Re: Problems with Adaptec 2940 as a module

Eric Youngdale (eric@sub2317.jic.com)
Mon, 21 Oct 1996 23:12:47 -0400


>It shows up a little more spectacularly with Red Hat Linux 4.0
>because we use initrd and SCSI modules to have a single boot
>image for everyone. Unfortunately, it took a while to track
>down, because at least some of these drivers appeared to work
>for some people and not others, and we're still working on getting
>more information. But I think we've got enough information at this
>point to indicate that this isn't a red herring.

If you could come up with a list of host adapters, amount of system
memory, and an indication of success/failure, it would be helpful. Also, for
those cases that fail, there should be the usual bootup messages that come
along with the manufacturer/model/etc of each device, and it would be helpful
to see whether this much is correct.

As I see it, there are two possible problems. One possibility is that
the bus isn't being scanned correctly, and as a result the devices are not
being found. The other possibility is that the bus is scanned correctly, but
the devices are not correctly attached to the high level drivers.

For the end user, the symptoms would be the same, namely an inability
to talk to a given device.

One other useful thing to try is to see whether the problem persists
if only the low-level driver is a module (i.e. the scsi core and the high
level drivers are not modules).

If there is anyone willing to try some test patches to the kernel,
I could easily add a bunch of debugging printk's to help narrow this down.

-Eric