1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
media / gpu / v4l2 / legacy / v4l2_video_decoder_backend_stateful.cc [blame]
// Copyright 2020 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "media/gpu/v4l2/legacy/v4l2_video_decoder_backend_stateful.h"
#include <cstddef>
#include <memory>
#include <optional>
#include <tuple>
#include <utility>
#include "base/containers/contains.h"
#include "base/functional/bind.h"
#include "base/logging.h"
#include "base/metrics/histogram_macros.h"
#include "base/sequence_checker.h"
#include "base/task/sequenced_task_runner.h"
#include "media/base/limits.h"
#include "media/base/platform_features.h"
#include "media/base/video_codecs.h"
#include "media/gpu/chromeos/dmabuf_video_frame_pool.h"
#include "media/gpu/chromeos/platform_video_frame_utils.h"
#include "media/gpu/chromeos/video_frame_resource.h"
#include "media/gpu/macros.h"
#include "media/gpu/v4l2/v4l2_device.h"
#include "media/gpu/v4l2/v4l2_vda_helpers.h"
#include "media/gpu/v4l2/v4l2_video_decoder_backend.h"
#include "media/gpu/v4l2/v4l2_vp9_helpers.h"
namespace media {
namespace {
bool IsVp9KSVCStream(VideoCodecProfile profile,
const DecoderBuffer& decoder_buffer) {
return VideoCodecProfileToVideoCodec(profile) == VideoCodec::kVP9 &&
decoder_buffer.side_data() &&
!decoder_buffer.side_data()->spatial_layers.empty();
}
bool IsVp9KSVCSupportedDriver(const std::string& driver_name) {
const std::string kVP9KSVCSupportedDrivers[] = {"qcom-venus"};
return base::Contains(kVP9KSVCSupportedDrivers, driver_name);
}
std::optional<uint8_t> V4L2PixelFormatToBitDepth(uint32_t v4l2_pixelformat) {
const auto fourcc = Fourcc::FromV4L2PixFmt(v4l2_pixelformat);
if (fourcc) {
return BitDepth(fourcc->ToVideoPixelFormat());
}
return std::nullopt;
}
} // namespace
V4L2StatefulVideoDecoderBackend::DecodeRequest::DecodeRequest(
scoped_refptr<DecoderBuffer> buf,
VideoDecoder::DecodeCB cb)
: buffer(std::move(buf)), decode_cb(std::move(cb)) {}
V4L2StatefulVideoDecoderBackend::DecodeRequest::DecodeRequest(DecodeRequest&&) =
default;
V4L2StatefulVideoDecoderBackend::DecodeRequest&
V4L2StatefulVideoDecoderBackend::DecodeRequest::operator=(DecodeRequest&&) =
default;
V4L2StatefulVideoDecoderBackend::DecodeRequest::~DecodeRequest() = default;
bool V4L2StatefulVideoDecoderBackend::DecodeRequest::IsCompleted() const {
return bytes_used == buffer->size();
}
V4L2StatefulVideoDecoderBackend::V4L2StatefulVideoDecoderBackend(
Client* const client,
scoped_refptr<V4L2Device> device,
VideoCodecProfile profile,
const VideoColorSpace& color_space,
scoped_refptr<base::SequencedTaskRunner> task_runner)
: V4L2VideoDecoderBackend(client, std::move(device)),
driver_name_(device_->GetDriverName()),
profile_(profile),
color_space_(color_space),
task_runner_(task_runner) {
DVLOGF(3);
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
weak_this_ = weak_this_factory_.GetWeakPtr();
}
V4L2StatefulVideoDecoderBackend::~V4L2StatefulVideoDecoderBackend() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
if (flush_cb_ || current_decode_request_ || !decode_request_queue_.empty()) {
VLOGF(1) << "Should not destroy backend during pending decode!";
}
struct v4l2_event_subscription sub;
memset(&sub, 0, sizeof(sub));
sub.type = V4L2_EVENT_SOURCE_CHANGE;
if (device_->Ioctl(VIDIOC_UNSUBSCRIBE_EVENT, &sub) != 0) {
VLOGF(1) << "Cannot unsubscribe to event";
}
}
bool V4L2StatefulVideoDecoderBackend::Initialize() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
if (!IsSupportedProfile(profile_)) {
VLOGF(1) << "Unsupported profile " << GetProfileName(profile_);
return false;
}
frame_splitter_ =
v4l2_vda_helpers::InputBufferFragmentSplitter::CreateFromProfile(
profile_);
if (!frame_splitter_) {
VLOGF(1) << "Failed to create frame splitter";
return false;
}
struct v4l2_event_subscription sub;
memset(&sub, 0, sizeof(sub));
sub.type = V4L2_EVENT_SOURCE_CHANGE;
if (device_->Ioctl(VIDIOC_SUBSCRIBE_EVENT, &sub) != 0) {
VLOGF(1) << "Cannot subscribe to event";
return false;
}
framerate_control_ = std::make_unique<V4L2FrameRateControl>(
base::BindRepeating(&V4L2Device::Ioctl, device_), task_runner_);
return true;
}
void V4L2StatefulVideoDecoderBackend::EnqueueDecodeTask(
scoped_refptr<DecoderBuffer> buffer,
VideoDecoder::DecodeCB decode_cb) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
if (!buffer->end_of_stream()) {
has_pending_requests_ = true;
}
decode_request_queue_.push(
DecodeRequest(std::move(buffer), std::move(decode_cb)));
DoDecodeWork();
}
void V4L2StatefulVideoDecoderBackend::DoDecodeWork() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
// Do not decode if a flush or resolution change is in progress.
if (!client_->IsDecoding())
return;
if (need_resume_resolution_change_) {
need_resume_resolution_change_ = false;
ChangeResolution();
if (!client_->IsDecoding())
return;
}
// Get a new decode request if none is in progress.
if (!current_decode_request_) {
// No more decode request, nothing to do for now.
if (decode_request_queue_.empty())
return;
auto decode_request = std::move(decode_request_queue_.front());
decode_request_queue_.pop();
// Need to flush?
if (decode_request.buffer->end_of_stream()) {
InitiateFlush(std::move(decode_request.decode_cb));
return;
}
// This is our new decode request.
current_decode_request_ = std::move(decode_request);
DCHECK_EQ(current_decode_request_->bytes_used, 0u);
if (IsVp9KSVCStream(profile_, *current_decode_request_->buffer)) {
if (!IsVp9KSVCSupportedDriver(driver_name_)) {
DLOG(ERROR) << driver_name_ << " doesn't support VP9 k-SVC decoding";
client_->OnBackendError();
return;
}
if (!AppendVP9SuperFrameIndex(current_decode_request_->buffer)) {
LOG(ERROR) << "Failed to append superframe index for VP9 k-SVC frame";
client_->OnBackendError();
return;
}
}
}
// Get a V4L2 buffer to copy the encoded data into.
if (!current_input_buffer_) {
current_input_buffer_ = input_queue_->GetFreeBuffer();
// We will be called again once an input buffer becomes available.
if (!current_input_buffer_)
return;
// Record timestamp of the input buffer so it propagates to the decoded
// frames.
// TODO(mcasas): Consider using TimeDeltaToTimeVal().
const struct timespec timespec =
current_decode_request_->buffer->timestamp().ToTimeSpec();
struct timeval timestamp = {
.tv_sec = timespec.tv_sec,
.tv_usec = timespec.tv_nsec / 1000,
};
current_input_buffer_->SetTimeStamp(timestamp);
const int64_t flat_timespec =
base::TimeDelta::FromTimeSpec(timespec).InMilliseconds();
encoding_timestamps_[flat_timespec] = base::TimeTicks::Now();
}
// From here on we have both a decode request and input buffer, so we can
// progress with decoding.
DCHECK(current_decode_request_.has_value());
DCHECK(current_input_buffer_.has_value());
const DecoderBuffer* current_buffer = current_decode_request_->buffer.get();
DCHECK_LT(current_decode_request_->bytes_used, current_buffer->size());
const uint8_t* const data =
current_buffer->data() + current_decode_request_->bytes_used;
const size_t data_size =
current_buffer->size() - current_decode_request_->bytes_used;
size_t bytes_to_copy = 0;
if (!frame_splitter_->AdvanceFrameFragment(data, data_size, &bytes_to_copy)) {
LOG(ERROR) << "Invalid bitstream detected.";
std::move(current_decode_request_->decode_cb)
.Run(DecoderStatus::Codes::kFailed);
current_decode_request_.reset();
current_input_buffer_.reset();
client_->OnBackendError();
return;
}
const size_t bytes_used = current_input_buffer_->GetPlaneBytesUsed(0);
if (bytes_used + bytes_to_copy > current_input_buffer_->GetPlaneSize(0)) {
LOG(ERROR) << "V4L2 buffer size is too small to contain a whole frame.";
std::move(current_decode_request_->decode_cb)
.Run(DecoderStatus::Codes::kFailed);
current_decode_request_.reset();
current_input_buffer_.reset();
client_->OnBackendError();
return;
}
uint8_t* dst =
static_cast<uint8_t*>(current_input_buffer_->GetPlaneMapping(0)) +
bytes_used;
memcpy(dst, data, bytes_to_copy);
current_input_buffer_->SetPlaneBytesUsed(0, bytes_used + bytes_to_copy);
current_decode_request_->bytes_used += bytes_to_copy;
// Release current_input_request_ if we reached its end.
if (current_decode_request_->IsCompleted()) {
std::move(current_decode_request_->decode_cb)
.Run(DecoderStatus::Codes::kOk);
current_decode_request_.reset();
}
// If we have a partial frame, wait before submitting it.
if (frame_splitter_->IsPartialFramePending()) {
VLOGF(4) << "Partial frame pending, not queueing any buffer now.";
return;
}
// The V4L2 input buffer contains a decodable entity, queue it.
if (!std::move(*current_input_buffer_).QueueMMap()) {
LOG(ERROR) << "Error while queuing input buffer!";
client_->OnBackendError();
}
current_input_buffer_.reset();
// If we can still progress on a decode request, do it.
if (current_decode_request_ || !decode_request_queue_.empty())
ScheduleDecodeWork();
}
void V4L2StatefulVideoDecoderBackend::ScheduleDecodeWork() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
task_runner_->PostTask(
FROM_HERE, base::BindOnce(&V4L2StatefulVideoDecoderBackend::DoDecodeWork,
weak_this_));
}
void V4L2StatefulVideoDecoderBackend::ProcessEventQueue() {
while (std::optional<struct v4l2_event> ev = device_->DequeueEvent()) {
if (ev->type == V4L2_EVENT_SOURCE_CHANGE &&
(ev->u.src_change.changes & V4L2_EVENT_SRC_CH_RESOLUTION)) {
ChangeResolution();
}
}
}
void V4L2StatefulVideoDecoderBackend::OnServiceDeviceTask(bool event) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
if (event)
ProcessEventQueue();
// We can enqueue dequeued output buffers immediately.
EnqueueOutputBuffers();
// Try to progress on our work since we may have dequeued input buffers.
DoDecodeWork();
}
void V4L2StatefulVideoDecoderBackend::EnqueueOutputBuffers() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
const v4l2_memory mem_type = output_queue_->GetMemoryType();
while (true) {
bool ret = false;
bool no_buffer = false;
std::optional<V4L2WritableBufferRef> buffer;
switch (mem_type) {
case V4L2_MEMORY_MMAP:
buffer = output_queue_->GetFreeBuffer();
if (!buffer) {
no_buffer = true;
break;
}
ret = std::move(*buffer).QueueMMap();
break;
case V4L2_MEMORY_DMABUF: {
scoped_refptr<FrameResource> frame = GetPoolVideoFrame();
// Running out of frame is not an error, we will be called again
// once frames are available.
if (!frame) {
return;
}
buffer = output_queue_->GetFreeBufferForFrame(frame->tracking_token());
if (!buffer) {
no_buffer = true;
break;
}
framerate_control_->AttachToFrameResource(frame);
ret = std::move(*buffer).QueueDMABuf(std::move(frame));
break;
}
default:
NOTREACHED();
}
// Running out of V4L2 buffers is not an error, so just exit the loop
// gracefully.
if (no_buffer)
break;
if (!ret) {
LOG(ERROR) << "Error while queueing output buffer!";
client_->OnBackendError();
}
}
DVLOGF(3) << output_queue_->QueuedBuffersCount() << "/"
<< output_queue_->AllocatedBuffersCount()
<< " output buffers queued";
}
scoped_refptr<FrameResource>
V4L2StatefulVideoDecoderBackend::GetPoolVideoFrame() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
DmabufVideoFramePool* pool = client_->GetVideoFramePool();
DCHECK_EQ(output_queue_->GetMemoryType(), V4L2_MEMORY_DMABUF);
DCHECK_NE(pool, nullptr);
scoped_refptr<FrameResource> frame = pool->GetFrame();
if (!frame) {
DVLOGF(3) << "No available VideoFrame for now";
// We will try again once a frame becomes available.
pool->NotifyWhenFrameAvailable(base::BindOnce(
base::IgnoreResult(&base::SequencedTaskRunner::PostTask), task_runner_,
FROM_HERE,
base::BindOnce(
base::IgnoreResult(
&V4L2StatefulVideoDecoderBackend::EnqueueOutputBuffers),
weak_this_)));
}
return frame;
}
// static
void V4L2StatefulVideoDecoderBackend::ReuseOutputBufferThunk(
scoped_refptr<base::SequencedTaskRunner> task_runner,
std::optional<base::WeakPtr<V4L2StatefulVideoDecoderBackend>> weak_this,
V4L2ReadableBufferRef buffer) {
DVLOGF(3);
DCHECK(weak_this);
if (task_runner->RunsTasksInCurrentSequence()) {
if (*weak_this)
(*weak_this)->ReuseOutputBuffer(std::move(buffer));
} else {
task_runner->PostTask(
FROM_HERE,
base::BindOnce(&V4L2StatefulVideoDecoderBackend::ReuseOutputBuffer,
*weak_this, std::move(buffer)));
}
}
void V4L2StatefulVideoDecoderBackend::ReuseOutputBuffer(
V4L2ReadableBufferRef buffer) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3) << "Reuse output buffer #" << buffer->BufferId();
// Lose reference to the buffer so it goes back to the free list.
buffer.reset();
// Enqueue the newly available buffer.
EnqueueOutputBuffers();
}
void V4L2StatefulVideoDecoderBackend::OnOutputBufferDequeued(
V4L2ReadableBufferRef buffer) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
// Zero-bytes buffers are returned as part of a flush and can be dismissed.
if (buffer->GetPlaneBytesUsed(0) > 0) {
// TODO(mcasas): Consider using TimeValToTimeDelta().
const struct timeval timeval = buffer->GetTimeStamp();
const struct timespec timespec = {
.tv_sec = timeval.tv_sec,
.tv_nsec = timeval.tv_usec * 1000,
};
const int64_t flat_timespec =
base::TimeDelta::FromTimeSpec(timespec).InMilliseconds();
// TODO(b/190615065) |flat_timespec| might be repeated with H.264
// bitstreams, investigate why, and change the if() to DCHECK().
if (base::Contains(encoding_timestamps_, flat_timespec)) {
UMA_HISTOGRAM_TIMES(
"Media.PlatformVideoDecoding.Decode",
base::TimeTicks::Now() - encoding_timestamps_[flat_timespec]);
encoding_timestamps_.erase(flat_timespec);
}
scoped_refptr<FrameResource> frame;
switch (output_queue_->GetMemoryType()) {
case V4L2_MEMORY_MMAP: {
// Wrap the frame into another one so we can be signaled when the
// consumer is done with it and reuse the V4L2 buffer.
scoped_refptr<FrameResource> origin_frame = buffer->GetFrameResource();
frame = origin_frame->CreateWrappingFrame();
frame->AddDestructionObserver(base::BindOnce(
&V4L2StatefulVideoDecoderBackend::ReuseOutputBufferThunk,
task_runner_, weak_this_, buffer));
break;
}
case V4L2_MEMORY_DMABUF:
// The frame from the frame pool that we passed to QueueDMABuf() has
// been decoded into. It can be output as-is.
frame = buffer->GetFrameResource();
break;
default:
NOTREACHED();
}
const base::TimeDelta timestamp = base::TimeDelta::FromTimeSpec(timespec);
// TODO(b/214190092): Get color space from the buffer.
client_->OutputFrame(std::move(frame), *visible_rect_, color_space_,
timestamp);
}
// We were waiting for the last buffer before a resolution change
// The order here is important! A flush event may come after a resolution
// change event (but not the opposite), so we must make sure both events
// are processed in the correct order.
if (buffer->IsLast()){
// Check that we don't have a resolution change event pending. If we do
// then this LAST buffer was related to it.
ProcessEventQueue();
if (resolution_change_cb_) {
std::move(resolution_change_cb_).Run();
} else if (flush_cb_) {
// We were waiting for a flush to complete, and received the last buffer.
CompleteFlush();
}
}
EnqueueOutputBuffers();
}
bool V4L2StatefulVideoDecoderBackend::InitiateFlush(
VideoDecoder::DecodeCB flush_cb) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
DCHECK(!flush_cb_);
// Submit any pending input buffer at the time of flush.
if (current_input_buffer_) {
if (!std::move(*current_input_buffer_).QueueMMap()) {
LOG(ERROR) << "Error while queuing input buffer!";
client_->OnBackendError();
}
current_input_buffer_.reset();
}
client_->InitiateFlush();
flush_cb_ = std::move(flush_cb);
// The stream could be stopped in the middle of the frame when the flush is
// being triggered. This makes sure there are no leftovers after the flush
// finishes.
frame_splitter_->Reset();
// Special case: if we haven't received any decoding request, we could
// complete the flush immediately.
if (!has_pending_requests_)
return CompleteFlush();
if (output_queue_->IsStreaming()) {
// If the CAPTURE queue is streaming, send the STOP command to the V4L2
// device. The device will let us know that the flush is completed by
// sending us a CAPTURE buffer with the LAST flag set.
return output_queue_->SendStopCommand();
} else {
// If the CAPTURE queue is not streaming, this means we received the flush
// request before the initial resolution has been established. The flush
// request will be processed in OnChangeResolutionDone(), when the CAPTURE
// queue starts streaming.
DVLOGF(2) << "Flush request to be processed after CAPTURE queue starts";
}
return true;
}
bool V4L2StatefulVideoDecoderBackend::CompleteFlush() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
DCHECK(flush_cb_);
// Signal that flush has properly been completed.
std::move(flush_cb_).Run(DecoderStatus::Codes::kOk);
// If CAPTURE queue is streaming, send the START command to the V4L2 device
// to signal that we are resuming decoding with the same state.
if (output_queue_->IsStreaming() && !output_queue_->SendStartCommand()) {
LOG(ERROR) << "Failed to issue START command";
std::move(flush_cb_).Run(DecoderStatus::Codes::kFailed);
client_->OnBackendError();
return false;
}
client_->CompleteFlush();
// Qualcomm venus stops capture queue after LAST buffer is dequeued and needs
// restarting to be ready for resume operation in case it was left in EOS
// state
client_->RestartStream();
// Resume decoding if data is available.
ScheduleDecodeWork();
has_pending_requests_ = false;
return true;
}
void V4L2StatefulVideoDecoderBackend::OnStreamStopped(bool stop_input_queue) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
// If we are resetting, also reset the splitter.
if (frame_splitter_ && stop_input_queue)
frame_splitter_->Reset();
}
void V4L2StatefulVideoDecoderBackend::ChangeResolution() {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
// Here we just query the new resolution, visible rect, and number of output
// buffers before asking the client to update the resolution.
auto format = output_queue_->GetFormat().first;
if (!format) {
LOG(ERROR) << "Unable to get format when changing resolution.";
client_->OnBackendError();
return;
}
const gfx::Size pic_size(format->fmt.pix_mp.width, format->fmt.pix_mp.height);
auto visible_rect = output_queue_->GetVisibleRect();
if (!visible_rect) {
LOG(ERROR) << "Unable to get visible rectangle when changing resolution.";
client_->OnBackendError();
return;
}
if (!gfx::Rect(pic_size).Contains(*visible_rect)) {
LOG(ERROR) << "Visible rectangle (" << visible_rect->ToString()
<< ") is not contained by the picture rectangle ("
<< gfx::Rect(pic_size).ToString() << ").";
client_->OnBackendError();
return;
}
const auto bit_depth =
V4L2PixelFormatToBitDepth(format->fmt.pix_mp.pixelformat);
if (!bit_depth) {
LOG(ERROR) << "Unable to determine bitdepth of format ";
client_->OnBackendError();
return;
}
// Estimate the amount of buffers needed for the CAPTURE queue and for codec
// reference requirements. For VP9 and AV1, the maximum number of reference
// frames is constant and 8 (for VP8 is 4); for H.264 and other ITU-T codecs,
// it depends on the bitstream. Here we query it from the driver anyway.
constexpr size_t kDefaultNumReferenceFrames = 8;
size_t num_codec_reference_frames = kDefaultNumReferenceFrames;
// On QC Venus, this control ranges between 1 and 32 at the time of writing.
auto ctrl = device_->GetCtrl(V4L2_CID_MIN_BUFFERS_FOR_CAPTURE);
if (ctrl) {
VLOGF(2) << "V4L2_CID_MIN_BUFFERS_FOR_CAPTURE = " << ctrl->value;
num_codec_reference_frames = std::max(
base::checked_cast<size_t>(ctrl->value), num_codec_reference_frames);
}
// Verify |num_codec_reference_frames| has a reasonable value. Anecdotally 16
// is the largest amount of reference frames seen, on an ITU-T H.264 test
// vector (CAPCM*1_Sand_E.h264).
CHECK_LE(num_codec_reference_frames, 32u);
// Signal that we are flushing and initiate the resolution change.
// Our flush will be done when we receive a buffer with the LAST flag on the
// CAPTURE queue.
client_->InitiateFlush();
DCHECK(!resolution_change_cb_);
resolution_change_cb_ = base::BindOnce(
&V4L2StatefulVideoDecoderBackend::ContinueChangeResolution, weak_this_,
pic_size, *visible_rect, num_codec_reference_frames, *bit_depth);
// ...that is, unless we are not streaming yet, in which case the resolution
// change can take place immediately.
if (!output_queue_->IsStreaming())
std::move(resolution_change_cb_).Run();
}
void V4L2StatefulVideoDecoderBackend::ContinueChangeResolution(
const gfx::Size& pic_size,
const gfx::Rect& visible_rect,
const size_t num_codec_reference_frames,
const uint8_t bit_depth) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
// Flush is done, but stay in flushing state and ask our client to set the new
// resolution.
client_->ChangeResolution(pic_size, visible_rect, num_codec_reference_frames,
bit_depth);
}
bool V4L2StatefulVideoDecoderBackend::ApplyResolution(
const gfx::Size& pic_size,
const gfx::Rect& visible_rect) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
// Use the visible rect for all new frames.
visible_rect_ = visible_rect;
return true;
}
void V4L2StatefulVideoDecoderBackend::OnChangeResolutionDone(CroStatus status) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3) << "status=" << static_cast<int>(status.code());
if (status == CroStatus::Codes::kResetRequired) {
need_resume_resolution_change_ = true;
return;
}
if (status != CroStatus::Codes::kOk) {
LOG(ERROR) << "Backend failure when changing resolution ("
<< static_cast<int>(status.code()) << ").";
client_->OnBackendError();
return;
}
// Flush can be considered completed on the client side.
client_->CompleteFlush();
// Enqueue all available output buffers now that they are allocated.
EnqueueOutputBuffers();
// If we had a flush request pending before the initial resolution change,
// process it now.
if (flush_cb_) {
DVLOGF(2) << "Processing pending flush request...";
client_->InitiateFlush();
if (!output_queue_->SendStopCommand()) {
return;
}
}
// Also try to progress on our work.
DoDecodeWork();
}
void V4L2StatefulVideoDecoderBackend::ClearPendingRequests(
DecoderStatus status) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DVLOGF(3);
if (resolution_change_cb_)
std::move(resolution_change_cb_).Run();
if (flush_cb_) {
std::move(flush_cb_).Run(status);
}
current_input_buffer_.reset();
if (current_decode_request_) {
std::move(current_decode_request_->decode_cb).Run(status);
current_decode_request_.reset();
}
while (!decode_request_queue_.empty()) {
std::move(decode_request_queue_.front().decode_cb).Run(status);
decode_request_queue_.pop();
}
has_pending_requests_ = false;
}
// TODO(b:149663704) move into helper function shared between both backends?
bool V4L2StatefulVideoDecoderBackend::IsSupportedProfile(
VideoCodecProfile profile) {
DCHECK_CALLED_ON_VALID_SEQUENCE(sequence_checker_);
DCHECK(device_);
if (supported_profiles_.empty()) {
const std::vector<uint32_t> kSupportedInputFourccs = {
V4L2_PIX_FMT_H264,
V4L2_PIX_FMT_VP8,
V4L2_PIX_FMT_VP9,
};
auto device = base::MakeRefCounted<V4L2Device>();
VideoDecodeAccelerator::SupportedProfiles profiles =
device->GetSupportedDecodeProfiles(kSupportedInputFourccs);
for (const auto& entry : profiles)
supported_profiles_.push_back(entry.profile);
}
return base::Contains(supported_profiles_, profile);
}
bool V4L2StatefulVideoDecoderBackend::StopInputQueueOnResChange() const {
return false;
}
size_t V4L2StatefulVideoDecoderBackend::GetNumOUTPUTQueueBuffers(
bool secure_mode) const {
CHECK(!secure_mode);
constexpr size_t kNumInputBuffers = 8;
return kNumInputBuffers;
}
} // namespace media