h264: move direct_spatial_mv_pred into the per-slice context
[ffmpeg.git] / libavcodec / h264.h
index 3abf895..a259a97 100644 (file)
 #define AVCODEC_H264_H
 
 #include "libavutil/intreadwrite.h"
-#include "dsputil.h"
 #include "cabac.h"
-#include "mpegvideo.h"
+#include "error_resilience.h"
+#include "get_bits.h"
+#include "h264chroma.h"
 #include "h264dsp.h"
 #include "h264pred.h"
+#include "h264qpel.h"
+#include "mpegutils.h"
+#include "parser.h"
+#include "qpeldsp.h"
 #include "rectangle.h"
+#include "videodsp.h"
 
-#define interlaced_dct interlaced_dct_is_a_bad_name
-#define mb_intra mb_intra_is_not_initialized_see_mb_type
+#define H264_MAX_PICTURE_COUNT 32
+#define H264_MAX_THREADS       16
 
-#define CHROMA_DC_COEFF_TOKEN_VLC_BITS 8
-#define COEFF_TOKEN_VLC_BITS           8
-#define TOTAL_ZEROS_VLC_BITS           9
-#define CHROMA_DC_TOTAL_ZEROS_VLC_BITS 3
-#define RUN_VLC_BITS                   3
-#define RUN7_VLC_BITS                  6
+#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
 
 /* Compiling in interlaced support reduces the speed
  * of progressive decoding by about 2%. */
  * The maximum number of slices supported by the decoder.
  * must be a power of 2
  */
-#define MAX_SLICES 16
+#define MAX_SLICES 32
 
 #ifdef ALLOW_INTERLACE
-#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 MB_MBAFF(h)    h->mb_mbaff
+#define MB_FIELD(h)    h->mb_field_decoding_flag
+#define FRAME_MBAFF(h) h->mb_aff_frame
+#define FIELD_PICTURE(h) (h->picture_structure != PICT_FRAME)
+#define LEFT_MBS 2
+#define LTOP     0
+#define LBOT     1
+#define LEFT(i)  (i)
 #else
-#define MB_MBAFF 0
-#define MB_FIELD 0
-#define FRAME_MBAFF 0
-#define FIELD_PICTURE 0
+#define MB_MBAFF(h)      0
+#define MB_FIELD(h)      0
+#define FRAME_MBAFF(h)   0
+#define FIELD_PICTURE(h) 0
 #undef  IS_INTERLACED
 #define IS_INTERLACED(mb_type) 0
+#define LEFT_MBS 1
+#define LTOP     0
+#define LBOT     0
+#define LEFT(i)  0
 #endif
-#define FIELD_OR_MBAFF_PICTURE (FRAME_MBAFF || FIELD_PICTURE)
+#define FIELD_OR_MBAFF_PICTURE(h) (FRAME_MBAFF(h) || FIELD_PICTURE(h))
 
 #ifndef CABAC
-#define CABAC h->pps.cabac
+#define CABAC(h) h->pps.cabac
 #endif
 
-#define CHROMA444 (h->sps.chroma_format_idc == 3)
+#define CHROMA422(h) (h->sps.chroma_format_idc == 2)
+#define CHROMA444(h) (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)
 
-/**
- * Value of Picture.reference when Picture is not a reference picture, but
- * is held for delayed output.
- */
-#define DELAYED_PIC_REF 4
-
-#define QP_MAX_NUM (51 + 2*6)           // The maximum supported qp
+#define QP_MAX_NUM (51 + 2 * 6)           // The maximum supported qp
 
 /* NAL unit types */
 enum {
-    NAL_SLICE=1,
-    NAL_DPA,
-    NAL_DPB,
-    NAL_DPC,
-    NAL_IDR_SLICE,
-    NAL_SEI,
-    NAL_SPS,
-    NAL_PPS,
-    NAL_AUD,
-    NAL_END_SEQUENCE,
-    NAL_END_STREAM,
-    NAL_FILLER_DATA,
-    NAL_SPS_EXT,
-    NAL_AUXILIARY_SLICE=19
+    NAL_SLICE           = 1,
+    NAL_DPA             = 2,
+    NAL_DPB             = 3,
+    NAL_DPC             = 4,
+    NAL_IDR_SLICE       = 5,
+    NAL_SEI             = 6,
+    NAL_SPS             = 7,
+    NAL_PPS             = 8,
+    NAL_AUD             = 9,
+    NAL_END_SEQUENCE    = 10,
+    NAL_END_STREAM      = 11,
+    NAL_FILLER_DATA     = 12,
+    NAL_SPS_EXT         = 13,
+    NAL_AUXILIARY_SLICE = 19,
+    NAL_FF_IGNORE       = 0xff0f001,
 };
 
 /**
  * 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_TYPE_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_FRAME_PACKING          = 45,  ///< frame packing arrangement
+    SEI_TYPE_DISPLAY_ORIENTATION    = 47,  ///< display orientation
 } SEI_Type;
 
 /**
@@ -147,8 +152,8 @@ typedef enum {
 /**
  * Sequence parameter set
  */
-typedef struct SPS{
-
+typedef struct SPS {
+    unsigned int sps_id;
     int profile_idc;
     int level_idc;
     int chroma_format_idc;
@@ -165,9 +170,11 @@ 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
+
+    /* those 4 are already in luma samples */
     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
@@ -184,7 +191,7 @@ 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;
@@ -194,20 +201,21 @@ 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
+    int new;                              ///< flag to keep track if the decoder context needs re-init due to changed SPS
+} 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
@@ -220,20 +228,20 @@ 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][64];  ///< pre-scaled (with chroma_qp_index_offset) version of qp_table
+    uint8_t chroma_qp_table[2][64]; ///< 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,
@@ -245,176 +253,242 @@ 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
 } MMCO;
 
-/**
- * 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
+typedef struct H264Picture {
+    struct AVFrame f;
+    ThreadFrame tf;
+
+    AVBufferRef *qscale_table_buf;
+    int8_t *qscale_table;
+
+    AVBufferRef *motion_val_buf[2];
+    int16_t (*motion_val[2])[2];
+
+    AVBufferRef *mb_type_buf;
+    uint32_t *mb_type;
+
+    AVBufferRef *hwaccel_priv_buf;
+    void *hwaccel_picture_private; ///< hardware accelerator private data
 
+    AVBufferRef *ref_index_buf[2];
+    int8_t *ref_index[2];
+
+    int field_poc[2];       ///< top/bottom POC
+    int poc;                ///< frame POC
+    int frame_num;          ///< frame_num (raw frame_num from slice header)
+    int mmco_reset;         /**< MMCO_RESET set this 1. Reordering code must
+                                 not mix pictures before and after MMCO_RESET. */
+    int pic_id;             /**< pic_num (short -> no wrap version of pic_num,
+                                 pic_num & max_pic_num; long -> long_pic_num) */
+    int long_ref;           ///< 1->long term reference 0->short term reference
+    int ref_poc[2][2][32];  ///< POCs of the frames used as reference (FIXME need per slice)
+    int ref_count[2][2];    ///< number of entries in ref_poc         (FIXME need per slice)
+    int mbaff;              ///< 1 -> MBAFF frame 0-> not MBAFF
+    int field_picture;      ///< whether or not picture was encoded in separate fields
+
+    int needs_realloc;      ///< picture needs to be reallocated (eg due to a frame size change)
+    int reference;
+    int recovered;          ///< picture at IDR or recovery point + recovery count
+} H264Picture;
+
+typedef struct H264SliceContext {
+    struct H264Context *h264;
+
+    int slice_num;
+    int slice_type;
+    int slice_type_nos;         ///< S free slice type (SI/SP are remapped to I/P)
+    int slice_type_fixed;
+
+    int qscale;
+    int chroma_qp[2];   // QPc
     int qp_thresh;      ///< QP threshold to skip loopfilter
 
+    // Weighted pred stuff
+    int use_weight;
+    int use_weight_chroma;
+    int luma_log2_weight_denom;
+    int chroma_log2_weight_denom;
+    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
+    // 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];
+
     int prev_mb_skipped;
     int next_mb_skipped;
 
-    //prediction stuff
     int chroma_pred_mode;
     int intra16x16_pred_mode;
 
+    int8_t intra4x4_pred_mode_cache[5 * 8];
+    int8_t(*intra4x4_pred_mode);
+
     int topleft_mb_xy;
     int top_mb_xy;
     int topright_mb_xy;
-    int left_mb_xy[2];
+    int left_mb_xy[LEFT_MBS];
 
     int topleft_type;
     int top_type;
     int topright_type;
-    int left_type[2];
+    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);
-    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];
+
+    ptrdiff_t mb_linesize;  ///< may be equal to s->linesize or s->linesize * 2, for mbaff
+    ptrdiff_t mb_uvlinesize;
 
     /**
-     * non zero coeff count cache.
-     * is 64 if not available.
+     * number of neighbors (top and/or left) that used 8x8 dct
      */
-    DECLARE_ALIGNED(8, uint8_t, non_zero_count_cache)[15*8];
+    int neighbor_transform_size;
 
-    uint8_t (*non_zero_count)[48];
+    int direct_spatial_mv_pred;
 
     /**
-     * Motion vector cache.
+     * non zero coeff count cache.
+     * is 64 if not available.
      */
-    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
+    DECLARE_ALIGNED(8, uint8_t, non_zero_count_cache)[15 * 8];
 
     /**
-     * is 1 if the specific list MV&references are set to 0,0,-2.
+     * Motion vector cache.
      */
-    int mv_cache_clean[2];
+    DECLARE_ALIGNED(16, int16_t, mv_cache)[2][5 * 8][2];
+    DECLARE_ALIGNED(8,  int8_t, ref_cache)[2][5 * 8];
 
-    /**
-     * number of neighbors (top and/or left) that used 8x8 dct
-     */
-    int neighbor_transform_size;
+    DECLARE_ALIGNED(8, uint16_t, sub_mb_type)[4];
+} H264SliceContext;
+
+/**
+ * H264Context
+ */
+typedef struct H264Context {
+    AVCodecContext *avctx;
+    VideoDSPContext vdsp;
+    H264DSPContext h264dsp;
+    H264ChromaContext h264chroma;
+    H264QpelContext h264qpel;
+    GetBitContext gb;
+    ERContext er;
+
+    H264Picture *DPB;
+    H264Picture *cur_pic_ptr;
+    H264Picture cur_pic;
+
+    H264SliceContext *slice_ctx;
+    int            nb_slice_ctx;
+
+    int pixel_shift;    ///< 0 for 8-bit H264, 1 for high-bit-depth H264
+
+    /* coded dimensions -- 16 * mb w/h */
+    int width, height;
+    ptrdiff_t linesize, uvlinesize;
+    int chroma_x_shift, chroma_y_shift;
+
+    int droppable;
+    int coded_picture_number;
+    int low_delay;
+
+    int context_initialized;
+    int flags;
+    int workaround_bugs;
+
+    int8_t(*intra4x4_pred_mode);
+    H264PredContext hpc;
+    uint8_t (*top_borders[2])[(16 * 3) * 2];
+
+    uint8_t (*non_zero_count)[48];
+
+#define LIST_NOT_USED -1 // FIXME rename?
+#define PART_NOT_AVAILABLE -2
 
     /**
      * 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 mb_linesize;   ///< may be equal to s->linesize or s->linesize*2, for mbaff
-    int mb_uvlinesize;
-
-    int emu_edge_width;
-    int emu_edge_height;
+    int b_stride;       // FIXME use s->b4_stride
 
     SPS sps; ///< current sps
+    PPS pps; ///< current pps
 
-    /**
-     * current pps
-     */
-    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
-    int slice_type;
-    int slice_type_nos;        ///< S free slice type (SI/SP are remapped to I/P)
-    int slice_type_fixed;
+    uint16_t *slice_table;      ///< slice_table_base + 2*mb_stride + 1
 
-    //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
-
-    DECLARE_ALIGNED(8, uint16_t, sub_mb_type)[4];
-
-    //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
-    int luma_weight[48][2][2];
-    int chroma_weight[48][2][2][2];
-    int implicit_weight[48][48][2];
+    int mb_mbaff;               ///< mb_aff_frame && mb_field_decoding_flag
+    int picture_structure;
+    int first_field;
 
-    int direct_spatial_mv_pred;
     int col_parity;
     int col_fieldoff;
-    int dist_scale_factor[16];
+    int dist_scale_factor[32];
     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
      */
-    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;
-    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
+    uint8_t *list_counts;               ///< Array of list_count per MB specifying the slice type
+    H264Picture 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
+    const uint8_t *intra_pcm_ptr;
+    DECLARE_ALIGNED(16, int16_t, mb)[16 * 48 * 2]; ///< as a dct coeffecient is int32_t in high depth, we need to reserve twice the space.
+    DECLARE_ALIGNED(16, int16_t, mb_luma_dc)[3][16 * 2];
+    int16_t 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];
@@ -431,17 +505,24 @@ typedef struct H264Context{
 
     int x264_build;
 
+    int mb_x, mb_y;
+    int resync_mb_x;
+    int resync_mb_y;
+    int mb_skip_run;
+    int mb_height, mb_width;
+    int mb_stride;
+    int mb_num;
     int mb_xy;
 
     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;
@@ -451,47 +532,49 @@ 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 bit_depth_luma;         ///< luma bit depth from sps to detect changes
+    int chroma_format_idc;      ///< chroma format from sps to detect changes
 
     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;
 
     int redundant_pic_count;
 
-    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 *next_output_pic;
+    H264Picture default_ref_list[2][32]; ///< base reference list for all slices of a coded picture
+    H264Picture *short_ref[32];
+    H264Picture *long_ref[32];
+    H264Picture *delayed_pic[MAX_DELAYED_PIC_COUNT + 2]; // FIXME size?
+    int last_pocs[MAX_DELAYED_PIC_COUNT];
+    H264Picture *next_output_pic;
     int outputed_poc;
     int next_outputed_poc;
 
@@ -500,17 +583,18 @@ typedef struct H264Context{
      */
     MMCO mmco[MAX_MMCO_COUNT];
     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;
 
     /**
-     * @defgroup multithreading Members for slice based multithreading
+     * @name Members for slice based multithreading
      * @{
      */
-    struct H264Context *thread_context[MAX_THREADS];
+    struct H264Context *thread_context[H264_MAX_THREADS];
 
     /**
      * current slice number, used to initalize slice_num of each thread/context
@@ -525,12 +609,16 @@ typedef struct H264Context{
      */
     int max_contexts;
 
+    int slice_context_count;
+
     /**
      *  1 if the single thread fallback warning has already been
      *  displayed, 0 otherwise.
      */
     int single_decode_warning;
 
+    enum AVPictureType pict_type;
+
     int last_slice_type;
     /** @} */
 
@@ -548,6 +636,21 @@ typedef struct H264Context{
     int prev_interlaced_frame;
 
     /**
+     * frame_packing_arrangment SEI message
+     */
+    int sei_frame_packing_present;
+    int frame_packing_arrangement_type;
+    int content_interpretation_type;
+    int quincunx_subsampling;
+
+    /**
+     * display orientation SEI message
+     */
+    int sei_display_orientation_present;
+    int sei_anticlockwise_rotation;
+    int sei_hflip, sei_vflip;
+
+    /**
      * Bit set of clock types for fields/frames in picture timing SEI message.
      * For each found ct_type, appropriate bit is set (e.g., bit 1 for
      * interlaced).
@@ -573,16 +676,49 @@ typedef struct H264Context{
      */
     int sei_recovery_frame_cnt;
 
-    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
+    /**
+     * recovery_frame is the frame_num at which the next frame should
+     * be fully constructed.
+     *
+     * Set to -1 when not expecting a recovery point.
+     */
+    int recovery_frame;
+
+/**
+ * We have seen an IDR, so all the following frames in coded order are correctly
+ * decodable.
+ */
+#define FRAME_RECOVERED_IDR  (1 << 0)
+/**
+ * Sufficient number of frames have been decoded since a SEI recovery point,
+ * so all the following frames in presentation order are correct.
+ */
+#define FRAME_RECOVERED_SEI  (1 << 1)
+
+    int frame_recovered;    ///< Initial frame has been completely recovered
+
 
     // Timestamp stuff
-    int sei_buffering_period_present;  ///< Buffering period SEI flag
-    int initial_cpb_removal_delay[32]; ///< Initial timestamps for CPBs
-}H264Context;
+    int sei_buffering_period_present;   ///< Buffering period SEI flag
+    int initial_cpb_removal_delay[32];  ///< Initial timestamps for CPBs
 
+    int cur_chroma_format_idc;
+    uint8_t *bipred_scratchpad;
+    uint8_t *edge_emu_buffer;
+    int16_t *dc_val_base;
 
-extern const uint8_t ff_h264_chroma_qp[3][QP_MAX_NUM+1]; ///< One chroma qp table for each supported bit depth (8, 9, 10).
+    AVBufferPool *qscale_table_pool;
+    AVBufferPool *mb_type_pool;
+    AVBufferPool *motion_val_pool;
+    AVBufferPool *ref_index_pool;
+
+    /* Motion Estimation */
+    qpel_mc_func (*qpel_put)[16];
+    qpel_mc_func (*qpel_avg)[16];
+} H264Context;
+
+extern const uint8_t ff_h264_chroma_qp[3][QP_MAX_NUM + 1]; ///< One chroma qp table for each supported bit depth (8, 9, 10).
+extern const uint16_t ff_h264_mb_sizes[4];
 
 /**
  * Decode SEI
@@ -608,20 +744,23 @@ 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);
+void ff_h264_free_context(H264Context *h);
 
 /**
  * Reconstruct bitstream slice_type.
  */
-int ff_h264_get_slice_type(const H264Context *h);
+int ff_h264_get_slice_type(const H264SliceContext *sl);
 
 /**
  * Allocate tables.
@@ -632,10 +771,10 @@ int ff_h264_alloc_tables(H264Context *h);
 /**
  * Fill the default_ref_list.
  */
-int ff_h264_fill_default_ref_list(H264Context *h);
+int ff_h264_fill_default_ref_list(H264Context *h, H264SliceContext *sl);
 
 int ff_h264_decode_ref_pic_list_reordering(H264Context *h);
-void ff_h264_fill_mbaff_ref_list(H264Context *h);
+void ff_h264_fill_mbaff_ref_list(H264Context *h, H264SliceContext *sl);
 void ff_h264_remove_all_refs(H264Context *h);
 
 /**
@@ -643,49 +782,56 @@ void ff_h264_remove_all_refs(H264Context *h);
  */
 int ff_h264_execute_ref_pic_marking(H264Context *h, MMCO *mmco, int mmco_count);
 
-int ff_h264_decode_ref_pic_marking(H264Context *h, GetBitContext *gb);
-
-void ff_generate_sliding_window_mmcos(H264Context *h);
+int ff_h264_decode_ref_pic_marking(H264Context *h, GetBitContext *gb,
+                                   int first_slice);
 
+int ff_generate_sliding_window_mmcos(H264Context *h, int first_slice);
 
 /**
- * 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);
+int ff_h264_check_intra4x4_pred_mode(H264Context *h, H264SliceContext *sl);
 
 /**
- * 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_intra_pred_mode(H264Context *h, int mode);
+int ff_h264_check_intra_pred_mode(H264Context *h, H264SliceContext *sl,
+                                  int mode, int is_chroma);
 
-void ff_h264_write_back_intra_pred_mode(H264Context *h);
-void ff_h264_hl_decode_mb(H264Context *h);
-int ff_h264_frame_start(H264Context *h);
+void ff_h264_hl_decode_mb(H264Context *h, H264SliceContext *sl);
 int ff_h264_decode_extradata(H264Context *h);
-av_cold int ff_h264_decode_init(AVCodecContext *avctx);
-av_cold int ff_h264_decode_end(AVCodecContext *avctx);
-av_cold void ff_h264_decode_init_vlc(void);
+int ff_h264_decode_init(AVCodecContext *avctx);
+void ff_h264_decode_init_vlc(void);
 
 /**
  * Decode a macroblock
- * @return 0 if OK, AC_ERROR / DC_ERROR / 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);
+int ff_h264_decode_mb_cavlc(H264Context *h, H264SliceContext *sl);
 
 /**
  * Decode a CABAC coded macroblock
- * @return 0 if OK, AC_ERROR / DC_ERROR / 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);
+int ff_h264_decode_mb_cabac(H264Context *h, H264SliceContext *sl);
 
-void ff_h264_init_cabac_states(H264Context *h);
+void ff_h264_init_cabac_states(H264Context *h, H264SliceContext *sl);
 
-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 h264_init_dequant_tables(H264Context *h);
 
-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_direct_dist_scale_factor(H264Context *const h);
+void ff_h264_direct_ref_list_init(H264Context *const h, H264SliceContext *sl);
+void ff_h264_pred_direct_motion(H264Context *const h, H264SliceContext *sl,
+                                int *mb_type);
+
+void ff_h264_filter_mb_fast(H264Context *h, H264SliceContext *sl, 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, H264SliceContext *sl, 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.
@@ -694,16 +840,15 @@ 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
@@ -728,600 +873,199 @@ 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
 }
 
 /**
- * gets the chroma qp.
+ * Get the chroma qp.
  */
-static 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];
 }
 
-static inline void pred_pskip_motion(H264Context * const h, int * const mx, int * const my);
-
-static void fill_decode_neighbors(H264Context *h, int mb_type){
-    MpegEncContext * const s = &h->s;
-    const int mb_xy= h->mb_xy;
-    int topleft_xy, top_xy, topright_xy, left_xy[2];
-    static const uint8_t left_block_options[4][32]={
-        {0,1,2,3,7,10,8,11,3+0*4, 3+1*4, 3+2*4, 3+3*4, 1+4*4, 1+8*4, 1+5*4, 1+9*4},
-        {2,2,3,3,8,11,8,11,3+2*4, 3+2*4, 3+3*4, 3+3*4, 1+5*4, 1+9*4, 1+5*4, 1+9*4},
-        {0,0,1,1,7,10,7,10,3+0*4, 3+0*4, 3+1*4, 3+1*4, 1+4*4, 1+8*4, 1+4*4, 1+8*4},
-        {0,2,0,2,7,10,7,10,3+0*4, 3+2*4, 3+0*4, 3+2*4, 1+4*4, 1+8*4, 1+4*4, 1+8*4}
-    };
-
-    h->topleft_partition= -1;
-
-    top_xy     = mb_xy  - (s->mb_stride << MB_FIELD);
-
-    /* Wow, what a mess, why didn't they simplify the interlacing & intra
-     * stuff, I can't imagine that these complex rules are worth it. */
-
-    topleft_xy = top_xy - 1;
-    topright_xy= top_xy + 1;
-    left_xy[1] = left_xy[0] = mb_xy-1;
-    h->left_block = left_block_options[0];
-    if(FRAME_MBAFF){
-        const int left_mb_field_flag     = IS_INTERLACED(s->current_picture.mb_type[mb_xy-1]);
-        const int curr_mb_field_flag     = IS_INTERLACED(mb_type);
-        if(s->mb_y&1){
-            if (left_mb_field_flag != curr_mb_field_flag) {
-                left_xy[1] = left_xy[0] = mb_xy - s->mb_stride - 1;
-                if (curr_mb_field_flag) {
-                    left_xy[1] += s->mb_stride;
-                    h->left_block = left_block_options[3];
-                } else {
-                    topleft_xy += s->mb_stride;
-                    // take top left mv from the middle of the mb, as opposed to all other modes which use the bottom right partition
-                    h->topleft_partition = 0;
-                    h->left_block = left_block_options[1];
-                }
-            }
-        }else{
-            if(curr_mb_field_flag){
-                topleft_xy  += s->mb_stride & (((s->current_picture.mb_type[top_xy - 1]>>7)&1)-1);
-                topright_xy += s->mb_stride & (((s->current_picture.mb_type[top_xy + 1]>>7)&1)-1);
-                top_xy      += s->mb_stride & (((s->current_picture.mb_type[top_xy    ]>>7)&1)-1);
-            }
-            if (left_mb_field_flag != curr_mb_field_flag) {
-                if (curr_mb_field_flag) {
-                    left_xy[1] += s->mb_stride;
-                    h->left_block = left_block_options[3];
-                } else {
-                    h->left_block = left_block_options[2];
-                }
-            }
-        }
-    }
-
-    h->topleft_mb_xy = topleft_xy;
-    h->top_mb_xy     = top_xy;
-    h->topright_mb_xy= topright_xy;
-    h->left_mb_xy[0] = left_xy[0];
-    h->left_mb_xy[1] = left_xy[1];
-    //FIXME do we need all in the context?
-
-    h->topleft_type = s->current_picture.mb_type[topleft_xy] ;
-    h->top_type     = s->current_picture.mb_type[top_xy]     ;
-    h->topright_type= s->current_picture.mb_type[topright_xy];
-    h->left_type[0] = s->current_picture.mb_type[left_xy[0]] ;
-    h->left_type[1] = s->current_picture.mb_type[left_xy[1]] ;
-
-    if(FMO){
-    if(h->slice_table[topleft_xy ] != h->slice_num) h->topleft_type = 0;
-    if(h->slice_table[top_xy     ] != h->slice_num) h->top_type     = 0;
-    if(h->slice_table[left_xy[0] ] != h->slice_num) h->left_type[0] = h->left_type[1] = 0;
-    }else{
-        if(h->slice_table[topleft_xy ] != h->slice_num){
-            h->topleft_type = 0;
-            if(h->slice_table[top_xy     ] != h->slice_num) h->top_type     = 0;
-            if(h->slice_table[left_xy[0] ] != h->slice_num) h->left_type[0] = h->left_type[1] = 0;
-        }
-    }
-    if(h->slice_table[topright_xy] != h->slice_num) h->topright_type= 0;
+/**
+ * Get the predicted intra4x4 prediction mode.
+ */
+static av_always_inline int pred_intra_mode(H264Context *h,
+                                            H264SliceContext *sl, int n)
+{
+    const int index8 = scan8[n];
+    const int left   = sl->intra4x4_pred_mode_cache[index8 - 1];
+    const int top    = sl->intra4x4_pred_mode_cache[index8 - 8];
+    const int min    = FFMIN(left, top);
+
+    tprintf(h->avctx, "mode:%d %d min:%d\n", left, top, min);
+
+    if (min < 0)
+        return DC_PRED;
+    else
+        return min;
 }
 
-static void fill_decode_caches(H264Context *h, int mb_type){
-    MpegEncContext * const s = &h->s;
-    int topleft_xy, top_xy, topright_xy, left_xy[2];
-    int topleft_type, top_type, topright_type, left_type[2];
-    const uint8_t * left_block= h->left_block;
-    int i;
-
-    topleft_xy   = h->topleft_mb_xy ;
-    top_xy       = h->top_mb_xy     ;
-    topright_xy  = h->topright_mb_xy;
-    left_xy[0]   = h->left_mb_xy[0] ;
-    left_xy[1]   = h->left_mb_xy[1] ;
-    topleft_type = h->topleft_type  ;
-    top_type     = h->top_type      ;
-    topright_type= h->topright_type ;
-    left_type[0] = h->left_type[0]  ;
-    left_type[1] = h->left_type[1]  ;
-
-    if(!IS_SKIP(mb_type)){
-        if(IS_INTRA(mb_type)){
-            int type_mask= h->pps.constrained_intra_pred ? IS_INTRA(-1) : -1;
-            h->topleft_samples_available=
-            h->top_samples_available=
-            h->left_samples_available= 0xFFFF;
-            h->topright_samples_available= 0xEEEA;
-
-            if(!(top_type & type_mask)){
-                h->topleft_samples_available= 0xB3FF;
-                h->top_samples_available= 0x33FF;
-                h->topright_samples_available= 0x26EA;
-            }
-            if(IS_INTERLACED(mb_type) != IS_INTERLACED(left_type[0])){
-                if(IS_INTERLACED(mb_type)){
-                    if(!(left_type[0] & type_mask)){
-                        h->topleft_samples_available&= 0xDFFF;
-                        h->left_samples_available&= 0x5FFF;
-                    }
-                    if(!(left_type[1] & type_mask)){
-                        h->topleft_samples_available&= 0xFF5F;
-                        h->left_samples_available&= 0xFF5F;
-                    }
-                }else{
-                    int left_typei = s->current_picture.mb_type[left_xy[0] + s->mb_stride];
-
-                    assert(left_xy[0] == left_xy[1]);
-                    if(!((left_typei & type_mask) && (left_type[0] & type_mask))){
-                        h->topleft_samples_available&= 0xDF5F;
-                        h->left_samples_available&= 0x5F5F;
-                    }
-                }
-            }else{
-                if(!(left_type[0] & type_mask)){
-                    h->topleft_samples_available&= 0xDF5F;
-                    h->left_samples_available&= 0x5F5F;
-                }
-            }
-
-            if(!(topleft_type & type_mask))
-                h->topleft_samples_available&= 0x7FFF;
-
-            if(!(topright_type & type_mask))
-                h->topright_samples_available&= 0xFBFF;
-
-            if(IS_INTRA4x4(mb_type)){
-                if(IS_INTRA4x4(top_type)){
-                    AV_COPY32(h->intra4x4_pred_mode_cache+4+8*0, h->intra4x4_pred_mode + h->mb2br_xy[top_xy]);
-                }else{
-                    h->intra4x4_pred_mode_cache[4+8*0]=
-                    h->intra4x4_pred_mode_cache[5+8*0]=
-                    h->intra4x4_pred_mode_cache[6+8*0]=
-                    h->intra4x4_pred_mode_cache[7+8*0]= 2 - 3*!(top_type & type_mask);
-                }
-                for(i=0; i<2; i++){
-                    if(IS_INTRA4x4(left_type[i])){
-                        int8_t *mode= h->intra4x4_pred_mode + h->mb2br_xy[left_xy[i]];
-                        h->intra4x4_pred_mode_cache[3+8*1 + 2*8*i]= mode[6-left_block[0+2*i]];
-                        h->intra4x4_pred_mode_cache[3+8*2 + 2*8*i]= mode[6-left_block[1+2*i]];
-                    }else{
-                        h->intra4x4_pred_mode_cache[3+8*1 + 2*8*i]=
-                        h->intra4x4_pred_mode_cache[3+8*2 + 2*8*i]= 2 - 3*!(left_type[i] & type_mask);
-                    }
-                }
-            }
-        }
-
+static av_always_inline void write_back_intra_pred_mode(H264Context *h,
+                                                        H264SliceContext *sl)
+{
+    int8_t *i4x4       = sl->intra4x4_pred_mode + h->mb2br_xy[h->mb_xy];
+    int8_t *i4x4_cache = sl->intra4x4_pred_mode_cache;
 
-/*
-0 . T T. T T T T
-1 L . .L . . . .
-2 L . .L . . . .
-3 . T TL . . . .
-4 L . .L . . . .
-5 L . .. . . . .
-*/
-//FIXME constraint_intra_pred & partitioning & nnz (let us hope this is just a typo in the spec)
-    if(top_type){
-        AV_COPY32(&h->non_zero_count_cache[4+8* 0], &h->non_zero_count[top_xy][4*3]);
-        if(CHROMA444){
-            AV_COPY32(&h->non_zero_count_cache[4+8* 5], &h->non_zero_count[top_xy][4* 7]);
-            AV_COPY32(&h->non_zero_count_cache[4+8*10], &h->non_zero_count[top_xy][4*11]);
-        }else{
-            AV_COPY32(&h->non_zero_count_cache[4+8* 5], &h->non_zero_count[top_xy][4* 5]);
-            AV_COPY32(&h->non_zero_count_cache[4+8*10], &h->non_zero_count[top_xy][4* 9]);
-        }
-    }else{
-        uint32_t top_empty = CABAC && !IS_INTRA(mb_type) ? 0 : 0x40404040;
-        AV_WN32A(&h->non_zero_count_cache[4+8* 0], top_empty);
-        AV_WN32A(&h->non_zero_count_cache[4+8* 5], top_empty);
-        AV_WN32A(&h->non_zero_count_cache[4+8*10], top_empty);
-    }
+    AV_COPY32(i4x4, i4x4_cache + 4 + 8 * 4);
+    i4x4[4] = i4x4_cache[7 + 8 * 3];
+    i4x4[5] = i4x4_cache[7 + 8 * 2];
+    i4x4[6] = i4x4_cache[7 + 8 * 1];
+}
 
-    for (i=0; i<2; i++) {
-        if(left_type[i]){
-            h->non_zero_count_cache[3+8* 1 + 2*8*i]= h->non_zero_count[left_xy[i]][left_block[8+0+2*i]];
-            h->non_zero_count_cache[3+8* 2 + 2*8*i]= h->non_zero_count[left_xy[i]][left_block[8+1+2*i]];
-            if(CHROMA444){
-                h->non_zero_count_cache[3+8* 6 + 2*8*i]= h->non_zero_count[left_xy[i]][left_block[8+0+2*i]+4*4];
-                h->non_zero_count_cache[3+8* 7 + 2*8*i]= h->non_zero_count[left_xy[i]][left_block[8+1+2*i]+4*4];
-                h->non_zero_count_cache[3+8*11 + 2*8*i]= h->non_zero_count[left_xy[i]][left_block[8+0+2*i]+8*4];
-                h->non_zero_count_cache[3+8*12 + 2*8*i]= h->non_zero_count[left_xy[i]][left_block[8+1+2*i]+8*4];
-            }else{
-                h->non_zero_count_cache[3+8* 6 +   8*i]= h->non_zero_count[left_xy[i]][left_block[8+4+2*i]];
-                h->non_zero_count_cache[3+8*11 +   8*i]= h->non_zero_count[left_xy[i]][left_block[8+5+2*i]];
-            }
-        }else{
-            h->non_zero_count_cache[3+8* 1 + 2*8*i]=
-            h->non_zero_count_cache[3+8* 2 + 2*8*i]=
-            h->non_zero_count_cache[3+8* 6 + 2*8*i]=
-            h->non_zero_count_cache[3+8* 7 + 2*8*i]=
-            h->non_zero_count_cache[3+8*11 + 2*8*i]=
-            h->non_zero_count_cache[3+8*12 + 2*8*i]= CABAC && !IS_INTRA(mb_type) ? 0 : 64;
-        }
+static av_always_inline void write_back_non_zero_count(H264Context *h,
+                                                       H264SliceContext *sl)
+{
+    const int mb_xy    = h->mb_xy;
+    uint8_t *nnz       = h->non_zero_count[mb_xy];
+    uint8_t *nnz_cache = sl->non_zero_count_cache;
+
+    AV_COPY32(&nnz[ 0], &nnz_cache[4 + 8 * 1]);
+    AV_COPY32(&nnz[ 4], &nnz_cache[4 + 8 * 2]);
+    AV_COPY32(&nnz[ 8], &nnz_cache[4 + 8 * 3]);
+    AV_COPY32(&nnz[12], &nnz_cache[4 + 8 * 4]);
+    AV_COPY32(&nnz[16], &nnz_cache[4 + 8 * 6]);
+    AV_COPY32(&nnz[20], &nnz_cache[4 + 8 * 7]);
+    AV_COPY32(&nnz[32], &nnz_cache[4 + 8 * 11]);
+    AV_COPY32(&nnz[36], &nnz_cache[4 + 8 * 12]);
+
+    if (!h->chroma_y_shift) {
+        AV_COPY32(&nnz[24], &nnz_cache[4 + 8 * 8]);
+        AV_COPY32(&nnz[28], &nnz_cache[4 + 8 * 9]);
+        AV_COPY32(&nnz[40], &nnz_cache[4 + 8 * 13]);
+        AV_COPY32(&nnz[44], &nnz_cache[4 + 8 * 14]);
     }
+}
 
-    if( CABAC ) {
-        // top_cbp
-        if(top_type) {
-            h->top_cbp = h->cbp_table[top_xy];
-        } else {
-            h->top_cbp = IS_INTRA(mb_type) ? 0x7CF : 0x00F;
-        }
-        // left_cbp
-        if (left_type[0]) {
-            h->left_cbp =   (h->cbp_table[left_xy[0]] & 0x7F0)
-                        |  ((h->cbp_table[left_xy[0]]>>(left_block[0]&(~1)))&2)
-                        | (((h->cbp_table[left_xy[1]]>>(left_block[2]&(~1)))&2) << 2);
+static av_always_inline void write_back_motion_list(H264Context *h,
+                                                    H264SliceContext *sl,
+                                                    int b_stride,
+                                                    int b_xy, int b8_xy,
+                                                    int mb_type, int list)
+{
+    int16_t(*mv_dst)[2] = &h->cur_pic.motion_val[list][b_xy];
+    int16_t(*mv_src)[2] = &sl->mv_cache[list][scan8[0]];
+    AV_COPY128(mv_dst + 0 * b_stride, mv_src + 8 * 0);
+    AV_COPY128(mv_dst + 1 * b_stride, mv_src + 8 * 1);
+    AV_COPY128(mv_dst + 2 * b_stride, mv_src + 8 * 2);
+    AV_COPY128(mv_dst + 3 * b_stride, mv_src + 8 * 3);
+    if (CABAC(h)) {
+        uint8_t (*mvd_dst)[2] = &h->mvd_table[list][FMO ? 8 * h->mb_xy
+                                                        : h->mb2br_xy[h->mb_xy]];
+        uint8_t(*mvd_src)[2]  = &h->mvd_cache[list][scan8[0]];
+        if (IS_SKIP(mb_type)) {
+            AV_ZERO128(mvd_dst);
         } else {
-            h->left_cbp = IS_INTRA(mb_type) ? 0x7CF : 0x00F;
+            AV_COPY64(mvd_dst, mvd_src + 8 * 3);
+            AV_COPY16(mvd_dst + 3 + 3, mvd_src + 3 + 8 * 0);
+            AV_COPY16(mvd_dst + 3 + 2, mvd_src + 3 + 8 * 1);
+            AV_COPY16(mvd_dst + 3 + 1, mvd_src + 3 + 8 * 2);
         }
     }
-    }
-
-    if(IS_INTER(mb_type) || (IS_DIRECT(mb_type) && h->direct_spatial_mv_pred)){
-        int list;
-        for(list=0; list<h->list_count; list++){
-            if(!USES_LIST(mb_type, list)){
-                /*if(!h->mv_cache_clean[list]){
-                    memset(h->mv_cache [list],  0, 8*5*2*sizeof(int16_t)); //FIXME clean only input? clean at all?
-                    memset(h->ref_cache[list], PART_NOT_AVAILABLE, 8*5*sizeof(int8_t));
-                    h->mv_cache_clean[list]= 1;
-                }*/
-                continue;
-            }
-            assert(!(IS_DIRECT(mb_type) && !h->direct_spatial_mv_pred));
-
-            h->mv_cache_clean[list]= 0;
-
-            if(USES_LIST(top_type, list)){
-                const int b_xy= h->mb2b_xy[top_xy] + 3*h->b_stride;
-                AV_COPY128(h->mv_cache[list][scan8[0] + 0 - 1*8], s->current_picture.motion_val[list][b_xy + 0]);
-                    h->ref_cache[list][scan8[0] + 0 - 1*8]=
-                    h->ref_cache[list][scan8[0] + 1 - 1*8]= s->current_picture.ref_index[list][4*top_xy + 2];
-                    h->ref_cache[list][scan8[0] + 2 - 1*8]=
-                    h->ref_cache[list][scan8[0] + 3 - 1*8]= s->current_picture.ref_index[list][4*top_xy + 3];
-            }else{
-                AV_ZERO128(h->mv_cache[list][scan8[0] + 0 - 1*8]);
-                AV_WN32A(&h->ref_cache[list][scan8[0] + 0 - 1*8], ((top_type ? LIST_NOT_USED : PART_NOT_AVAILABLE)&0xFF)*0x01010101);
-            }
-
-            if(mb_type & (MB_TYPE_16x8|MB_TYPE_8x8)){
-            for(i=0; i<2; i++){
-                int cache_idx = scan8[0] - 1 + i*2*8;
-                if(USES_LIST(left_type[i], list)){
-                    const int b_xy= h->mb2b_xy[left_xy[i]] + 3;
-                    const int b8_xy= 4*left_xy[i] + 1;
-                    AV_COPY32(h->mv_cache[list][cache_idx  ], s->current_picture.motion_val[list][b_xy + h->b_stride*left_block[0+i*2]]);
-                    AV_COPY32(h->mv_cache[list][cache_idx+8], s->current_picture.motion_val[list][b_xy + h->b_stride*left_block[1+i*2]]);
-                        h->ref_cache[list][cache_idx  ]= s->current_picture.ref_index[list][b8_xy + (left_block[0+i*2]&~1)];
-                        h->ref_cache[list][cache_idx+8]= s->current_picture.ref_index[list][b8_xy + (left_block[1+i*2]&~1)];
-                }else{
-                    AV_ZERO32(h->mv_cache [list][cache_idx  ]);
-                    AV_ZERO32(h->mv_cache [list][cache_idx+8]);
-                    h->ref_cache[list][cache_idx  ]=
-                    h->ref_cache[list][cache_idx+8]= (left_type[i]) ? LIST_NOT_USED : PART_NOT_AVAILABLE;
-                }
-            }
-            }else{
-                if(USES_LIST(left_type[0], list)){
-                    const int b_xy= h->mb2b_xy[left_xy[0]] + 3;
-                    const int b8_xy= 4*left_xy[0] + 1;
-                    AV_COPY32(h->mv_cache[list][scan8[0] - 1], s->current_picture.motion_val[list][b_xy + h->b_stride*left_block[0]]);
-                    h->ref_cache[list][scan8[0] - 1]= s->current_picture.ref_index[list][b8_xy + (left_block[0]&~1)];
-                }else{
-                    AV_ZERO32(h->mv_cache [list][scan8[0] - 1]);
-                    h->ref_cache[list][scan8[0] - 1]= left_type[0] ? LIST_NOT_USED : PART_NOT_AVAILABLE;
-                }
-            }
-
-            if(USES_LIST(topright_type, list)){
-                const int b_xy= h->mb2b_xy[topright_xy] + 3*h->b_stride;
-                AV_COPY32(h->mv_cache[list][scan8[0] + 4 - 1*8], s->current_picture.motion_val[list][b_xy]);
-                h->ref_cache[list][scan8[0] + 4 - 1*8]= s->current_picture.ref_index[list][4*topright_xy + 2];
-            }else{
-                AV_ZERO32(h->mv_cache [list][scan8[0] + 4 - 1*8]);
-                h->ref_cache[list][scan8[0] + 4 - 1*8]= topright_type ? LIST_NOT_USED : PART_NOT_AVAILABLE;
-            }
-            if(h->ref_cache[list][scan8[0] + 4 - 1*8] < 0){
-                if(USES_LIST(topleft_type, list)){
-                    const int b_xy = h->mb2b_xy [topleft_xy] + 3 + h->b_stride + (h->topleft_partition & 2*h->b_stride);
-                    const int b8_xy= 4*topleft_xy + 1 + (h->topleft_partition & 2);
-                    AV_COPY32(h->mv_cache[list][scan8[0] - 1 - 1*8], s->current_picture.motion_val[list][b_xy]);
-                    h->ref_cache[list][scan8[0] - 1 - 1*8]= s->current_picture.ref_index[list][b8_xy];
-                }else{
-                    AV_ZERO32(h->mv_cache[list][scan8[0] - 1 - 1*8]);
-                    h->ref_cache[list][scan8[0] - 1 - 1*8]= topleft_type ? LIST_NOT_USED : PART_NOT_AVAILABLE;
-                }
-            }
-
-            if((mb_type&(MB_TYPE_SKIP|MB_TYPE_DIRECT2)) && !FRAME_MBAFF)
-                continue;
-
-            if(!(mb_type&(MB_TYPE_SKIP|MB_TYPE_DIRECT2))) {
-            h->ref_cache[list][scan8[4 ]] =
-            h->ref_cache[list][scan8[12]] = PART_NOT_AVAILABLE;
-            AV_ZERO32(h->mv_cache [list][scan8[4 ]]);
-            AV_ZERO32(h->mv_cache [list][scan8[12]]);
-
-            if( CABAC ) {
-                /* XXX beurk, Load mvd */
-                if(USES_LIST(top_type, list)){
-                    const int b_xy= h->mb2br_xy[top_xy];
-                    AV_COPY64(h->mvd_cache[list][scan8[0] + 0 - 1*8], h->mvd_table[list][b_xy + 0]);
-                }else{
-                    AV_ZERO64(h->mvd_cache[list][scan8[0] + 0 - 1*8]);
-                }
-                if(USES_LIST(left_type[0], list)){
-                    const int b_xy= h->mb2br_xy[left_xy[0]] + 6;
-                    AV_COPY16(h->mvd_cache[list][scan8[0] - 1 + 0*8], h->mvd_table[list][b_xy - left_block[0]]);
-                    AV_COPY16(h->mvd_cache[list][scan8[0] - 1 + 1*8], h->mvd_table[list][b_xy - left_block[1]]);
-                }else{
-                    AV_ZERO16(h->mvd_cache [list][scan8[0] - 1 + 0*8]);
-                    AV_ZERO16(h->mvd_cache [list][scan8[0] - 1 + 1*8]);
-                }
-                if(USES_LIST(left_type[1], list)){
-                    const int b_xy= h->mb2br_xy[left_xy[1]] + 6;
-                    AV_COPY16(h->mvd_cache[list][scan8[0] - 1 + 2*8], h->mvd_table[list][b_xy - left_block[2]]);
-                    AV_COPY16(h->mvd_cache[list][scan8[0] - 1 + 3*8], h->mvd_table[list][b_xy - left_block[3]]);
-                }else{
-                    AV_ZERO16(h->mvd_cache [list][scan8[0] - 1 + 2*8]);
-                    AV_ZERO16(h->mvd_cache [list][scan8[0] - 1 + 3*8]);
-                }
-                AV_ZERO16(h->mvd_cache [list][scan8[4 ]]);
-                AV_ZERO16(h->mvd_cache [list][scan8[12]]);
-                if(h->slice_type_nos == AV_PICTURE_TYPE_B){
-                    fill_rectangle(&h->direct_cache[scan8[0]], 4, 4, 8, MB_TYPE_16x16>>1, 1);
-
-                    if(IS_DIRECT(top_type)){
-                        AV_WN32A(&h->direct_cache[scan8[0] - 1*8], 0x01010101u*(MB_TYPE_DIRECT2>>1));
-                    }else if(IS_8X8(top_type)){
-                        int b8_xy = 4*top_xy;
-                        h->direct_cache[scan8[0] + 0 - 1*8]= h->direct_table[b8_xy + 2];
-                        h->direct_cache[scan8[0] + 2 - 1*8]= h->direct_table[b8_xy + 3];
-                    }else{
-                        AV_WN32A(&h->direct_cache[scan8[0] - 1*8], 0x01010101*(MB_TYPE_16x16>>1));
-                    }
-
-                    if(IS_DIRECT(left_type[0]))
-                        h->direct_cache[scan8[0] - 1 + 0*8]= MB_TYPE_DIRECT2>>1;
-                    else if(IS_8X8(left_type[0]))
-                        h->direct_cache[scan8[0] - 1 + 0*8]= h->direct_table[4*left_xy[0] + 1 + (left_block[0]&~1)];
-                    else
-                        h->direct_cache[scan8[0] - 1 + 0*8]= MB_TYPE_16x16>>1;
-
-                    if(IS_DIRECT(left_type[1]))
-                        h->direct_cache[scan8[0] - 1 + 2*8]= MB_TYPE_DIRECT2>>1;
-                    else if(IS_8X8(left_type[1]))
-                        h->direct_cache[scan8[0] - 1 + 2*8]= h->direct_table[4*left_xy[1] + 1 + (left_block[2]&~1)];
-                    else
-                        h->direct_cache[scan8[0] - 1 + 2*8]= MB_TYPE_16x16>>1;
-                }
-            }
-            }
-            if(FRAME_MBAFF){
-#define MAP_MVS\
-                    MAP_F2F(scan8[0] - 1 - 1*8, topleft_type)\
-                    MAP_F2F(scan8[0] + 0 - 1*8, top_type)\
-                    MAP_F2F(scan8[0] + 1 - 1*8, top_type)\
-                    MAP_F2F(scan8[0] + 2 - 1*8, top_type)\
-                    MAP_F2F(scan8[0] + 3 - 1*8, top_type)\
-                    MAP_F2F(scan8[0] + 4 - 1*8, topright_type)\
-                    MAP_F2F(scan8[0] - 1 + 0*8, left_type[0])\
-                    MAP_F2F(scan8[0] - 1 + 1*8, left_type[0])\
-                    MAP_F2F(scan8[0] - 1 + 2*8, left_type[1])\
-                    MAP_F2F(scan8[0] - 1 + 3*8, left_type[1])
-                if(MB_FIELD){
-#define MAP_F2F(idx, mb_type)\
-                    if(!IS_INTERLACED(mb_type) && h->ref_cache[list][idx] >= 0){\
-                        h->ref_cache[list][idx] <<= 1;\
-                        h->mv_cache[list][idx][1] /= 2;\
-                        h->mvd_cache[list][idx][1] >>=1;\
-                    }
-                    MAP_MVS
-#undef MAP_F2F
-                }else{
-#define MAP_F2F(idx, mb_type)\
-                    if(IS_INTERLACED(mb_type) && h->ref_cache[list][idx] >= 0){\
-                        h->ref_cache[list][idx] >>= 1;\
-                        h->mv_cache[list][idx][1] <<= 1;\
-                        h->mvd_cache[list][idx][1] <<= 1;\
-                    }
-                    MAP_MVS
-#undef MAP_F2F
-                }
-            }
-        }
-    }
-
-        h->neighbor_transform_size= !!IS_8x8DCT(top_type) + !!IS_8x8DCT(left_type[0]);
-}
-
-/**
- * gets the predicted intra4x4 prediction mode.
- */
-static inline int pred_intra_mode(H264Context *h, int n){
-    const int index8= scan8[n];
-    const int left= h->intra4x4_pred_mode_cache[index8 - 1];
-    const int top = h->intra4x4_pred_mode_cache[index8 - 8];
-    const int min= FFMIN(left, top);
-
-    tprintf(h->s.avctx, "mode:%d %d min:%d\n", left ,top, min);
-
-    if(min<0) return DC_PRED;
-    else      return min;
-}
 
-static inline void write_back_non_zero_count(H264Context *h){
-    const int mb_xy= h->mb_xy;
-
-    AV_COPY32(&h->non_zero_count[mb_xy][ 0], &h->non_zero_count_cache[4+8* 1]);
-    AV_COPY32(&h->non_zero_count[mb_xy][ 4], &h->non_zero_count_cache[4+8* 2]);
-    AV_COPY32(&h->non_zero_count[mb_xy][ 8], &h->non_zero_count_cache[4+8* 3]);
-    AV_COPY32(&h->non_zero_count[mb_xy][12], &h->non_zero_count_cache[4+8* 4]);
-    AV_COPY32(&h->non_zero_count[mb_xy][16], &h->non_zero_count_cache[4+8* 6]);
-    AV_COPY32(&h->non_zero_count[mb_xy][20], &h->non_zero_count_cache[4+8* 7]);
-    AV_COPY32(&h->non_zero_count[mb_xy][32], &h->non_zero_count_cache[4+8*11]);
-    AV_COPY32(&h->non_zero_count[mb_xy][36], &h->non_zero_count_cache[4+8*12]);
-
-    if(CHROMA444){
-        AV_COPY32(&h->non_zero_count[mb_xy][24], &h->non_zero_count_cache[4+8* 8]);
-        AV_COPY32(&h->non_zero_count[mb_xy][28], &h->non_zero_count_cache[4+8* 9]);
-        AV_COPY32(&h->non_zero_count[mb_xy][40], &h->non_zero_count_cache[4+8*13]);
-        AV_COPY32(&h->non_zero_count[mb_xy][44], &h->non_zero_count_cache[4+8*14]);
+    {
+        int8_t *ref_index = &h->cur_pic.ref_index[list][b8_xy];
+        int8_t *ref_cache = sl->ref_cache[list];
+        ref_index[0 + 0 * 2] = ref_cache[scan8[0]];
+        ref_index[1 + 0 * 2] = ref_cache[scan8[4]];
+        ref_index[0 + 1 * 2] = ref_cache[scan8[8]];
+        ref_index[1 + 1 * 2] = ref_cache[scan8[12]];
     }
 }
 
-static inline void write_back_motion(H264Context *h, int mb_type){
-    MpegEncContext * const s = &h->s;
-    const int b_xy = 4*s->mb_x + 4*s->mb_y*h->b_stride; //try mb2b(8)_xy
-    const int b8_xy= 4*h->mb_xy;
-    int list;
-
-    if(!USES_LIST(mb_type, 0))
-        fill_rectangle(&s->current_picture.ref_index[0][b8_xy], 2, 2, 2, (uint8_t)LIST_NOT_USED, 1);
-
-    for(list=0; list<h->list_count; list++){
-        int y, b_stride;
-        int16_t (*mv_dst)[2];
-        int16_t (*mv_src)[2];
-
-        if(!USES_LIST(mb_type, list))
-            continue;
-
-        b_stride = h->b_stride;
-        mv_dst   = &s->current_picture.motion_val[list][b_xy];
-        mv_src   = &h->mv_cache[list][scan8[0]];
-        for(y=0; y<4; y++){
-            AV_COPY128(mv_dst + y*b_stride, mv_src + 8*y);
-        }
-        if( CABAC ) {
-            uint8_t (*mvd_dst)[2] = &h->mvd_table[list][FMO ? 8*h->mb_xy : h->mb2br_xy[h->mb_xy]];
-            uint8_t (*mvd_src)[2] = &h->mvd_cache[list][scan8[0]];
-            if(IS_SKIP(mb_type))
-                AV_ZERO128(mvd_dst);
-            else{
-            AV_COPY64(mvd_dst, mvd_src + 8*3);
-                AV_COPY16(mvd_dst + 3 + 3, mvd_src + 3 + 8*0);
-                AV_COPY16(mvd_dst + 3 + 2, mvd_src + 3 + 8*1);
-                AV_COPY16(mvd_dst + 3 + 1, mvd_src + 3 + 8*2);
-            }
-        }
-
-        {
-            int8_t *ref_index = &s->current_picture.ref_index[list][b8_xy];
-            ref_index[0+0*2]= h->ref_cache[list][scan8[0]];
-            ref_index[1+0*2]= h->ref_cache[list][scan8[4]];
-            ref_index[0+1*2]= h->ref_cache[list][scan8[8]];
-            ref_index[1+1*2]= h->ref_cache[list][scan8[12]];
-        }
+static av_always_inline void write_back_motion(H264Context *h,
+                                               H264SliceContext *sl,
+                                               int mb_type)
+{
+    const int b_stride      = h->b_stride;
+    const int b_xy  = 4 * h->mb_x + 4 * h->mb_y * h->b_stride; // try mb2b(8)_xy
+    const int b8_xy = 4 * h->mb_xy;
+
+    if (USES_LIST(mb_type, 0)) {
+        write_back_motion_list(h, sl, b_stride, b_xy, b8_xy, mb_type, 0);
+    } else {
+        fill_rectangle(&h->cur_pic.ref_index[0][b8_xy],
+                       2, 2, 2, (uint8_t)LIST_NOT_USED, 1);
     }
-
-    if(h->slice_type_nos == AV_PICTURE_TYPE_B && CABAC){
-        if(IS_8X8(mb_type)){
-            uint8_t *direct_table = &h->direct_table[4*h->mb_xy];
-            direct_table[1] = h->sub_mb_type[1]>>1;
-            direct_table[2] = h->sub_mb_type[2]>>1;
-            direct_table[3] = h->sub_mb_type[3]>>1;
+    if (USES_LIST(mb_type, 1))
+        write_back_motion_list(h, sl, b_stride, b_xy, b8_xy, mb_type, 1);
+
+    if (sl->slice_type_nos == AV_PICTURE_TYPE_B && CABAC(h)) {
+        if (IS_8X8(mb_type)) {
+            uint8_t *direct_table = &h->direct_table[4 * h->mb_xy];
+            direct_table[1] = sl->sub_mb_type[1] >> 1;
+            direct_table[2] = sl->sub_mb_type[2] >> 1;
+            direct_table[3] = sl->sub_mb_type[3] >> 1;
         }
     }
 }
 
-static inline int get_dct8x8_allowed(H264Context *h){
-    if(h->sps.direct_8x8_inference_flag)
-        return !(AV_RN64A(h->sub_mb_type) & ((MB_TYPE_16x8|MB_TYPE_8x16|MB_TYPE_8x8                )*0x0001000100010001ULL));
+static av_always_inline int get_dct8x8_allowed(H264Context *h, H264SliceContext *sl)
+{
+    if (h->sps.direct_8x8_inference_flag)
+        return !(AV_RN64A(sl->sub_mb_type) &
+                 ((MB_TYPE_16x8 | MB_TYPE_8x16 | MB_TYPE_8x8) *
+                  0x0001000100010001ULL));
     else
-        return !(AV_RN64A(h->sub_mb_type) & ((MB_TYPE_16x8|MB_TYPE_8x16|MB_TYPE_8x8|MB_TYPE_DIRECT2)*0x0001000100010001ULL));
+        return !(AV_RN64A(sl->sub_mb_type) &
+                 ((MB_TYPE_16x8 | MB_TYPE_8x16 | MB_TYPE_8x8 | MB_TYPE_DIRECT2) *
+                  0x0001000100010001ULL));
 }
 
-/**
- * decodes a P_SKIP or B_SKIP macroblock
- */
-static void av_unused decode_mb_skip(H264Context *h){
-    MpegEncContext * const s = &h->s;
-    const int mb_xy= h->mb_xy;
-    int mb_type=0;
+int ff_h264_field_end(H264Context *h, int in_setup);
 
-    memset(h->non_zero_count[mb_xy], 0, 48);
+int ff_h264_ref_picture(H264Context *h, H264Picture *dst, H264Picture *src);
+void ff_h264_unref_picture(H264Context *h, H264Picture *pic);
 
-    if(MB_FIELD)
-        mb_type|= MB_TYPE_INTERLACED;
+int ff_h264_context_init(H264Context *h);
+int ff_h264_set_parameter_from_sps(H264Context *h);
 
-    if( h->slice_type_nos == AV_PICTURE_TYPE_B )
-    {
-        // just for fill_caches. pred_direct_motion will set the real mb_type
-        mb_type|= MB_TYPE_L0L1|MB_TYPE_DIRECT2|MB_TYPE_SKIP;
-        if(h->direct_spatial_mv_pred){
-            fill_decode_neighbors(h, mb_type);
-        fill_decode_caches(h, mb_type); //FIXME check what is needed and what not ...
-        }
-        ff_h264_pred_direct_motion(h, &mb_type);
-        mb_type|= MB_TYPE_SKIP;
-    }
-    else
-    {
-        int mx, my;
-        mb_type|= MB_TYPE_16x16|MB_TYPE_P0L0|MB_TYPE_P1L0|MB_TYPE_SKIP;
-
-        fill_decode_neighbors(h, mb_type);
-        fill_decode_caches(h, mb_type); //FIXME check what is needed and what not ...
-        pred_pskip_motion(h, &mx, &my);
-        fill_rectangle(&h->ref_cache[0][scan8[0]], 4, 4, 8, 0, 1);
-        fill_rectangle(  h->mv_cache[0][scan8[0]], 4, 4, 8, pack16to32(mx,my), 4);
-    }
+void ff_h264_draw_horiz_band(H264Context *h, int y, int height);
+int ff_init_poc(H264Context *h, int pic_field_poc[2], int *pic_poc);
+int ff_pred_weight_table(H264Context *h, H264SliceContext *sl);
+int ff_set_ref_count(H264Context *h, H264SliceContext *sl);
 
-    write_back_motion(h, mb_type);
-    s->current_picture.mb_type[mb_xy]= mb_type;
-    s->current_picture.qscale_table[mb_xy]= s->qscale;
-    h->slice_table[ mb_xy ]= h->slice_num;
-    h->prev_mb_skipped= 1;
-}
+int ff_h264_decode_slice_header(H264Context *h, H264SliceContext *sl, H264Context *h0);
+int ff_h264_execute_decode_slices(H264Context *h, unsigned context_count);
+int ff_h264_update_thread_context(AVCodecContext *dst,
+                                  const AVCodecContext *src);
+
+void ff_h264_flush_change(H264Context *h);
 
-#include "h264_mvpred.h" //For pred_pskip_motion()
+void ff_h264_free_tables(H264Context *h, int free_rbsp);
 
 #endif /* AVCODEC_H264_H */