DAC960 + sym53c8xx conflict

DAC960 + sym53c8xx conflict

[Michael]

I've got a 4-processor Intel SC450NX-based system with motherboard
Sym53c896 dual controller driven a Mylex Acceleraid 200. The motherboard
also has a sym53c810a which drives a Sony SDT-9000 tape drive. This
combo has always worked well until I tried updating to a 2.6 kernel.

When I load 2.6 (tried 2.6. - 2.6.8-rc4 (mm tree too)) every works OK
until I load the sym53c8xx module. When that module is loaded all access
to the drives connected to the DAC960 vanishes. I get no errors and the
system is not locked up. Any attempt to access the disks on the DAC960
simply hangs.

Any ideas? I hacked teh sym driver to remove all support for the 896
chip, but it didn't.

-- 
Michael <mozstuff@yahoo.com>

Re: DAC960 + sym53c8xx conflict

[Matthew Wilcox]

On Sun, Aug 15, 2004 at 10:53:07PM -0500, Michael wrote:
> I've got a 4-processor Intel SC450NX-based system with motherboard
> Sym53c896 dual controller driven a Mylex Acceleraid 200. The motherboard
> also has a sym53c810a which drives a Sony SDT-9000 tape drive. This
> combo has always worked well until I tried updating to a 2.6 kernel.
> 
> When I load 2.6 (tried 2.6. - 2.6.8-rc4 (mm tree too)) every works OK
> until I load the sym53c8xx module. When that module is loaded all access
> to the drives connected to the DAC960 vanishes. I get no errors and the
> system is not locked up. Any attempt to access the disks on the DAC960
> simply hangs.
> 
> Any ideas? I hacked teh sym driver to remove all support for the 896
> chip, but it didn't.

On 2.4, did you get the message:

        printf_info("%s: not initializing, driven by RAID controller.\n",
                                            sym_name(device));

-- 
"Next the statesmen will invent cheap lies, putting the blame upon 
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince 
himself that the war is just, and will thank God for the better sleep 
he enjoys after this process of grotesque self-deception." -- Mark Twain

Re: DAC960 + sym53c8xx conflict

[Matthew Wilcox]

On Mon, Aug 16, 2004 at 06:30:47PM -0500, Michael wrote:
> On 2.6 I get the message you list as well. I thought maybe 2.6 wasn't
> actually recognizing the raid controller, so I disabled support for the
> 896 chip in the driver so it gives an "unsupported" message, but I still
> lose the dac960 drives.

Crap, I see the problem.  When we fail initialisation, we disable the
device, so DAC960 stops working.  Either we need to not disable the device
when exiting or we need to claim the sym2 device in the dac960 driver
so the sym2 driver never gets called.

I'm going to be away till the end of the week, but for the moment, deleting

        pci_release_regions(pdev);
and
        pci_disable_device(pdev);

from sym2_probe() in drivers/scsi/sym53c8xx_2/sym_glue.c should fix
your problem.

-- 
"Next the statesmen will invent cheap lies, putting the blame upon 
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince 
himself that the war is just, and will thank God for the better sleep 
he enjoys after this process of grotesque self-deception." -- Mark Twain