Re: boot fail with large RAM

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

On Fri, Nov 20, 2009 at 07:58:52AM +0000, Matthias Scheler wrote:
> Well, I never tried that because I didn't know what the problem was.
> And as the machine is a server I don't like fiddling with it.
> 
> But it seems I submitted a PR which includes a stack trace:
> 
>       http://gnats.netbsd.org/41133

The real problem here seems to be that the kernel can go tsleep
with the kernel lock held. This is bad.

> 
> It was definitely trying to call bus_dmamem_alloc().

From the stack trace it would be looping in uvm_pglistalloc_simple().
This could be exactly the same case as the one I did get; where
a DEBUG kernel would flood the console with "pglistalloc waiting".
So my patch could workaround this problem.

-- 
Manuel Bouyer, LIP6, Universite Paris VI.           
Manuel.Bouyer%lip6.fr@localhost
     NetBSD: 26 ans d'experience feront toujours la difference
--