[2.6 patch] fix aic7xxx_old.c for !PCI

[2.6 patch] fix aic7xxx_old.c for !PCI

[Adrian Bunk]

I got the following compile error in 2.6.6-mm1 (but it's not specific to 
-mm) with CONFIG_PCI=n:

<--  snip  -->

...
  CC      drivers/scsi/aic7xxx_old.o
drivers/scsi/aic7xxx_old.c: In function `aic7xxx_release':
drivers/scsi/aic7xxx_old.c:10971: warning: implicit declaration of function `pci_release_regions'
...
  LD      .tmp_vmlinux1
drivers/built-in.o(.text+0x2bf0ff): In function `aic7xxx_release':
: undefined reference to `pci_release_regions'
make: *** [.tmp_vmlinux1] Error 1

<--  snip  -->

The patch below fixes this issue.

Please apply
Adrian

--- linux-2.6.6-mm1-full/drivers/scsi/aic7xxx_old.c.old	2004-05-13 01:23:42.000000000 +0200
+++ linux-2.6.6-mm1-full/drivers/scsi/aic7xxx_old.c	2004-05-13 01:25:05.000000000 +0200
@@ -9675,7 +9675,9 @@
           found++;
 	  continue;
 skip_pci_controller:
+#ifdef CONFIG_PCI
 	  pci_release_regions(temp_p->pdev);
+#endif
 	  kfree(temp_p);
         }  /* Found an Adaptec PCI device. */
         else /* Well, we found one, but we couldn't get any memory */
@@ -10967,8 +10969,10 @@
 #endif /* MMAPIO */
   if(!p->pdev)
     release_region(p->base, MAXREG - MINREG);
+#ifdef CONFIG_PCI
   else
     pci_release_regions(p->pdev);
+#endif
   prev = NULL;
   next = first_aic7xxx;
   while(next != NULL)

Re: [2.6 patch] fix aic7xxx_old.c for !PCI

[Arjan van de Ven]

[-- Attachment #1: Type: text/plain, Size: 235 bytes --]

On Thu, 2004-05-13 at 01:55, Adrian Bunk wrote:
> I got the following compile error in 2.6.6-mm1 (but it's not specific to 
> -mm) with CONFIG_PCI=n:

or how about just providing a dummy pci_release_regions() for the !PCI
case ?

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [2.6 patch] fix aic7xxx_old.c for !PCI

[Adrian Bunk]

On Thu, May 13, 2004 at 09:41:53AM +0200, Arjan van de Ven wrote:
> On Thu, 2004-05-13 at 01:55, Adrian Bunk wrote:
> > I got the following compile error in 2.6.6-mm1 (but it's not specific to 
> > -mm) with CONFIG_PCI=n:
> 
> or how about just providing a dummy pci_release_regions() for the !PCI
> case ?

There's no dummy pci_request_regions(), and it might be hard to find a 
return value for a dummy pci_request_regions() that covers all cases.

I'm not religious about this issue, but a dummy pci_release_regions() 
without a dummy pci_request_regions() is a bit asymmetric.

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed