[PATCH 04/11] video/aperture: use generic code to figure out the vga default device

[PATCH 04/11] video/aperture: use generic code to figure out the vga default device

[Daniel Vetter]

Since vgaarb has been promoted to be a core piece of the pci subsystem
we don't have to open code random guesses anymore, we actually know
this in a platform agnostic way, and there's no need for an x86
specific hack. See also 1d38fe6ee6a8 ("PCI/VGA: Move vgaarb to
drivers/pci")

This should not result in any functional change, and the non-x86
multi-gpu pci systems are probably rare enough to not matter (I don't
know of any tbh). But it's a nice cleanup, so let's do it.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: linux-pci@vger.kernel.org
---
 drivers/video/aperture.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
index 41e77de1ea82..3d8c925c7365 100644
--- a/drivers/video/aperture.c
+++ b/drivers/video/aperture.c
@@ -324,13 +324,11 @@ EXPORT_SYMBOL(aperture_remove_conflicting_devices);
  */
 int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *name)
 {
-	bool primary = false;
+	bool primary;
 	resource_size_t base, size;
 	int bar, ret;
 
-#ifdef CONFIG_X86
-	primary = pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW;
-#endif
+	primary = pdev == vga_default_device();
 
 	for (bar = 0; bar < PCI_STD_NUM_BARS; ++bar) {
 		if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM))
-- 
2.39.0

Re: [PATCH 04/11] video/aperture: use generic code to figure out the vga default device

[Thomas Zimmermann]

[-- Attachment #1.1: Type: text/plain, Size: 2779 bytes --]

Hi

Am 11.01.23 um 16:41 schrieb Daniel Vetter:
> Since vgaarb has been promoted to be a core piece of the pci subsystem
> we don't have to open code random guesses anymore, we actually know
> this in a platform agnostic way, and there's no need for an x86
> specific hack. See also 1d38fe6ee6a8 ("PCI/VGA: Move vgaarb to
> drivers/pci")
> 
> This should not result in any functional change, and the non-x86
> multi-gpu pci systems are probably rare enough to not matter (I don't
> know of any tbh). But it's a nice cleanup, so let's do it.
> 
> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
> Cc: Thomas Zimmermann <tzimmermann@suse.de>
> Cc: Javier Martinez Canillas <javierm@redhat.com>
> Cc: Helge Deller <deller@gmx.de>
> Cc: linux-fbdev@vger.kernel.org
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Cc: linux-pci@vger.kernel.org
> ---
>   drivers/video/aperture.c | 6 ++----
>   1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
> index 41e77de1ea82..3d8c925c7365 100644
> --- a/drivers/video/aperture.c
> +++ b/drivers/video/aperture.c
> @@ -324,13 +324,11 @@ EXPORT_SYMBOL(aperture_remove_conflicting_devices);
>    */
>   int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *name)
>   {
> -	bool primary = false;
> +	bool primary;
>   	resource_size_t base, size;
>   	int bar, ret;
>   
> -#ifdef CONFIG_X86
> -	primary = pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW;
> -#endif
> +	primary = pdev == vga_default_device();

vga_default_device() is part of vgaarb and can return NULL. [1] That new 
test is likely to be incorrect on many systems.

I suggest to implement a helper like fb_is_primary_device() on x86: it 
uses the default VGA if set, or falls back to the original test. [2]

It's noteworthy that on most architectures, fb_is_primary_device() 
returns 0. But at least on Sparc [3] and some Parisc [4] machines, it 
does not.

I've long wanted to rework this helper anyway. So this is a good 
opportunity.

Best regards
Thomas

[1] 
https://elixir.bootlin.com/linux/latest/source/include/linux/vgaarb.h#L69
[2] 
https://elixir.bootlin.com/linux/latest/source/arch/x86/video/fbdev.c#L14
[3] 
https://elixir.bootlin.com/linux/latest/source/arch/sparc/include/asm/fb.h#L18
[4] 
https://elixir.bootlin.com/linux/latest/source/drivers/video/console/sticore.c#L1153


>   
>   	for (bar = 0; bar < PCI_STD_NUM_BARS; ++bar) {
>   		if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM))

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

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 840 bytes --]

Re: [PATCH 04/11] video/aperture: use generic code to figure out the vga default device

[Daniel Vetter]

On Wed, Jan 11, 2023 at 04:59:30PM +0100, Thomas Zimmermann wrote:
> Hi
> 
> Am 11.01.23 um 16:41 schrieb Daniel Vetter:
> > Since vgaarb has been promoted to be a core piece of the pci subsystem
> > we don't have to open code random guesses anymore, we actually know
> > this in a platform agnostic way, and there's no need for an x86
> > specific hack. See also 1d38fe6ee6a8 ("PCI/VGA: Move vgaarb to
> > drivers/pci")
> > 
> > This should not result in any functional change, and the non-x86
> > multi-gpu pci systems are probably rare enough to not matter (I don't
> > know of any tbh). But it's a nice cleanup, so let's do it.
> > 
> > Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
> > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > Cc: Javier Martinez Canillas <javierm@redhat.com>
> > Cc: Helge Deller <deller@gmx.de>
> > Cc: linux-fbdev@vger.kernel.org
> > Cc: Bjorn Helgaas <bhelgaas@google.com>
> > Cc: linux-pci@vger.kernel.org
> > ---
> >   drivers/video/aperture.c | 6 ++----
> >   1 file changed, 2 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
> > index 41e77de1ea82..3d8c925c7365 100644
> > --- a/drivers/video/aperture.c
> > +++ b/drivers/video/aperture.c
> > @@ -324,13 +324,11 @@ EXPORT_SYMBOL(aperture_remove_conflicting_devices);
> >    */
> >   int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *name)
> >   {
> > -	bool primary = false;
> > +	bool primary;
> >   	resource_size_t base, size;
> >   	int bar, ret;
> > -#ifdef CONFIG_X86
> > -	primary = pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW;
> > -#endif
> > +	primary = pdev == vga_default_device();
> 
> vga_default_device() is part of vgaarb and can return NULL. [1] That new
> test is likely to be incorrect on many systems.
> 
> I suggest to implement a helper like fb_is_primary_device() on x86: it uses
> the default VGA if set, or falls back to the original test. [2]
> 
> It's noteworthy that on most architectures, fb_is_primary_device() returns
> 0. But at least on Sparc [3] and some Parisc [4] machines, it does not.

Afaik these neither do legacy vga nor sysfb, so we should be fine. I'll
augment the commit message.

> I've long wanted to rework this helper anyway. So this is a good
> opportunity.

Hm yeah that sounds like a good thing to copy. I'm honestly not sure
whether it's needed, but at least it shouldn't hurt. I thought at least
that the boot default device should be set on all pci architectures.

I'll also reorder this with the previous patch to avoid the compile fail.
-Daniel

> 
> Best regards
> Thomas
> 
> [1]
> https://elixir.bootlin.com/linux/latest/source/include/linux/vgaarb.h#L69
> [2]
> https://elixir.bootlin.com/linux/latest/source/arch/x86/video/fbdev.c#L14
> [3] https://elixir.bootlin.com/linux/latest/source/arch/sparc/include/asm/fb.h#L18
> [4] https://elixir.bootlin.com/linux/latest/source/drivers/video/console/sticore.c#L1153
> 
> 
> >   	for (bar = 0; bar < PCI_STD_NUM_BARS; ++bar) {
> >   		if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM))
> 
> -- 
> Thomas Zimmermann
> Graphics Driver Developer
> SUSE Software Solutions Germany GmbH
> Maxfeldstr. 5, 90409 Nürnberg, Germany
> (HRB 36809, AG Nürnberg)
> Geschäftsführer: Ivo Totev




-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch