cbs.mak is meant to contain tests strictly for the CBS framework, not for any
bsf that happens to use it under the hood.
Signed-off-by: James Almer <jamrial@gmail.com>
On Windows, static libraries are typically named with a .lib extension.
An exception to this is MinGW targets, which are treated as a distinct
target-os.
Using Windows-style naming allows `clang` to be used as the linker
driver, instead of invoking link or lld-link directly. The latter
approach requires manually specifying standard libraries when compiling
with `clang` rather than `clang-cl`, and manually specifying standard
libraries may be error-prone or incomplete.
Using Windows-style naming allows Clang to be used as the linker driver,
instead of invoking link or lld-link directly. The latter approach
requires manually specifying standard libraries, which may be
error-prone or incomplete.
This change also improves support for LTO and sanitizer builds, where
it's significantly easier to let the compiler driver manage the
necessary linker flags.
It fixes issues where Clang is asked to link `-lavcodec`, which gets
passed to the linker as avcodec.lib, resulting in an error like:
lld-link: error: could not open 'avcodec.lib': no such file or directory
This happens because `libavcodec.a` was unexpectedly generated, not
`avcodec.lib` expected by tooling.
Additionally, using `clang` (not `clang-cl`) is simplified, as it does
not use autolinking like MSVC/clang-cl does. Now `--ld=clang` can be
used to add all the required libraries. Previously, building with
`clang` was only possible by using `--ld=lld-link` and manually
specifying system dependencies in `extra-ldflags`.
Note that those changes doesn't affect mingw build. MSVC builds will
produce .lib static libraries now, but the linking process itself is not
affected, because filenames are passed directly.
To summarize in Windows non-mingw builds:
Static builds now produce `lib/avcodec.lib` instead of `lib/libavcodec.a`.
Shared builds remain unchanged, producing `bin/avcodec.lib` together with
`bin/avcodec-62.dll`.
This also removes setting LD_LIB from Win32/64 target as there is one
type of .lib in practice. We cannot build both shared and static at the
same time as noted by the next line.
Signed-off-by: Kacper Michajłow <kasper93@gmail.com>
Remove redundant av_freep() to avoid double free since task will be freed in dnn_free_model_tf() after the success of ff_queue_push_back().
Fixes: af052f9066 ("lavfi/dnn: fix mem leak in TF backend error handle")
Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
The code writing in the buffer has a wide range of error checks
which simply leave it partly uninitialized.
Initializing it on allocation ensures no sensitive data leaks and that
bugs are more reliably reproduceable
Fixes: use of uninitialized memory
Fixes: 435225510/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_DXV_DEC_fuzzer-4521918634196992
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Dithering relies on a 8 line dithering table and the code always uses it from
the beginning. So in order to make dithering independent from height of the
slices used we must enforce a 8 line alignment.
Fixes issue #20071.
Signed-off-by: Marton Balint <cus@passwd.hu>
You can technically feed the decoder a single packet with more than one AU in
it, even if a demuxer is unlikely to ever export such a thing.
Given the parser does no packetization to ensure we get a single AU, in order
to not discard any frame beyond the first, the decoupled input/output API
should be used.
Signed-off-by: James Almer <jamrial@gmail.com>
If a single fragment contains more than one frame unit, the tile information stored
in the private context will only correspond to one of them.
Fixes: crash (out of array access)
Fixes: 435489659/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APV_fuzzer-6194885205229568
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: James Almer <jamrial@gmail.com>
Fixes: out of array access
Fixes: BIGSLEEP-436511754/testcase.exr
Found-by: Google Big Sleep
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
If we use the private BlackDetectContext for logging, the filter ID does
not get printed alongside the filter name. Using the parent AVFilterContext
makes this consistent with other filters.
When called from avformat_find_stream_info(), we are only interested in
decoding SOF fields.
This patch makes the decoder skip all other fields (including SOI, SOS,
and EOI). This also prevents the decoder from incorrectly printing the
warning "EOI missing, emulating" (which is the case since 2ae82788).
Fixes: #20119
pl_frame_mix_current() will return NULL if all frames are in the future,
but when libplacebo is using a frame mixer with a radius greater than 1,
future frames are expected to already be partially renderered. Instead, use
pl_frame_mix_nearest(), which is guaranteed to give us a valid frame for any
nonempty frame mix.
Fixes: 3091bca3ed
Previously, these tests failed when running on Windows, if the
system is configured with a time zone east of Greenwich, i.e.
with a positive GMT offset.
The muxer converts the creation_date given by the user using
av_parse_time to unix time, as a time_t. The creation_date is
interpreted as a local time, i.e. according to the current time
zone. (This time_t value is then converted back to a broken out
local time form with localtime_r.)
The given reference date/time, "1970-01-01T00:00:00", is the
origin point for unix time, corresponding to time_t zero. However
when interpreted as local time, this doesn't map to exactly zero.
Time zones east of Greenwich reached this time a number of hours
before the point of zero time_t - so the corresponding time_t
value essentially is minus the GMT offset, in seconds.
Windows mktime returns an error, returning (time_t)-1, when given
such a "struct tm", while e.g. glibc mktime happily returns a
negative time_t. av_parse_time doesn't check the return value of
mktime for potential errors.
This is observable with the following test snippet:
struct tm tm = { 0 };
tm.tm_year = 70;
tm.tm_isdst = -1;
tm.tm_mday = 1;
tm.tm_hour = 0;
time_t t = mktime(&tm);
printf("%d-%02d-%02d %02d:%02d:%02d\n", tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday, tm.tm_hour, tm.tm_min, tm.tm_sec);
printf("t %d\n", (int)t);
By varying the value of tm_hour and the system time zone, one
can observe that Windows mktime returns -1 for all time_t values
that would have been negative.
This range limit is also documented by Microsoft in detail at
https://learn.microsoft.com/en-us/cpp/c-runtime-library/reference/mktime-mktime32-mktime64.
To avoid the issue, pick a different, arbitrary reference time,
which should have a nonnegative time_t for all time zones.
