This currently happens by accident in a few filters that use
ff_set_common_(samplerates|channel_layouts) like afir (if the response
option is set) or agraphmonitor (due to the default code in
avfiltergraph.c). So change those functions to make sure it does no
longer happen.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The ff_set_common_(formats|channel_layouts|samplerates) have to free
their list in case it doesn't have an owner; therefore they tracked
whether they attached it to an owner. But the list's refcount already
contains such a counter, so we don't have to keep track of whether we
have attached the list to an owner.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It is unnecessary as the number of static inputs and outputs can now
be directly read via AVFilter.nb_(in|out)puts.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It is intended as replacement for avfilter_pad_count(). In contrast to
the latter, it avoids a loop.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Up until now, an AVFilter's lists of input and output AVFilterPads
were terminated by a sentinel and the only way to get the length
of these lists was by using avfilter_pad_count(). This has two
drawbacks: first, sizeof(AVFilterPad) is not negligible
(i.e. 64B on 64bit systems); second, getting the size involves
a function call instead of just reading the data.
This commit therefore changes this. The sentinels are removed and new
private fields nb_inputs and nb_outputs are added to AVFilter that
contain the number of elements of the respective AVFilterPad array.
Given that AVFilter.(in|out)puts are the only arrays of zero-terminated
AVFilterPads an API user has access to (AVFilterContext.(in|out)put_pads
are not zero-terminated and they already have a size field) the argument
to avfilter_pad_count() is always one of these lists, so it just has to
find the filter the list belongs to and read said number. This is slower
than before, but a replacement function that just reads the internal numbers
that users are expected to switch to will be added soon; and furthermore,
avfilter_pad_count() is probably never called in hot loops anyway.
This saves about 49KiB from the binary; notice that these sentinels are
not in .bss despite being zeroed: they are in .data.rel.ro due to the
non-sentinels.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Besides being nicer code this also has the advantage of not making
assumptions about the internal implementation: While it is documented
that the AVFilter.inputs and AVFilter.outputs arrays are terminated
by a zeroed sentinel, one is not allowed to infer that one can just
check avfilter_pad_get_name(padarray, i) to see whether one has reached
the sentinel:
It could be that the pointer to the string is contained
in a different structure than AVFilterPad that needs to be accessed
first: return pad->struct->string.
It could be that for small strings an internal buffer in
AVFilterPad is used (to avoid a relocation) whereas for longer strings
an external string is used; this is useful to avoid relocations:
return pad->string_ptr ? pad->string_ptr : pad->interal_string
Or it could be that the name has a default value:
return pad->name ? pad->name : "default"
(This actually made sense for us because the name of most of our
AVFilterPads is just "default"; doing so would save lots of relocations.)
The only thing one is allowed to infer from the existence of the
sentinel is that one is allowed to use avfilter_pad_count() to get
the number of pads. Therefore it is used.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Use an uint32_t for the NAL unit size of an AVC H.264 NAL unit instead
of an int as a left shift of a signed value is undefined behaviour
if the result doesn't fit into the target type.
Also make the log message never output negative lengths.
Fixes: left shift of 16711968 by 8 places cannot be represented in type 'int'
Fixes: 36601/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_H264_fuzzer-6581933285965824
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
because subtitles streams will be written to webvtt m3u8 list
so the stream index should minus subtitles streams count when subtitle
between audio and video streams.
testcase:
before patch:
ffmpeg -i input -map 0🅰️0 -map 0:s:0 -map 0✌️0 -f hls aaaa.m3u8
will EXC_BAD_ACCESS
after patch:
ffmpeg -i input -map 0🅰️0 -map 0:s:0 -map 0✌️0 -f hls aaaa.m3u8
will ok
Signed-off-by: Steven Liu <liuqi05@kuaishou.com>
"frag_start" is redundant, and every occurance can be replaced with cluster[0].dts - start_dts
The proof of no behaviour changes: (All line number below is based on commit bff7d662d7)
"frag_start" is read at 4 place (with all possible call stacks):
mov_write_packet
...
mov_flush_fragment
mov_write_moof_tag
mov_write_moof_tag_internal
mov_write_traf_tag
mov_write_tfxd_tag (#1)
mov_write_tfdt_tag (#2)
mov_add_tfra_entries (#3)
mov_write_sidx_tags
mov_write_sidx_tag (#4)
mov_write_trailer
mov_auto_flush_fragment
mov_flush_fragment
... (#1#2#3#4)
mov_write_sidx_tags
mov_write_sidx_tag (#4)
shift_data
compute_sidx_size
get_sidx_size
mov_write_sidx_tags
mov_write_sidx_tag (#4)
All read happens in "mov_write_trailer" and "mov_write_moof_tag". So we need to prove no behaviour change in these two
functions.
Condition 1: for every track that have "trk->entry == 0", trk->frag_start == trk->track_duration.
Condition 2: for every track that have "trk->entry > 0", trk->frag_start == trk->cluster[0].dts - trk->start_dts.
Definition 1: "Before flush" means just before the invocation of "mov_flush_fragment", except for the auto-flush case in
"mov_write_single_packet", which means before L5934.
Lemma 1: If Condition 1 & 2 is true before flush, Condition 1 & 2 is still true after "mov_flush_fragment" returns.
Proof:
No update to the tracks that have "trk->entry == 0" before flushing, so we only consider tracks that have "trk->entry > 0":
Case 1: !moov_written and moov will be written in this iteration
trk->entry = 0 L5366
trk->frag_start == trk->cluster[0].dts - trk->start_dts Lemma condition
trk->frag_start += trk->start_dts + trk->track_duration - trk->cluster[0].dts; L5363
So trk->entry == 0 && trk->frag_start == trk->track_duration
Case 2: !moov_written and moov will NOT be written in this iteration
nothing changed
Case 3: moov_written
trk->entry = 0 L5445
trk->frag_start == trk->cluster[0].dts - trk->start_dts Lemma condition
trk->frag_start += trk->start_dts + trk->track_duration - trk->cluster[0].dts; L5444
So trk->entry == 0 && trk->frag_start == trk->track_duration
Note that trk->track_duration may be updated for the tracks that have "trk->entry > 0" (mov_write_moov_tag will
update track_duration of "tmcd" track, but it must have 1 entry). But in all case, trk->frag_start is also updated
to consider the new value.
Lemma 2: If Condition 1 & 2 is true before "ff_mov_write_packet" invocation, Condition 1 & 2 is still true after it returns.
Proof:
Only the track corresponding to the pkt is updated, and no update to relevant variables if trk->entry > 0 before invocation.
So we only need to prove "trk->frag_start == trk->cluster[0].dts - trk->start_dts" after trk->entry increase from 0 to 1.
Case 1: trk->start_dts == AV_NOPTS_VALUE
Case 1.1: trk->frag_discont && use_editlist
trk->cluster[0].dts = pkt->dts at L5741
trk->frag_start = pkt->pts at L5785
trk->start_dts = pkt->dts - pkt->pts at L5786
So trk->frag_start == trk->cluster[0].dts - trk->start_dts
Case 1.2: trk->frag_discont && !use_editlist
trk->cluster[0].dts = pkt->dts at L5741
trk->frag_start = pkt->dts at L5790
trk->start_dts = 0 at L5791
So trk->frag_start == trk->cluster[0].dts - trk->start_dts
Case 1.3: !trk->frag_discont
trk->cluster[0].dts = pkt->dts at L5741
trk->frag_start = 0 init
trk->start_dts = pkt->dts at L5779
So trk->frag_start == trk->cluster[0].dts - trk->start_dts
Case 2: trk->start_dts != AV_NOPTS_VALUE
Case 2.1: trk->frag_discont
trk->cluster[0].dts = pkt->dts at L5741
trk->frag_start = pkt->dts - trk->start_dts at L5763
So trk->frag_start == trk->cluster[0].dts - trk->start_dts
Case 2.2: !trk->frag_discont
trk->cluster[0].dts = trk->start_dts + trk->track_duration at L5749
trk->track_duration == trk->frag_start Lemma condition
So trk->frag_start == trk->cluster[0].dts - trk->start_dts
Lemma 3: Condition 1 & 2 is true in all case before and after "ff_mov_write_packet" invocation, before flush and after
"mov_flush_fragment" returns.
Proof: All updates to relevant variable happen either in "ff_mov_write_packet", or during flush. And Condition 1 & 2
is true initially. So with lemma 1 & 2, we can prove this use induction.
Noticed that all read of "frag_start" only happen in "trk->entry > 0" branch. Now we need to prove Condition 2 is true
before each read.
Because no update to variables relevant to Condition 2 between "before flush" and "mov_write_moof_tag" invocation, we
can conclude Condition 2 is true before every invocation of "mov_write_moof_tag". No behaviour change in
"mov_write_moof_tag" is proved.
In "mov_write_trailer", No update to relevant variables after the last flush and before the invocation of
"mov_write_sidx_tag". So no behaviour change to "mov_write_trailer" is proved.
Q.E.D.
Signed-off-by: Hu Weiwen <sehuww@mail.scut.edu.cn>
Signed-off-by: Martin Storsjö <martin@martin.st>
track->mdat_buf can be not NULL while the track is still empty if the
last packet write failed.
Signed-off-by: Hu Weiwen <sehuww@mail.scut.edu.cn>
Signed-off-by: Martin Storsjö <martin@martin.st>
Unlike libx264, libx265 does not have a separate "unspecified"/"auto"
default for color range, so we do always have to specify it.
Thus, we are required to handle the RGB case on the libavcodec
side to enable the correct value to be written out in in case
of RGB content with unspecified color range being received.
In other words:
1. If the user has set color range specifically, follow that.
2. If the user has not set color range specifically, set full
range by default in case of RGB and YUVJ pixel formats.
By default the x264 full range flag is set to -1. By not setting
it to something else, we can let libx264 handle the RGB case.
Additionally, change the preference order to user-specified range
first, and then any fall-back logic left for the YUVJ pix_fmts.
Fixes the capture part of #9374
It will be useful in the future when more flags are added.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
These fields are mutually exclusive, so putting them in a union
is possible and makes AVFilterPad smaller.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
av_frame_copy_props() already copies pts.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
In particular, document that av_opt_copy() always disentangles
allocated options even on error; this guarantee is needed to e.g.
properly free duplicated thread contexts in libavcodec on error.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Currently user may use '-init_hw_device type=name' to initialize a hw
device, however the key parameter is ignored when use '-init_hw_device
type=name,key=value'. After applying this patch, user may set key
parameter if needed.
Reviewed-by: Soft Works <softworkz@hotmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
The last init_opaque callback has been removed in commit
07ffdedf784e86b88074d8d3e08e55752869562a; the opaque argument has been
always NULL since 0acf7e268b.
Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The code for inserting inpads can't be reached by ff_vsrc_openclsrc
(unsurprising given that it is a source filter), so it didn't get
the flag.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Also adapt some FATE tests to already cover this.
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
AVFrame.metadata is always owned by its AVFrame, it is not shared
in the first place, so one does not need to make the frame writable
to modify it.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Andreas Rheinhardt
Paul B Mahol
Nicolas George
Christopher Degawa
Steven Liu
Hu Weiwen
Jan Ekström
Haihao Xiang
Michael Niedermayer
Limin Wang