@@ -38,6 +38,7 @@

 #if CONFIG_POSTPROC

 #include "libpostproc/postprocess.h"

 #endif

+#include "libavutil/avassert.h"

 #include "libavutil/avstring.h"

 #include "libavutil/mathematics.h"

 #include "libavutil/parseutils.h"

@@ -1075,7 +1076,7 @@ int check_stream_specifier(AVFormatContext *s, AVStream *st, const char *spec)

         case 's': type = AVMEDIA_TYPE_SUBTITLE;   break;

         case 'd': type = AVMEDIA_TYPE_DATA;       break;

         case 't': type = AVMEDIA_TYPE_ATTACHMENT; break;

-        default: abort(); // never reached, silence warning

+        default:  av_assert0(0);

         }

         if (type != st->codec->codec_type)

             return 0;

@@ -2662,7 +2662,7 @@ case $target_os in

         SHFLAGS='-shared -Wl,-h,$$(@F)'

         enabled x86 && SHFLAGS="-mimpure-text $SHFLAGS"

         network_extralibs="-lsocket -lnsl"

-        add_cppflags -D__EXTENSIONS__

+        add_cppflags -D__EXTENSIONS__ -D_XOPEN_SOURCE=600

         # When using suncc to build, the Solaris linker will mark

         # an executable with each instruction set encountered by

         # the Solaris assembler.  As our libraries contain their own

@@ -393,7 +393,7 @@ static inline int range_get_symbol(APEContext *ctx,

 }

 /** @} */ // group rangecoder

-static inline void update_rice(APERice *rice, int x)

+static inline void update_rice(APERice *rice, unsigned int x)

 {

     int lim = rice->k ? (1 << (rice->k + 4)) : 0;

     rice->ksum += ((x + 1) / 2) - ((rice->ksum + 16) >> 5);

@@ -406,7 +406,7 @@ static inline void update_rice(APERice *rice, int x)

 static inline int ape_decode_value(APEContext *ctx, APERice *rice)

 {

-    int x, overflow;

+    unsigned int x, overflow;

     if (ctx->fileversion < 3990) {

         int tmpk;

@@ -95,6 +95,7 @@ endfunc

 .endm

 .macro  pixels16_y2     rnd=1, avg=0

+        sub             r3,  r3,  #2

         vld1.64         {q0},     [r1], r2

         vld1.64         {q1},     [r1], r2

 1:      subs            r3,  r3,  #2

@@ -114,10 +115,25 @@ endfunc

         vst1.64         {q2},     [r0,:128], r2

         vst1.64         {q3},     [r0,:128], r2

         bne             1b

+

+        avg             q2,  q0,  q1

+        vld1.64         {q0},     [r1], r2

+        avg             q3,  q0,  q1

+  .if \avg

+        vld1.8          {q8},     [r0,:128], r2

+        vld1.8          {q9},     [r0,:128]

+        vrhadd.u8       q2,  q2,  q8

+        vrhadd.u8       q3,  q3,  q9

+        sub             r0,  r0,  r2

+  .endif

+        vst1.64         {q2},     [r0,:128], r2

+        vst1.64         {q3},     [r0,:128], r2

+

         bx              lr

 .endm

 .macro  pixels16_xy2    rnd=1, avg=0

+        sub             r3,  r3,  #2

         vld1.64         {d0-d2},  [r1], r2

         vld1.64         {d4-d6},  [r1], r2

   .ifeq \rnd

@@ -173,6 +189,42 @@ endfunc

         vaddl.u8        q11, d3,  d5

         vst1.64         {q15},    [r0,:128], r2

         bgt             1b

+

+        vld1.64         {d0-d2},  [r1], r2

+        vadd.u16        q12, q8,  q9

+  .ifeq \rnd

+        vadd.u16        q12, q12, q13

+  .endif

+        vext.8          q15, q0,  q1,  #1

+        vadd.u16        q1 , q10, q11

+        shrn            d28, q12, #2

+  .ifeq \rnd

+        vadd.u16        q1,  q1,  q13

+  .endif

+        shrn            d29, q1,  #2

+  .if \avg

+        vld1.8          {q8},     [r0,:128]

+        vrhadd.u8       q14, q14, q8

+  .endif

+        vaddl.u8        q8,  d0,  d30

+        vaddl.u8        q10, d1,  d31

+        vst1.64         {q14},    [r0,:128], r2

+        vadd.u16        q12, q8,  q9

+  .ifeq \rnd

+        vadd.u16        q12, q12, q13

+  .endif

+        vadd.u16        q0,  q10, q11

+        shrn            d30, q12, #2

+  .ifeq \rnd

+        vadd.u16        q0,  q0,  q13

+  .endif

+        shrn            d31, q0,  #2

+  .if \avg

+        vld1.8          {q9},     [r0,:128]

+        vrhadd.u8       q15, q15, q9

+  .endif

+        vst1.64         {q15},    [r0,:128], r2

+

         bx              lr

 .endm

@@ -228,6 +280,7 @@ endfunc

 .endm

 .macro  pixels8_y2      rnd=1, avg=0

+        sub             r3,  r3,  #2

         vld1.64         {d0},     [r1], r2

         vld1.64         {d1},     [r1], r2

 1:      subs            r3,  r3,  #2

@@ -246,10 +299,24 @@ endfunc

         vst1.64         {d4},     [r0,:64], r2

         vst1.64         {d5},     [r0,:64], r2

         bne             1b

+

+        avg             d4,  d0,  d1

+        vld1.64         {d0},     [r1], r2

+        avg             d5,  d0,  d1

+  .if \avg

+        vld1.8          {d2},     [r0,:64], r2

+        vld1.8          {d3},     [r0,:64]

+        vrhadd.u8       q2,  q2,  q1

+        sub             r0,  r0,  r2

+  .endif

+        vst1.64         {d4},     [r0,:64], r2

+        vst1.64         {d5},     [r0,:64], r2

+

         bx              lr

 .endm

 .macro  pixels8_xy2     rnd=1, avg=0

+        sub             r3,  r3,  #2

         vld1.64         {q0},     [r1], r2

         vld1.64         {q1},     [r1], r2

   .ifeq \rnd

@@ -291,6 +358,31 @@ endfunc

         vaddl.u8        q9,  d2,  d6

         vst1.64         {d7},     [r0,:64], r2

         bgt             1b

+

+        vld1.64         {q0},     [r1], r2

+        vadd.u16        q10, q8,  q9

+        vext.8          d4,  d0,  d1,  #1

+  .ifeq \rnd

+        vadd.u16        q10, q10, q11

+  .endif

+        vaddl.u8        q8,  d0,  d4

+        shrn            d5,  q10, #2

+        vadd.u16        q10, q8,  q9

+  .if \avg

+        vld1.8          {d7},     [r0,:64]

+        vrhadd.u8       d5,  d5,  d7

+  .endif

+  .ifeq \rnd

+        vadd.u16        q10, q10, q11

+  .endif

+        vst1.64         {d5},     [r0,:64], r2

+        shrn            d7,  q10, #2

+  .if \avg

+        vld1.8          {d5},     [r0,:64]

+        vrhadd.u8       d7,  d7,  d5

+  .endif

+        vst1.64         {d7},     [r0,:64], r2

+

         bx              lr

 .endm

@@ -37,14 +37,14 @@

 #include "rectangle.h"

 #define interlaced_dct interlaced_dct_is_a_bad_name

-#define mb_intra mb_intra_is_not_initialized_see_mb_type

+#define mb_intra       mb_intra_is_not_initialized_see_mb_type

-#define MAX_SPS_COUNT 32

-#define MAX_PPS_COUNT 256

+#define MAX_SPS_COUNT          32

+#define MAX_PPS_COUNT         256

-#define MAX_MMCO_COUNT 66

+#define MAX_MMCO_COUNT         66

-#define MAX_DELAYED_PIC_COUNT 16

+#define MAX_DELAYED_PIC_COUNT  16

 #define MAX_MBPAIR_SIZE (256*1024) // a tighter bound could be calculated if someone cares about a few bytes

@@ -61,25 +61,25 @@

 #define MAX_SLICES 16

 #ifdef ALLOW_INTERLACE

-#define MB_MBAFF h->mb_mbaff

-#define MB_FIELD h->mb_field_decoding_flag

+#define MB_MBAFF    h->mb_mbaff

+#define MB_FIELD    h->mb_field_decoding_flag

 #define FRAME_MBAFF h->mb_aff_frame

 #define FIELD_PICTURE (s->picture_structure != PICT_FRAME)

 #define LEFT_MBS 2

-#define LTOP 0

-#define LBOT 1

-#define LEFT(i) (i)

+#define LTOP     0

+#define LBOT     1

+#define LEFT(i)  (i)

 #else

-#define MB_MBAFF 0

-#define MB_FIELD 0

-#define FRAME_MBAFF 0

+#define MB_MBAFF      0

+#define MB_FIELD      0

+#define FRAME_MBAFF   0

 #define FIELD_PICTURE 0

 #undef  IS_INTERLACED

 #define IS_INTERLACED(mb_type) 0

 #define LEFT_MBS 1

-#define LTOP 0

-#define LBOT 0

-#define LEFT(i) 0

+#define LTOP     0

+#define LBOT     0

+#define LEFT(i)  0

 #endif

 #define FIELD_OR_MBAFF_PICTURE (FRAME_MBAFF || FIELD_PICTURE)

@@ -91,9 +91,9 @@

 #define CHROMA422 (h->sps.chroma_format_idc == 2)

 #define CHROMA444 (h->sps.chroma_format_idc == 3)

-#define EXTENDED_SAR          255

+#define EXTENDED_SAR       255

-#define MB_TYPE_REF0       MB_TYPE_ACPRED //dirty but it fits in 16 bit

+#define MB_TYPE_REF0       MB_TYPE_ACPRED // dirty but it fits in 16 bit

 #define MB_TYPE_8x8DCT     0x01000000

 #define IS_REF0(a)         ((a) & MB_TYPE_REF0)

 #define IS_8x8DCT(a)       ((a) & MB_TYPE_8x8DCT)

@@ -108,7 +108,7 @@

 /* NAL unit types */

 enum {

-    NAL_SLICE=1,

+    NAL_SLICE = 1,

     NAL_DPA,

     NAL_DPB,

     NAL_DPC,

@@ -121,17 +121,17 @@ enum {

     NAL_END_STREAM,

     NAL_FILLER_DATA,

     NAL_SPS_EXT,

-    NAL_AUXILIARY_SLICE=19

+    NAL_AUXILIARY_SLICE = 19

 };

 /**

  * SEI message types

  */

 typedef enum {

-    SEI_BUFFERING_PERIOD             =  0, ///< buffering period (H.264, D.1.1)

-    SEI_TYPE_PIC_TIMING              =  1, ///< picture timing

-    SEI_TYPE_USER_DATA_UNREGISTERED  =  5, ///< unregistered user data

-    SEI_TYPE_RECOVERY_POINT          =  6  ///< recovery point (frame # to decoder sync)

+    SEI_BUFFERING_PERIOD            = 0,   ///< buffering period (H.264, D.1.1)

+    SEI_TYPE_PIC_TIMING             = 1,   ///< picture timing

+    SEI_TYPE_USER_DATA_UNREGISTERED = 5,   ///< unregistered user data

+    SEI_TYPE_RECOVERY_POINT         = 6    ///< recovery point (frame # to decoder sync)

 } SEI_Type;

 /**

@@ -152,8 +152,7 @@ typedef enum {

 /**

  * Sequence parameter set

  */

-typedef struct SPS{

-

+typedef struct SPS {

     int profile_idc;

     int level_idc;

     int chroma_format_idc;

@@ -170,9 +169,9 @@ typedef struct SPS{

     int mb_width;                      ///< pic_width_in_mbs_minus1 + 1

     int mb_height;                     ///< pic_height_in_map_units_minus1 + 1

     int frame_mbs_only_flag;

-    int mb_aff;                        ///<mb_adaptive_frame_field_flag

+    int mb_aff;                        ///< mb_adaptive_frame_field_flag

     int direct_8x8_inference_flag;

-    int crop;                   ///< frame_cropping_flag

+    int crop;                          ///< frame_cropping_flag

     unsigned int crop_left;            ///< frame_cropping_rect_left_offset

     unsigned int crop_right;           ///< frame_cropping_rect_right_offset

     unsigned int crop_top;             ///< frame_cropping_rect_top_offset

@@ -189,7 +188,7 @@ typedef struct SPS{

     uint32_t num_units_in_tick;

     uint32_t time_scale;

     int fixed_frame_rate_flag;

-    short offset_for_ref_frame[256]; //FIXME dyn aloc?

+    short offset_for_ref_frame[256]; // FIXME dyn aloc?

     int bitstream_restriction_flag;

     int num_reorder_frames;

     int scaling_matrix_present;

@@ -199,20 +198,20 @@ typedef struct SPS{

     int vcl_hrd_parameters_present_flag;

     int pic_struct_present_flag;

     int time_offset_length;

-    int cpb_cnt;                       ///< See H.264 E.1.2

-    int initial_cpb_removal_delay_length; ///< initial_cpb_removal_delay_length_minus1 +1

-    int cpb_removal_delay_length;      ///< cpb_removal_delay_length_minus1 + 1

-    int dpb_output_delay_length;       ///< dpb_output_delay_length_minus1 + 1

-    int bit_depth_luma;                ///< bit_depth_luma_minus8 + 8

-    int bit_depth_chroma;              ///< bit_depth_chroma_minus8 + 8

-    int residual_color_transform_flag; ///< residual_colour_transform_flag

-    int constraint_set_flags;          ///< constraint_set[0-3]_flag

-}SPS;

+    int cpb_cnt;                          ///< See H.264 E.1.2

+    int initial_cpb_removal_delay_length; ///< initial_cpb_removal_delay_length_minus1 + 1

+    int cpb_removal_delay_length;         ///< cpb_removal_delay_length_minus1 + 1

+    int dpb_output_delay_length;          ///< dpb_output_delay_length_minus1 + 1

+    int bit_depth_luma;                   ///< bit_depth_luma_minus8 + 8

+    int bit_depth_chroma;                 ///< bit_depth_chroma_minus8 + 8

+    int residual_color_transform_flag;    ///< residual_colour_transform_flag

+    int constraint_set_flags;             ///< constraint_set[0-3]_flag

+} SPS;

 /**

  * Picture parameter set

  */

-typedef struct PPS{

+typedef struct PPS {

     unsigned int sps_id;

     int cabac;                  ///< entropy_coding_mode_flag

     int pic_order_present;      ///< pic_order_present_flag

@@ -225,20 +224,20 @@ typedef struct PPS{

     int init_qs;                ///< pic_init_qs_minus26 + 26

     int chroma_qp_index_offset[2];

     int deblocking_filter_parameters_present; ///< deblocking_filter_parameters_present_flag

-    int constrained_intra_pred; ///< constrained_intra_pred_flag

-    int redundant_pic_cnt_present; ///< redundant_pic_cnt_present_flag

-    int transform_8x8_mode;     ///< transform_8x8_mode_flag

+    int constrained_intra_pred;     ///< constrained_intra_pred_flag

+    int redundant_pic_cnt_present;  ///< redundant_pic_cnt_present_flag

+    int transform_8x8_mode;         ///< transform_8x8_mode_flag

     uint8_t scaling_matrix4[6][16];

     uint8_t scaling_matrix8[6][64];

     uint8_t chroma_qp_table[2][QP_MAX_NUM+1];  ///< pre-scaled (with chroma_qp_index_offset) version of qp_table

     int chroma_qp_diff;

-}PPS;

+} PPS;

 /**

  * Memory management control operation opcode.

  */

-typedef enum MMCOOpcode{

-    MMCO_END=0,

+typedef enum MMCOOpcode {

+    MMCO_END = 0,

     MMCO_SHORT2UNUSED,

     MMCO_LONG2UNUSED,

     MMCO_SHORT2LONG,

@@ -250,7 +249,7 @@ typedef enum MMCOOpcode{

 /**

  * Memory management control operation.

  */

-typedef struct MMCO{

+typedef struct MMCO {

     MMCOOpcode opcode;

     int short_pic_num;  ///< pic_num without wrapping (pic_num & max_pic_num)

     int long_arg;       ///< index, pic_num, or num long refs depending on opcode

@@ -259,18 +258,18 @@ typedef struct MMCO{

 /**

  * H264Context

  */

-typedef struct H264Context{

+typedef struct H264Context {

     MpegEncContext s;

     H264DSPContext h264dsp;

     int pixel_shift;    ///< 0 for 8-bit H264, 1 for high-bit-depth H264

-    int chroma_qp[2]; //QPc

+    int chroma_qp[2];   // QPc

     int qp_thresh;      ///< QP threshold to skip loopfilter

     int prev_mb_skipped;

     int next_mb_skipped;

-    //prediction stuff

+    // prediction stuff

     int chroma_pred_mode;

     int intra16x16_pred_mode;

@@ -284,32 +283,32 @@ typedef struct H264Context{

     int topright_type;

     int left_type[LEFT_MBS];

-    const uint8_t * left_block;

+    const uint8_t *left_block;

     int topleft_partition;

-    int8_t intra4x4_pred_mode_cache[5*8];

-    int8_t (*intra4x4_pred_mode);

+    int8_t intra4x4_pred_mode_cache[5 * 8];

+    int8_t(*intra4x4_pred_mode);

     H264PredContext hpc;

     unsigned int topleft_samples_available;

     unsigned int top_samples_available;

     unsigned int topright_samples_available;

     unsigned int left_samples_available;

-    uint8_t (*top_borders[2])[(16*3)*2];

+    uint8_t (*top_borders[2])[(16 * 3) * 2];

     /**

      * non zero coeff count cache.

      * is 64 if not available.

      */

-    DECLARE_ALIGNED(8, uint8_t, non_zero_count_cache)[15*8];

+    DECLARE_ALIGNED(8, uint8_t, non_zero_count_cache)[15 * 8];

     uint8_t (*non_zero_count)[48];

     /**

      * Motion vector cache.

      */

-    DECLARE_ALIGNED(16, int16_t, mv_cache)[2][5*8][2];

-    DECLARE_ALIGNED(8, int8_t, ref_cache)[2][5*8];

-#define LIST_NOT_USED -1 //FIXME rename?

+    DECLARE_ALIGNED(16, int16_t, mv_cache)[2][5 * 8][2];

+    DECLARE_ALIGNED(8, int8_t, ref_cache)[2][5 * 8];

+#define LIST_NOT_USED -1 // FIXME rename?

 #define PART_NOT_AVAILABLE -2

     /**

@@ -321,13 +320,13 @@ typedef struct H264Context{

      * block_offset[ 0..23] for frame macroblocks

      * block_offset[24..47] for field macroblocks

      */

-    int block_offset[2*(16*3)];

+    int block_offset[2 * (16 * 3)];

-    uint32_t *mb2b_xy; //FIXME are these 4 a good idea?

+    uint32_t *mb2b_xy;  // FIXME are these 4 a good idea?

     uint32_t *mb2br_xy;

-    int b_stride; //FIXME use s->b4_stride

+    int b_stride;       // FIXME use s->b4_stride

-    int mb_linesize;   ///< may be equal to s->linesize or s->linesize*2, for mbaff

+    int mb_linesize;    ///< may be equal to s->linesize or s->linesize * 2, for mbaff

     int mb_uvlinesize;

     int emu_edge_width;

@@ -338,32 +337,32 @@ typedef struct H264Context{

     /**

      * current pps

      */

-    PPS pps; //FIXME move to Picture perhaps? (->no) do we need that?

+    PPS pps; // FIXME move to Picture perhaps? (->no) do we need that?

-    uint32_t dequant4_buffer[6][QP_MAX_NUM+1][16]; //FIXME should these be moved down?

-    uint32_t dequant8_buffer[6][QP_MAX_NUM+1][64];

-    uint32_t (*dequant4_coeff[6])[16];

-    uint32_t (*dequant8_coeff[6])[64];

+    uint32_t dequant4_buffer[6][QP_MAX_NUM + 1][16]; // FIXME should these be moved down?

+    uint32_t dequant8_buffer[6][QP_MAX_NUM + 1][64];

+    uint32_t(*dequant4_coeff[6])[16];

+    uint32_t(*dequant8_coeff[6])[64];

     int slice_num;

-    uint16_t *slice_table;     ///< slice_table_base + 2*mb_stride + 1

+    uint16_t *slice_table;      ///< slice_table_base + 2*mb_stride + 1

     int slice_type;

-    int slice_type_nos;        ///< S free slice type (SI/SP are remapped to I/P)

+    int slice_type_nos;         ///< S free slice type (SI/SP are remapped to I/P)

     int slice_type_fixed;

-    //interlacing specific flags

+    // interlacing specific flags

     int mb_aff_frame;

     int mb_field_decoding_flag;

-    int mb_mbaff;              ///< mb_aff_frame && mb_field_decoding_flag

+    int mb_mbaff;               ///< mb_aff_frame && mb_field_decoding_flag

     DECLARE_ALIGNED(8, uint16_t, sub_mb_type)[4];

-    //Weighted pred stuff

+    // Weighted pred stuff

     int use_weight;

     int use_weight_chroma;

     int luma_log2_weight_denom;

     int chroma_log2_weight_denom;

-    //The following 2 can be changed to int8_t but that causes 10cpu cycles speedloss

+    // The following 2 can be changed to int8_t but that causes 10cpu cycles speedloss

     int luma_weight[48][2][2];

     int chroma_weight[48][2][2][2];

     int implicit_weight[48][48][2];

@@ -373,48 +372,48 @@ typedef struct H264Context{

     int col_fieldoff;

     int dist_scale_factor[16];

     int dist_scale_factor_field[2][32];

-    int map_col_to_list0[2][16+32];

-    int map_col_to_list0_field[2][2][16+32];

+    int map_col_to_list0[2][16 + 32];

+    int map_col_to_list0_field[2][2][16 + 32];

     /**

      * num_ref_idx_l0/1_active_minus1 + 1

      */

-    uint8_t *list_counts;            ///< Array of list_count per MB specifying the slice type

-    unsigned int ref_count[2];   ///< counts frames or fields, depending on current mb mode

+    unsigned int ref_count[2];          ///< counts frames or fields, depending on current mb mode

     unsigned int list_count;

-    Picture ref_list[2][48];         /**< 0..15: frame refs, 16..47: mbaff field refs.

-                                          Reordered version of default_ref_list

-                                          according to picture reordering in slice header */

-    int ref2frm[MAX_SLICES][2][64];  ///< reference to frame number lists, used in the loop filter, the first 2 are for -2,-1

+    uint8_t *list_counts;               ///< Array of list_count per MB specifying the slice type

+    Picture ref_list[2][48];            /**< 0..15: frame refs, 16..47: mbaff field refs.

+                                         *   Reordered version of default_ref_list

+                                         *   according to picture reordering in slice header */

+    int ref2frm[MAX_SLICES][2][64];     ///< reference to frame number lists, used in the loop filter, the first 2 are for -2,-1

-    //data partitioning

+    // data partitioning

     GetBitContext intra_gb;

     GetBitContext inter_gb;

     GetBitContext *intra_gb_ptr;

     GetBitContext *inter_gb_ptr;

-    DECLARE_ALIGNED(16, DCTELEM, mb)[16*48*2]; ///< as a dct coeffecient is int32_t in high depth, we need to reserve twice the space.

-    DECLARE_ALIGNED(16, DCTELEM, mb_luma_dc)[3][16*2];

-    DCTELEM mb_padding[256*2];        ///< as mb is addressed by scantable[i] and scantable is uint8_t we can either check that i is not too large or ensure that there is some unused stuff after mb

+    DECLARE_ALIGNED(16, DCTELEM, mb)[16 * 48 * 2]; ///< as a dct coeffecient is int32_t in high depth, we need to reserve twice the space.

+    DECLARE_ALIGNED(16, DCTELEM, mb_luma_dc)[3][16 * 2];

+    DCTELEM mb_padding[256 * 2];        ///< as mb is addressed by scantable[i] and scantable is uint8_t we can either check that i is not too large or ensure that there is some unused stuff after mb

     /**

      * Cabac

      */

     CABACContext cabac;

-    uint8_t      cabac_state[1024];

+    uint8_t cabac_state[1024];

-    /* 0x100 -> non null luma_dc, 0x80/0x40 -> non null chroma_dc (cb/cr), 0x?0 -> chroma_cbp(0,1,2), 0x0? luma_cbp */

-    uint16_t     *cbp_table;

+    /* 0x100 -> non null luma_dc, 0x80/0x40 -> non null chroma_dc (cb/cr), 0x?0 -> chroma_cbp(0, 1, 2), 0x0? luma_cbp */

+    uint16_t *cbp_table;

     int cbp;

     int top_cbp;

     int left_cbp;

     /* chroma_pred_mode for i4x4 or i16x16, else 0 */

-    uint8_t     *chroma_pred_mode_table;

-    int         last_qscale_diff;

-    uint8_t     (*mvd_table[2])[2];

-    DECLARE_ALIGNED(16, uint8_t, mvd_cache)[2][5*8][2];

-    uint8_t     *direct_table;

-    uint8_t     direct_cache[5*8];

+    uint8_t *chroma_pred_mode_table;

+    int last_qscale_diff;

+    uint8_t (*mvd_table[2])[2];

+    DECLARE_ALIGNED(16, uint8_t, mvd_cache)[2][5 * 8][2];

+    uint8_t *direct_table;

+    uint8_t direct_cache[5 * 8];

     uint8_t zigzag_scan[16];

     uint8_t zigzag_scan8x8[64];

@@ -435,13 +434,13 @@ typedef struct H264Context{

     int is_complex;

-    //deblock

-    int deblocking_filter;         ///< disable_deblocking_filter_idc with 1<->0

+    // deblock

+    int deblocking_filter;          ///< disable_deblocking_filter_idc with 1 <-> 0

     int slice_alpha_c0_offset;

     int slice_beta_offset;

-//=============================================================

-    //Things below are not used in the MB or more inner code

+    // =============================================================

+    // Things below are not used in the MB or more inner code

     int nal_ref_idc;

     int nal_unit_type;

@@ -451,37 +450,36 @@ typedef struct H264Context{

     /**

      * Used to parse AVC variant of h264

      */

-    int is_avc; ///< this flag is != 0 if codec is avc1

-    int nal_length_size; ///< Number of bytes used for nal length (1, 2 or 4)

-    int got_first; ///< this flag is != 0 if we've parsed a frame

+    int is_avc;           ///< this flag is != 0 if codec is avc1

+    int nal_length_size;  ///< Number of bytes used for nal length (1, 2 or 4)

+    int got_first;        ///< this flag is != 0 if we've parsed a frame

     SPS *sps_buffers[MAX_SPS_COUNT];

     PPS *pps_buffers[MAX_PPS_COUNT];

-    int dequant_coeff_pps;     ///< reinit tables when pps changes

+    int dequant_coeff_pps;      ///< reinit tables when pps changes

     uint16_t *slice_table_base;

-

-    //POC stuff

+    // POC stuff

     int poc_lsb;

     int poc_msb;

     int delta_poc_bottom;

     int delta_poc[2];

     int frame_num;

-    int prev_poc_msb;             ///< poc_msb of the last reference pic for POC type 0

-    int prev_poc_lsb;             ///< poc_lsb of the last reference pic for POC type 0

-    int frame_num_offset;         ///< for POC type 2

-    int prev_frame_num_offset;    ///< for POC type 2

-    int prev_frame_num;           ///< frame_num of the last pic for POC type 1/2

+    int prev_poc_msb;           ///< poc_msb of the last reference pic for POC type 0

+    int prev_poc_lsb;           ///< poc_lsb of the last reference pic for POC type 0

+    int frame_num_offset;       ///< for POC type 2

+    int prev_frame_num_offset;  ///< for POC type 2

+    int prev_frame_num;         ///< frame_num of the last pic for POC type 1/2

     /**

-     * frame_num for frames or 2*frame_num+1 for field pics.

+     * frame_num for frames or 2 * frame_num + 1 for field pics.

      */

     int curr_pic_num;

     /**

-     * max_frame_num or 2*max_frame_num for field pics.

+     * max_frame_num or 2 * max_frame_num for field pics.

      */

     int max_pic_num;

@@ -490,7 +488,7 @@ typedef struct H264Context{

     Picture *short_ref[32];

     Picture *long_ref[32];

     Picture default_ref_list[2][32]; ///< base reference list for all slices of a coded picture

-    Picture *delayed_pic[MAX_DELAYED_PIC_COUNT+2]; //FIXME size?

+    Picture *delayed_pic[MAX_DELAYED_PIC_COUNT + 2]; // FIXME size?

     int last_pocs[MAX_DELAYED_PIC_COUNT];

     Picture *next_output_pic;

     int outputed_poc;

@@ -503,10 +501,10 @@ typedef struct H264Context{

     int mmco_index;

     int mmco_reset;

-    int long_ref_count;  ///< number of actual long term references

-    int short_ref_count; ///< number of actual short term references

+    int long_ref_count;     ///< number of actual long term references

+    int short_ref_count;    ///< number of actual short term references

-    int          cabac_init_idc;

+    int cabac_init_idc;

     /**

      * @name Members for slice based multithreading

@@ -582,12 +580,12 @@ typedef struct H264Context{

      */

     int recovery_frame;

-    int luma_weight_flag[2];   ///< 7.4.3.2 luma_weight_lX_flag

-    int chroma_weight_flag[2]; ///< 7.4.3.2 chroma_weight_lX_flag

+    int luma_weight_flag[2];    ///< 7.4.3.2 luma_weight_lX_flag

+    int chroma_weight_flag[2];  ///< 7.4.3.2 chroma_weight_lX_flag

     // Timestamp stuff

-    int sei_buffering_period_present;  ///< Buffering period SEI flag

-    int initial_cpb_removal_delay[32]; ///< Initial timestamps for CPBs

+    int sei_buffering_period_present;   ///< Buffering period SEI flag

+    int initial_cpb_removal_delay[32];  ///< Initial timestamps for CPBs

     int cur_chroma_format_idc;

@@ -598,10 +596,9 @@ typedef struct H264Context{

     uint8_t parse_history[4];

     int parse_history_count;

     int parse_last_mb;

-}H264Context;

-

+} H264Context;

-extern const uint8_t ff_h264_chroma_qp[5][QP_MAX_NUM+1]; ///< One chroma qp table for each possible bit depth (8-12).

+extern const uint8_t ff_h264_chroma_qp[5][QP_MAX_NUM + 1]; ///< One chroma qp table for each possible bit depth (8-12).

 extern const uint16_t ff_h264_mb_sizes[4];

 /**

@@ -628,13 +625,16 @@ int ff_h264_decode_picture_parameter_set(H264Context *h, int bit_length);

  * Decode a network abstraction layer unit.

  * @param consumed is the number of bytes used as input

  * @param length is the length of the array

- * @param dst_length is the number of decoded bytes FIXME here or a decode rbsp tailing?

+ * @param dst_length is the number of decoded bytes FIXME here

+ *                   or a decode rbsp tailing?

  * @return decoded bytes, might be src+1 if no escapes

  */

-const uint8_t *ff_h264_decode_nal(H264Context *h, const uint8_t *src, int *dst_length, int *consumed, int length);

+const uint8_t *ff_h264_decode_nal(H264Context *h, const uint8_t *src,

+                                  int *dst_length, int *consumed, int length);

 /**

- * Free any data that may have been allocated in the H264 context like SPS, PPS etc.

+ * Free any data that may have been allocated in the H264 context

+ * like SPS, PPS etc.

  */

 av_cold void ff_h264_free_context(H264Context *h);

@@ -667,12 +667,16 @@ int ff_h264_decode_ref_pic_marking(H264Context *h, GetBitContext *gb);

 void ff_generate_sliding_window_mmcos(H264Context *h);

-

 /**

- * Check if the top & left blocks are available if needed & change the dc mode so it only uses the available blocks.

+ * Check if the top & left blocks are available if needed & change the

+ * dc mode so it only uses the available blocks.

  */

 int ff_h264_check_intra4x4_pred_mode(H264Context *h);

+/**

+ * Check if the top & left blocks are available if needed & change the

+ * dc mode so it only uses the available blocks.

+ */

 int ff_h264_check_intra_pred_mode(H264Context *h, int mode, int is_chroma);

 void ff_h264_hl_decode_mb(H264Context *h);

@@ -683,24 +687,28 @@ av_cold void ff_h264_decode_init_vlc(void);

 /**

  * Decode a macroblock

- * @return 0 if OK, ER_AC_ERROR / ER_DC_ERROR / ER_MV_ERROR if an error is noticed

+ * @return 0 if OK, ER_AC_ERROR / ER_DC_ERROR / ER_MV_ERROR on error

  */

 int ff_h264_decode_mb_cavlc(H264Context *h);

 /**

  * Decode a CABAC coded macroblock

- * @return 0 if OK, ER_AC_ERROR / ER_DC_ERROR / ER_MV_ERROR if an error is noticed

+ * @return 0 if OK, ER_AC_ERROR / ER_DC_ERROR / ER_MV_ERROR on error

  */

 int ff_h264_decode_mb_cabac(H264Context *h);

 void ff_h264_init_cabac_states(H264Context *h);

-void ff_h264_direct_dist_scale_factor(H264Context * const h);

-void ff_h264_direct_ref_list_init(H264Context * const h);

-void ff_h264_pred_direct_motion(H264Context * const h, int *mb_type);

+void ff_h264_direct_dist_scale_factor(H264Context *const h);

+void ff_h264_direct_ref_list_init(H264Context *const h);

+void ff_h264_pred_direct_motion(H264Context *const h, int *mb_type);

-void ff_h264_filter_mb_fast( H264Context *h, int mb_x, int mb_y, uint8_t *img_y, uint8_t *img_cb, uint8_t *img_cr, unsigned int linesize, unsigned int uvlinesize);

-void ff_h264_filter_mb( H264Context *h, int mb_x, int mb_y, uint8_t *img_y, uint8_t *img_cb, uint8_t *img_cr, unsigned int linesize, unsigned int uvlinesize);

+void ff_h264_filter_mb_fast(H264Context *h, int mb_x, int mb_y,

+                            uint8_t *img_y, uint8_t *img_cb, uint8_t *img_cr,

+                            unsigned int linesize, unsigned int uvlinesize);

+void ff_h264_filter_mb(H264Context *h, int mb_x, int mb_y,

+                       uint8_t *img_y, uint8_t *img_cb, uint8_t *img_cr,

+                       unsigned int linesize, unsigned int uvlinesize);

 /**

  * Reset SEI values at the beginning of the frame.

@@ -709,16 +717,15 @@ void ff_h264_filter_mb( H264Context *h, int mb_x, int mb_y, uint8_t *img_y, uint

  */

 void ff_h264_reset_sei(H264Context *h);

-

 /*

-o-o o-o

- / / /

-o-o o-o

- ,---'

-o-o o-o

- / / /

-o-o o-o

-*/

+ * o-o o-o

+ *  / / /

+ * o-o o-o

+ *  ,---'

+ * o-o o-o

+ *  / / /

+ * o-o o-o

+ */

 /* Scan8 organization:

  *    0 1 2 3 4 5 6 7

@@ -743,156 +750,173 @@ o-o o-o

 #define LUMA_DC_BLOCK_INDEX   48

 #define CHROMA_DC_BLOCK_INDEX 49

-//This table must be here because scan8[constant] must be known at compiletime

-static const uint8_t scan8[16*3 + 3]={

- 4+ 1*8, 5+ 1*8, 4+ 2*8, 5+ 2*8,

- 6+ 1*8, 7+ 1*8, 6+ 2*8, 7+ 2*8,

- 4+ 3*8, 5+ 3*8, 4+ 4*8, 5+ 4*8,

- 6+ 3*8, 7+ 3*8, 6+ 4*8, 7+ 4*8,

- 4+ 6*8, 5+ 6*8, 4+ 7*8, 5+ 7*8,

- 6+ 6*8, 7+ 6*8, 6+ 7*8, 7+ 7*8,

- 4+ 8*8, 5+ 8*8, 4+ 9*8, 5+ 9*8,

- 6+ 8*8, 7+ 8*8, 6+ 9*8, 7+ 9*8,

- 4+11*8, 5+11*8, 4+12*8, 5+12*8,

- 6+11*8, 7+11*8, 6+12*8, 7+12*8,

- 4+13*8, 5+13*8, 4+14*8, 5+14*8,

- 6+13*8, 7+13*8, 6+14*8, 7+14*8,

- 0+ 0*8, 0+ 5*8, 0+10*8

+// This table must be here because scan8[constant] must be known at compiletime

+static const uint8_t scan8[16 * 3 + 3] = {

+    4 +  1 * 8, 5 +  1 * 8, 4 +  2 * 8, 5 +  2 * 8,

+    6 +  1 * 8, 7 +  1 * 8, 6 +  2 * 8, 7 +  2 * 8,

+    4 +  3 * 8, 5 +  3 * 8, 4 +  4 * 8, 5 +  4 * 8,

+    6 +  3 * 8, 7 +  3 * 8, 6 +  4 * 8, 7 +  4 * 8,

+    4 +  6 * 8, 5 +  6 * 8, 4 +  7 * 8, 5 +  7 * 8,

+    6 +  6 * 8, 7 +  6 * 8, 6 +  7 * 8, 7 +  7 * 8,

+    4 +  8 * 8, 5 +  8 * 8, 4 +  9 * 8, 5 +  9 * 8,

+    6 +  8 * 8, 7 +  8 * 8, 6 +  9 * 8, 7 +  9 * 8,

+    4 + 11 * 8, 5 + 11 * 8, 4 + 12 * 8, 5 + 12 * 8,

+    6 + 11 * 8, 7 + 11 * 8, 6 + 12 * 8, 7 + 12 * 8,

+    4 + 13 * 8, 5 + 13 * 8, 4 + 14 * 8, 5 + 14 * 8,

+    6 + 13 * 8, 7 + 13 * 8, 6 + 14 * 8, 7 + 14 * 8,

+    0 +  0 * 8, 0 +  5 * 8, 0 + 10 * 8

 };

-static av_always_inline uint32_t pack16to32(int a, int b){

+static av_always_inline uint32_t pack16to32(int a, int b)

+{

 #if HAVE_BIGENDIAN

-   return (b&0xFFFF) + (a<<16);

+    return (b & 0xFFFF) + (a << 16);

 #else

-   return (a&0xFFFF) + (b<<16);

+    return (a & 0xFFFF) + (b << 16);

 #endif

 }

-static av_always_inline uint16_t pack8to16(int a, int b){

+static av_always_inline uint16_t pack8to16(int a, int b)

+{

 #if HAVE_BIGENDIAN

-   return (b&0xFF) + (a<<8);

+    return (b & 0xFF) + (a << 8);

 #else

-   return (a&0xFF) + (b<<8);

+    return (a & 0xFF) + (b << 8);

 #endif

 }

 /**

  * Get the chroma qp.

  */

-static av_always_inline int get_chroma_qp(H264Context *h, int t, int qscale){

+static av_always_inline int get_chroma_qp(H264Context *h, int t, int qscale)

+{

     return h->pps.chroma_qp_table[t][qscale];

 }

 /**