diff options
Diffstat (limited to 'encode.c')
| -rw-r--r-- | encode.c | 279 |
1 files changed, 201 insertions, 78 deletions
@@ -15,6 +15,8 @@ * along with streamer. If not, see <https://www.gnu.org/licenses/>. */ +#include "encode.h" + #include <assert.h> #include <errno.h> #include <fcntl.h> @@ -28,7 +30,6 @@ #include <va/va_drmcommon.h> #include "bitstream.h" -#include "encode.h" #include "gpu.h" #include "hevc.h" #include "toolbox/utils.h" @@ -44,10 +45,9 @@ struct EncodeContext { VADisplay va_display; VAConfigID va_config_id; - struct { - bool packed_header_sequence; - bool packed_header_slice; - } codec_quirks; + uint32_t va_packed_headers; + VAConfigAttribValEncHEVCFeatures va_hevc_features; + VAConfigAttribValEncHEVCBlockSizes va_hevc_block_sizes; VAContextID va_context_id; VASurfaceID input_surface_id; @@ -116,34 +116,134 @@ static void OnVaLogMessage(void* context, const char* message) { LOG("%.*s", (int)len, message); } -static bool InitializeCodecQuirks(struct EncodeContext* encode_context) { - bool result = false; - VAProfile dummy_profile; - VAEntrypoint dummy_entrypoint; - int num_attribs = vaMaxNumConfigAttributes(encode_context->va_display); - VAConfigAttrib* attrib_list = - malloc((size_t)num_attribs * sizeof(VAConfigAttrib)); - VAStatus status = vaQueryConfigAttributes( - encode_context->va_display, encode_context->va_config_id, &dummy_profile, - &dummy_entrypoint, attrib_list, &num_attribs); +static bool InitializeCodecCaps(struct EncodeContext* encode_context) { + VAConfigAttrib attrib_list[] = { + {.type = VAConfigAttribEncPackedHeaders}, + {.type = VAConfigAttribEncHEVCFeatures}, + {.type = VAConfigAttribEncHEVCBlockSizes}, + }; + VAStatus status = vaGetConfigAttributes( + encode_context->va_display, VAProfileHEVCMain, VAEntrypointEncSlice, + attrib_list, LENGTH(attrib_list)); if (status != VA_STATUS_SUCCESS) { - LOG("Failed to query va config attributes (%s)", VaErrorString(status)); - goto rollback_attrib_list; + LOG("Failed to get va config attributes (%s)", VaErrorString(status)); + return false; } - for (int i = 0; i < num_attribs; i++) { - if (attrib_list[i].type == VAConfigAttribEncPackedHeaders) { - encode_context->codec_quirks.packed_header_sequence = - !!(attrib_list[i].value & VA_ENC_PACKED_HEADER_SEQUENCE); - encode_context->codec_quirks.packed_header_slice = - !!(attrib_list[i].value & VA_ENC_PACKED_HEADER_SLICE); - } + if (attrib_list[0].value == VA_ATTRIB_NOT_SUPPORTED) { + LOG("VAConfigAttribEncPackedHeaders is not supported"); + } else { + LOG("VAConfigAttribEncPackedHeaders is 0x%08x", attrib_list[0].value); + encode_context->va_packed_headers = attrib_list[0].value; + } + + if (attrib_list[1].value == VA_ATTRIB_NOT_SUPPORTED) { + LOG("VAConfigAttribEncHEVCFeatures is not supported"); + encode_context->va_hevc_features = (VAConfigAttribValEncHEVCFeatures){ + .bits = + { + // mburakov: ffmpeg hardcodes these for i965 Skylake driver. + .separate_colour_planes = 0, // Table 6-1 + .scaling_lists = 0, // No scaling lists + .amp = 1, // hardcoded + .sao = 0, // hardcoded + .pcm = 0, // hardcoded + .temporal_mvp = 0, // hardcoded + .strong_intra_smoothing = 0, // TODO + .dependent_slices = 0, // No slice segments + .sign_data_hiding = 0, // TODO + .constrained_intra_pred = 0, // TODO + .transform_skip = 0, // defaulted + .cu_qp_delta = 0, // Fixed quality + .weighted_prediction = 0, // TODO + .transquant_bypass = 0, // TODO + .deblocking_filter_disable = 0, // TODO + }, + }; + } else { + LOG("VAConfigAttribEncHEVCFeatures is 0x%08x", attrib_list[1].value); + encode_context->va_hevc_features.value = attrib_list[1].value; + } + + if (attrib_list[2].value == VA_ATTRIB_NOT_SUPPORTED) { + LOG("VAConfigAttribEncHEVCBlockSizes is not supported"); + encode_context->va_hevc_block_sizes = (VAConfigAttribValEncHEVCBlockSizes){ + .bits = + { + // mburakov: ffmpeg hardcodes these for i965 Skylake driver. + .log2_max_coding_tree_block_size_minus3 = 2, // hardcoded + .log2_min_coding_tree_block_size_minus3 = 0, // TODO + .log2_min_luma_coding_block_size_minus3 = 0, // hardcoded + .log2_max_luma_transform_block_size_minus2 = 3, // hardcoded + .log2_min_luma_transform_block_size_minus2 = 0, // hardcoded + .max_max_transform_hierarchy_depth_inter = 3, // hardcoded + .min_max_transform_hierarchy_depth_inter = 0, // defaulted + .max_max_transform_hierarchy_depth_intra = 3, // hardcoded + .min_max_transform_hierarchy_depth_intra = 0, // defaulted + .log2_max_pcm_coding_block_size_minus3 = 0, // TODO + .log2_min_pcm_coding_block_size_minus3 = 0, // TODO + }, + }; + } else { + LOG("VAConfigAttribEncHEVCBlockSizes is 0x%08x", attrib_list[2].value); + encode_context->va_hevc_block_sizes.value = attrib_list[2].value; } - result = true; -rollback_attrib_list: - free(attrib_list); - return result; +#ifndef NDEBUG + const typeof(encode_context->va_hevc_features.bits)* features_bits = + &encode_context->va_hevc_features.bits; + const typeof(encode_context->va_hevc_block_sizes.bits)* block_sizes_bits = + &encode_context->va_hevc_block_sizes.bits; + LOG("VAConfigAttribEncHEVCFeatures dump:" + "\n\tseparate_colour_planes = %u" + "\n\tscaling_lists = %u" + "\n\tamp = %u" + "\n\tsao = %u" + "\n\tpcm = %u" + "\n\ttemporal_mvp = %u" + "\n\tstrong_intra_smoothing = %u" + "\n\tdependent_slices = %u" + "\n\tsign_data_hiding = %u" + "\n\tconstrained_intra_pred = %u" + "\n\ttransform_skip = %u" + "\n\tcu_qp_delta = %u" + "\n\tweighted_prediction = %u" + "\n\ttransquant_bypass = %u" + "\n\tdeblocking_filter_disable = %u", + features_bits->separate_colour_planes, features_bits->scaling_lists, + features_bits->amp, features_bits->sao, features_bits->pcm, + features_bits->temporal_mvp, features_bits->strong_intra_smoothing, + features_bits->dependent_slices, features_bits->sign_data_hiding, + features_bits->constrained_intra_pred, features_bits->transform_skip, + features_bits->cu_qp_delta, features_bits->weighted_prediction, + features_bits->transquant_bypass, + features_bits->deblocking_filter_disable); + LOG("VAConfigAttribEncHEVCBlockSizes dump:" + "\n\tlog2_max_coding_tree_block_size_minus3 = %u" + "\n\tlog2_min_coding_tree_block_size_minus3 = %u" + "\n\tlog2_min_luma_coding_block_size_minus3 = %u" + "\n\tlog2_max_luma_transform_block_size_minus2 = %u" + "\n\tlog2_min_luma_transform_block_size_minus2 = %u" + "\n\tmax_max_transform_hierarchy_depth_inter = %u" + "\n\tmin_max_transform_hierarchy_depth_inter = %u" + "\n\tmax_max_transform_hierarchy_depth_intra = %u" + "\n\tmin_max_transform_hierarchy_depth_intra = %u" + "\n\tlog2_max_pcm_coding_block_size_minus3 = %u" + "\n\tlog2_min_pcm_coding_block_size_minus3 = %u", + block_sizes_bits->log2_max_coding_tree_block_size_minus3, + block_sizes_bits->log2_min_coding_tree_block_size_minus3, + block_sizes_bits->log2_min_luma_coding_block_size_minus3, + block_sizes_bits->log2_max_luma_transform_block_size_minus2, + block_sizes_bits->log2_min_luma_transform_block_size_minus2, + block_sizes_bits->max_max_transform_hierarchy_depth_inter, + block_sizes_bits->min_max_transform_hierarchy_depth_inter, + block_sizes_bits->max_max_transform_hierarchy_depth_intra, + block_sizes_bits->min_max_transform_hierarchy_depth_intra, + block_sizes_bits->log2_max_pcm_coding_block_size_minus3, + block_sizes_bits->log2_min_pcm_coding_block_size_minus3); +#endif + + return true; } static struct GpuFrame* VaSurfaceToGpuFrame(VADisplay va_display, @@ -193,6 +293,18 @@ release_planes: static void InitializeSeqHeader(struct EncodeContext* encode_context, uint16_t pic_width_in_luma_samples, uint16_t pic_height_in_luma_samples) { + const typeof(encode_context->va_hevc_features.bits)* features_bits = + &encode_context->va_hevc_features.bits; + const typeof(encode_context->va_hevc_block_sizes.bits)* block_sizes_bits = + &encode_context->va_hevc_block_sizes.bits; + + uint8_t log2_diff_max_min_luma_coding_block_size = + block_sizes_bits->log2_max_coding_tree_block_size_minus3 - + block_sizes_bits->log2_min_luma_coding_block_size_minus3; + uint8_t log2_diff_max_min_transform_block_size = + block_sizes_bits->log2_max_luma_transform_block_size_minus2 - + block_sizes_bits->log2_min_luma_transform_block_size_minus2; + encode_context->seq = (VAEncSequenceParameterBufferHEVC){ .general_profile_idc = 1, // Main profile .general_level_idc = 120, // Level 4 @@ -201,7 +313,7 @@ static void InitializeSeqHeader(struct EncodeContext* encode_context, .intra_period = 120, // Where this one comes from? .intra_idr_period = 120, // Each I frame is an IDR frame .ip_period = 1, // No B-frames - .bits_per_second = 0, // TODO (investigate) + .bits_per_second = 0, // TODO .pic_width_in_luma_samples = pic_width_in_luma_samples, .pic_height_in_luma_samples = pic_height_in_luma_samples, @@ -212,27 +324,31 @@ static void InitializeSeqHeader(struct EncodeContext* encode_context, .separate_colour_plane_flag = 0, // Table 6-1 .bit_depth_luma_minus8 = 0, // 8 bpp luma .bit_depth_chroma_minus8 = 0, // 8 bpp chroma - .scaling_list_enabled_flag = 0, // defaulted + .scaling_list_enabled_flag = 0, // No scaling lists .strong_intra_smoothing_enabled_flag = 0, // defaulted - // mburakov: ffmpeg hardcodes these for i965 Skylake driver. - .amp_enabled_flag = 1, // TODO (quirks) - .sample_adaptive_offset_enabled_flag = 0, // TODO (quirks) - .pcm_enabled_flag = 0, // TODO (quirks) - .pcm_loop_filter_disabled_flag = 0, // defaulted - .sps_temporal_mvp_enabled_flag = 0, // TODO (quirks) + .amp_enabled_flag = features_bits->amp, + .sample_adaptive_offset_enabled_flag = features_bits->sao, + .pcm_enabled_flag = features_bits->pcm, + .pcm_loop_filter_disabled_flag = 0, // defaulted + .sps_temporal_mvp_enabled_flag = features_bits->temporal_mvp, .low_delay_seq = 1, // No B-frames .hierachical_flag = 0, // defaulted }, - // mburakov: ffmpeg hardcodes these for i965 Skylake driver. - .log2_min_luma_coding_block_size_minus3 = 0, // TODO (quirks) - .log2_diff_max_min_luma_coding_block_size = 2, // TODO (quirks) - .log2_min_transform_block_size_minus2 = 0, // hardcoded - .log2_diff_max_min_transform_block_size = 3, // hardcoded - .max_transform_hierarchy_depth_inter = 3, // hardcoded - .max_transform_hierarchy_depth_intra = 3, // hardcoded + .log2_min_luma_coding_block_size_minus3 = + block_sizes_bits->log2_min_luma_coding_block_size_minus3, + .log2_diff_max_min_luma_coding_block_size = + log2_diff_max_min_luma_coding_block_size, + .log2_min_transform_block_size_minus2 = + block_sizes_bits->log2_min_luma_transform_block_size_minus2, + .log2_diff_max_min_transform_block_size = + log2_diff_max_min_transform_block_size, + .max_transform_hierarchy_depth_inter = + block_sizes_bits->max_max_transform_hierarchy_depth_inter, + .max_transform_hierarchy_depth_intra = + block_sizes_bits->max_max_transform_hierarchy_depth_intra, .pcm_sample_bit_depth_luma_minus1 = 0, // defaulted .pcm_sample_bit_depth_chroma_minus1 = 0, // defaulted @@ -245,7 +361,7 @@ static void InitializeSeqHeader(struct EncodeContext* encode_context, .aspect_ratio_info_present_flag = 0, // defaulted .neutral_chroma_indication_flag = 0, // defaulted .field_seq_flag = 0, // defaulted - .vui_timing_info_present_flag = 1, // hardcoded + .vui_timing_info_present_flag = 0, // No timing .bitstream_restriction_flag = 1, // hardcoded .tiles_fixed_structure_flag = 0, // defaulted .motion_vectors_over_pic_boundaries_flag = 1, // hardcoded @@ -254,8 +370,8 @@ static void InitializeSeqHeader(struct EncodeContext* encode_context, .log2_max_mv_length_vertical = 15, // hardcoded }, - .vui_num_units_in_tick = 1, // TODO (investigate) - .vui_time_scale = 60, // TODO (investigate) + .vui_num_units_in_tick = 0, // No timing + .vui_time_scale = 0, // No timing .min_spatial_segmentation_idc = 0, // defaulted .max_bytes_per_pic_denom = 0, // hardcoded .max_bits_per_min_cu_denom = 0, // hardcoded @@ -270,6 +386,8 @@ static void InitializeSeqHeader(struct EncodeContext* encode_context, static void InitializePicHeader(struct EncodeContext* encode_context) { const typeof(encode_context->seq.seq_fields.bits)* seq_bits = &encode_context->seq.seq_fields.bits; + const typeof(encode_context->va_hevc_features.bits)* features_bits = + &encode_context->va_hevc_features.bits; uint8_t collocated_ref_pic_index = seq_bits->sps_temporal_mvp_enabled_flag ? 0 : 0xff; @@ -313,7 +431,7 @@ static void InitializePicHeader(struct EncodeContext* encode_context) { .dependent_slice_segments_enabled_flag = 0, // defaulted .sign_data_hiding_enabled_flag = 0, // defaulted .constrained_intra_pred_flag = 0, // defaulted - .transform_skip_enabled_flag = 0, // TODO (quirks) + .transform_skip_enabled_flag = features_bits->transform_skip, .cu_qp_delta_enabled_flag = 0, // Fixed quality .weighted_pred_flag = 0, // defaulted .weighted_bipred_flag = 0, // defaulted @@ -323,9 +441,9 @@ static void InitializePicHeader(struct EncodeContext* encode_context) { .loop_filter_across_tiles_enabled_flag = 0, // No tiles .pps_loop_filter_across_slices_enabled_flag = 1, // hardcoded - .scaling_list_data_present_flag = 0, // defaulted + .scaling_list_data_present_flag = 0, // No scaling lists - .screen_content_flag = 0, // TODO (investigate) + .screen_content_flag = 0, // TODO .enable_gpu_weighted_prediction = 0, // hardcoded .no_output_of_prior_pics_flag = 0, // hardcoded }, @@ -345,10 +463,18 @@ static void InitializePicHeader(struct EncodeContext* encode_context) { } } -static void InitializeSliceHeader(struct EncodeContext* encode_context, - uint32_t num_ctu_in_slice) { +static void InitializeSliceHeader(struct EncodeContext* encode_context) { const typeof(encode_context->seq.seq_fields.bits)* seq_bits = &encode_context->seq.seq_fields.bits; + const typeof(encode_context->va_hevc_block_sizes.bits)* block_sizes_bits = + &encode_context->va_hevc_block_sizes.bits; + + uint32_t ctu_size = + 1 << (block_sizes_bits->log2_max_coding_tree_block_size_minus3 + 3); + uint32_t slice_block_rows = + (encode_context->height + ctu_size - 1) / ctu_size; + uint32_t slice_block_cols = (encode_context->width + ctu_size - 1) / ctu_size; + uint32_t num_ctu_in_slice = slice_block_rows * slice_block_cols; encode_context->slice = (VAEncSliceParameterBufferHEVC){ .slice_segment_address = 0, // No slice segments @@ -477,28 +603,25 @@ struct EncodeContext* EncodeContextCreate(struct GpuContext* gpu_context, goto rollback_va_display; } - if (!InitializeCodecQuirks(encode_context)) { - LOG("Failed to initialize codec quirks"); + if (!InitializeCodecCaps(encode_context)) { + LOG("Failed to initialize codec caps"); goto rollback_va_config_id; } - // TODO(mburakov): ffmpeg attempts to deduce this. + // mburakov: Intel fails badly when min_cb_size value is not set to 16 and + // log2_min_luma_coding_block_size_minus3 is not set to zero. Judging from + // ffmpeg code, calculating one from another should work on other platforms, + // but I hardcoded it instead since AMD is fine with alignment on 16 anyway. static const uint32_t min_cb_size = 16; - uint32_t width_in_cb = (width + min_cb_size - 1) / min_cb_size; - uint32_t height_in_cb = (height + min_cb_size - 1) / min_cb_size; - - // TODO(mburakov): ffmpeg attempts to deduce this. - static const uint32_t slice_block_size = 32; - uint32_t slice_block_rows = - (encode_context->width + slice_block_size - 1) / slice_block_size; - uint32_t slice_block_cols = - (encode_context->height + slice_block_size - 1) / slice_block_size; - uint32_t num_ctu_in_slice = slice_block_rows * slice_block_cols; + uint32_t aligned_width = + (encode_context->width + min_cb_size - 1) & ~(min_cb_size - 1); + uint32_t aligned_height = + (encode_context->height + min_cb_size - 1) & ~(min_cb_size - 1); - status = vaCreateContext( - encode_context->va_display, encode_context->va_config_id, - (int)(width_in_cb * min_cb_size), (int)(height_in_cb * min_cb_size), - VA_PROGRESSIVE, NULL, 0, &encode_context->va_context_id); + status = + vaCreateContext(encode_context->va_display, encode_context->va_config_id, + (int)aligned_width, (int)aligned_height, VA_PROGRESSIVE, + NULL, 0, &encode_context->va_context_id); if (status != VA_STATUS_SUCCESS) { LOG("Failed to create va context (%s)", VaErrorString(status)); goto rollback_va_config_id; @@ -520,11 +643,10 @@ struct EncodeContext* EncodeContextCreate(struct GpuContext* gpu_context, goto rollback_input_surface_id; } - status = - vaCreateSurfaces(encode_context->va_display, VA_RT_FORMAT_YUV420, - width_in_cb * min_cb_size, height_in_cb * min_cb_size, - encode_context->recon_surface_ids, - LENGTH(encode_context->recon_surface_ids), NULL, 0); + status = vaCreateSurfaces(encode_context->va_display, VA_RT_FORMAT_YUV420, + aligned_width, aligned_height, + encode_context->recon_surface_ids, + LENGTH(encode_context->recon_surface_ids), NULL, 0); if (status != VA_STATUS_SUCCESS) { LOG("Failed to create va recon surfaces (%s)", VaErrorString(status)); goto rollback_gpu_frame; @@ -541,10 +663,10 @@ struct EncodeContext* EncodeContextCreate(struct GpuContext* gpu_context, goto rollback_recon_surface_ids; } - InitializeSeqHeader(encode_context, (uint16_t)(width_in_cb * min_cb_size), - (uint16_t)(height_in_cb * min_cb_size)); + InitializeSeqHeader(encode_context, (uint16_t)aligned_width, + (uint16_t)aligned_height); InitializePicHeader(encode_context); - InitializeSliceHeader(encode_context, num_ctu_in_slice); + InitializeSliceHeader(encode_context); return encode_context; rollback_recon_surface_ids: @@ -669,7 +791,8 @@ bool EncodeContextEncodeFrame(struct EncodeContext* encode_context, int fd) { goto rollback_buffers; } - if (encode_context->codec_quirks.packed_header_sequence && idr) { + if (idr && + (encode_context->va_packed_headers & VA_ENC_PACKED_HEADER_SEQUENCE)) { char buffer[256]; struct Bitstream bitstream = { .data = buffer, @@ -722,7 +845,7 @@ bool EncodeContextEncodeFrame(struct EncodeContext* encode_context, int fd) { encode_context->slice.slice_type = idr ? I : P; encode_context->slice.ref_pic_list0[0] = encode_context->pic.reference_frames[0]; - if (encode_context->codec_quirks.packed_header_slice) { + if (encode_context->va_packed_headers & VA_ENC_PACKED_HEADER_SLICE) { char buffer[256]; struct Bitstream bitstream = { .data = buffer, |