Hi
Am 05.04.24 um 10:34 schrieb Hogander, Jouni:
[...]
diff --git a/drivers/gpu/drm/i915/i915_driver.c
b/drivers/gpu/drm/i915/i915_driver.c
index e0f13c62a1832..69178b73845e1 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -816,6 +816,8 @@ int i915_driver_probe(struct pci_dev *pdev, const
struct pci_device_id *ent)
i915->do_release = true;
+ intel_fbdev_setup(i915);
+
This doesn't work for Xe. I propose you move it to
drivers/gpu/drm/i915/display/intel_display_dirver.c:intel_display_drive
r_probe? Otherwise patch looks ok to me.
The client setup needs to run after the DRM device has been registered,
so _probe is too early. intel_display_driver_register() looks like a
possible place to put this. That would be symmetric to unregistering in
intel_display_driver_unregister().
Best regards
Thomas
BR,
Jouni Högander
return 0;
out_cleanup_gem:
diff --git a/drivers/gpu/drm/xe/display/xe_display.c
b/drivers/gpu/drm/xe/display/xe_display.c
index cdbc3f04c80a7..ca5cbe1d8a03b 100644
--- a/drivers/gpu/drm/xe/display/xe_display.c
+++ b/drivers/gpu/drm/xe/display/xe_display.c
@@ -214,9 +214,7 @@ void xe_display_fini(struct xe_device *xe)
if (!xe->info.enable_display)
return;
- /* poll work can call into fbdev, hence clean that up
afterwards */
intel_hpd_poll_fini(xe);
- intel_fbdev_fini(xe);
intel_hdcp_component_fini(xe);
intel_audio_deinit(xe);
--
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstrasse 146, 90461 Nuernberg, Germany
GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman
HRB 36809 (AG Nuernberg)
