X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=x264.h;h=9b6d8f41d2e527e6d9ba141a3de738b24f3ab0ec;hb=5c13589be828b524100c787057d6bef77898c657;hp=8dbada85efa3259fa11c5c94ec7e1cf88caa97d6;hpb=ae289e6f03b76afa8736806e683349e8e59fcc93;p=x264 diff --git a/x264.h b/x264.h index 8dbada85..9b6d8f41 100644 --- a/x264.h +++ b/x264.h @@ -1,7 +1,7 @@ /***************************************************************************** * x264.h: x264 public header ***************************************************************************** - * Copyright (C) 2003-2012 x264 project + * Copyright (C) 2003-2015 x264 project * * Authors: Laurent Aimar * Loren Merritt @@ -28,8 +28,12 @@ #ifndef X264_X264_H #define X264_X264_H -#if !defined(_STDINT_H) && !defined(_STDINT_H_) && \ - !defined(_INTTYPES_H) && !defined(_INTTYPES_H_) +#ifdef __cplusplus +extern "C" { +#endif + +#if !defined(_STDINT_H) && !defined(_STDINT_H_) && !defined(_STDINT_H_INCLUDED) && !defined(_STDINT) &&\ + !defined(_SYS_STDINT_H_) && !defined(_INTTYPES_H) && !defined(_INTTYPES_H_) && !defined(_INTTYPES) # ifdef _MSC_VER # pragma message("You must include stdint.h or inttypes.h before x264.h") # else @@ -41,7 +45,18 @@ #include "x264_config.h" -#define X264_BUILD 120 +#define X264_BUILD 148 + +/* Application developers planning to link against a shared library version of + * libx264 from a Microsoft Visual Studio or similar development environment + * will need to define X264_API_IMPORTS before including this header. + * This clause does not apply to MinGW, similar development environments, or non + * Windows platforms. */ +#ifdef X264_API_IMPORTS +#define X264_API __declspec(dllimport) +#else +#define X264_API +#endif /* x264_t: * opaque handler for encoder */ @@ -79,7 +94,7 @@ enum nal_priority_e * All data returned in an x264_nal_t, including the data in p_payload, is no longer * valid after the next call to x264_encoder_encode. Thus it must be used or copied * before calling x264_encoder_encode or x264_encoder_headers again. */ -typedef struct +typedef struct x264_nal_t { int i_ref_idc; /* nal_priority_e */ int i_type; /* nal_unit_type_e */ @@ -87,54 +102,70 @@ typedef struct int i_first_mb; /* If this NAL is a slice, the index of the first MB in the slice. */ int i_last_mb; /* If this NAL is a slice, the index of the last MB in the slice. */ - /* Size of payload in bytes. */ + /* Size of payload (including any padding) in bytes. */ int i_payload; /* If param->b_annexb is set, Annex-B bytestream with startcode. * Otherwise, startcode is replaced with a 4-byte size. * This size is the size used in mp4/similar muxing; it is equal to i_payload-4 */ uint8_t *p_payload; + + /* Size of padding in bytes. */ + int i_padding; } x264_nal_t; /**************************************************************************** * Encoder parameters ****************************************************************************/ -/* CPU flags - */ -#define X264_CPU_CACHELINE_32 0x0000001 /* avoid memory loads that span the border between two cachelines */ -#define X264_CPU_CACHELINE_64 0x0000002 /* 32/64 is the size of a cacheline in bytes */ -#define X264_CPU_ALTIVEC 0x0000004 -#define X264_CPU_MMX 0x0000008 -#define X264_CPU_MMX2 0x0000010 /* MMX2 aka MMXEXT aka ISSE */ +/* CPU flags */ + +/* x86 */ +#define X264_CPU_CMOV 0x0000001 +#define X264_CPU_MMX 0x0000002 +#define X264_CPU_MMX2 0x0000004 /* MMX2 aka MMXEXT aka ISSE */ #define X264_CPU_MMXEXT X264_CPU_MMX2 -#define X264_CPU_SSE 0x0000020 -#define X264_CPU_SSE2 0x0000040 -#define X264_CPU_SSE2_IS_SLOW 0x0000080 /* avoid most SSE2 functions on Athlon64 */ -#define X264_CPU_SSE2_IS_FAST 0x0000100 /* a few functions are only faster on Core2 and Phenom */ -#define X264_CPU_SSE3 0x0000200 -#define X264_CPU_SSSE3 0x0000400 -#define X264_CPU_SHUFFLE_IS_FAST 0x0000800 /* Penryn, Nehalem, and Phenom have fast shuffle units */ -#define X264_CPU_STACK_MOD4 0x0001000 /* if stack is only mod4 and not mod16 */ -#define X264_CPU_SSE4 0x0002000 /* SSE4.1 */ -#define X264_CPU_SSE42 0x0004000 /* SSE4.2 */ -#define X264_CPU_SSE_MISALIGN 0x0008000 /* Phenom support for misaligned SSE instruction arguments */ -#define X264_CPU_LZCNT 0x0010000 /* Phenom support for "leading zero count" instruction. */ -#define X264_CPU_ARMV6 0x0020000 -#define X264_CPU_NEON 0x0040000 /* ARM NEON */ -#define X264_CPU_FAST_NEON_MRC 0x0080000 /* Transfer from NEON to ARM register is fast (Cortex-A9) */ -#define X264_CPU_SLOW_CTZ 0x0100000 /* BSR/BSF x86 instructions are really slow on some CPUs */ -#define X264_CPU_SLOW_ATOM 0x0200000 /* The Atom just sucks */ -#define X264_CPU_AVX 0x0400000 /* AVX support: requires OS support even if YMM registers - * aren't used. */ -#define X264_CPU_XOP 0x0800000 /* AMD XOP */ -#define X264_CPU_FMA4 0x1000000 /* AMD FMA4 */ -#define X264_CPU_AVX2 0x2000000 /* AVX2 */ -#define X264_CPU_FMA3 0x4000000 /* Intel FMA3 */ -#define X264_CPU_BMI1 0x8000000 /* BMI1 */ -#define X264_CPU_BMI2 0x10000000 /* BMI2 */ -#define X264_CPU_TBM 0x20000000 /* AMD TBM */ - -/* Analyse flags - */ +#define X264_CPU_SSE 0x0000008 +#define X264_CPU_SSE2 0x0000010 +#define X264_CPU_SSE3 0x0000020 +#define X264_CPU_SSSE3 0x0000040 +#define X264_CPU_SSE4 0x0000080 /* SSE4.1 */ +#define X264_CPU_SSE42 0x0000100 /* SSE4.2 */ +#define X264_CPU_LZCNT 0x0000200 /* Phenom support for "leading zero count" instruction. */ +#define X264_CPU_AVX 0x0000400 /* AVX support: requires OS support even if YMM registers aren't used. */ +#define X264_CPU_XOP 0x0000800 /* AMD XOP */ +#define X264_CPU_FMA4 0x0001000 /* AMD FMA4 */ +#define X264_CPU_FMA3 0x0002000 /* FMA3 */ +#define X264_CPU_AVX2 0x0004000 /* AVX2 */ +#define X264_CPU_BMI1 0x0008000 /* BMI1 */ +#define X264_CPU_BMI2 0x0010000 /* BMI2 */ +/* x86 modifiers */ +#define X264_CPU_CACHELINE_32 0x0020000 /* avoid memory loads that span the border between two cachelines */ +#define X264_CPU_CACHELINE_64 0x0040000 /* 32/64 is the size of a cacheline in bytes */ +#define X264_CPU_SSE2_IS_SLOW 0x0080000 /* avoid most SSE2 functions on Athlon64 */ +#define X264_CPU_SSE2_IS_FAST 0x0100000 /* a few functions are only faster on Core2 and Phenom */ +#define X264_CPU_SLOW_SHUFFLE 0x0200000 /* The Conroe has a slow shuffle unit (relative to overall SSE performance) */ +#define X264_CPU_STACK_MOD4 0x0400000 /* if stack is only mod4 and not mod16 */ +#define X264_CPU_SLOW_CTZ 0x0800000 /* BSR/BSF x86 instructions are really slow on some CPUs */ +#define X264_CPU_SLOW_ATOM 0x1000000 /* The Atom is terrible: slow SSE unaligned loads, slow + * SIMD multiplies, slow SIMD variable shifts, slow pshufb, + * cacheline split penalties -- gather everything here that + * isn't shared by other CPUs to avoid making half a dozen + * new SLOW flags. */ +#define X264_CPU_SLOW_PSHUFB 0x2000000 /* such as on the Intel Atom */ +#define X264_CPU_SLOW_PALIGNR 0x4000000 /* such as on the AMD Bobcat */ + +/* PowerPC */ +#define X264_CPU_ALTIVEC 0x0000001 + +/* ARM and AArch64 */ +#define X264_CPU_ARMV6 0x0000001 +#define X264_CPU_NEON 0x0000002 /* ARM NEON */ +#define X264_CPU_FAST_NEON_MRC 0x0000004 /* Transfer from NEON to ARM register is fast (Cortex-A9) */ +#define X264_CPU_ARMV8 0x0000008 + +/* MIPS */ +#define X264_CPU_MSA 0x0000001 /* MIPS MSA */ + +/* Analyse flags */ #define X264_ANALYSE_I4x4 0x0001 /* Analyse i4x4 */ #define X264_ANALYSE_I8x8 0x0002 /* Analyse i8x8 (requires 8x8 transform) */ #define X264_ANALYSE_PSUB16x16 0x0010 /* Analyse p16x8, p8x16 and p8x8 */ @@ -159,6 +190,7 @@ typedef struct #define X264_AQ_NONE 0 #define X264_AQ_VARIANCE 1 #define X264_AQ_AUTOVARIANCE 2 +#define X264_AQ_AUTOVARIANCE_BIASED 3 #define X264_B_ADAPT_NONE 0 #define X264_B_ADAPT_FAST 1 #define X264_B_ADAPT_TRELLIS 2 @@ -177,9 +209,10 @@ static const char * const x264_b_pyramid_names[] = { "none", "strict", "normal", static const char * const x264_overscan_names[] = { "undef", "show", "crop", 0 }; static const char * const x264_vidformat_names[] = { "component", "pal", "ntsc", "secam", "mac", "undef", 0 }; static const char * const x264_fullrange_names[] = { "off", "on", 0 }; -static const char * const x264_colorprim_names[] = { "", "bt709", "undef", "", "bt470m", "bt470bg", "smpte170m", "smpte240m", "film", 0 }; -static const char * const x264_transfer_names[] = { "", "bt709", "undef", "", "bt470m", "bt470bg", "smpte170m", "smpte240m", "linear", "log100", "log316", 0 }; -static const char * const x264_colmatrix_names[] = { "GBR", "bt709", "undef", "", "fcc", "bt470bg", "smpte170m", "smpte240m", "YCgCo", 0 }; +static const char * const x264_colorprim_names[] = { "", "bt709", "undef", "", "bt470m", "bt470bg", "smpte170m", "smpte240m", "film", "bt2020", 0 }; +static const char * const x264_transfer_names[] = { "", "bt709", "undef", "", "bt470m", "bt470bg", "smpte170m", "smpte240m", "linear", "log100", "log316", + "iec61966-2-4", "bt1361e", "iec61966-2-1", "bt2020-10", "bt2020-12", 0 }; +static const char * const x264_colmatrix_names[] = { "GBR", "bt709", "undef", "", "fcc", "bt470bg", "smpte170m", "smpte240m", "YCgCo", "bt2020nc", "bt2020c", 0 }; static const char * const x264_nal_hrd_names[] = { "none", "vbr", "cbr", 0 }; /* Colorspace type */ @@ -188,15 +221,17 @@ static const char * const x264_nal_hrd_names[] = { "none", "vbr", "cbr", 0 }; #define X264_CSP_I420 0x0001 /* yuv 4:2:0 planar */ #define X264_CSP_YV12 0x0002 /* yvu 4:2:0 planar */ #define X264_CSP_NV12 0x0003 /* yuv 4:2:0, with one y plane and one packed u+v */ -#define X264_CSP_I422 0x0004 /* yuv 4:2:2 planar */ -#define X264_CSP_YV16 0x0005 /* yvu 4:2:2 planar */ -#define X264_CSP_NV16 0x0006 /* yuv 4:2:2, with one y plane and one packed u+v */ -#define X264_CSP_I444 0x0007 /* yuv 4:4:4 planar */ -#define X264_CSP_YV24 0x0008 /* yvu 4:4:4 planar */ -#define X264_CSP_BGR 0x0009 /* packed bgr 24bits */ -#define X264_CSP_BGRA 0x000a /* packed bgr 32bits */ -#define X264_CSP_RGB 0x000b /* packed rgb 24bits */ -#define X264_CSP_MAX 0x000c /* end of list */ +#define X264_CSP_NV21 0x0004 /* yuv 4:2:0, with one y plane and one packed v+u */ +#define X264_CSP_I422 0x0005 /* yuv 4:2:2 planar */ +#define X264_CSP_YV16 0x0006 /* yvu 4:2:2 planar */ +#define X264_CSP_NV16 0x0007 /* yuv 4:2:2, with one y plane and one packed u+v */ +#define X264_CSP_V210 0x0008 /* 10-bit yuv 4:2:2 packed in 32 */ +#define X264_CSP_I444 0x0009 /* yuv 4:4:4 planar */ +#define X264_CSP_YV24 0x000a /* yvu 4:4:4 planar */ +#define X264_CSP_BGR 0x000b /* packed bgr 24bits */ +#define X264_CSP_BGRA 0x000c /* packed bgr 32bits */ +#define X264_CSP_RGB 0x000d /* packed rgb 24bits */ +#define X264_CSP_MAX 0x000e /* end of list */ #define X264_CSP_VFLIP 0x1000 /* the csp is vertically flipped */ #define X264_CSP_HIGH_DEPTH 0x2000 /* the csp has a depth of 16 bits per pixel component */ @@ -208,7 +243,7 @@ static const char * const x264_nal_hrd_names[] = { "none", "vbr", "cbr", 0 }; #define X264_TYPE_BREF 0x0004 /* Non-disposable B-frame */ #define X264_TYPE_B 0x0005 #define X264_TYPE_KEYFRAME 0x0006 /* IDR or I depending on b_open_gop option */ -#define IS_X264_TYPE_I(x) ((x)==X264_TYPE_I || (x)==X264_TYPE_IDR) +#define IS_X264_TYPE_I(x) ((x)==X264_TYPE_I || (x)==X264_TYPE_IDR || (x)==X264_TYPE_KEYFRAME) #define IS_X264_TYPE_B(x) ((x)==X264_TYPE_B || (x)==X264_TYPE_BREF) /* Log level */ @@ -230,7 +265,7 @@ static const char * const x264_nal_hrd_names[] = { "none", "vbr", "cbr", 0 }; /* Zones: override ratecontrol or other options for specific sections of the video. * See x264_encoder_reconfig() for which options can be changed. * If zones overlap, whichever comes later in the list takes precedence. */ -typedef struct +typedef struct x264_zone_t { int i_start, i_end; /* range of frame numbers */ int b_force_qp; /* whether to use qp vs bitrate factor */ @@ -243,7 +278,8 @@ typedef struct x264_param_t { /* CPU flags */ unsigned int cpu; - int i_threads; /* encode multiple frames in parallel */ + int i_threads; /* encode multiple frames in parallel */ + int i_lookahead_threads; /* multiple threads for lookahead analysis */ int b_sliced_threads; /* Whether to use slice-based threading. */ int b_deterministic; /* whether to allow non-deterministic optimizations when threaded */ int b_cpu_independent; /* force canonical behavior rather than cpu-dependent optimal algorithms */ @@ -296,6 +332,7 @@ typedef struct x264_param_t int i_bframe_pyramid; /* Keep some B-frames as references: 0=off, 1=strict hierarchical, 2=normal */ int b_open_gop; int b_bluray_compat; + int i_avcintra_class; int b_deblocking_filter; int i_deblocking_filter_alphac0; /* [-6, 6] -6 light filter, 6 strong */ @@ -308,7 +345,7 @@ typedef struct x264_param_t int b_constrained_intra; int i_cqm_preset; - char *psz_cqm_file; /* JM format */ + char *psz_cqm_file; /* filename (in UTF-8) of CQM file, JM format */ uint8_t cqm_4iy[16]; /* used only if i_cqm_preset == X264_CQM_CUSTOM */ uint8_t cqm_4py[16]; uint8_t cqm_4ic[16]; @@ -322,8 +359,8 @@ typedef struct x264_param_t void (*pf_log)( void *, int i_level, const char *psz, va_list ); void *p_log_private; int i_log_level; - int b_visualize; - char *psz_dump_yuv; /* filename for reconstructed frames */ + int b_full_recon; /* fully reconstruct frames, even when not necessary for encoding. Implied by psz_dump_yuv */ + char *psz_dump_yuv; /* filename (in UTF-8) for reconstructed frames */ /* Encoder analyser parameters */ struct @@ -352,6 +389,9 @@ typedef struct x264_param_t float f_psy_trellis; /* Psy trellis strength */ int b_psy; /* Toggle all psy optimizations */ + int b_mb_info; /* Use input mb_info data in x264_picture_t */ + int b_mb_info_update; /* Update the values in mb_info according to the results of encoding. */ + /* the deadzone size that will be used in luma quantization */ int i_luma_deadzone[2]; /* {inter, intra} */ @@ -379,6 +419,10 @@ typedef struct x264_param_t float f_ip_factor; float f_pb_factor; + /* VBV filler: force CBR VBV and use filler bytes to ensure hard-CBR. + * Implied by NAL-HRD CBR. */ + int b_filler; + int i_aq_mode; /* psy adaptive QP. (X264_AQ_*) */ float f_aq_strength; int b_mb_tree; /* Macroblock-tree ratecontrol. */ @@ -386,9 +430,9 @@ typedef struct x264_param_t /* 2pass */ int b_stat_write; /* Enable stat writing in psz_stat_out */ - char *psz_stat_out; + char *psz_stat_out; /* output filename (in UTF-8) of the 2pass stats file */ int b_stat_read; /* Read stat from psz_stat_in and use it */ - char *psz_stat_in; + char *psz_stat_in; /* input filename (in UTF-8) of the 2pass stats file */ /* 2pass params (same as ffmpeg ones) */ float f_qcompress; /* 0.0 => cbr, 1.0 => constant qp */ @@ -448,10 +492,23 @@ typedef struct x264_param_t int b_fake_interlaced; + /* Don't optimize header parameters based on video content, e.g. ensure that splitting an input video, compressing + * each part, and stitching them back together will result in identical SPS/PPS. This is necessary for stitching + * with container formats that don't allow multiple SPS/PPS. */ + int b_stitchable; + + int b_opencl; /* use OpenCL when available */ + int i_opencl_device; /* specify count of GPU devices to skip, for CLI users */ + void *opencl_device_id; /* pass explicit cl_device_id as void*, for API users */ + char *psz_clbin_file; /* filename (in UTF-8) of the compiled OpenCL kernel cache file */ + /* Slicing parameters */ int i_slice_max_size; /* Max size per slice in bytes; includes estimated NAL overhead. */ int i_slice_max_mbs; /* Max number of MBs per slice; overrides i_slice_count. */ + int i_slice_min_mbs; /* Min number of MBs per slice */ int i_slice_count; /* Number of slices per frame: forces rectangular slices. */ + int i_slice_count_max; /* Absolute cap on slices per frame; stops applying slice-max-size + * and slice-max-mbs if this is reached. */ /* Optional callback for freeing this x264_param_t when it is done being used. * Only used when the x264_param_t sits in memory for an indefinite period of time, @@ -465,7 +522,7 @@ typedef struct x264_param_t * is done encoding. * * This callback MUST do the following in order to work correctly: - * 1) Have available an output buffer of at least size nal->i_payload*3/2 + 5 + 16. + * 1) Have available an output buffer of at least size nal->i_payload*3/2 + 5 + 64. * 2) Call x264_nal_encode( h, dst, nal ), where dst is the output buffer. * After these steps, the content of nal is valid and can be used in the same way as if * the NAL unit were output by x264_encoder_encode. @@ -487,8 +544,13 @@ typedef struct x264_param_t * the calling application is expected to acquire all output NALs through the callback. * * It is generally sensible to combine this callback with a use of slice-max-mbs or - * slice-max-size. */ - void (*nalu_process) ( x264_t *h, x264_nal_t *nal ); + * slice-max-size. + * + * The opaque pointer is the opaque pointer from the input frame associated with this + * NAL unit. This helps distinguish between nalu_process calls from different sources, + * e.g. if doing multiple encodes in one process. + */ + void (*nalu_process) ( x264_t *h, x264_nal_t *nal, void *opaque ); } x264_param_t; void x264_nal_encode( x264_t *h, uint8_t *dst, x264_nal_t *nal ); @@ -497,12 +559,12 @@ void x264_nal_encode( x264_t *h, uint8_t *dst, x264_nal_t *nal ); * H.264 level restriction information ****************************************************************************/ -typedef struct +typedef struct x264_level_t { int level_idc; int mbps; /* max macroblock processing rate (macroblocks/sec) */ int frame_size; /* max frame size (macroblocks) */ - int dpb; /* max decoded picture buffer (bytes) */ + int dpb; /* max decoded picture buffer (mbs) */ int bitrate; /* max bitrate (kbit/sec) */ int cpb; /* max vbv buffer (kbit) */ int mv_range; /* max vertical mv component range (pixels) */ @@ -515,7 +577,7 @@ typedef struct } x264_level_t; /* all of the levels defined in the standard, terminated by .level_idc=0 */ -extern const x264_level_t x264_levels[]; +X264_API extern const x264_level_t x264_levels[]; /**************************************************************************** * Basic parameter handling functions @@ -612,14 +674,14 @@ int x264_param_apply_profile( x264_param_t *, const char *profile ); * (16-x264_bit_depth) bits to be zero. * Note: The flag X264_CSP_HIGH_DEPTH must be used to specify the * colorspace depth as well. */ -extern const int x264_bit_depth; +X264_API extern const int x264_bit_depth; /* x264_chroma_format: * Specifies the chroma formats that x264 supports encoding. When this * value is non-zero, then it represents a X264_CSP_* that is the only * chroma format that x264 supports encoding. If the value is 0 then * there are no restrictions. */ -extern const int x264_chroma_format; +X264_API extern const int x264_chroma_format; enum pic_struct_e { @@ -634,7 +696,7 @@ enum pic_struct_e PIC_STRUCT_TRIPLE = 9, // triple frame }; -typedef struct +typedef struct x264_hrd_t { double cpb_initial_arrival_time; double cpb_final_arrival_time; @@ -652,14 +714,14 @@ typedef struct * Payloads are written first in order of input, apart from in the case when HRD * is enabled where payloads are written after the Buffering Period SEI. */ -typedef struct +typedef struct x264_sei_payload_t { int payload_size; int payload_type; uint8_t *payload; } x264_sei_payload_t; -typedef struct +typedef struct x264_sei_t { int num_payloads; x264_sei_payload_t *payloads; @@ -667,7 +729,7 @@ typedef struct void (*sei_free)( void* ); } x264_sei_t; -typedef struct +typedef struct x264_image_t { int i_csp; /* Colorspace */ int i_plane; /* Number of image planes */ @@ -675,29 +737,67 @@ typedef struct uint8_t *plane[4]; /* Pointers to each plane */ } x264_image_t; -typedef struct +typedef struct x264_image_properties_t { + /* All arrays of data here are ordered as follows: + * each array contains one offset per macroblock, in raster scan order. In interlaced + * mode, top-field MBs and bottom-field MBs are interleaved at the row level. + * Macroblocks are 16x16 blocks of pixels (with respect to the luma plane). For the + * purposes of calculating the number of macroblocks, width and height are rounded up to + * the nearest 16. If in interlaced mode, height is rounded up to the nearest 32 instead. */ + /* In: an array of quantizer offsets to be applied to this image during encoding. * These are added on top of the decisions made by x264. * Offsets can be fractional; they are added before QPs are rounded to integer. * Adaptive quantization must be enabled to use this feature. Behavior if quant - * offsets differ between encoding passes is undefined. - * - * Array contains one offset per macroblock, in raster scan order. In interlaced - * mode, top-field MBs and bottom-field MBs are interleaved at the row level. */ + * offsets differ between encoding passes is undefined. */ float *quant_offsets; /* In: optional callback to free quant_offsets when used. * Useful if one wants to use a different quant_offset array for each frame. */ void (*quant_offsets_free)( void* ); + + /* In: optional array of flags for each macroblock. + * Allows specifying additional information for the encoder such as which macroblocks + * remain unchanged. Usable flags are listed below. + * x264_param_t.analyse.b_mb_info must be set to use this, since x264 needs to track + * extra data internally to make full use of this information. + * + * Out: if b_mb_info_update is set, x264 will update this array as a result of encoding. + * + * For "MBINFO_CONSTANT", it will remove this flag on any macroblock whose decoded + * pixels have changed. This can be useful for e.g. noting which areas of the + * frame need to actually be blitted. Note: this intentionally ignores the effects + * of deblocking for the current frame, which should be fine unless one needs exact + * pixel-perfect accuracy. + * + * Results for MBINFO_CONSTANT are currently only set for P-frames, and are not + * guaranteed to enumerate all blocks which haven't changed. (There may be false + * negatives, but no false positives.) + */ + uint8_t *mb_info; + /* In: optional callback to free mb_info when used. */ + void (*mb_info_free)( void* ); + + /* The macroblock is constant and remains unchanged from the previous frame. */ + #define X264_MBINFO_CONSTANT (1<<0) + /* More flags may be added in the future. */ + + /* Out: SSIM of the the frame luma (if x264_param_t.b_ssim is set) */ + double f_ssim; + /* Out: Average PSNR of the frame (if x264_param_t.b_psnr is set) */ + double f_psnr_avg; + /* Out: PSNR of Y, U, and V (if x264_param_t.b_psnr is set) */ + double f_psnr[3]; + + /* Out: Average effective CRF of the encoded frame */ + double f_crf_avg; } x264_image_properties_t; -typedef struct +typedef struct x264_picture_t { /* In: force picture type (if not auto) * If x264 encoding parameters are violated in the forcing of picture types, * x264 will correct the input picture type and log a warning. - * The quality of frametype decisions may suffer if a great deal of fine-grained - * mixing of auto and forced frametypes is done. * Out: type of the picture encoded */ int i_type; /* In: force quantizer for != X264_QP_AUTO */ @@ -721,9 +821,13 @@ typedef struct of H.264 itself; in this case, the caller must force an IDR frame if it needs the changed parameter to apply immediately. */ x264_param_t *param; - /* In: raw data */ + /* In: raw image data */ + /* Out: reconstructed image data. x264 may skip part of the reconstruction process, + e.g. deblocking, in frames where it isn't necessary. To force complete + reconstruction, at a small speed cost, set b_full_recon. */ x264_image_t img; - /* In: optional information to modify encoder decisions for this frame */ + /* In: optional information to modify encoder decisions for this frame + * Out: information about the encoded frame */ x264_image_properties_t prop; /* Out: HRD timing information. Output only when i_nal_hrd is set. */ x264_hrd_t hrd_timing; @@ -769,7 +873,13 @@ x264_t *x264_encoder_open( x264_param_t * ); * due to delay, this may not be the next frame passed to encoder_encode. * if the change should apply to some particular frame, use x264_picture_t->param instead. * returns 0 on success, negative on parameter validation error. - * not all parameters can be changed; see the actual function for a detailed breakdown. */ + * not all parameters can be changed; see the actual function for a detailed breakdown. + * + * since not all parameters can be changed, moving from preset to preset may not always + * fully copy all relevant parameters, but should still work usably in practice. however, + * more so than for other presets, many of the speed shortcuts used in ultrafast cannot be + * switched out of; using reconfig to switch between ultrafast and other presets is not + * recommended without a more fine-grained breakdown of parameters to take this into account. */ int x264_encoder_reconfig( x264_t *, x264_param_t * ); /* x264_encoder_parameters: * copies the current internal set of parameters to the pointer provided @@ -782,13 +892,15 @@ void x264_encoder_parameters( x264_t *, x264_param_t * ); /* x264_encoder_headers: * return the SPS and PPS that will be used for the whole stream. * *pi_nal is the number of NAL units outputted in pp_nal. + * returns the number of bytes in the returned NALs. * returns negative on error. * the payloads of all output NALs are guaranteed to be sequential in memory. */ int x264_encoder_headers( x264_t *, x264_nal_t **pp_nal, int *pi_nal ); /* x264_encoder_encode: * encode one picture. * *pi_nal is the number of NAL units outputted in pp_nal. - * returns negative on error, zero if no NAL units returned. + * returns the number of bytes in the returned NALs. + * returns negative on error and zero if no NAL units returned. * the payloads of all output NALs are guaranteed to be sequential in memory. */ int x264_encoder_encode( x264_t *, x264_nal_t **pp_nal, int *pi_nal, x264_picture_t *pic_in, x264_picture_t *pic_out ); /* x264_encoder_close: @@ -840,4 +952,8 @@ void x264_encoder_intra_refresh( x264_t * ); * Returns 0 on success, negative on failure. */ int x264_encoder_invalidate_reference( x264_t *, int64_t pts ); +#ifdef __cplusplus +} +#endif + #endif