Subject: Re: Crash in sbp2_abort()?
To: None <firstname.lastname@example.org>
From: Peter Seebach <email@example.com>
Date: 05/18/2003 20:22:18
In message <200305182225.h4IMPYKa011457@guild.plethora.net>, Peter Seebach writ
>Anyone out there using ieee1394 drives successfully in -current? I have
>a drive which used to work fine, but today, I get page fault panics in
This is stranger than I thought.
Trace says sbp2_abort was called from sbp2_free, called from sbpscsi_match.
In sbp2_abort, the parameter orb seems to have pointers to bogus data:
&orb->sbp2 = 0xc0fd801c
&orb->sbp2->orbs = 0xdeadbf33
This suggests that orb->sbp2 is very close to 0xdeadbeef - which shouldn't
happen. I have checked every point at which a new orb is allocated in sbp2.c,
and none of them seem to be forgetting to initialize it...
Just my luck this should turn up on the day when I have a drive failure and
need to use an external drive right this minute.