Port-i386 archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: SPL NOT ZERO ON SYSCALL ENTRY




To: Matthias Drochner <M.Drochner%fz-juelich.de@localhost>

Subject: Re: SPL NOT ZERO ON SYSCALL ENTRY

From: Bill Stouder-Studenmund <wrstuden%netbsd.org@localhost>

Date: Tue, 11 Mar 2008 14:29:31 -0700


On Tue, Mar 11, 2008 at 08:33:31PM +0100, Matthias Drochner wrote:
> 
> Just got that panic, on a dual-core i386.
> It happened during heavy file system activity while
> accelerated X apps were active. I've seen a couple
> of DDB traps in that situation in recent days, but
> was never able to get a coredump before.
> 
> What could cause that panic? splxxx() use in interrupt
> context?

My guess is the problem was a missing splx() in an error-handling case. 
You got busy-enough to trigger a transient error, and so you found a bad 
error-handlng path. Somewhere. :-(

Take care,

Bill

Attachment: pgpG_FSH39NwT.pgp
Description: PGP signature


Follow-Ups:

Re: SPL NOT ZERO ON SYSCALL ENTRY
From: Matthias Drochner


References:

SPL NOT ZERO ON SYSCALL ENTRY
From: Matthias Drochner




Prev by Date: SPL NOT ZERO ON SYSCALL ENTRY

Next by Date: Re: Shuffling INSTALL kernels

Previous by Thread: SPL NOT ZERO ON SYSCALL ENTRY

Next by Thread: Re: SPL NOT ZERO ON SYSCALL ENTRY

Indexes:

reverse Date

reverse Thread

Old Index



Home | Main Index | Thread Index | Old Index