Since the actual max length of the jitter buffer is restricted by
max_delay, this shouldn't harm the overall latency (assuming that
max_delay is set properly), while allowing packet reordering with
a larger number of packets (which may be required with high bitrate
video).
Signed-off-by: Martin Storsjö <martin@martin.st>
The previous restriction was partially designed to fix certain
(broken) samples from bug 215. There should be no restriction on the
number of keyframes per fragment or trun.
The spec suggests that all frames lacking MOV_FRAG_SAMPLE_FLAG_IS_NON_SYNC
are key frames, but we require the flag MOV_FRAG_SAMPLE_FLAG_DEPENDS_YES
to be unset as well. This works for (possibly broken) media that never
sets the NON_SYNC flag and should also be correct for any spec-compliant
file.
For files that never set either of the flags, all samples are marked
as keyframes.
Signed-off-by: Martin Storsjö <martin@martin.st>
Previously most of the error paths leaked.
Also add FF_CODEC_CAP_INIT_THREADSAFE while adding caps_internal;
this decoder wrapper doesn't have any static data that is initialized.
Signed-off-by: Martin Storsjö <martin@martin.st>
For ADTS streams, the output format (number of channels, frame size)
can change at any point (with the latest version of fdk-aac, the decoder
seems to change format after a handful of frames, not outputting the
right format immediately, for cases that worked fine with the earlier
version of the lib).
Previously, the decoder decoded straight into the output frame once the
number of channels and frame size was known. This obviously does not
work if the number of channels or frame size changes.
The alternative would be to allocate the AVFrame with the maximum number
of channels and frame size, and change them afterward decoding into it,
but that may cause confusion to users e.g. of the get_buffer callback.
This solution should be more robust.
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
In the latest version of fdk-aac, the decoder can output up to 8
channels; take this into account when preallocating buffers that
need to fit the output from any packet.
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
Don't try to detect the float ABI by checking at the toolchain
name or by trying to assemble and link files with eabi_attributes.
This fixes the float ABI detection when building using clang
with -fembed-bitcode, where the current eabi_attributes check
accidentally passes.
This issue was pointed out by James Howe <james.howe@hp.com>.
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
The double meaning of the faststart flag (moving a moov atom
to the start of files, making them streamable, for non-fragmented
files, vs inserting a global sidx index at the start of files
for fragmented files) is confusing - see 40ed1cbf1 for
explanation of its origins.
Since the second meaning of the flag hasn't been part of any
libav release yet, just rename it to get rid of the confusion
without any extra deprecation (which wouldn't get rid of the
potential confusion, of users adding -movflags faststart
even for fragmented files, where it isn't needed for making
them "streamable").
This gets back the old behaviour, where -movflags faststart
doesn't have any effect for fragmented files.
Signed-off-by: Martin Storsjö <martin@martin.st>
For fragmented files with non-empty moov, with a fragment index
(sidx), place the index after the initial moov/mdat pair.
Previously, for this pathological case, the index was written
at the start of the file.
Signed-off-by: Martin Storsjö <martin@martin.st>
The same field is also used for writing the sidx index header,
for fragmented files, when the faststart flag is used.
Signed-off-by: Martin Storsjö <martin@martin.st>
This fixes crashes with pathological cases when trying to write
a sidx index (via the -movflags faststart option, in combination
with fragmenting options), when no fragments actually have been
written. (This is possible if the empty_moov flag isn't used,
so that all actual packet data is written in the moov/mdat pair,
and no moof/mdat pairs have been written.)
In these pathological cases, no sidx should be written at all.
Signed-off-by: Martin Storsjö <martin@martin.st>
Allow $ as character anywhere within normal RTSP replies - both
within the lines, and as the first character of RTSP header lines.
(The existing old comment indicated that an inline packet could
start at any line within a RTSP reply header, but that doesn't
sound valid to me, and I'm not sure if the existing code
handled that correctly either.)
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
configure does check for isatty, and checkasm properly checks
HAVE_ISATTY, but on some platforms (e.g. WinRT), io.h needs to be
included for isatty to be available.
Signed-off-by: Martin Storsjö <martin@martin.st>
This avoids having to manually set _WIN32_WINNT in --extra-cflags
when targeting these API families, which only was necessary to
work around configure setting _WIN32_WINNT to an older version
by default.
Signed-off-by: Martin Storsjö <martin@martin.st>
If _WIN32_WINNT is unset, we force it to a new enough value to
make sure the necessary definitions are visible.
When targeting Windows Phone or Windows RT, _WIN32_WINNT should
be at least 0x0602 - otherwise the windows headers themselves
can cause errors (which technically are bugs in the headers).
Raising this value here shouldn't hurt; the alternative would
be to not touch it at all if WINAPI_FAMILY is set to phone/app,
or to force setting it to 0x0602 in configure if unset (for phone/app).
Signed-off-by: Martin Storsjö <martin@martin.st>
This macro identifies whether VFPv3 is available; MSVC defaults
to hardfloat (except for older MSVC versions for CE, targeting
ARMv4).
Signed-off-by: Martin Storsjö <martin@martin.st>
This is only necessary on MSVC 2010/2012 (and possibly on some
ICL versions). This both avoids an extra hack on newer MSVC
versions, and better documents what the extra compiler option
is used for.
Signed-off-by: Martin Storsjö <martin@martin.st>
Most of the fate-dds-* and fate-txd-* tests already
output into the same pixel format regardless of
platform endianness, so there's no need to force
conversion to another format.
This fixes the tests fate-txd-16bpp, fate-txd-odd,
fate-dds-rgb16, fate-dds-rgb24 and fate-dds-xrgb on
big endian, where the tests seem to fail due to issues
with certain conversion codepaths in swscale.
Those conversion codepaths should of course be fixed, but
the individual decoder tests should use as little extra
conversion steps as possible.
Signed-off-by: Martin Storsjö <martin@martin.st>
These are only necessary once/if avconv gets support for this hwaccel.
While that obviously is desireable, we don't have it yet, and they
currently only are a distraction.
Signed-off-by: Martin Storsjö <martin@martin.st>
The only need to be listed if they are to be used in ifdefs from
within the code - config items used as dependencies only within
configure don't need to be exported.
Signed-off-by: Martin Storsjö <martin@martin.st>
This fixes dxva2 detection (i.e. correctly realizes that it isn't
available) for WinRT, where dxva2api.h does exist, but these definitions
are omitted (when targeting the API subsets).
Ideally we should rather check for e.g. DXVA2_ConfigPictureDecode,
but configure might fail to find that definition due to _WIN32_WINNT
not being set to the right value during configure. (libavcodec/dxva2.h
manually overrides the _WIN32_WINNT define.)
This allows removing hardcoded --disable-dxva2 from such build
configurations.
Signed-off-by: Martin Storsjö <martin@martin.st>
mpz_import and mpz_export were added in GMP 4.1, in 2002.
This simplifies the DH code by clarifying that it only uses pure
bignum functions, no other parts of nettle/hogweed.
Signed-off-by: Martin Storsjö <martin@martin.st>
There was a misunderstanding betewen bits and bytes for the parameter
value for generating random big numbers.
Signed-off-by: Martin Storsjö <martin@martin.st>
Since the underlying URLContext read functions are used,
they handle interruption, without having to handle it at
this level.
Signed-off-by: Martin Storsjö <martin@martin.st>
This avoids hijacking the fd, by reading using the normal
URLContext functions instead. This allowing reading data that has
been buffered in the underlying URLContext.
This avoids using the libraries own send functions that can
cause SIGPIPE.
The fd is still used for polling the lowlevel socket, for
waiting for retries.
Signed-off-by: Martin Storsjö <martin@martin.st>
This makes sure that the time + duration of the first segment
matches the start time of the next segment for e.g. AAC audio
with encoder delay.
Signed-off-by: Martin Storsjö <martin@martin.st>
$cc is the compiler requested as main target compiler, while $_cc
is the actual tool tested in the probe function right now (which
can also be e.g. the host compiler).
Signed-off-by: Martin Storsjö <martin@martin.st>
This is printed in a separate subgroup "displaymatrix" inside a
new group named "sidedata". The subgroup has got two values,
"rotation" (which is the parsed rotation from the matrix) and
"matrix" containing the full actual values.
Signed-off-by: Martin Storsjö <martin@martin.st>
The option is enabled by default, but can be disabled.
If this is enabled, such side data isn't copied into the output stream
(except when doing stream copy).
Signed-off-by: Martin Storsjö <martin@martin.st>
This is enabled by default, but can be disbled via the -noautorotate
option.
Based on a patch by Clément Bœsch.
Signed-off-by: Martin Storsjö <martin@martin.st>
This removes the error logging added in 4e54432164.
This avoids warnings about "Invalid interval start specification 'now'"
for live rtsp streams.
We only try to parse some of the many valid values for time ranges
in RTSP - the other ones are fully valid but not interesting for the
use case in rtsp.c, so we shouldn't warn about them.
(Parsing the time ranges is needed to allow seeking, but e.g. setting
the current realtime clock for the start time doesn't make sense.
av_parse_time has got a different mode for parsing absolute times
as well, which can handle the special case "now", but that doesn't
make much sense for this particular use in rtsp.c.)
Signed-off-by: Martin Storsjö <martin@martin.st>
Make sure we don't buffer up more than max_delay worth of data
before writing a PES packet, even if pes_payload_size is set to
a larger value.
Signed-off-by: Martin Storsjö <martin@martin.st>
These are essential allowing QuickTime to keep detecting content
as slow-motion - this allows preserving them on stream copy.
Signed-off-by: Martin Storsjö <martin@martin.st>
Since the duration is compared to the tfra durations/intervals which
are expressed in pts, calculate that here as well.
Signed-off-by: Martin Storsjö <martin@martin.st>
For strict CFR, they should be pretty much equal, but if the stream
is VFR, there can be a sometimes significant difference.
Calculate the pts duration separately, used in sidx atoms and for
tfrf/tfxd boxes in smooth streaming ismv files.
Also make sure to reduce the duration of sidx entries according to
edit lists.
Signed-off-by: Martin Storsjö <martin@martin.st>
Adjusting it is only necessary when a sidx/tfrf/tfxd atom already has
been written for the previous fragment (since the sidx/tfrf/tfxd atoms
include the duration between the first pts of the previous fragment, to
the first pts of the new fragment).
Signed-off-by: Martin Storsjö <martin@martin.st>
When automatically flushing fragments based on set conditions
(fragmentation on keyframes, after some interval or byte size),
we already have the next packet for one stream - use this for setting
the duration of the last packet in the flushed fragment correctly.
This avoids having to adjust the timestamp of the first packet in
the new fragment since the last duration was unknown.
Unfortunately, this only works for automatic flushing (not for
caller-triggered flushing, like in the dash muxer), and only for the
one single track that triggered the flushing. The duration of the
last sample in all other tracks still is dependent on AVPacket
duration (or heuristics).
Signed-off-by: Martin Storsjö <martin@martin.st>
This avoids that the mp4 muxer does a similar heuristic, adjusting
the timestamps in a way that the dash muxer doesn't know the actual
timestamps written to the file in the end. By making sure that the
mp4 muxer internal heuristic isn't applied, we know the exact
timestamps written to file, so that the timestamps in manifest match
the files.
Signed-off-by: Martin Storsjö <martin@martin.st>
Even if this is a guess, it is way better than writing a zero duration
of the last sample in a fragment (because if the duration is zero,
the first sample of the next fragment will have the same timestamp
as the last sample in the previous one).
Since we normally don't require libavformat muxer users to set
the duration field in AVPacket, we probably can't strictly require
it here either, so don't log this as a strict warning, only as info.
Signed-off-by: Martin Storsjö <martin@martin.st>
Add a missing AVClass member, check whether localaddr is null.
(Previously, localaddr was always a local stack buffer, while it
now also can be an avoption string which can be null.)
This fixes crashes when not passing any localaddr parameter, since
66028b7ba.
Signed-off-by: Martin Storsjö <martin@martin.st>
This is incompatible with the omit_tfhd_offset flag (writing
position independent fragments with interleaving requires the
default_base_moof flag).
This makes the moof atoms slightly bigger, but can be better for
playback (improving locality of sample data in the mdat).
Signed-off-by: Martin Storsjö <martin@martin.st>
This is needed if all the data for one track isn't continuous
within the mdat. Normally we make sure all the data for one
track is continuous, but in new cases we will need to have
the samples interleaved.
Signed-off-by: Martin Storsjö <martin@martin.st>
This way, the caller doesn't need to coordinate setting the option
after the moov atom has been written. The downside is that it is
no longer possible to use the option for checking whether the moov
atom already has been written, but a caller is able to keep track
of that by other means anyway.
Signed-off-by: Martin Storsjö <martin@martin.st>
The previous use of the mov->fragments field, for determining whether
written packets were part of the first fragment or not, didn't
work as intended when using the empty_moov flag.
Signed-off-by: Martin Storsjö <martin@martin.st>
By making sure we at each time only have one pointer set, either a
local variable or one in the context, we avoid potential double frees
in the cleanup routines. If chain->rtp_ctx is set, it is closed by
calling avformat_write_trailer, but that shouldn't be called unless
avformat_write_header succeeded.
This issue was pointed out by Andreas Cadhalpun.
Signed-off-by: Martin Storsjö <martin@martin.st>
This avoids assuming that e.g. audio samples are marked as
sync samples.
This allows omitting the sample flags from trun, if the default
flags happen to be right for all the samples.
Signed-off-by: Martin Storsjö <martin@martin.st>
When all the codepaths using manually set .arch/.fpu code is
behind runtime detection, the elf attributes should be suppressed.
This allows tools to know that the final built binary doesn't
strictly require these extensions.
Signed-off-by: Martin Storsjö <martin@martin.st>
a876585215 had the unintended side effect of returning AVERROR(ENOMEM)
when track->entry is zero, while the code intentionally wants to
continue in that case.
Signed-off-by: Martin Storsjö <martin@martin.st>
The original flags variable contains rtpdec flags, while the
rmflags variable contains RM flag bits which have a completely
different definition.
Signed-off-by: Martin Storsjö <martin@martin.st>
The only case where RTP_FLAG_KEY actually is needed is
in RDT, where such a flag needs to be passed via the
rtpdec parse function's flags parameter.
Signed-off-by: Martin Storsjö <martin@martin.st>
Nothing in the framework nor in the rest of the depacketizer actually
uses this flag - the chained demuxer sets the keyframe flag properly on
demuxed packets already.
Signed-off-by: Martin Storsjö <martin@martin.st>
This fixes an oversight in 96084251, in a refactoring done on top
of Gilles' original patch.
Pointed out by Gilles Chanteperdrix.
Signed-off-by: Martin Storsjö <martin@martin.st>
This reverts commit 4abfa387b8.
This commit broke playback of fragmented mp4 files with b-frames.
While investigating this, it turned out that the general framework
isn't ready for a PTS-based index yet. Revert this change until
a better thought out solution is in place.
Signed-off-by: Martin Storsjö <martin@martin.st>
Instead check the timestamps while muxing, to avoid buffering a
too long timestamp range into one single packet.
This makes the AMR and AAC packetization slightly less efficient,
since we set a possibly unnecessarily high max_frames_per_packet.
(These packetizers end up doing a memmove of the TOC bytes if
sending a packet before max_frames_per_packet is achieved, and
we end up setting max_frames_per_packet to a value that should
be high enough for most uses.)
All packetizers that use max_frames_per_packet now set it either
to a default value, or to a value calculated based on other
parameters, so none of them rely on the previous default setting.
For iLBC, copy one frame at a time, to allow checking the timestamp
range for each of them - basically doing potentially multiple
loops to simplify the code instead of trying to calculate the
number of frames to buffer while honoring s1->max_delay.
This is in preparation for reducing the coupling between libavformat
and libavcodec, by not having the muxers use the encoder field
frame_size (which may not be available during e.g. stream copy).
Signed-off-by: Martin Storsjö <martin@martin.st>
Factorize out the s->num_frames check at the start of the if statements,
simplifying adding more alternative causes for sending the buffered
frames.
Signed-off-by: Martin Storsjö <martin@martin.st>
This avoids having to jump to the defaultcase in the switch. Manually
override the stream time base back to 90 kHz for the few audio codecs
that don't use the sample rate as time base (mp2, mp3).
Signed-off-by: Martin Storsjö <martin@martin.st>
This doesn't fix any bug, but makes the code simpler for later
patches, and more straightforward to read as is.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö