Re: Problems with kernel

From: Amon Ott <ao@rsbac.org>
Subject: Re: Problems with kernel
Date: Fri, 23 Mar 2001 10:49:52 +0100

Next Article (by Subject): Re: Problems with kernel Jörgen Sigvardsson
Previous Article (by Subject): Re: Problems with kernel Bencsath Boldizsar
Top of Thread: Problems with kernel inger@linux.ru.net
Next in Thread: Re: Problems with kernel Jörgen Sigvardsson
Articles sorted by: [Date] [Author] [Subject]

On Fre, 23 Mär 2001 Bencsath Boldizsar wrote:
> Ok, then kmalloc is not unstable ;-) I just asked because the hangs with
> the "unable to handle kernel paging request". Is it possible that
> initiation kmalloces go up to 128k which is the limit of kmalloc? 

Whenever we might reach this limit, there is a check that uses vmalloc, if size
is > RSBAC_MAX_KMALLOC or kmalloc failed.

Anyway, kmalloc for >128K results in a kernel panic.

The limit is easy to reach when writing a list with several thousand entries.
BTW, if even vmalloc fails, write_fd_lists complains and falls back to
partial_write_fd_list, which makes a list copy and writes the items one by one.
So your attributes should always get written, as long as the system is alive.

> Anyway, I'll check the vmalloc thing.

Good.

> (I looked in other kernel parts, reiserfs uses it's own reiserfs_kmalloc,
> which counts how many bytes where kmalloced. It could be useful to us too
> to get clean picture about memory state...)

Yes, agreed. I will put it into my to-do list for 1.1.2.

Amon.
-
To unsubscribe from the rsbac list, send a mail to
majordomo@rsbac.org with
unsubscribe rsbac
as single line in the body.

Go to Compuniverse LWGate Home Page.