f31522130f
Build system (5 gaps hardened): - COOKBOOK_OFFLINE defaults to true (fork-mode) - normalize_patch handles diff -ruN format - New 'repo validate-patches' command (25/25 relibc patches) - 14 patched Qt/Wayland/display recipes added to protected list - relibc archive regenerated with current patch chain Boot fixes (fixable): - Full ISO EFI partition: 16 MiB → 1 MiB (matches mini, BIOS hardcoded 2 MiB offset) - D-Bus system bus: absolute /usr/bin/dbus-daemon path (was skipped) - redbear-sessiond: absolute /usr/bin/redbear-sessiond path (was skipped) - daemon framework: silenced spurious INIT_NOTIFY warnings for oneshot_async services (P0-daemon-silence-init-notify.patch) - udev-shim: demoted INIT_NOTIFY warning to INFO (expected for oneshot_async) - relibc: comprehensive named semaphores (sem_open/close/unlink) replacing upstream todo!() stubs - greeterd: Wayland socket timeout 15s → 30s (compositor DRM wait) - greeter-ui: built and linked (header guard unification, sem_compat stubs removed) - mc: un-ignored in both configs, fixed glib/libiconv/pcre2 transitive deps - greeter config: removed stale keymapd dependency from display/greeter services - prefix toolchain: relibc headers synced, _RELIBC_STDLIB_H guard unified Unfixable (diagnosed, upstream): - i2c-hidd: abort on no-I2C-hardware (QEMU) — process::exit → relibc abort - kded6/greeter-ui: page fault 0x8 — Qt library null deref - Thread panics fd != -1 — Rust std library on Redox - DHCP timeout / eth0 MAC — QEMU user-mode networking - hwrngd/thermald — no hardware RNG/thermal in VM - live preload allocation — BIOS memory fragmentation, continues on demand
To test whether QScreen properties are updated properly when the screen actually changes, you will need to run some kind of control panel to make changes, and this test program at the same time. E.g. on Linux, you can use xrandr with various parameters on the command line, but there is also a nice GUI called arandr which will probably work on any distro. Real-world users would probably use the Gnome or KDE control panels, so that's also a good way to test. On OSX you can make changes in System Preferences | Displays, and you can also configure it to put a "monitors" icon on the menubar with a drop-down menu for convenience. On Windows you can right-click on the desktop to get display settings. Note that on Linux, if you have one graphics card with two outputs, typically the two monitors connected to the outputs are combined into a single virtual "screen", but each screen has multiple outputs. In that case there will be a unique QScreen for each output, and they will be virtual siblings. The virtual geometry depends on how you arrange the monitors (second one is to the right, or above the first one, for example). This test app will create two windows, and will center one each screen, by setting the geometry. Alternatively you can configure xorg.conf to create separate screens for each graphics card; then the mouse cursor can move between the screens, but application windows cannot: each app needs to be started on the screen that you want to run it on (by specifying e.g. DISPLAY=:0.1 for the second screen), or the application has to set the desired screen via QWindow::setScreen() before showing the window. The physical size of the screen is considered to be a constant. This can create discrepancies in DPI when orientation is changed, or when the screen is actually a VNC server and you change the resolution. So maybe QScreen::physicalSize should also have a notifier, but that doesn't physically make sense except when the screen is virtual. Another case is running two separate X servers on two graphics cards. In that case they really do not know about each other, even at the xlib/xcb level, so this test is irrelevant. You can run the test independently on each X server, but you will just get one QScreen instance on each.