You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

169 lines
6.6 KiB

/*
* VP9 HW decode acceleration through VA API
*
* Copyright (C) 2015 Timo Rothenpieler <timo@rothenpieler.org>
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "libavutil/pixdesc.h"
#include "vaapi_internal.h"
#include "vp9.h"
static void fill_picture_parameters(AVCodecContext *avctx,
const VP9SharedContext *h,
VADecPictureParameterBufferVP9 *pp)
{
const AVPixFmtDescriptor *pixdesc = av_pix_fmt_desc_get(avctx->sw_pix_fmt);
int i;
pp->frame_width = avctx->width;
pp->frame_height = avctx->height;
pp->frame_header_length_in_bytes = h->h.uncompressed_header_size;
pp->first_partition_size = h->h.compressed_header_size;
pp->profile = h->h.profile;
pp->filter_level = h->h.filter.level;
pp->sharpness_level = h->h.filter.sharpness;
pp->log2_tile_rows = h->h.tiling.log2_tile_rows;
pp->log2_tile_columns = h->h.tiling.log2_tile_cols;
pp->pic_fields.bits.subsampling_x = pixdesc->log2_chroma_w;
pp->pic_fields.bits.subsampling_y = pixdesc->log2_chroma_h;
pp->pic_fields.bits.frame_type = !h->h.keyframe;
pp->pic_fields.bits.show_frame = !h->h.invisible;
pp->pic_fields.bits.error_resilient_mode = h->h.errorres;
pp->pic_fields.bits.intra_only = h->h.intraonly;
pp->pic_fields.bits.allow_high_precision_mv = h->h.keyframe ? 0 : h->h.highprecisionmvs;
pp->pic_fields.bits.mcomp_filter_type = h->h.filtermode ^ (h->h.filtermode <= 1);
pp->pic_fields.bits.frame_parallel_decoding_mode = h->h.parallelmode;
pp->pic_fields.bits.reset_frame_context = h->h.resetctx;
pp->pic_fields.bits.refresh_frame_context = h->h.refreshctx;
pp->pic_fields.bits.frame_context_idx = h->h.framectxid;
pp->pic_fields.bits.segmentation_enabled = h->h.segmentation.enabled;
pp->pic_fields.bits.segmentation_temporal_update = h->h.segmentation.temporal;
pp->pic_fields.bits.segmentation_update_map = h->h.segmentation.update_map;
pp->pic_fields.bits.last_ref_frame = h->h.refidx[0];
pp->pic_fields.bits.last_ref_frame_sign_bias = h->h.signbias[0];
pp->pic_fields.bits.golden_ref_frame = h->h.refidx[1];
pp->pic_fields.bits.golden_ref_frame_sign_bias = h->h.signbias[1];
pp->pic_fields.bits.alt_ref_frame = h->h.refidx[2];
pp->pic_fields.bits.alt_ref_frame_sign_bias = h->h.signbias[2];
pp->pic_fields.bits.lossless_flag = h->h.lossless;
for (i = 0; i < 7; i++)
pp->mb_segment_tree_probs[i] = h->h.segmentation.prob[i];
if (h->h.segmentation.temporal) {
for (i = 0; i < 3; i++)
pp->segment_pred_probs[i] = h->h.segmentation.pred_prob[i];
} else {
memset(pp->segment_pred_probs, 255, sizeof(pp->segment_pred_probs));
}
for (i = 0; i < 8; i++) {
if (h->refs[i].f->buf[0]) {
pp->reference_frames[i] = ff_vaapi_get_surface_id(h->refs[i].f);
} else {
pp->reference_frames[i] = VA_INVALID_ID;
}
}
}
static int vaapi_vp9_start_frame(AVCodecContext *avctx,
av_unused const uint8_t *buffer,
av_unused uint32_t size)
{
const VP9SharedContext *h = avctx->priv_data;
FFVAContext * const vactx = ff_vaapi_get_context(avctx);
VADecPictureParameterBufferVP9 *pic_param;
vactx->slice_param_size = sizeof(VASliceParameterBufferVP9);
pic_param = ff_vaapi_alloc_pic_param(vactx, sizeof(VADecPictureParameterBufferVP9));
if (!pic_param)
return -1;
fill_picture_parameters(avctx, h, pic_param);
return 0;
}
static int vaapi_vp9_end_frame(AVCodecContext *avctx)
{
FFVAContext * const vactx = ff_vaapi_get_context(avctx);
const VP9SharedContext *h = avctx->priv_data;
int ret;
ret = ff_vaapi_commit_slices(vactx);
if (ret < 0)
goto finish;
ret = ff_vaapi_render_picture(vactx, ff_vaapi_get_surface_id(h->frames[CUR_FRAME].tf.f));
if (ret < 0)
goto finish;
finish:
ff_vaapi_common_end_frame(avctx);
return ret;
}
static int vaapi_vp9_decode_slice(AVCodecContext *avctx,
const uint8_t *buffer,
uint32_t size)
{
FFVAContext * const vactx = ff_vaapi_get_context(avctx);
const VP9SharedContext *h = avctx->priv_data;
VASliceParameterBufferVP9 *slice_param;
int i;
slice_param = (VASliceParameterBufferVP9*)ff_vaapi_alloc_slice(vactx, buffer, size);
if (!slice_param)
return -1;
for (i = 0; i < 8; i++) {
slice_param->seg_param[i].segment_flags.fields.segment_reference_enabled = h->h.segmentation.feat[i].ref_enabled;
slice_param->seg_param[i].segment_flags.fields.segment_reference = h->h.segmentation.feat[i].ref_val;
slice_param->seg_param[i].segment_flags.fields.segment_reference_skipped = h->h.segmentation.feat[i].skip_enabled;
memcpy(slice_param->seg_param[i].filter_level, h->h.segmentation.feat[i].lflvl, sizeof(slice_param->seg_param[i].filter_level));
slice_param->seg_param[i].luma_dc_quant_scale = h->h.segmentation.feat[i].qmul[0][0];
slice_param->seg_param[i].luma_ac_quant_scale = h->h.segmentation.feat[i].qmul[0][1];
slice_param->seg_param[i].chroma_dc_quant_scale = h->h.segmentation.feat[i].qmul[1][0];
slice_param->seg_param[i].chroma_ac_quant_scale = h->h.segmentation.feat[i].qmul[1][1];
}
return 0;
}
AVHWAccel ff_vp9_vaapi_hwaccel = {
.name = "vp9_vaapi",
.type = AVMEDIA_TYPE_VIDEO,
.id = AV_CODEC_ID_VP9,
.pix_fmt = AV_PIX_FMT_VAAPI,
.start_frame = vaapi_vp9_start_frame,
.end_frame = vaapi_vp9_end_frame,
.decode_slice = vaapi_vp9_decode_slice,
.init = ff_vaapi_context_init,
.uninit = ff_vaapi_context_fini,
.priv_data_size = sizeof(FFVAContext),
};