lookups with 2.1.8x

[Robert G. Brown]

Regarding the 2.1.X lockup when scanning the onboard 2940 on a P6DLS:

I have a bit more data for anyone listening (possibly Doug Ledford?).  I
put a bunch of printk's in aic7xxx_handle_scsiint and in aic7xxx_queue
(where the last message I get before the hang is generated.  As far as I
can tell, the SCB is built correctly and the queues are constructed and
run without incident -- the kernel returns from the routine still
unhung, as far as I can tell.  However, I don't know enough (yet) about
the scsi probe and initialization to tell where this is called from or
what is done next to start putting printk's in there.  Since the NEXT
line that comes out of a normal aic7xxx initialization is (e.g., from
2.0.33 boot):

(scsi0:0:0:0) Using wide (16 bit) transfers.
(scsi0:0:0:0) Synchronous at 20.0MHz, offset 8.
  Vendor: SEAGATE   Model: ST19101W          Rev: 0014
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sda at scsi0, channel 0, id 0, lun 0
(scsi0:0:1:0) Using wide (16 bit) transfers.
(scsi0:0:1:0) Synchronous at 20.0MHz, offset 8.
  Vendor: SEAGATE   Model: ST34371W          Rev: 0338
...

generated in aic7xxx_parse_msg() called (I believe) from
aic7xxx_handle_scsiint(), I expected to get some response from the
latter, but it was not invoked at all.  So, somewhere between and the
bus_width switch in the controller, 2.1.X goes off to neverland.  From
the observation that this appears to occur for more than one kind of
SCSI controller (a buslogic, I believe, had the same problem with
2.1.(x>82)) it is probably in a non-driver specific section of the scsi
initialization process.  If you give me a hint where to look, I'll dump
a bunch more printk's in and try to narrow it down some more.

   rgb

Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu





To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message