On Feb 2, 2007, at 6:42 PM, Wakko Warner wrote:
Andrew Morton wrote:
Yes, getting the oops traces will help, thanks. And confirmation
on a more recent kernel would be good.
Here's what I get. I used netconsole so whatever was logged prior
to it
starting was lost.
The PC is a suprtmicro x5da8 with an onboard dual channel AHA-39320
u320
controller. I have a dual channel AHA-39160 u160 and a dual channel
AHA-2940U/UW (ch0/internal is wide/narrow, ch1/externel is narrow).
I have used an x6-class Supermicro motherboard with the Adaptec u320
controller and I had problems hot-swapping drives with 2.6.18. It
seemed that the bus reset that the backplane processor generated
caused trouble for the driver, killing the SCSI bus. 2.6.16 and
2.6.17 locked up the kernel in the case of hot-swapping drives.
I switched to 2.6.19.2 and things are better. I did find that a card
dump is produced when hot-inserting a drive, so it is way noisier
than I think it should be, but it continues to operate and life goes
on, which is much better behavior than 2.6.16, 2.6.17 or 2.6.18.
I thought it was because I had option roms turned off, but when I
turned
them on, it still has problems. What's odd is the fact that if I
boot with init=/bin/sh, modprobe aic7xxx, it works fine and I can
exec init and it works fine.
I don't know what is up with that, but based on what I have seen I
would recommend using 2.6.19.x instead of 2.6.18 for systems using
the aic79xx driver.
--
Mark Rustad, MRustad@xxxxxxxxx
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
