manfred-h
accepted
request 6594
1 day ago
(revision 14)
- Update to version 1.1.3:
+ This release completes high bit depth decoding (up to 16 bit)
and repairs cross-component prediction, which had corrupted
4:4:4 Range Extensions streams since v1.0.17. It is ABI- and
API-compatible with v1.1.2 and a drop-in replacement; no
functions or enum values were added.
+ High bit depth decoding:
- Inter prediction above 12 bit was wrong from the first P/B
picture on. The fractional sample interpolation used shift1 =
BitDepth-8 instead of the Range Extensions' Min(4,
BitDepth-8), and the intermediate prediction samples were
kept in int16_t although they need max(14, BitDepth+2) bits.
The 16-bit kernels are now templated on the intermediate
sample type and use int32_t above 12 bit; 8- to 12-bit
streams are unaffected and keep their SSE paths. Explicit
weighted bi-prediction was broken in a second way: the
shortcut for identical motion vectors checked
weighted_pred_flag, which governs P slices, instead of
weighted_bipred_flag.
- Together with the fixes in v1.1.1 and v1.1.2 this completes
high bit depth support. Output is bit-exact against the HM 18
reference decoder at 10, 12, 13, 14 and 16 bit, for mixed
luma/chroma bit depths and for 4:2:2, 4:4:4 and 4:0:0. Above
12 bit motion compensation runs the scalar kernels; 8-bit
performance is unchanged.
+ Range Extensions:
- Cross-component prediction corrupted every chroma block it
was applied to, a regression since v1.0.17. A cast added to
silence undefined behaviour on a left shift turned the
following arithmetic shift into a logical one, so every
negative luma residual became a large positive value. This
affects 4:4:4 streams only, but at every bit depth including
8 bit.
- cabac_bypass_alignment_enabled_flag was parsed and then
ignored, so a stream using it desynchronized CABAC and
decoded to garbage without any diagnostic. It is now
implemented as specified in 9.3.4.3.6, and was the last Range
Extensions tool still missing. Both fixes are bit-exact
against HM 18.
manfred-h
accepted
request 6389
6 months ago
(revision 11)
New upstream release (with CVE fixes).
Note the aarch64 builds seems a bit flaky, kicking off a new rebuild let the builds compile without errors, so not sure whats going on there.
manfred-h
accepted
request 6074
about 2 years ago
(revision 9)
- Update to version 1.0.15:
* A couple of bug fixes, including the following CVEs:
- CVE-2023-49465
- CVE-2023-49467
- CVE-2023-49468
- Changes from version 1.0.14:
* This fixes build-time SSE detection when using the CMake build
system. No other changes than that. You don't need to update if
you are using the autotools build system.
- Changes from version 1.0.13:
* This release fixes among other smaller issues the following
crashes:
- SEGV:occured in function main at dec265.cc
- Memory allocation failed in function main at dec265.cc
- Buffer over-read causes segmentation fault in
pic_parameter_set::dump
- Potential segmentation fault due to incorrect realloc in CABA
C_encoder_bitstream::check_size_and_resize (unused function)
- SEGV in libde265 in slice_segment_header::dump_slice_segment_
header
- Libde265 v1.0.12 was discovered that requested allocation
size exceeds maximum supported size of 0x10000000000
- heap-buffer-overflow in derive_spatial_luma_vector_prediction
(...)
manfred.h
accepted
request 5861
almost 3 years ago
(revision 8)
- Update to version 1.0.12:
* Fix CVE-2023-27102
* Fix CVE-2023-27103
manfred.h
accepted
request 5670
over 3 years ago
(revision 6)
New upstream release - check results in staging before ack!
Olaf Hering (olh)
accepted
request 5286
over 5 years ago
(revision 5)
- Update to version 1.0.8
* read HDR parameters from VUI header
* fix reading matrix_coeffs from VUI header
version 1.0.7:
* add API to read VUI color information from bitstream
version 1.0.6:
* fix race in static mutex initialization
* fix missing headers to install in CMakeLists.txt
Olaf Hering (olh)
accepted
request 4056
almost 9 years ago
(revision 2)
Not sure this is really needed, but doing it just for safety as we do build -32 bits libs for gst-bad
Olaf Hering (olh)
accepted
request 4052
almost 9 years ago
(revision 1)
New package, gst-plugins-bad can use this as a builddep. Subbing here at packman as I'm 99% sure legal would reject this one.