Subject: Re: ciss(9), aac(9) and Xen
To: None <tech-kern@netbsd.org>
From: Tonnerre LOMBARD <tonnerre@thebsh.sygroup.ch>
List: tech-kern
Date: 03/17/2006 09:45:21
--GYaKytDE8aa4+VVK
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Salut,
If I assign 1GB of RAM to the VM, I get a more valuable stack trace for
ciss(9) under XEN0:
ciss0 at pci5 dev 1 function 0: Compaq Smart Array 641
ciss0: interrupting at irq 48, event channel 12
(At this point, sc=0xC1A12600, sc->ccbs=0xCB597000)
uvm_fault(0xc06731a0, 0x6e624000, 0, 2) -> 0xe
kernel: page fault trap, code=0
Stopped in pid 0.1 (swapper) at netbsd:ciss_cmd+0x2be: movl %esi,0x58(%ebx)
ciss_cmd(cb616e00, 1, 3, c07ffa98, 0) at netbsd:ciss_cmd+0x2be
ciss_inq(c1a12600, cb617000, 1000, 0, d) at netbsd:ciss_inq+0x6b
ciss_attach(c1a12600, c1a1261c, c0654f40, c1a1261c, 10) at netbsd:ciss_attach+0x53
...
However, this points us to
ciss.c:513: ccb1 = (struct ciss_ccb *) (sc->ccbs + (id >> 2) * sc->ccblen);
I'm going to check lengths again, but this time it should be well within
bounds...
Any ideas?
Tonnerre
--GYaKytDE8aa4+VVK
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.1 (NetBSD)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=9PSY
-----END PGP SIGNATURE-----
--GYaKytDE8aa4+VVK--