FFmpeg/libavcodec/dxva2_vp9.c
Andreas Rheinhardt 7bd3b73716 avcodec/vp9: Switch to ProgressFrames
This already fixes a race in the vp9-encparams test. In this test,
side data is added to the current frame after having been decoded
(and therefore after ff_thread_finish_setup() has been called).
Yet the update_thread_context callback called ff_thread_ref_frame()
and therefore av_frame_ref() with this frame as source frame and
the ensuing read was unsynchronised with adding the side data,
i.e. there was a data race.

By switching to the ProgressFrame API the implicit av_frame_ref()
is removed and the race fixed except if this frame is later reused by
a show-existing-frame which uses an explicit av_frame_ref().
The vp9-encparams test does not cover this, so this commit
already fixes all the races in this test.

This decoder kept multiple references to the same ThreadFrames
in the same context and therefore had lots of implicit av_frame_ref()
even when decoding single-threaded. This incurred lots of small
allocations: When decoding an ordinary 10s video in single-threaded
mode the number of allocations reported by Valgrind went down
from 57,814 to 20,908; for 10 threads it went down from 84,223 to
21,901.

Reviewed-by: Anton Khirnov <anton@khirnov.net>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2024-04-19 13:18:04 +02:00

361 lines
14 KiB
C

/*
* DXVA2 VP9 HW acceleration.
*
* copyright (c) 2015 Hendrik Leppkes
*
* 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 "config_components.h"
#include "libavutil/avassert.h"
#include "libavutil/pixdesc.h"
#include "dxva2_internal.h"
#include "hwaccel_internal.h"
#include "vp9shared.h"
struct vp9_dxva2_picture_context {
DXVA_PicParams_VP9 pp;
DXVA_Slice_VPx_Short slice;
const uint8_t *bitstream;
unsigned bitstream_size;
};
static void fill_picture_entry(DXVA_PicEntry_VPx *pic,
unsigned index, unsigned flag)
{
av_assert0((index & 0x7f) == index && (flag & 0x01) == flag);
pic->bPicEntry = index | (flag << 7);
}
int ff_dxva2_vp9_fill_picture_parameters(const AVCodecContext *avctx, AVDXVAContext *ctx,
DXVA_PicParams_VP9 *pp)
{
const VP9SharedContext *h = avctx->priv_data;
const AVPixFmtDescriptor *pixdesc = av_pix_fmt_desc_get(avctx->sw_pix_fmt);
int i;
if (!pixdesc)
return -1;
memset(pp, 0, sizeof(*pp));
pp->profile = h->h.profile;
pp->wFormatAndPictureInfoFlags = ((h->h.keyframe == 0) << 0) |
((h->h.invisible == 0) << 1) |
(h->h.errorres << 2) |
(pixdesc->log2_chroma_w << 3) | /* subsampling_x */
(pixdesc->log2_chroma_h << 4) | /* subsampling_y */
(0 << 5) | /* extra_plane */
(h->h.refreshctx << 6) |
(h->h.parallelmode << 7) |
(h->h.intraonly << 8) |
(h->h.framectxid << 9) |
(h->h.resetctx << 11) |
((h->h.keyframe ? 0 : h->h.highprecisionmvs) << 13) |
(0 << 14); /* ReservedFormatInfo2Bits */
pp->width = avctx->width;
pp->height = avctx->height;
pp->BitDepthMinus8Luma = pixdesc->comp[0].depth - 8;
pp->BitDepthMinus8Chroma = pixdesc->comp[1].depth - 8;
/* swap 0/1 to match the reference */
pp->interp_filter = h->h.filtermode ^ (h->h.filtermode <= 1);
pp->Reserved8Bits = 0;
for (i = 0; i < 8; i++) {
if (h->refs[i].f) {
fill_picture_entry(&pp->ref_frame_map[i], ff_dxva2_get_surface_index(avctx, ctx, h->refs[i].f, 0), 0);
pp->ref_frame_coded_width[i] = h->refs[i].f->width;
pp->ref_frame_coded_height[i] = h->refs[i].f->height;
} else
pp->ref_frame_map[i].bPicEntry = 0xFF;
}
for (i = 0; i < 3; i++) {
uint8_t refidx = h->h.refidx[i];
if (h->refs[refidx].f)
fill_picture_entry(&pp->frame_refs[i], ff_dxva2_get_surface_index(avctx, ctx, h->refs[refidx].f, 0), 0);
else
pp->frame_refs[i].bPicEntry = 0xFF;
pp->ref_frame_sign_bias[i + 1] = h->h.signbias[i];
}
fill_picture_entry(&pp->CurrPic, ff_dxva2_get_surface_index(avctx, ctx, h->frames[CUR_FRAME].tf.f, 1), 0);
pp->filter_level = h->h.filter.level;
pp->sharpness_level = h->h.filter.sharpness;
pp->wControlInfoFlags = (h->h.lf_delta.enabled << 0) |
(h->h.lf_delta.updated << 1) |
(h->h.use_last_frame_mvs << 2) |
(0 << 3); /* ReservedControlInfo5Bits */
for (i = 0; i < 4; i++)
pp->ref_deltas[i] = h->h.lf_delta.ref[i];
for (i = 0; i < 2; i++)
pp->mode_deltas[i] = h->h.lf_delta.mode[i];
pp->base_qindex = h->h.yac_qi;
pp->y_dc_delta_q = h->h.ydc_qdelta;
pp->uv_dc_delta_q = h->h.uvdc_qdelta;
pp->uv_ac_delta_q = h->h.uvac_qdelta;
/* segmentation data */
pp->stVP9Segments.wSegmentInfoFlags = (h->h.segmentation.enabled << 0) |
(h->h.segmentation.update_map << 1) |
(h->h.segmentation.temporal << 2) |
(h->h.segmentation.absolute_vals << 3) |
(0 << 4); /* ReservedSegmentFlags4Bits */
for (i = 0; i < 7; i++)
pp->stVP9Segments.tree_probs[i] = h->h.segmentation.prob[i];
if (h->h.segmentation.temporal)
for (i = 0; i < 3; i++)
pp->stVP9Segments.pred_probs[i] = h->h.segmentation.pred_prob[i];
else
memset(pp->stVP9Segments.pred_probs, 255, sizeof(pp->stVP9Segments.pred_probs));
for (i = 0; i < 8; i++) {
pp->stVP9Segments.feature_mask[i] = (h->h.segmentation.feat[i].q_enabled << 0) |
(h->h.segmentation.feat[i].lf_enabled << 1) |
(h->h.segmentation.feat[i].ref_enabled << 2) |
(h->h.segmentation.feat[i].skip_enabled << 3);
pp->stVP9Segments.feature_data[i][0] = h->h.segmentation.feat[i].q_val;
pp->stVP9Segments.feature_data[i][1] = h->h.segmentation.feat[i].lf_val;
pp->stVP9Segments.feature_data[i][2] = h->h.segmentation.feat[i].ref_val;
pp->stVP9Segments.feature_data[i][3] = 0; /* no data for skip */
}
pp->log2_tile_cols = h->h.tiling.log2_tile_cols;
pp->log2_tile_rows = h->h.tiling.log2_tile_rows;
pp->uncompressed_header_size_byte_aligned = h->h.uncompressed_header_size;
pp->first_partition_size = h->h.compressed_header_size;
pp->StatusReportFeedbackNumber = 1 + DXVA_CONTEXT_REPORT_ID(avctx, ctx)++;
return 0;
}
static void fill_slice_short(DXVA_Slice_VPx_Short *slice,
unsigned position, unsigned size)
{
memset(slice, 0, sizeof(*slice));
slice->BSNALunitDataLocation = position;
slice->SliceBytesInBuffer = size;
slice->wBadSliceChopping = 0;
}
static int commit_bitstream_and_slice_buffer(AVCodecContext *avctx,
DECODER_BUFFER_DESC *bs,
DECODER_BUFFER_DESC *sc)
{
const VP9SharedContext *h = avctx->priv_data;
AVDXVAContext *ctx = DXVA_CONTEXT(avctx);
struct vp9_dxva2_picture_context *ctx_pic = h->frames[CUR_FRAME].hwaccel_picture_private;
void *dxva_data_ptr;
uint8_t *dxva_data;
unsigned dxva_size;
unsigned padding;
unsigned type;
#if CONFIG_D3D11VA
if (ff_dxva2_is_d3d11(avctx)) {
type = D3D11_VIDEO_DECODER_BUFFER_BITSTREAM;
if (FAILED(ID3D11VideoContext_GetDecoderBuffer(D3D11VA_CONTEXT(ctx)->video_context,
D3D11VA_CONTEXT(ctx)->decoder,
type,
&dxva_size, &dxva_data_ptr)))
return -1;
}
#endif
#if CONFIG_DXVA2
if (avctx->pix_fmt == AV_PIX_FMT_DXVA2_VLD) {
type = DXVA2_BitStreamDateBufferType;
if (FAILED(IDirectXVideoDecoder_GetBuffer(DXVA2_CONTEXT(ctx)->decoder,
type,
&dxva_data_ptr, &dxva_size)))
return -1;
}
#endif
dxva_data = dxva_data_ptr;
if (ctx_pic->slice.SliceBytesInBuffer > dxva_size) {
av_log(avctx, AV_LOG_ERROR, "Failed to build bitstream");
return -1;
}
memcpy(dxva_data, ctx_pic->bitstream, ctx_pic->slice.SliceBytesInBuffer);
padding = FFMIN(128 - ((ctx_pic->slice.SliceBytesInBuffer) & 127), dxva_size - ctx_pic->slice.SliceBytesInBuffer);
if (padding > 0) {
memset(dxva_data + ctx_pic->slice.SliceBytesInBuffer, 0, padding);
ctx_pic->slice.SliceBytesInBuffer += padding;
}
#if CONFIG_D3D11VA
if (ff_dxva2_is_d3d11(avctx))
if (FAILED(ID3D11VideoContext_ReleaseDecoderBuffer(D3D11VA_CONTEXT(ctx)->video_context, D3D11VA_CONTEXT(ctx)->decoder, type)))
return -1;
#endif
#if CONFIG_DXVA2
if (avctx->pix_fmt == AV_PIX_FMT_DXVA2_VLD)
if (FAILED(IDirectXVideoDecoder_ReleaseBuffer(DXVA2_CONTEXT(ctx)->decoder, type)))
return -1;
#endif
#if CONFIG_D3D11VA
if (ff_dxva2_is_d3d11(avctx)) {
D3D11_VIDEO_DECODER_BUFFER_DESC *dsc11 = bs;
memset(dsc11, 0, sizeof(*dsc11));
dsc11->BufferType = type;
dsc11->DataSize = ctx_pic->slice.SliceBytesInBuffer;
dsc11->NumMBsInBuffer = 0;
type = D3D11_VIDEO_DECODER_BUFFER_SLICE_CONTROL;
}
#endif
#if CONFIG_DXVA2
if (avctx->pix_fmt == AV_PIX_FMT_DXVA2_VLD) {
DXVA2_DecodeBufferDesc *dsc2 = bs;
memset(dsc2, 0, sizeof(*dsc2));
dsc2->CompressedBufferType = type;
dsc2->DataSize = ctx_pic->slice.SliceBytesInBuffer;
dsc2->NumMBsInBuffer = 0;
type = DXVA2_SliceControlBufferType;
}
#endif
return ff_dxva2_commit_buffer(avctx, ctx, sc,
type,
&ctx_pic->slice, sizeof(ctx_pic->slice), 0);
}
static int dxva2_vp9_start_frame(AVCodecContext *avctx,
av_unused const uint8_t *buffer,
av_unused uint32_t size)
{
const VP9SharedContext *h = avctx->priv_data;
AVDXVAContext *ctx = DXVA_CONTEXT(avctx);
struct vp9_dxva2_picture_context *ctx_pic = h->frames[CUR_FRAME].hwaccel_picture_private;
if (!DXVA_CONTEXT_VALID(avctx, ctx))
return -1;
av_assert0(ctx_pic);
/* Fill up DXVA_PicParams_VP9 */
if (ff_dxva2_vp9_fill_picture_parameters(avctx, ctx, &ctx_pic->pp) < 0)
return -1;
ctx_pic->bitstream_size = 0;
ctx_pic->bitstream = NULL;
return 0;
}
static int dxva2_vp9_decode_slice(AVCodecContext *avctx,
const uint8_t *buffer,
uint32_t size)
{
const VP9SharedContext *h = avctx->priv_data;
struct vp9_dxva2_picture_context *ctx_pic = h->frames[CUR_FRAME].hwaccel_picture_private;
unsigned position;
if (!ctx_pic->bitstream)
ctx_pic->bitstream = buffer;
ctx_pic->bitstream_size += size;
position = buffer - ctx_pic->bitstream;
fill_slice_short(&ctx_pic->slice, position, size);
return 0;
}
static int dxva2_vp9_end_frame(AVCodecContext *avctx)
{
VP9SharedContext *h = avctx->priv_data;
struct vp9_dxva2_picture_context *ctx_pic = h->frames[CUR_FRAME].hwaccel_picture_private;
int ret;
if (ctx_pic->bitstream_size <= 0)
return -1;
ret = ff_dxva2_common_end_frame(avctx, h->frames[CUR_FRAME].tf.f,
&ctx_pic->pp, sizeof(ctx_pic->pp),
NULL, 0,
commit_bitstream_and_slice_buffer);
return ret;
}
#if CONFIG_VP9_DXVA2_HWACCEL
const FFHWAccel ff_vp9_dxva2_hwaccel = {
.p.name = "vp9_dxva2",
.p.type = AVMEDIA_TYPE_VIDEO,
.p.id = AV_CODEC_ID_VP9,
.p.pix_fmt = AV_PIX_FMT_DXVA2_VLD,
.init = ff_dxva2_decode_init,
.uninit = ff_dxva2_decode_uninit,
.start_frame = dxva2_vp9_start_frame,
.decode_slice = dxva2_vp9_decode_slice,
.end_frame = dxva2_vp9_end_frame,
.frame_params = ff_dxva2_common_frame_params,
.frame_priv_data_size = sizeof(struct vp9_dxva2_picture_context),
.priv_data_size = sizeof(FFDXVASharedContext),
};
#endif
#if CONFIG_VP9_D3D11VA_HWACCEL
const FFHWAccel ff_vp9_d3d11va_hwaccel = {
.p.name = "vp9_d3d11va",
.p.type = AVMEDIA_TYPE_VIDEO,
.p.id = AV_CODEC_ID_VP9,
.p.pix_fmt = AV_PIX_FMT_D3D11VA_VLD,
.init = ff_dxva2_decode_init,
.uninit = ff_dxva2_decode_uninit,
.start_frame = dxva2_vp9_start_frame,
.decode_slice = dxva2_vp9_decode_slice,
.end_frame = dxva2_vp9_end_frame,
.frame_params = ff_dxva2_common_frame_params,
.frame_priv_data_size = sizeof(struct vp9_dxva2_picture_context),
.priv_data_size = sizeof(FFDXVASharedContext),
};
#endif
#if CONFIG_VP9_D3D11VA2_HWACCEL
const FFHWAccel ff_vp9_d3d11va2_hwaccel = {
.p.name = "vp9_d3d11va2",
.p.type = AVMEDIA_TYPE_VIDEO,
.p.id = AV_CODEC_ID_VP9,
.p.pix_fmt = AV_PIX_FMT_D3D11,
.init = ff_dxva2_decode_init,
.uninit = ff_dxva2_decode_uninit,
.start_frame = dxva2_vp9_start_frame,
.decode_slice = dxva2_vp9_decode_slice,
.end_frame = dxva2_vp9_end_frame,
.frame_params = ff_dxva2_common_frame_params,
.frame_priv_data_size = sizeof(struct vp9_dxva2_picture_context),
.priv_data_size = sizeof(FFDXVASharedContext),
};
#endif