Re: libATA PATA status report, new patch

[Alan Cox <alan@lxorguk.ukuu.org.uk>]

On Mer, 2006-02-08 at 20:28 +0200, Meelis Roos wrote:
> A new kind of crash on ALi chipset laptop (Toshiba Satellite 1800), 
> lspci below. I could not use netconsole on this laptop since yenta is 
> initialized after netconsole and so my cardbus NIC is not usable for 
> netconsole. I captured the details using a digital camera, in 2 
> different vga resolutions (to see different parts of it):
> 
> http://www.cs.ut.ee/~mroos/alicrash1.png
> http://www.cs.ut.ee/~mroos/alicrash2.png


Thanks: Utterly dumb bug made while converting to the newer refcounting
PCI API

--- drivers/scsi/pata_ali.c~	2006-02-09 15:17:06.430326208 +0000
+++ drivers/scsi/pata_ali.c	2006-02-09 15:17:06.431326056 +0000
@@ -528,7 +528,7 @@
 		pci_write_config_byte(pdev, 0x4B, tmp | 0x08);
 	}
 		
-	north = pci_get_slot(0, PCI_DEVFN(0,0));
+	north = pci_get_slot(pdev->bus, PCI_DEVFN(0,0));
 	isa_bridge = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, NULL);
 	
 	if(north && north->vendor == PCI_VENDOR_ID_AL) {