Builder ffmpegsos-solaris10-i386 Build #14308

Results:

Failed

SourceStamp:

Project	ffmpeg
Repository	https://git.ffmpeg.org/ffmpeg.git
Branch	master
Revision	1b5041c785f1e735e14ef2db2c98a786ef180954
Changes	14 changes

BuildSlave:

unstable10x

Reason:

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

Steps and Logfiles:

git updating ( 6 secs )
1. stdio
shell
1. - no logs -
shell_1
1. - no logs -
shell_2
1. - no logs -
shell_3
1. - no logs -
shell_4
1. - no logs -
shell_5
1. - no logs -

Build Properties:

Name	Value	Source
branch	master	Build
builddir	/export/home/buildbot/slave/ffmpegsos-solaris10-i386	slave
buildername	ffmpegsos-solaris10-i386	Builder
buildnumber	14308	Build
codebase		Build
project	ffmpeg	Build
repository	https://git.ffmpeg.org/ffmpeg.git	Build
revision	1b5041c785f1e735e14ef2db2c98a786ef180954	Build
scheduler	schedule-ffmpegsos-solaris10-i386	Scheduler
slavename	unstable10x	BuildSlave
workdir	/export/home/buildbot/slave/ffmpegsos-solaris10-i386	slave (deprecated)

Forced Build Properties:

Name	Label	Value

Responsible Users:

Lynne
devohnoyoudont@lynne.ee

Timing:

Start	Sun Sep 20 08:44:41 2026
End	Sun Sep 20 08:44:47 2026
Elapsed	6 secs

All Changes:

:

Change #282274

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 07:45:46
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	daa210cfd7b166e01a231264ad3bdec281b8788d

Comments

h264: correct the comment on the field POCs of frame_num gap dummies
Hwaccels submit the field POCs of a picture, not its frame POC.

Changed files

libavcodec/h264_slice.c

Change #282275

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 07:45:50
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	0ae15fffefd66d78828902176da25d19ae652a0e

Comments

vulkan_decode: do not bind the current picture twice when its slot is active
When decoding the second field of a pair, the picture is already bound as
an active reference at its DPB slot, since the first field made it one.
Appending it again as an inactive slot duplicates the picture resource
within the coding scope, which is not allowed
(VUID-vkCmdBeginVideoCodingKHR-pPictureResource-07238).

Changed files

libavcodec/vulkan_decode.c

Change #282276

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 07:45:52
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	2f0c34b10e02df29a707cdc4d16a29044fbbe2bd

Comments

vulkan_h264: only bind reference fields that were decoded into their slot
A Vulkan DPB slot only becomes active once a decode operation sets it up,
and from then on holds exactly the picture resource that was decoded into
it. The H.264 decoder however hands us references that never went through
the hwaccel: the dummies it synthesizes to fill frame_num gaps and to stand
in for missing references, and fields lost to a seek. Binding those as
active slots fails validation (VUIDs 07239, 07265, 07266, 07151), and since
the slot lookup for long-term references only searched the first 16
pictures and fell back to slot 0, a resource could end up bound twice
(07238, 07264).

Track, per picture, which fields were actually decoded into its slot, and
bind only the references, and only the fields of them, that were:
vk_h264_fill_pict returns 1 when it bound the picture and 0 when it cannot
be bound, and the reference loops advance by that. Use the picture's index
in the DPB as its slot directly instead of searching for it, and describe
the setup slot by the structure of the picture being decoded rather than by
its reference marking, which does not describe the slot's contents while
decoding the second field of a pair.

Changed files

libavcodec/vulkan_h264.c

Change #282277

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 07:45:56
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	82a2875192b6e6d45ee3ed8d0baf03e91dcca532

Comments

vulkan_h264: give each picture a persistent DPB slot instead of its DPB index
A frame_num gap dummy shares the duplicated picture's hwaccel state, and
with it the DPB slot that picture was decoded into, which can differ from
the dummy's own index into h->DPB: the dummy and the picture it duplicates
are distinct H264Picture entries. Deriving the Vulkan slot from the DPB
index, as before, could then bind the dummy at an index nothing was ever
decoded into, or one currently in use by an unrelated picture.

Assign the slot once, when a picture is first decoded, to the lowest index
no current reference holds, and keep it for as long as the picture (or a
dummy sharing its state) remains referenced. ff_ctz(0) is undefined, so
guard the case where every slot is taken rather than fall into it.

A slot can still end up requested twice in the same scope: once for the
original picture, once for a dummy sharing its state. Detect that by
identity on the shared hwaccel state rather than by DPB index, since the
two no longer coincide.

Changed files

libavcodec/vulkan_h264.c

Change #282284

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 08:13:01
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	9a69dc787de7ec6a5a17c74247eaadd46b7a5066

Comments

vulkan/prores_idct: clip values to maximum allowed range
This matches the software decoder.

Changed files

libavcodec/vulkan/prores_idct.comp.glsl

Change #282286

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 09:45:55
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	8cf5b8b0b335ba8d524a14542027c3593e6d3bfe

Comments

vulkan_prores: undo quant matrix permutation
The quantization matrix gets immediately permuted for the assembly.
But the Vulkan decoder uses it in regular standard order.

Changed files

libavcodec/vulkan_prores.c

Change #282288

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 09:45:57
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	083bf5cecb1bdbe3d5f9ba28e558d38511000874

Comments

vulkan_prores_raw: undo potential quant matrix permutation
Same as last commit, except this is done just in case someone
introduces a ProRes RAW iDCT.

Changed files

libavcodec/vulkan_prores_raw.c

Change #282289

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 09:45:59
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	3110b21964fa8317abd1a33975ae6233a30c1e7c

Comments

vulkan_apv: remove unnecessary image clear
We moved to using a dedicated buffer for the intermediate data.

Changed files

libavcodec/vulkan/apv_decode.comp.glsl
libavcodec/vulkan_apv.c

Change #282290

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 09:46:03
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	fe06105c8a9ab4e873efdbea7cd00da1d0e1295c

Comments

prores_raw: parse the frame before honouring skip_frame
The decoder returned before parsing anything when skip_frame was
AVDISCARD_ALL, yet advertises FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM, so it
never set the pixel format and dimensions while probing. Probing gives up
after probesize, and ProRes RAW frames are larger than the default, so
ffmpeg failed to open such files at all with "Could not find codec
parameters ... unspecified pixel format". Parse up to the tile table, which
sets every stream parameter, before skipping the frame.

Changed files

libavcodec/prores_raw.c

Change #282291

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 09:46:07
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	bc1cf21643f6cb63c0756268f61bf33ebd0b8712

Comments

prores_raw: end a component at 32 zero bits, like the reference decoder
The reference decoder reads every symbol from a 32-bit window that is
zero-filled past the end of the component's data, and stops decoding the
component as soon as that window is all zero. Both our decoders checked for
remaining bits instead, and near the end of the data decoded a zero symbol
from a short window and stored spurious coefficients of magnitude 1 at the
end of the last blocks of a tile. The software decoder also ran one
coefficient past the end of the block array. Decode the way the reference
does, in both, so their coefficients match it, and each other.

Changed files

libavcodec/prores_raw.c
libavcodec/vulkan/prores_raw_decode.comp.glsl

Change #282292

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 10:19:11
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	e9e698f482589285049a58ee7a3286f9af37405c

Comments

vulkan: clamp GET_BITS_SMEM refills to the end of the bitstream
FILL_SMEM read up to 16*GET_BITS_SMEM bytes past the end of the buffer.
At the default depth of 4 that is exactly AV_INPUT_BUFFER_PADDING_SIZE,
so it only worked by coincidence; deeper windows faulted.

Clamp lines past the end onto the last line; nothing reads them. Issuing
each line as its own load is also ~3% faster than one indexed fetch.

Changed files

libavcodec/vulkan/common.glsl

Change #282293

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 10:19:16
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	e99556134446a7672915a82c6a6a425eb875f908

Comments

vulkan_prores_raw: pick the bitstream prefetch depth per device
With the packet host-mapped, the decode shader is bound by fetch
latency over the bus: 5.49 ms per frame for the VLD alone, versus
0.21 ms from a device-local buffer.

Make GET_BITS_SMEM a specialization constant: 64 lines (1 KB) per
invocation on discrete GPUs reading host-mapped packets, 4 otherwise.
Integrated GPUs get slower with every extra line.

5.7K ProRes RAW on an RX 6900 XT: 145 -> 249 fps (non-HQ), 98 -> 154
fps (HQ), bit-identical output. Intel Iris Xe unchanged.

Changed files

libavcodec/vulkan/prores_raw_decode.comp.glsl
libavcodec/vulkan_prores_raw.c

Change #282294

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 10:19:22
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	e3654ccab4e19ffbbaf1a73b534ad44bdfcaac72

Comments

vulkan/ffv1: initialize the version 4 slice header fields on older streams
slice_reset_contexts, remap and remap_count were only set on version 4
streams. The slice context buffer is pooled and never cleared, so on
older streams the reset shader read stale memory as the reset flag and
wiped the range coder state carried over from the previous frame.

Non-keyframes of pre-version-4 streams intermittently decoded to
garbage, with the slice CRCs still passing.

Changed files

libavcodec/vulkan/ffv1_dec_setup.comp.glsl

Change #282295

Category	ffmpeg
Changed by	Lynne <devohnoyoudont@lynne.ee>
Changed at	Sun 20 Sep 2026 10:19:28
Repository	https://git.ffmpeg.org/ffmpeg.git
Project	ffmpeg
Branch	master
Revision	1b5041c785f1e735e14ef2db2c98a786ef180954

Comments

vulkan: keep the iDCT scale factors separable
The 8x8 scale table was 64 specialization constants indexed per
coefficient. The index is divergent, so nothing folds and every
coefficient costs a load from the constant-data segment.

The table is an outer product, s[y]*s[x], so pass the eight 1D factors
instead: the x factor hoists out of the coefficient loop and the y
factor folds once the loop is unrolled. ProRes RAW's gain of 8 becomes
a literal in its shader.

Reassociation moves a few samples in ten thousand by an LSB or so,
well below the existing float iDCT error; PSNR versus software is
unchanged.

Changed files

libavcodec/vulkan/apv_idct.comp.glsl
libavcodec/vulkan/dct.glsl
libavcodec/vulkan/prores_idct.comp.glsl
libavcodec/vulkan/prores_raw_idct.comp.glsl
libavcodec/vulkan_apv.c
libavcodec/vulkan_prores.c
libavcodec/vulkan_prores_raw.c