Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 20 Jul 2004 00:54:34 -0600
From:      Scott Long <scottl@freebsd.org>
To:        Daniel Eriksson <daniel_k_eriksson@telia.com>
Cc:        freebsd-current@freebsd.org
Subject:   Re: HEADS UP [Re: thread+preemption stability improvement]
Message-ID:  <40FCC1AA.9060609@freebsd.org>
In-Reply-To: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA0VcX9IoJqUaXPS8MjT1PdsKAAAAQAAAAdmIYthj07UGjiabwmpYACgEAAAAA@telia.com>
References:  <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA0VcX9IoJqUaXPS8MjT1PdsKAAAAQAAAAdmIYthj07UGjiabwmpYACgEAAAAA@telia.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Daniel Eriksson wrote:

>>I have set aside some time on the server (normally a busy 
>>production-type
>>server) to try a few different configs (SHED_ULE instead of SHED_4BSD,
>>turning off apic, removing polling) to see if any of them 
>>will have any
>>impact.
> 
> 
> I have just tried a number of different combinations of sheduler, acpi,
> apic, polling, ...
> 
> No matter what I do, the only way to prevent errors like the one below seems
> to be to run with preemption turned off.
> 
> Interrupt storm detected on "irq20: atapci4"; throttling interrupt source
> ad20: FAILURE - already active DMA on this device
> ad20: setting up DMA failed
> 
> See my recent mail on this list entitled "Kernel preemption leads to SATA
> instability (how to debug?)" for machine configuration and further
> information.
> 
> /Daniel Eriksson
> 

Thanks for the report.  I'm sure that there is a bug of some sort in the
ATA driver that is being exposed by PREEMPTION, but I'm also going to
consider turning off preemption in the next day or two since this isn't
the only lingering problem.

Scott



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?40FCC1AA.9060609>