Re: pkg/35416: agp driver hangs on ALi/ULi M1689 chipset

[Nicolas Joly <njoly%pasteur.fr@localhost>]

The following reply was made to PR kern/35416; it has been noted by GNATS.

From: Nicolas Joly <njoly%pasteur.fr@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: kern-bug-people%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
        netbsd-bugs%netbsd.org@localhost, gary%duzan.org@localhost
Subject: Re: pkg/35416: agp driver hangs on ALi/ULi M1689 chipset
Date: Sat, 5 Apr 2008 15:17:58 +0200

 On Sat, Apr 05, 2008 at 11:45:01AM +0000, Martin Husemann wrote:
 >  On Sat, Apr 05, 2008 at 07:23:40AM -0400, Gary Duzan wrote:
 >  >    I believe so. However, the problem isn't the size, it is that
 >  > the wrong driver attaches, and I assume the aperture size is being
 >  > pulled from the wrong place.
 >  
 >  The check for zero size should be added there, however the right fix is
 >  to add an agp_match function in agp.c:
 [...]
 
 Can you try this patch ?
 
 Index: sys/dev/pci/agp.c
 ===================================================================
 RCS file: /cvsroot/src/sys/dev/pci/agp.c,v
 retrieving revision 1.55
 diff -u -p -r1.55 agp.c
 --- sys/dev/pci/agp.c  29 Feb 2008 06:13:39 -0000      1.55
 +++ sys/dev/pci/agp.c  5 Apr 2008 13:10:34 -0000
 @@ -119,6 +119,11 @@ const struct agp_product {
        int             (*ap_match)(const struct pci_attach_args *);
        int             (*ap_attach)(struct device *, struct device *, void *);
  } agp_products[] = {
 +#if NAGP_AMD64 > 0
 +      { PCI_VENDOR_ALI,       PCI_PRODUCT_ALI_M1689,
 +        agp_amd64_match,      agp_amd64_attach },
 +#endif
 +
  #if NAGP_ALI > 0
        { PCI_VENDOR_ALI,       -1,
          NULL,                 agp_ali_attach },
 
 
 -- 
 Nicolas Joly
 
 Biological Software and Databanks.
 Institut Pasteur, Paris.