Date: Mon, 4 May 1998 12:21:58 -0500 (EST) From: "John S. Dyson" <dyson@FreeBSD.ORG> To: adkin003@tc.umn.edu (dave adkins) Cc: dyson@FreeBSD.ORG, adkin003@tc.umn.edu, freebsd-current@FreeBSD.ORG Subject: Re: cvs commit: src/sys/kern sysv_shm.c src/sys/vm swap_pager.c vm_fault.c vm_map.c Message-ID: <199805041721.MAA00588@dyson.iquest.net> In-Reply-To: <Pine.NEB.3.96.980504115659.1943A-100000@samthedog> from dave adkins at "May 4, 98 12:07:33 pm"
index | next in thread | previous in thread | raw e-mail
dave adkins said:
>
>
> On Mon, 4 May 1998, John S. Dyson wrote:
>
> > What you are seeing is a "DIAGNOSTIC" panic that really isn't
> > indicating a problem. I am reworking the issue, but for now, just
> > disable diagnostic. I'll be committing a fix in an hour.
> >
>
> John,
>
> It panics without the DIAGNOSTIC message at the line:
>
> if ((robject->handle == NULL) &&
> (robject->type == OBJT_DEFAULT ||
> robject->type == OBJT_SWAP)) {
>
> trying to derefence robject->handle.
>
> I put in an additional test to prevent the dereference through NULL:
>
> if ( robject && (robject->handle == NULL) &&
> (robject->type == OBJT_DEFAULT ||
> robject->type == OBJT_SWAP)) {
>
>
> and have been running with it for a few hours. I know its not a real fix,
> just a little more info on where I'm failing.
>
Cool, I have committed a fix. Your handle idea is good, but I can imagine
times that the handle check isn't sufficient (note that the split code
can do damage to the shm structure.) Instead of depending on the handle,
I have added a flag. You have done good detective work, and -current should
be okay as of now. (I still have a softupdates problem, that I broke again,
so will be fixing that unrelated problem.)
--
John | Never try to teach a pig to sing,
dyson@freebsd.org | it just makes you look stupid,
jdyson@nc.com | and it irritates the pig.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199805041721.MAA00588>