An alternative to requiring the display is to mark the attached display as empty (no available CRTC, no outputs) and let the tests skip if they required any output. This allows the binary to open the display once in its global fixture, even if it doesn't use the display in every subtest and so avoid skipping the subtests that didn't require the display. Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> --- lib/igt_kms.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lib/igt_kms.c b/lib/igt_kms.c index 0e6f91475..b38d64415 100644 --- a/lib/igt_kms.c +++ b/lib/igt_kms.c @@ -1857,7 +1857,8 @@ void igt_display_init(igt_display_t *display, int drm_fd) display->drm_fd = drm_fd; resources = drmModeGetResources(display->drm_fd); - igt_require(resources); + if (!resources) + goto out; /* * We cache the number of pipes, that number is a physical limit of the @@ -2005,6 +2006,7 @@ void igt_display_init(igt_display_t *display, int drm_fd) igt_display_reset(display); igt_kms_disallow_hotplug(drm_fd); +out: LOG_UNINDENT(display); } -- 2.19.0 _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx