[PATCH] drm/amdgpu: only remove existing FBs for devices with displays

[Christian König]

Am 16.01.24 um 15:39 schrieb Alex Deucher:
> Seems calling drm_aperture_remove_conflicting_pci_framebuffers()
> will take away the apertures for unrelated devices on some kernel
> versions.  E.g., calling this on a PCIe accelerator with no display
> IP may take the apertures away from the actual PCIe display adapter
> on the system, breaking efifb, depending on the kernel version.
>
> Just do this if there is display IP present.

I would have checked the PCI device type instead, because a system BIOS 
most likely has no idea that a VGA device doesn't has a connector.

> Signed-off-by: Alex Deucher <alexander.deucher at amd.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 10 ++++++----
>   1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 62772b58ef3d..353c38f008e8 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -4056,10 +4056,12 @@ int amdgpu_device_init(struct amdgpu_device *adev,
>   
>   	amdgpu_device_set_mcbp(adev);
>   
> -	/* Get rid of things like offb */
> -	r = drm_aperture_remove_conflicting_pci_framebuffers(adev->pdev, &amdgpu_kms_driver);
> -	if (r)
> -		return r;
> +	if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_DCE)) {

This certainly worth a comment why we do this.

Apart from that I'm not sure we should upstream this, the customer 
kernel is most likely just missing this fix here:

commit 5ae3716cfdcd286268133867f67d0803847acefc
Author: Daniel Vetter <daniel.vetter at ffwll.ch>
Date:   Thu Apr 6 15:21:07 2023 +0200

     video/aperture: Only remove sysfb on the default vga pci device

     Instead of calling aperture_remove_conflicting_devices() to remove the
     conflicting devices, just call to aperture_detach_devices() to detach
     the device that matches the same PCI BAR / aperture range. Since the
     former is just a wrapper of the latter plus a sysfb_disable() call,
     and now that's done in this function but only for the primary devices.

     This fixes a regression introduced by commit ee7a69aa38d8 ("fbdev:
     Disable sysfb device registration when removing conflicting FBs"),
     where we remove the sysfb when loading a driver for an unrelated pci
     device, resulting in the user losing their efifb console or similar.

     Note that in practice this only is a problem with the nvidia blob,
     because that's the only gpu driver people might install which does not
     come with an fbdev driver of it's own. For everyone else the real gpu
     driver will restore a working console.

     Also note that in the referenced bug there's confusion that this same
     bug also happens on amdgpu. But that was just another amdgpu specific
     regression, which just happened to happen at roughly the same time and
     with the same user-observable symptoms. That bug is fixed now, see
     https://bugzilla.kernel.org/show_bug.cgi?id=216331#c15

Regards,
Christian.

> +		/* Get rid of things like offb */
> +		r = drm_aperture_remove_conflicting_pci_framebuffers(adev->pdev, &amdgpu_kms_driver);
> +		if (r)
> +			return r;
> +	}
>   
>   	/* Enable TMZ based on IP_VERSION */
>   	amdgpu_gmc_tmz_set(adev);