Re: Linux 2.4.6-ac4

[Udo A. Steinberg]

"Udo A. Steinberg" wrote:
> 
> Alan Cox wrote:
> 
> > 2.4.6-ac4
> > o       Update VIA southbridge bug fix to VIA provided  (me)
> >         workaround.
> >         | Except we apply it even when no sblive is
> >         | present
> 
> 
> I have a VIA 686a chipset and the workaround is being enabled nonetheless.
> I guess that's not quite correct. ac3 did the right thing, ac4 doesn't.

I kinda see why (relevant part of -ac4 patch):
The underlined piece won't ever execute, and the range doesn't reflect the
comment above.

+       p=pci_find_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686, NULL);
+       if(p!=NULL)
+       {
+               pci_read_config_byte(p, PCI_CLASS_REVISION, &rev);
+               /* 0x40 - 0x4f == 686B, 0x10 - 0x2f == 686A; thanks Dan Hollis
+               /* Check for buggy part revisions */
+               if (rev < 0x40 && rev > 0x42)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+                       return;
+       }
+       else
+       {
+               p = pci_find_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231, NULL);
+               if(p==NULL)     /* No problem parts */
+                       return;
+               pci_read_config_byte(p, PCI_CLASS_REVISION, &rev);
+               /* Check for buggy part revisions */
+               if (rev < 0x10 && rev > 0x12)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+                       return;
+       }