Re: 2.6.25: random stalls on certain hardware - regression?

[Michael Tokarev <mjt@tls.msk.ru>]

Michael Tokarev wrote at Thu, 10 Jul 2008 11:04:55 +0400:

[almost a month-ago email/thread...]

> Marcin Slusarz wrote:
>> On Wed, Jul 09, 2008 at 01:56:12AM +0400, Michael Tokarev wrote:
> 
> [random complete hangs on 2.6.25 but not 2.6.24]
> 
>> Definitely try 2.6.26-rc9 (which will soon become 2.6.26, so people will
> 
> Ok, tried 2.6.26-rc9 today night - after disabling some drivers which
> were not compilable and fixing some places in core code.  It does the
> same thing - i.e. it locks up the same way as 2.6.25 did, and locks up
> hard as before (no keyboard and network working).

Ok, I finally tried current 2.6.26 [.2] on this same machine, -- it's
up and running since last Friday, so almost 4 days already, and it works
so far, no lock-ups for now.  2.6.26-pre definitely didn't work - max
uptime was ~12 hours.  I can't say the problem has been fixed for sure,
maybe it's just a {bad|good} luck, will see how it will work in next days.

And btw, nmi-watchdog didn't help, -- with it enabled, the machine locked
up the same way as before -- completely silent, nothing's on the screen etc.
(And no, it's not a hardware problem, since all kernels up to 2.6.24 works
just fine, under any loads.  Or should not be hardware, anyway)

/mjt