Skip to content

Name the 14-inch MacBook Pro M4 panel: 10-bit, Display P3 - #134

Draft
IronsideXXVI wants to merge 4 commits into
omacom:mainfrom
IronsideXXVI:panel-line-14-macbook-pro-m4
Draft

IronsideXXVI wants to merge 4 commits into
omacom:mainfrom
IronsideXXVI:panel-line-14-macbook-pro-m4

Conversation

@IronsideXXVI

Copy link
Copy Markdown

What changes:

  • Names the 14-inch MacBook Pro panel in displays.conf with the runtime's new panel directive: connector eDP, mode 3024x1964, bitdepth=10 cm=dp3.
  • Hyprland drives the built-in panel at 8 bits per channel and as an sRGB display, its defaults for every output. Apple's display controller takes a 10-bit frame from it and shows SDR frames in the panel's own gamut, Display P3 here, so sRGB sent unconverted shows oversaturated; that is why Omarchy gives Apple Silicon displays the EDID preset. On this panel dp3 fits better: it maps into exact Display P3 rather than the EDID's rounded primaries, and where the panel's EDID lists BT.2020 colorimetry, as this Mac's kernel's does once it offers HDR, Hyprland switches the connector to BT.2020 for the EDID preset.
  • Omarchy then gives that panel's monitor rules both settings and leaves external monitors alone; a bitdepth or cm set in hypr/monitors.lua wins. A runtime without the directive ignores the line.
  • omarchy-mac 0.2.7.

Depends on omacom/omarchy#18582, which adds default/hypr/panel-defaults.lua, the reader of the directive. The integration test runs that file against the staged displays.conf, so test/integration/runtime points at that pull request's head commit for now and moves to its merge commit once it merges. This stays a draft until then. #131 also moves omarchy-mac to 0.2.7; whichever merges second takes the next version.

Verification:

  • On the 14-inch MacBook Pro M4, which we are bringing up as a supported Mac, Hyprland then hands the display controller XRGB2101010 frames, and with cm = "dp3" the connector stays on its default colorspace and the driver tells the display controller the frames are Display P3. Whether the panel link itself carries 10 bits can't be read from Linux; macOS reports 8 bits with dithering for this panel.
  • omarchy-mac/test/desktop-setup-test.sh learns the directive and checks that it names a panel display-cutouts.json describes; it passes. test/integration/apple-platform-hooks-test.sh runs the runtime's panel-defaults.lua against the staged file; CI runs it against the pinned runtime.

Other Macs: only this panel is listed, the one panel this has been checked on. Another panel joins once a report shows the same on it.

Hyprland drives the built-in panel at 8 bits per channel and as an sRGB
display, its defaults for every output. Apple's display controller takes
a 10-bit frame from it and shows SDR frames in the panel's own gamut,
Display P3 here, so sRGB sent unconverted shows oversaturated; that is
why Omarchy gives Apple Silicon displays the EDID preset. On this panel
dp3 fits better: it maps into exact Display P3 rather than the EDID's
rounded primaries, and where the panel's EDID lists BT.2020 colorimetry,
as the J604 kernel's does once it offers HDR, Hyprland switches the
connector to BT.2020 for the EDID preset. Name the panel in
displays.conf with the runtime's new panel directive: connector eDP,
mode 3024x1964, bitdepth=10 cm=dp3. Omarchy then gives that panel's
monitor rules both settings and leaves external monitors alone; a
bitdepth or cm set in hypr/monitors.lua wins.

Only the 14-inch MacBook Pro panel is listed: it is the one panel this
has been checked on, a MacBook Pro 14-inch (M4, J604). There Hyprland
then hands the display controller XRGB2101010 frames, and with cm =
"dp3" the connector stays on its default colorspace and the J604 driver
tells the display controller the frames are Display P3. Whether the
panel link itself carries 10 bits can't be read from Linux; macOS
reports 8 bits with dithering for this panel. Another panel joins once a
report shows the same on it. The package's displays.conf grammar test
learns the directive and checks that it names a panel
display-cutouts.json describes; an integration test runs the runtime's
panel-defaults.lua against the staged file, so test/integration/runtime
moves to the omacom/omarchy commit that adds it. A runtime without the
directive ignores the line.
The omacom/omarchy pull request that adds default/hypr/panel-defaults.lua
gained two commits: description rules for a named panel now get its
settings, and its tests moved into hyprland-monitor-settings-test.sh.
Point the runtime at its new head, 37fc7a9ffda6; it moves to the merge
commit once that pull request merges.
The omacom/omarchy pull request that adds default/hypr/panel-defaults.lua
gained a commit that keeps a description rule shared with an external
display off that display. Point the runtime at its new head, cc96ed111309;
it moves to the merge commit once that pull request merges.
The omacom/omarchy pull request that adds default/hypr/panel-defaults.lua
gained a commit that follows the latest monitor rule naming the panel, by
port or by description, as Hyprland does. Point the runtime at its new
head, 52d735da59e7; it moves to the merge commit once that pull request
merges.
IronsideXXVI added a commit to IronsideXXVI/omarchy-j604 that referenced this pull request Nov 12, 2026
What the J604 runs now: kernel 1.848, mesa-aurora and vulkan-aurora
1.8139 (the 1.8134 content, rebuilt so the repository matches what is
installed after a rollback), omarchy-dev 1.3 (the 1.1 content), stock
Hyprland. The mesa T8132 patch drops a trailer line; its change is the
same.

UPSTREAM.md: the new heads of aurora-silicon/linux#1494, #1495 and #1496
(now with Dylan's Signed-off-by), omacom/omarchy#18582 and
omacom/omarchy-mac-pkgs#134. omacom/omarchy#18228 is a live release item
again: on 1.848 two NVMe namespaces fall back to mq-deadline on every
boot.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant