] + 64) >> 7]
-
-#define VP8_EPEL_H(SIZE, TAPS) \
-static void put_vp8_epel ## SIZE ## _h ## TAPS ## _c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int mx, int my) \
-{ \
-    const uint8_t *filter = subpel_filters[mx-1]; \
-    const uint8_t *cm = ff_crop_tab + MAX_NEG_CROP; \
-    int x, y; \
-\
-    for (y = 0; y < h; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            dst[x] = FILTER_ ## TAPS ## TAP(src, filter, 1); \
-        dst += dststride; \
-        src += srcstride; \
-    } \
+#define FILTER_6TAP(src, F, stride)                                           \
+    cm[(F[2] * src[x + 0 * stride] - F[1] * src[x - 1 * stride] +             \
+        F[0] * src[x - 2 * stride] + F[3] * src[x + 1 * stride] -             \
+        F[4] * src[x + 2 * stride] + F[5] * src[x + 3 * stride] + 64) >> 7]
+
+#define FILTER_4TAP(src, F, stride)                                           \
+    cm[(F[2] * src[x + 0 * stride] - F[1] * src[x - 1 * stride] +             \
+        F[3] * src[x + 1 * stride] - F[4] * src[x + 2 * stride] + 64) >> 7]
+
+#define VP8_EPEL_H(SIZE, TAPS)                                                \
+static void put_vp8_epel ## SIZE ## _h ## TAPS ## _c(uint8_t *dst,            \
+                                                     ptrdiff_t dststride,     \
+                                                     uint8_t *src,            \
+                                                     ptrdiff_t srcstride,     \
+                                                     int h, int mx, int my)   \
+{                                                                             \
+    const uint8_t *filter = subpel_filters[mx - 1];                           \
+    const uint8_t *cm     = ff_crop_tab + MAX_NEG_CROP;                       \
+    int x, y;                                                                 \
+    for (y = 0; y < h; y++) {                                                 \
+        for (x = 0; x < SIZE; x++)                                            \
+            dst[x] = FILTER_ ## TAPS ## TAP(src, filter, 1);                  \
+        dst += dststride;                                                     \
+        src += srcstride;                                                     \
+    }                                                                         \
 }
-#define VP8_EPEL_V(SIZE, TAPS) \
-static void put_vp8_epel ## SIZE ## _v ## TAPS ## _c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int mx, int my) \
-{ \
-    const uint8_t *filter = subpel_filters[my-1]; \
-    const uint8_t *cm = ff_crop_tab + MAX_NEG_CROP; \
-    int x, y; \
-\
-    for (y = 0; y < h; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            dst[x] = FILTER_ ## TAPS ## TAP(src, filter, srcstride); \
-        dst += dststride; \
-        src += srcstride; \
-    } \
+
+#define VP8_EPEL_V(SIZE, TAPS)                                                \
+static void put_vp8_epel ## SIZE ## _v ## TAPS ## _c(uint8_t *dst,            \
+                                                     ptrdiff_t dststride,     \
+                                                     uint8_t *src,            \
+                                                     ptrdiff_t srcstride,     \
+                                                     int h, int mx, int my)   \
+{                                                                             \
+    const uint8_t *filter = subpel_filters[my - 1];                           \
+    const uint8_t *cm     = ff_crop_tab + MAX_NEG_CROP;                       \
+    int x, y;                                                                 \
+    for (y = 0; y < h; y++) {                                                 \
+        for (x = 0; x < SIZE; x++)                                            \
+            dst[x] = FILTER_ ## TAPS ## TAP(src, filter, srcstride);          \
+        dst += dststride;                                                     \
+        src += srcstride;                                                     \
+    }                                                                         \
 }
-#define VP8_EPEL_HV(SIZE, HTAPS, VTAPS) \
-static void put_vp8_epel ## SIZE ## _h ## HTAPS ## v ## VTAPS ## _c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int mx, int my) \
-{ \
-    const uint8_t *filter = subpel_filters[mx-1]; \
-    const uint8_t *cm = ff_crop_tab + MAX_NEG_CROP; \
-    int x, y; \
-    uint8_t tmp_array[(2*SIZE+VTAPS-1)*SIZE]; \
-    uint8_t *tmp = tmp_array; \
-    src -= (2-(VTAPS==4))*srcstride; \
-\
-    for (y = 0; y < h+VTAPS-1; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            tmp[x] = FILTER_ ## HTAPS ## TAP(src, filter, 1); \
-        tmp += SIZE; \
-        src += srcstride; \
-    } \
-\
-    tmp = tmp_array + (2-(VTAPS==4))*SIZE; \
-    filter = subpel_filters[my-1]; \
-\
-    for (y = 0; y < h; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            dst[x] = FILTER_ ## VTAPS ## TAP(tmp, filter, SIZE); \
-        dst += dststride; \
-        tmp += SIZE; \
-    } \
+
+#define VP8_EPEL_HV(SIZE, HTAPS, VTAPS)                                       \
+static void                                                                   \
+put_vp8_epel ## SIZE ## _h ## HTAPS ## v ## VTAPS ## _c(uint8_t *dst,         \
+                                                        ptrdiff_t dststride,  \
+                                                        uint8_t *src,         \
+                                                        ptrdiff_t srcstride,  \
+                                                        int h, int mx,        \
+                                                        int my)               \
+{                                                                             \
+    const uint8_t *filter = subpel_filters[mx - 1];                           \
+    const uint8_t *cm     = ff_crop_tab + MAX_NEG_CROP;                       \
+    int x, y;                                                                 \
+    uint8_t tmp_array[(2 * SIZE + VTAPS - 1) * SIZE];                         \
+    uint8_t *tmp = tmp_array;                                                 \
+    src -= (2 - (VTAPS == 4)) * srcstride;                                    \
+                                                                              \
+    for (y = 0; y < h + VTAPS - 1; y++) {                                     \
+        for (x = 0; x < SIZE; x++)                                            \
+            tmp[x] = FILTER_ ## HTAPS ## TAP(src, filter, 1);                 \
+        tmp += SIZE;                                                          \
+        src += srcstride;                                                     \
+    }                                                                         \
+    tmp    = tmp_array + (2 - (VTAPS == 4)) * SIZE;                           \
+    filter = subpel_filters[my - 1];                                          \
+                                                                              \
+    for (y = 0; y < h; y++) {                                                 \
+        for (x = 0; x < SIZE; x++)                                            \
+            dst[x] = FILTER_ ## VTAPS ## TAP(tmp, filter, SIZE);              \
+        dst += dststride;                                                     \
+        tmp += SIZE;                                                          \
+    }                                                                         \
 }
 
 VP8_EPEL_H(16, 4)
@@ -519,6 +549,7 @@ VP8_EPEL_V(4,  4)
 VP8_EPEL_V(16, 6)
 VP8_EPEL_V(8,  6)
 VP8_EPEL_V(4,  6)
+
 VP8_EPEL_HV(16, 4, 4)
 VP8_EPEL_HV(8,  4, 4)
 VP8_EPEL_HV(4,  4, 4)
@@ -532,73 +563,77 @@ VP8_EPEL_HV(16, 6, 6)
 VP8_EPEL_HV(8,  6, 6)
 VP8_EPEL_HV(4,  6, 6)
 
-#define VP8_BILINEAR(SIZE) \
-static void put_vp8_bilinear ## SIZE ## _h_c(uint8_t *dst, ptrdiff_t dstride, uint8_t *src, ptrdiff_t sstride, int h, int mx, int my) \
-{ \
-    int a = 8-mx, b = mx; \
-    int x, y; \
-\
-    for (y = 0; y < h; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            dst[x] = (a*src[x] + b*src[x+1] + 4) >> 3; \
-        dst += dstride; \
-        src += sstride; \
-    } \
-} \
-static void put_vp8_bilinear ## SIZE ## _v_c(uint8_t *dst, ptrdiff_t dstride, uint8_t *src, ptrdiff_t sstride, int h, int mx, int my) \
-{ \
-    int c = 8-my, d = my; \
-    int x, y; \
-\
-    for (y = 0; y < h; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            dst[x] = (c*src[x] + d*src[x+sstride] + 4) >> 3; \
-        dst += dstride; \
-        src += sstride; \
-    } \
-} \
-\
-static void put_vp8_bilinear ## SIZE ## _hv_c(uint8_t *dst, ptrdiff_t dstride, uint8_t *src, ptrdiff_t sstride, int h, int mx, int my) \
-{ \
-    int a = 8-mx, b = mx; \
-    int c = 8-my, d = my; \
-    int x, y; \
-    uint8_t tmp_array[(2*SIZE+1)*SIZE]; \
-    uint8_t *tmp = tmp_array; \
-\
-    for (y = 0; y < h+1; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            tmp[x] = (a*src[x] + b*src[x+1] + 4) >> 3; \
-        tmp += SIZE; \
-        src += sstride; \
-    } \
-\
-    tmp = tmp_array; \
-\
-    for (y = 0; y < h; y++) { \
-        for (x = 0; x < SIZE; x++) \
-            dst[x] = (c*tmp[x] + d*tmp[x+SIZE] + 4) >> 3; \
-        dst += dstride; \
-        tmp += SIZE; \
-    } \
+#define VP8_BILINEAR(SIZE)                                                    \
+static void put_vp8_bilinear ## SIZE ## _h_c(uint8_t *dst, ptrdiff_t dstride, \
+                                             uint8_t *src, ptrdiff_t sstride, \
+                                             int h, int mx, int my)           \
+{                                                                             \
+    int a = 8 - mx, b = mx;                                                   \
+    int x, y;                                                                 \
+    for (y = 0; y < h; y++) {                                                 \
+        for (x = 0; x < SIZE; x++)                                            \
+            dst[x] = (a * src[x] + b * src[x + 1] + 4) >> 3;                  \
+        dst += dstride;                                                       \
+        src += sstride;                                                       \
+    }                                                                         \
+}                                                                             \
+                                                                              \
+static void put_vp8_bilinear ## SIZE ## _v_c(uint8_t *dst, ptrdiff_t dstride, \
+                                             uint8_t *src, ptrdiff_t sstride, \
+                                             int h, int mx, int my)           \
+{                                                                             \
+    int c = 8 - my, d = my;                                                   \
+    int x, y;                                                                 \
+    for (y = 0; y < h; y++) {                                                 \
+        for (x = 0; x < SIZE; x++)                                            \
+            dst[x] = (c * src[x] + d * src[x + sstride] + 4) >> 3;            \
+        dst += dstride;                                                       \
+        src += sstride;                                                       \
+    }                                                                         \
+}                                                                             \
+                                                                              \
+static void put_vp8_bilinear ## SIZE ## _hv_c(uint8_t *dst,                   \
+                                              ptrdiff_t dstride,              \
+                                              uint8_t *src,                   \
+                                              ptrdiff_t sstride,              \
+                                              int h, int mx, int my)          \
+{                                                                             \
+    int a = 8 - mx, b = mx;                                                   \
+    int c = 8 - my, d = my;                                                   \
+    int x, y;                                                                 \
+    uint8_t tmp_array[(2 * SIZE + 1) * SIZE];                                 \
+    uint8_t *tmp = tmp_array;                                                 \
+    for (y = 0; y < h + 1; y++) {                                             \
+        for (x = 0; x < SIZE; x++)                                            \
+            tmp[x] = (a * src[x] + b * src[x + 1] + 4) >> 3;                  \
+        tmp += SIZE;                                                          \
+        src += sstride;                                                       \
+    }                                                                         \
+    tmp = tmp_array;                                                          \
+    for (y = 0; y < h; y++) {                                                 \
+        for (x = 0; x < SIZE; x++)                                            \
+            dst[x] = (c * tmp[x] + d * tmp[x + SIZE] + 4) >> 3;               \
+        dst += dstride;                                                       \
+        tmp += SIZE;                                                          \
+    }                                                                         \
 }
 
 VP8_BILINEAR(16)
 VP8_BILINEAR(8)
 VP8_BILINEAR(4)
 
-#define VP8_MC_FUNC(IDX, SIZE) \
-    dsp->put_vp8_epel_pixels_tab[IDX][0][0] = put_vp8_pixels ## SIZE ## _c; \
-    dsp->put_vp8_epel_pixels_tab[IDX][0][1] = put_vp8_epel ## SIZE ## _h4_c; \
-    dsp->put_vp8_epel_pixels_tab[IDX][0][2] = put_vp8_epel ## SIZE ## _h6_c; \
-    dsp->put_vp8_epel_pixels_tab[IDX][1][0] = put_vp8_epel ## SIZE ## _v4_c; \
+#define VP8_MC_FUNC(IDX, SIZE)                                                \
+    dsp->put_vp8_epel_pixels_tab[IDX][0][0] = put_vp8_pixels ## SIZE ## _c;   \
+    dsp->put_vp8_epel_pixels_tab[IDX][0][1] = put_vp8_epel ## SIZE ## _h4_c;  \
+    dsp->put_vp8_epel_pixels_tab[IDX][0][2] = put_vp8_epel ## SIZE ## _h6_c;  \
+    dsp->put_vp8_epel_pixels_tab[IDX][1][0] = put_vp8_epel ## SIZE ## _v4_c;  \
     dsp->put_vp8_epel_pixels_tab[IDX][1][1] = put_vp8_epel ## SIZE ## _h4v4_c; \
     dsp->put_vp8_epel_pixels_tab[IDX][1][2] = put_vp8_epel ## SIZE ## _h6v4_c; \
-    dsp->put_vp8_epel_pixels_tab[IDX][2][0] = put_vp8_epel ## SIZE ## _v6_c; \
+    dsp->put_vp8_epel_pixels_tab[IDX][2][0] = put_vp8_epel ## SIZE ## _v6_c;  \
     dsp->put_vp8_epel_pixels_tab[IDX][2][1] = put_vp8_epel ## SIZE ## _h4v6_c; \
     dsp->put_vp8_epel_pixels_tab[IDX][2][2] = put_vp8_epel ## SIZE ## _h6v6_c
 
-#define VP8_BILINEAR_MC_FUNC(IDX, SIZE) \
+#define VP8_BILINEAR_MC_FUNC(IDX, SIZE)                                       \
     dsp->put_vp8_bilinear_pixels_tab[IDX][0][0] = put_vp8_pixels ## SIZE ## _c; \
     dsp->put_vp8_bilinear_pixels_tab[IDX][0][1] = put_vp8_bilinear ## SIZE ## _h_c; \
     dsp->put_vp8_bilinear_pixels_tab[IDX][0][2] = put_vp8_bilinear ## SIZE ## _h_c; \