Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 24 Oct 1996 12:52:31 -0500 (EST)
From:      John Dyson <>
To: (T. William Wells)
Subject:   Re: more about system hangs
Message-ID:  <>
In-Reply-To: <> from "T. William Wells" at Oct 24, 96 01:10:13 pm

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help
> (It appears that the list is dropping my messages randomly.  Here
> is a resend.)
> I've been tracing down the hangs and it appears to be within
> kmem_alloc, called from pmap_pinit. I don't think I can trace it
> down further, as there's no way to stick debugs in this code.
> (Debugs that won't bring the system to its knees anyway....)
> What I'm wondering is this: since it's allocating out of
> kernel_map is it possible to run out of space there? And doesn't
> this code simply wait for memory, rather than panicing? If so,
> that would account for my hangs. Is there a way I can check for
> this from ddb or by adding debugs? Other suggestions?
Got your message :-).  kernel_map cannot easily run out of space,
but there is opportunity for problems if the system cannot allocate
map entries or page-table pages.  You have narrowed the problem
down ALOT...  Will study the problem further, but if you get
any more info -- let me/us know!!!


Want to link to this message? Use this URL: <>