DRI not working

Alex mailinglist at ahhyes.net
Thu Oct 2 13:56:28 UTC 2008


Robert Noland wrote:

> Ok, can you guys give this patch a try?  I think the memory failures are
> un-related and we probably need to try and figure that out separately.
> This patch should be rejecting the second head during the probe phase,
> so drm should not report *anything* about it.
>
> One of you said something about X also reporting both heads?  If that is
> true and remains so with this patch, could you send me X logs?  I'll try
> and get someone to look into the Intel 2d driver and see what is going
> on with it.
>
> This patch is against -CURRENT.  If I need one for STABLE, let me know
> and I'll work out the difference.
>
> robert.
>  
> 
  
Hi Robert,

The patch did not apply cleanly. I cvsup'd my entire source tree to 
RELENG_7 and the patch still wouldn't apply, so had to resort to 
inserting the code myself (perhaps I need a patch for STABLE)

* Added

#include <dev/pci/pcireg.h>

to drmP.h

* Added the following code to drm_probe() in drm_drv.c

	if (pci_get_class(dev) != PCIC_DISPLAY
	    || pci_get_subclass(dev) != PCIS_DISPLAY_VGA)
		return ENXIO;


Kernel build is now broken, code in drm_drv.c throws undefined references to 
PCIC_DISPLAY


Any ideas? 





More information about the freebsd-x11 mailing list