Re: [PATCH] klogd core

From: Martin Schulze <joey_at_infodrom.org>
Date: Sat, 5 Jul 2008 09:14:57 +0200

Sean Young wrote:
> On Fri, Jul 04, 2008 at 04:47:38PM +0200, Martin Schulze wrote:
> > Sean Young wrote:
> > > On Fri, Jul 04, 2008 at 04:16:31PM +0200, Martin Schulze wrote:
> > > > Sean Young wrote:
> > > > > With /proc/kallsyms disabled in the kernel, klogd crashes. fclose(NULL) is
> > > > > executed.
> > > >
> > > > Hmbf! Do you know since which version?
> > >
> > > Looking at the history, I assume klogd 1.5.
> >
> > Err... sorry, I was more interested in the kernel version. :)
>
> At least since 2.6.20 there is option CONFIG_KALLSYMS which can only be
> disabled if CONFIG_EMBEDDED is enabled. Not having your symbols available
> is pretty useless as you don't get function names in Oopses, etc. I was

That makes sense. However, don't you run busybox-syslogd in embedded
environments?

The symbols help explain oopses. klogd and an external debugger are
able to decode the addresses with help of the symbol table.

Regards,

        Joey

-- 
Testing? What's that? If it compiles, it is good, if it boots up, it is perfect.
Received on Sat Jul 05 2008 - 09:14:57 CEST

This archive was generated by hypermail 2.2.0 : Sat Jul 05 2008 - 09:17:06 CEST