Conflict when loading initio driver

Conflict when loading initio driver

[Scott Simpson]

 
  
  
  
  
  
  
  
   <!--
#toc,
.toc,
.mw-warning {
	border: 1px solid #aaa;
	background-color: #f9f9f9;
	padding: 5px;
	font-size: 95%;
}
#toc h2,
.toc h2 {
	display: inline;
	border: none;
	padding: 0;
	font-size: 100%;
	font-weight: bold;
}
#toc #toctitle,
.toc #toctitle,
#toc .toctitle,
.toc .toctitle {
	text-align: center;
}
#toc ul,
.toc ul {
	list-style-type: none;
	list-style-image: none;
	margin-left: 0;
	padding-left: 0;
	text-align: left;
}
#toc ul ul,
.toc ul ul {
	margin: 0 0 0 2em;
}
#toc .toctoggle,
.toc .toctoggle {
	font-size: 94%;
}@media print, projection, embossed {
	body {
		padding-top:1in;
		padding-bottom:1in;
		padding-left:1in;
		padding-right:1in;
	}
}
body {
	font-family:'Times New Roman';
	color:#000000;
	widows:2;
	font-style:normal;
	text-indent:0in;
	font-variant:normal;
	font-size:12pt;
	text-decoration:none;
	font-weight:normal;
	text-align:left;
}
table {
}
td {
	border-collapse:collapse;
	text-align:left;
	vertical-align:top;
}
p, h1, h2, h3, li {
	color:#000000;
	font-family:'Times New Roman';
	font-size:12pt;
	text-align:left;
	vertical-align:normal;
}
     -->
  
 
  
   SuSE 10.3 (new release candidate), 2.6.22.5-29 kernel, SCSI initio driver



   

   I have a scanner connected to a Initio INI-950 SCSI card and I recently upgraded from SuSE 10.2 to 10.3. The new kernel doesn't see any of my devices. I get the following in /var/log/messages:



   

   Sep 30 09:05:13 r2d2 kernel: ACPI: PCI Interrupt 0000:00:0a.0[A] -> GSI 17 (level, low) -> IRQ 16

   Sep 30 09:05:13 r2d2 kernel: initio: I/O port range 0x0 is busy.

   Sep 30 09:05:13 r2d2 kernel: ACPI: PCI interrupt for device 0000:00:0a.0 disabled



   

   so there is some conflict of some kind. How would I go about reconciling this? Thanks.

   


  

 



       
____________________________________________________________________________________
Pinpoint customers who are looking for what you sell. 
http://searchmarketing.yahoo.com/

Re: Conflict when loading initio driver

[Alan Cox]

>    I have a scanner connected to a Initio INI-950 SCSI card and I recently upgraded from SuSE 10.2 to 10.3. The new kernel doesn't see any of my devices. I get the following in /var/log/messages:
> 
> 
> 
>    
> 
>    Sep 30 09:05:13 r2d2 kernel: ACPI: PCI Interrupt 0000:00:0a.0[A] -> GSI 17 (level, low) -> IRQ 16
> 
>    Sep 30 09:05:13 r2d2 kernel: initio: I/O port range 0x0 is busy.
> 
>    Sep 30 09:05:13 r2d2 kernel: ACPI: PCI interrupt for device 0000:00:0a.0 disabled

Humm not a collision - thats a bug in the driver updating. Looks like the
changes I made and combined with Christoph's lost a line somewhere when I
was merging it all. Try the following

Signed-off-by: Alan Cox <alan@redhat.com>

diff -u --new-file --exclude-from /usr/src/exclude --recursive linux.vanilla-2.6.23rc8-mm1/drivers/scsi/initio.c linux-2.6.23rc8-mm1/drivers/scsi/initio.c
--- linux.vanilla-2.6.23rc8-mm1/drivers/scsi/initio.c	2007-09-26 16:46:54.000000000 +0100
+++ linux-2.6.23rc8-mm1/drivers/scsi/initio.c	2007-09-30 22:22:03.839113616 +0100
@@ -2867,6 +2867,7 @@
 	}
 	host = (struct initio_host *)shost->hostdata;
 	memset(host, 0, sizeof(struct initio_host));
+	host->addr = pci_resource_start(pdev, 0);
 
 	if (!request_region(host->addr, 256, "i91u")) {
 		printk(KERN_WARNING "initio: I/O port range 0x%x is busy.\n", host->addr);

Re: Conflict when loading initio driver

[Scott Simpson]

Alan Cox wrote:
> Humm not a collision - thats a bug in the driver updating. Looks like the
> changes I made and combined with Christoph's lost a line somewhere when I
> was merging it all. Try the following
>
> Signed-off-by: Alan Cox <alan@redhat.com>
>
> diff -u --new-file --exclude-from /usr/src/exclude --recursive linux.vanilla-2.6.23rc8-mm1/drivers/scsi/initio.c linux-2.6.23rc8-mm1/drivers/scsi/initio.c
> --- linux.vanilla-2.6.23rc8-mm1/drivers/scsi/initio.c	2007-09-26 16:46:54.000000000 +0100
> +++ linux-2.6.23rc8-mm1/drivers/scsi/initio.c	2007-09-30 22:22:03.839113616 +0100
> @@ -2867,6 +2867,7 @@
>  	}
>  	host = (struct initio_host *)shost->hostdata;
>  	memset(host, 0, sizeof(struct initio_host));
> +	host->addr = pci_resource_start(pdev, 0);
>
>  	if (!request_region(host->addr, 256, "i91u")) {
>  		printk(KERN_WARNING "initio: I/O port range 0x%x is busy.\n", host->addr);

I tried this fix on my SuSE 10.3 system (2.6.22.5-29 kernel) and it
didn't work. The system froze on boot. I think there might be more to
it.
     Scott

Re: Conflict when loading initio driver

[Alan Cox]

> > +	host->addr = pci_resource_start(pdev, 0);
> >
> >  	if (!request_region(host->addr, 256, "i91u")) {
> >  		printk(KERN_WARNING "initio: I/O port range 0x%x is busy.\n", host->addr);
> 
> I tried this fix on my SuSE 10.3 system (2.6.22.5-29 kernel) and it
> didn't work. The system froze on boot. I think there might be more to
> it.

Can you stick some printk statements in so we can see where it hangs
roughly ? 

Re: Conflict when loading initio driver

[Scott Simpson]

OK. I did some debugging in this initio.c driver. I put printk
statements at the beginning of each routine and I got the following
output:

initio_se2_rd
initio_se2_instr
(The previous two routines repeat a bunch of times then I get:)
initio_stop_bm
i91u: PCI Base=0xEC00, IRQ=0, BIOS=0x0000, SCSI ID=7
i91u: Reset SCSI Bus...
initio_reset_scsi
scsi2: Initio INI-9X00U/UW SCSI device driver
i91u_queuecommand
initio_alloc_scb

So it seems to be dieing in initio_alloc_scb. I think it is dieing on
the first spin lock if that makes any sense, but it could be dieing
somewhere else in that routine. Hope this helps.