To answer your question publicly, GLib test suite fails by design if some log is reported using warning or higher level without being explicitly handled
-- You received this bug notification because you are a member of Desktop Packages, which is subscribed to gtk4 in Ubuntu. https://bugs.launchpad.net/bugs/2077313 Title: GTK 4.15.5: memorytexture, dmabufformats and scaling tests fail Status in GTK+: New Status in gtk4 package in Ubuntu: Fix Released Bug description: MERGE REQUEST https://salsa.debian.org/gnome- team/gtk4/-/merge_requests/25 UPSTREAM BUG: https://gitlab.gnome.org/GNOME/gtk/-/issues/6947 These GTK 4.15.5 tests fail: ---> gtk:gdk / memorytexture ERROR 0.33s killed by signal 5 SIGTRAP gtk:gdk / dmabufformats ERROR 0.21s killed by signal 5 SIGTRAP gtk:gsk / scaling ERROR 0.25s killed by signal 5 SIGTRAP <--- as per https://launchpad.net/~jbicha/+archive/ubuntu/arch/+sourcepub/16394764/+listing- archive-extra. The full log of memorytexture is [1]. Its backtrace is ---> #0 0x00007ffff7ecf5fa in g_log_writer_default () at /lib/x86_64-linux-gnu/libglib-2.0.so.0 #1 0x00007ffff7ec6d64 in g_log_structured_array () at /lib/x86_64-linux-gnu/libglib-2.0.so.0 #2 0x00007ffff7ec6fb8 in g_log_structured_standard () at /lib/x86_64-linux-gnu/libglib-2.0.so.0 #3 0x00007ffff78c0db7 in gdk_vulkan_handle_result (res=<optimized out>, called_function=0x7ffff7a10955 "vkCreateInstance") at ../../../gdk/gdkvulkancontextprivate.h:68 #4 gdk_vulkan_handle_result (res=<optimized out>, called_function=0x7ffff7a10955 "vkCreateInstance") at ../../../gdk/gdkvulkancontextprivate.h:63 #5 gdk_display_create_vulkan_instance (display=0x5555555a14f0, error=<optimized out>) at ../../../gdk/gdkvulkancontext.c:1675 #6 0x00007ffff7961ef3 in gdk_display_init_vulkan (display=0x5555555a14f0, error=0x0) at ../../../gdk/gdkvulkancontext.c:1771 #7 gsk_vulkan_device_get_for_display (display=0x5555555a14f0, error=0x0) at ../../../gsk/gpu/gskvulkandevice.c:421 #8 0x00007ffff794c717 in gsk_gpu_renderer_realize (renderer=0x555560defa00, display=0x5555555a14f0, surface=0x0, error=0x0) at ../../../gsk/gpu/gskgpurenderer.c:213 #9 0x00007ffff78d863b in gsk_renderer_do_realize (renderer=0x555560defa00, display=display@entry=0x5555555a14f0, surface=surface@entry=0x0, error=error@entry=0x0) at ../../../gsk/gskrenderer.c:271 #10 0x00007ffff78df4c5 in gsk_renderer_realize_for_display (renderer=<optimized out>, display=display@entry=0x5555555a14f0, error=error@entry=0x0) at ../../../gsk/gskrenderer.c:357 #11 0x0000555555556a41 in main (argc=<optimized out>, argv=<optimized out>) at ../../../testsuite/gdk/memorytexture.c:1587 <--- gdkvulkancontextprivate.h:68 at #3 is nothing but the warning: g_warning ("%s(): %s (%d)", called_function, gdk_vulkan_strerror (res), res); So I must conclude the test is failing because the warning is being called, as if G_DEBUG=fatal-warnings had been passed, which it is however not. Open questions are: 1. Why is this causing a SIGTRAP if no fatal-warnings is enabled? 2. Why do other tests (e.g. 511 gtk:reftest / reftest actionbar.ui) that raise the same warning not fail? Anyway, those three tests pass with vulkan-disable (number of skipped remains unaltered) as a matter of fact. I'm proposing that as a merge request[2] in Debian's bug tracking system. [1] https://l.termbin.com/a0e2 [2] https://salsa.debian.org/gnome-team/gtk4/-/merge_requests/25 To manage notifications about this bug go to: https://bugs.launchpad.net/gtk/+bug/2077313/+subscriptions -- Mailing list: https://launchpad.net/~desktop-packages Post to : desktop-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~desktop-packages More help : https://help.launchpad.net/ListHelp
