Hi

Am 11.07.22 um 13:13 schrieb Javier Martinez Canillas:

On 7/7/22 17:39, Thomas Zimmermann wrote:

quoted

Remove firmware devices on the PCI bus, by calling
aperture_remove_conflicting_pci_devices() in the probe function of
each related fbdev driver. iSo far, most of these drivers did not
remove conflicting VESA or EFI devices, or outride failed for
resource conflicts (i.e., matroxfb.) This must have been broken
for quite some time.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
---

[...]

quoted

@@ -949,6 +950,10 @@ static int ark_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)
  	int rc;
  	u8 regval;
  
+	rc = aperture_remove_conflicting_pci_devices(dev, "arkfb");
+	if (rc < 0)
+		return rc;
+

I wonder if we could think of a trick to avoid open coding the same check in
all drivers. Maybe a combination of using KBUILD_MODNAME for the name and
a probe callback wrapper or something ?

Originally, I tried to hack this into register_framebuffer(), where the 
removal was originally located. But that's too late as there is 
conflicting resource access in most driver's probe function. The top of 
the probe function is the only place where this call can realistically 
be located. It's the same rule as with DRM drivers: the firmware driver 
needs to be gone before the native driver touches hardware.

Best regards
Thomas

But probably not worth to invest more in the fbdev drivers and could be done
as a follow-up anyways if someone feels like it.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>

-- 
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
(HRB 36809, AG Nürnberg)
Geschäftsführer: Ivo Totev