Re: ongoing major problems with NetBSD-5 and LOCKDEBUG on multi-core system (mfi(4) related?)

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Sun, Jan 22, 2012 at 03:34:48PM -0800, Greg A. Woods wrote:
> Thanks, I finally got around to re-adding the KERNEL_LOCK() hooks
> to mfi.c, but I did it quite differently than the original two
> patches which did it, and instead I moved the splbio()/splx() calls into
> the function where they're really needed, as was done in OpenBSD, since
> that code can, I think, be called by other than bio(4) and
> sysmon_envstat(9).  I'll attach my current changes, which include the
> missing bug fixes from OpenBSD, as well as my changes to splbio()/splx()

You're now mixing splbio/splx with mutexes, I don't think this is correct.

> 
> 
> > BTW, I run several dell PE2950 with this code and I don't have
> > problems.
> 
> I'm more or less certain now that mfi(4) is not the cause of the
> problems I have been seeing (the ones first reported in PR# kern/45827)
> 
> 
> Are you running kernels from the netbsd-5 branch on any of them?
> 
> Kernels with LOCKDEBUG (and DIAGNOSTIC and DEBUG)?

I have in the past; once the kernel was stable I removed them for
performances.

> 
> 
> I'm still stuck on how to use GDB to look at the different CPU stack
> backtraces.

No idea, I usually do it from ddb.

-- 
Manuel Bouyer <bouyer%antioche.eu.org@localhost>
     NetBSD: 26 ans d'experience feront toujours la difference
--