marfrit/marfrit-packages
0
0
Fork 1
Code Issues 2 Pull Requests Actions Packages Projects Releases Wiki Activity

daedalus-v4l2{,-dkms}: 481279c -> 3dd0eb0 (DAEMON-PPS close) #49

Merged
marfrit merged 1 commits from claude-noether/marfrit-packages:noether/daedalus-bump-3dd0eb0 into main 2026-05-20 16:54:11 +00:00
Conversation 0 Commits 1 Files Changed 6
+43 -12
claude-noether commented 2026-05-20 16:53:29 +00:00
Contributor
Copy Link
Copy Source

Follow-up to daedalus-v4l2 PR #1 (merged as 3dd0eb0).

Upstream landed daemon-side synthesis of H.264 SPS/PPS NAL units from the V4L2 stateless control structs. This bumps the package pin so both the userspace daemon and the kernel module pick up the change in lockstep — the wire protocol grew a new DAEDALUS_REQ_FLAG_H264_META bit + struct daedalus_h264_meta, so kernel and daemon MUST match. Both arch + debian variants bumped.

Effect on higgs

ffmpeg -hwaccel vaapi -i h264_test.mp4 should produce a successful decode through the daedalus daemon instead of failing with 'non-existing PPS 0 referenced'. Daemon journal should show:

decoder: opened h264 context
decoder: h264 prepended SPS=NB PPS=MB slice=KB
decoder: OK WxH fmt=0 (yuv420p) fnv1a=0x...

Versions

Both packages: pkgver 0.1.0.r20.3dd0eb0 (count from git rev-list), pkgrel reset to 1 (new upstream pin).

VP9 / AV1 behaviour unchanged.

CI

The daedalus-v4l2-debian + daedalus-v4l2-dkms-debian jobs both moved to actrunner-debian-aarch64-bohr in PR #47, so this will build on the native trixie runner. Kernel module compile via DKMS will be the first kernel-side test of the new code (boltzmann doesn't have aarch64 kernel-headers locally) — if v4l2_ctrl_find / ctrl->p_cur.p_h264_* API differs from what was assumed, the DKMS build on bohr or the install on higgs will catch it.

Generated with Claude Code

Follow-up to daedalus-v4l2 PR #1 (merged as 3dd0eb0). Upstream landed daemon-side synthesis of H.264 SPS/PPS NAL units from the V4L2 stateless control structs. This bumps the package pin so both the userspace daemon and the kernel module pick up the change in lockstep — the wire protocol grew a new DAEDALUS_REQ_FLAG_H264_META bit + struct daedalus_h264_meta, so kernel and daemon MUST match. Both arch + debian variants bumped. ### Effect on higgs ffmpeg -hwaccel vaapi -i h264_test.mp4 should produce a successful decode through the daedalus daemon instead of failing with 'non-existing PPS 0 referenced'. Daemon journal should show: ``` decoder: opened h264 context decoder: h264 prepended SPS=NB PPS=MB slice=KB decoder: OK WxH fmt=0 (yuv420p) fnv1a=0x... ``` ### Versions Both packages: pkgver 0.1.0.r20.3dd0eb0 (count from git rev-list), pkgrel reset to 1 (new upstream pin). VP9 / AV1 behaviour unchanged. ### CI The daedalus-v4l2-debian + daedalus-v4l2-dkms-debian jobs both moved to actrunner-debian-aarch64-bohr in PR #47, so this will build on the native trixie runner. Kernel module compile via DKMS will be the first kernel-side test of the new code (boltzmann doesn't have aarch64 kernel-headers locally) — if v4l2_ctrl_find / ctrl->p_cur.p_h264_* API differs from what was assumed, the DKMS build on bohr or the install on higgs will catch it. Generated with Claude Code
claude-noether added 1 commit 2026-05-20 16:53:30 +00:00
daedalus-v4l2{,-dkms}: 481279c -> 3dd0eb0 (DAEMON-PPS close) b16a3f1a77 
Upstream PR #1 landed daemon-side synthesis of H.264 SPS/PPS NAL
units from the V4L2 stateless control structs.  This bumps the
package pin so both the userspace daemon and the kernel module
pick up the change in lockstep (wire protocol grew a new
DAEDALUS_REQ_FLAG_H264_META bit + struct daedalus_h264_meta —
kernel and daemon MUST match).

Effect on Pi 5 / CM5 hosts (higgs): ffmpeg -hwaccel vaapi against
H.264 sources will actually decode through the daedalus daemon
instead of failing with "non-existing PPS 0 referenced".

Both arch + debian packages: pkgver 0.1.0.r20.3dd0eb0 (count from
git rev-list), pkgrel reset to 1 (new upstream pin).

VP9 / AV1 paths unchanged.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
marfrit merged commit 29cc145d44 into main 2026-05-20 16:54:11 +00:00
marfrit deleted branch noether/daedalus-bump-3dd0eb0 2026-05-20 16:54:11 +00:00
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
marfrit (Markus Fritsche)
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: marfrit/marfrit-packages#49
Delete Branch "claude-noether/marfrit-packages:noether/daedalus-bump-3dd0eb0"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?