Home - Waterfall Grid T-Grid Console Builders Recent Builds Buildslaves Changesources - JSON API - About

Builder ffmpegsos-solaris10-i386 Build #14520

Results:

Failed

SourceStamp:

Projectffmpeg
Repositoryhttps://git.ffmpeg.org/ffmpeg.git
Branchmaster
Revisionaf25036ac09146b5187adfc247a643b3f993545c
Changes5 changes

BuildSlave:

unstable10x

Reason:

The SingleBranchScheduler scheduler named 'schedule-ffmpegsos-solaris10-i386' triggered this build

Steps and Logfiles:

  1. git updating ( 10 secs )
    1. stdio
  2. shell  
    1. - no logs -
  3. shell_1  
    1. - no logs -
  4. shell_2  
    1. - no logs -
  5. shell_3  
    1. - no logs -
  6. shell_4  
    1. - no logs -
  7. shell_5  
    1. - no logs -

Build Properties:

NameValueSource
branch master Build
builddir /export/home/buildbot/slave/ffmpegsos-solaris10-i386 slave
buildername ffmpegsos-solaris10-i386 Builder
buildnumber 14520 Build
codebase Build
project ffmpeg Build
repository https://git.ffmpeg.org/ffmpeg.git Build
revision af25036ac09146b5187adfc247a643b3f993545c Build
scheduler schedule-ffmpegsos-solaris10-i386 Scheduler
slavename unstable10x BuildSlave
workdir /export/home/buildbot/slave/ffmpegsos-solaris10-i386 slave (deprecated)

Forced Build Properties:

NameLabelValue

Responsible Users:

  1. Devin Heitmueller

Timing:

StartThu Oct 15 19:34:55 2026
EndThu Oct 15 19:35:06 2026
Elapsed10 secs

All Changes:

:

  1. Change #286274

    Category ffmpeg
    Changed by Devin Heitmueller <dheitmuellerohnoyoudont@ltnglobal.com>
    Changed at Thu 15 Oct 2026 20:46:02
    Repository https://git.ffmpeg.org/ffmpeg.git
    Project ffmpeg
    Branch master
    Revision 9d6c50f258a94911969924098e20c32b613b5487

    Comments

    avcodec/bitpacked: refactor to support assembly optimizations
Move the packed YUV422P10 unpacking loop behind a function pointer and keep the existing C implementation as the default.

Process multiples of eight through the DSP-style interface while retaining the C path for any remaining pixels. This provides a directly testable interface for later architecture-specific implementations without changing decoded output.

    Changed files

    • libavcodec/bitpacked_dec.c
    • libavcodec/bitpacked_dec.h
    • libavcodec/bitpacked_dec_init.h

  2. Change #286276

    Category ffmpeg
    Changed by Devin Heitmueller <dheitmuellerohnoyoudont@ltnglobal.com>
    Changed at Thu 15 Oct 2026 20:46:02
    Repository https://git.ffmpeg.org/ffmpeg.git
    Project ffmpeg
    Branch master
    Revision 821902ddf67e9a1e8c9c5da78d172074c23585d9

    Comments

    checkasm: add bitpacked decoder test
Compare packed input and all three output planes across the NEON block sizes, scalar-tail widths, and a 3840-pixel benchmark case.

    Changed files

    • tests/checkasm/Makefile
    • tests/checkasm/bitpackeddec.c
    • tests/checkasm/checkasm.c
    • tests/checkasm/checkasm.h
    • tests/fate/checkasm.mak

  3. Change #286277

    Category ffmpeg
    Changed by Devin Heitmueller <dheitmuellerohnoyoudont@ltnglobal.com>
    Changed at Thu 15 Oct 2026 20:46:02
    Repository https://git.ffmpeg.org/ffmpeg.git
    Project ffmpeg
    Branch master
    Revision 098dd9f46a4962bfe6c50bbb7bc0d05db1a39843

    Comments

    avcodec/bitpacked: add AArch64 NEON unpacking
Unpack packed 10-bit YUV 4:2:2 in 16-pixel NEON blocks and unroll the
main loop to 32 pixels. Keep exact-width source loads and use the scalar
implementation for decoder tails smaller than eight pixels.

Enable the optimized implementation on little-endian AArch64. Retain the
C fallback on big-endian AArch64, where the byte layout expected by this
implementation does not apply.

The implementation passed 300 consecutive checkasm repetitions and
produced bit-exact output in a 100-frame UHD decode.

Thanks to Martin Storsjö for feedback and suggestions for improvement.

Correctness command:
./tests/checkasm/checkasm --test=bitpackeddec --repeat=300 66000

Benchmark command:
./tests/checkasm/checkasm --bench --test=bitpackeddec \
    --function=bitpacked_unpack_yuv422p10 --duration=1000000 95000

Apple M4 Mac mini, Apple clang 17.0.0:
bitpacked_unpack_yuv422p10_c:       876.8 ns
bitpacked_unpack_yuv422p10_neon:    227.2 ns (3.86x)

Clang auto-vectorizes the C reference with NEON, so the result measures
the hand-written implementation against compiler-generated SIMD rather than
against purely scalar machine code.

    Changed files

    • libavcodec/aarch64/Makefile
    • libavcodec/aarch64/bitpacked_init_aarch64.c
    • libavcodec/aarch64/bitpacked_neon.S
    • libavcodec/bitpacked_dec.h
    • libavcodec/bitpacked_dec_init.h

  4. Change #286278

    Category ffmpeg
    Changed by Devin Heitmueller <dheitmuellerohnoyoudont@ltnglobal.com>
    Changed at Thu 15 Oct 2026 20:46:02
    Repository https://git.ffmpeg.org/ffmpeg.git
    Project ffmpeg
    Branch master
    Revision 508b82c605b0e53ffcb3c92a3ceac225cbf32f79

    Comments

    avcodec/bitpacked: add x86 SIMD unpacking
Add runtime-dispatched SSSE3 and AVX2 implementations for packed
10-bit YUV 4:2:2. The SSSE3 loop follows the negative-width output-offset
pattern used by v210. The AVX2 loop processes 16 pixels at a time and has
an internal eight-pixel tail.

Both implementations use exact-width input loads and therefore do not
rely on padded input or legal overread. They passed 300 consecutive checkasm
seeds.

Benchmark command:
./tests/checkasm/checkasm --bench --test=bitpackeddec \
    --duration=1000000 51000

Intel Xeon E-2356G, GCC 8.5, NASM 2.15.03:
bitpacked_unpack_yuv422p10_c:      13516.7
bitpacked_unpack_yuv422p10_ssse3:   3415.8 (3.96x)
bitpacked_unpack_yuv422p10_avx2:    2183.1 (6.19x)

    Changed files

    • libavcodec/bitpacked_dec.h
    • libavcodec/bitpacked_dec_init.h
    • libavcodec/x86/Makefile
    • libavcodec/x86/bitpacked.asm
    • libavcodec/x86/bitpacked_init.c

  5. Change #286279

    Category ffmpeg
    Changed by Devin Heitmueller <dheitmuellerohnoyoudont@ltnglobal.com>
    Changed at Thu 15 Oct 2026 20:46:02
    Repository https://git.ffmpeg.org/ffmpeg.git
    Project ffmpeg
    Branch master
    Revision af25036ac09146b5187adfc247a643b3f993545c

    Comments

    avcodec/bitpacked: add AVX-512ICL unpacking
Use a masked exact-width load and VBMI byte permutations to expand each
40-byte group, then compact the planar Y, U, and V outputs. Handle an
eight-pixel prefix separately so the hot loop can use the negative-width
output-offset pattern from v210.

The AVX-512ICL feature tier is required for vpermb. CPUs with baseline
AVX-512 but without VBMI continue to use AVX2. Reuse one result register for
all three output planes so the function also assembles for 32-bit x86.

The implementation passed 300 consecutive checkasm seeds and assembled
as ELF, Windows COFF, and Mach-O in both 32- and 64-bit modes.

Benchmark command:
./tests/checkasm/checkasm --bench --test=bitpackeddec \
    --duration=1000000 60300

Intel Xeon E-2356G, GCC 8.5, NASM 2.15.03:
bitpacked_unpack_yuv422p10_c:          13502.8
bitpacked_unpack_yuv422p10_ssse3:       3436.6 ( 3.93x)
bitpacked_unpack_yuv422p10_avx2:        2222.8 ( 6.07x)
bitpacked_unpack_yuv422p10_avx512icl:   1131.5 (11.93x)

    Changed files

    • libavcodec/x86/bitpacked.asm
    • libavcodec/x86/bitpacked_init.c
BuildBot (0.8.14) working for the OpenCSW Buildbot project.
Page built: Thu 15 Oct 2026 22:00:58 (CEST)