[techapostle]

What changed

Opening Controls (Super+Escape) on a desktop whose external monitor is driven over DDC/CI made every output stutter. ControlsLevels.qml ran ddcutil getvcp 10 the moment the panel opened. That call takes about 0.3 s, nearly all of it kernel time on NVIDIA I2C, and it overlapped the panel's overlay layers mapping.

Measured with a separate 120 Hz frame-probe client on a second display:

Case	Frame stalls per open
Stock	two: ~45 ms at +90 ms, 80-200 ms at +220-320 ms
getvcp run on its own	none
Brightness drawer opened while the panel is already mapped (two reads)	none
Bare fullscreen overlay mapped	none
This change (250 ms delay)	none over 6 warm opens

So the stall needs DDC traffic overlapping the map. The qs process peaks at about 70% of one core for about 0.4 s on a cold open and stays low on warm opens, so this is not a CPU or GPU load spike.

The open path now starts the read through a 250 ms timer, and closing stops a pending read. The stall only overlaps the first ~100 ms while the layers map: delays of 100 ms, 250 ms, enterDuration, enterDuration + 200 and enterDuration + 400 all showed no stalls over three warm opens each, so a fixed 250 ms keeps margin without tying the read to the motion speed. The read still runs on every open and nothing runs while the panel is closed, which keeps the docs/sidebars.md rule that display probing runs only while its surface is active. The trade-off: the live level lands about 0.5 s after opening instead of about 0.35 s. On a warm reopen the slider shows the last known value until then; on a cold open it reads as unavailable until then.

Separate from this change: the first cold open after a shell reload still showed one ~47 ms stall from building the surface. It did not recur in later runs.

Area

ryoku

How it was tested

Arch, Hyprland 0.57.0, NVIDIA, external monitor on DDC bus 1, Ryoku 1.0.7 (f7197d7), with this change applied to the live shell:

• Opened and closed Controls repeatedly with ryoku-shell quicksettings while the frame probe recorded. Stalls on open went from two per open to none on warm opens.
• The external brightness slider still picks up the monitor's live level after opening.
• bin/ryoku-dev-lint-qml and bin/ryoku-dev-verify-qml-bindings pass. qmllint shows only the unqualified SidebarState access the file already has.

Checklist

• One logical change, with a clear [area] scope: summary commit subject.
• Matching CHANGELOG.md updated in the area I touched.
• The git hooks pass locally; I did not use --no-verify.
• Lua parses (luac -p), shell scripts pass bash -n, and QML passes
  qmllint where applicable.
• No duplicated config, no dead code, no commented-out code, no em-dash.
• Docs updated if behavior or layout changed.