Sometimes, HTTP sources require a lot of seeking during probing / header
parsing (especially for formats like MXF). Currently, we need to completely
tear down and re-establish the connection most times this happens, which puts
a lot of stress on the network stack and also results in transmission of
possibly many unnecessary bytes.
This patch adds an option to allow FFmpeg to request partial ranges during
the initialization stage. This is done until the initial request size is fully
read, after which we fall back to the normal behavior (i.e. infinite streaming
via an unbounded request).
The usefulness of this is limited without also specifying -multiple_requests 1,
since otherwise there is little point to requesting partial ranges to begin
with. (However, it is semantically independent, so we keep it that way.)
When the previous reply was a partial response (e.g. due to a seek to the
end of the file), and the remaining data from that partial response is
below the short seek size threshold, we can serve this seek by just draining
that data and re-using the existing connection.
This can currently only happen when using keep-alive connections
(-multiple_requests 1) and seeking from the end of the file to somewhere
else, in which case the file's tail can be drained and the connection re-used.
Under other circumstances, however, we still need to force a reconnection,
because we do not yet send partial range requests. (This will be changed in the
following commit)
We need to take special care not to regress the existing fallback logic
for when `http_open_cnx` fails, so here is a quick case analysis:
non-drain path:
- identical to the current
soft drain fails: (ffurl_read error path)
- s->hd = old_hd = NULL
- http_open_cnx() always opens a new connection
- on failure, old buffer is restored and s->hd remains NULL
soft drain succeeds, http_open_cnx() fails:
- s->hd is set to NULL by http_open_cnx() failure path
- old_hd was never set, so remains NULL
- old buffer is restored, s->hd remains NULL
In either case, the outcome that any (previously valid) buffer is left as-is,
the offset is unchanged, and the connection ends up closed (s->hd == NULL).
This is okay to do after the previous change to http_buf_read, which allows
it to internally re-open the connection if needed.
If the Content-Range indicates a smaller range than what we expected,
we should send a new request for the remainder before attempting to read
more.
Again, this commit is theoretically non-functional on its own, since any
conforming HTTP server should give us the entire range we asked for in the
first place, but it is semantically independent from and prepares us for the
following changes.
This could conceivably happen currently if the user tries reading more
bytes after the last chunk has already been received. In this case,
we currently segfault - but simply returning AVERROR(EIO) seems more
reasonable and lets the higher end retry the connection in this case.
In the event that the range returned is smaller than the true filesize, we
should only expect to receive that many bytes - not the entire rest of the
file.
This commit is theoretically non-functional on its own, since any conforming
HTTP server will always return us the full file range, but I wanted to split
it off from the subsequent changes in order to make review easier.
This fails to consider the case of whence == SEEK_END and the resulting
offset happening to exactly match the current position.
Reorder the check to compute the target position first, then compare.
This reverts commit aa20d7b3e8.
Adding these avpriv functions is absolutely overblown: Muxers
can get the desired duration in a few lines themselves.
In particular, using the parse functions from this file
necessitated parsing the extradata (and entailed exporting
the parsing function), although it was only used to know
whether the frames are self-delimiting, but everything of
interest to a muxer does not depend on this at all.
The commit to be reverted also made several structures
part of the ABI, which should be avoided in general.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
This avoids avpriv functions from lavc/opus/parse.c
(which parse way more than we need, necessitating
parsing the extradata).
It furthermore makes the output of the muxer consistent,
i.e. no longer depending upon whether the Opus parser
or decoder are enabled (the avpriv functions would just
return AVERROR(ENOSYS)).
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Before this patch, the last packet in the affected fate test would be written
without a BlockDuration element despite the packet's duration being shorter
than the Opus frame size.
Signed-off-by: James Almer <jamrial@gmail.com>
Following up on b613eebe78, if a demuxer that exports complete frames sets a
duration, don't overwrite it from the output of the parser.
Signed-off-by: James Almer <jamrial@gmail.com>
We did check video codecs but not audio
Fixes: Assertion failure (on codec_id) in parser.c
Fixes: 472097507/clusterfuzz-testcase-minimized-ffmpeg_dem_LIVE_FLV_fuzzer-6016386662203392
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
dxv assumes that newly reallocated memory in tex_data is not uninitialized
thus we have to do that too in case of reallocation in ff_lzf_uncompress()
Fixes: 475000819/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DXV_DEC_fuzzer-5571269310611456
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
size represents the output size
randomly changing it but not reseting it on errors leaks uninitialized memory.
Fixes: 475000819/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DXV_DEC_fuzzer-5571269310611456
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
In the case where the last sorted value was -1u and we where on the first
pass of run1 we failed to fill the last few values of bitmap
No real world testcase is known
Fixes: use of uninitialized memory
Fixes: 460333808/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_FFV1_fuzzer-6370167888347136
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The ldrsw instruction requires immediate offset with # prefix.
This fixes the syntax error introduced in commit 26752368f0
(aarch64/h26x: Add put_hevc_pel_bi_w_pixels) where the
load_bi_w_pixels_param macro was added.
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
The 'sws_dither' and 'alphablend' options access 'SwsDither' and
'SwsAlphaBlend' enum fields as integers. This is unsafe when the
code is compiled with -fshort-enums, as the enum size might be
smaller than an int.
Since the 'dither' and 'alpha_blend' struct members are part of the
public API, their types cannot be easily changed.
To ensure safe integer access and maintain ABI compatibility across
different compiler settings, a MAX_ENUM value is added to force the
enums to a 32-bit underlying type.
Fix#20712
When print_sdp() fails during rtp streaming, sch_start()
returns error code without calling wirte_trailer, causing
a memory leak.
The avformat_free_context() will call the mux->deinit, so
we should add a rtp_deinit() to free all internal allocations.
Signed-off-by: Jack Lau <jacklau1222gm@gmail.com>
The -sources and -sinks options were defined with OPT_FUNC_ARG flag,
which requires an argument. This caused "Missing argument for option
'sources'" error when running ffprobe -sources without arguments.
Removing OPT_FUNC_ARG flag allows these options to work without
arguments, listing all available devices, while still supporting
optional device name argument for filtering specific devices.
The `sub_packet` index in `QDM2Context` was not reset to 0 when
`qdm2_decode_frame` started processing a new packet. If an error
occurred during the decoding of a previous packet, `sub_packet` would
retain a non-zero value.
In subsequent calls to `qdm2_decode_frame` with a new packet, this
non-zero `sub_packet` value caused `qdm2_decode` to skip
`qdm2_decode_super_block`. This function is responsible for initializing
packet lists with pointers to the current packet's data. Skipping it led
to the use of stale pointers from the previous (freed) packet, resulting
in a heap-use-after-free vulnerability.
This patch explicitly resets `s->sub_packet = 0` at the beginning of
`qdm2_decode_frame`, ensuring correct initialization for each new
packet.
Fixes: OSS-Fuzz issue 476179569
(https://issues.oss-fuzz.com/issues/476179569).
Niklas Haas