linux-kernel - Re: [PATCH] CONFIG_RELOCATABLE modpost fix

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <1155107198.4341.87.camel@localhost>
Date:	Wed, 09 Aug 2006 16:06:38 +0900
From:	Magnus Damm <magnus@...inux.co.jp>
To:	Sam Ravnborg <sam@...nborg.org>
Cc:	linux-kernel@...r.kernel.org, fastboot@...ts.osdl.org,
	ebiederm@...ssion.com
Subject: Re: [PATCH] CONFIG_RELOCATABLE modpost fix

On Wed, 2006-08-09 at 08:29 +0200, Sam Ravnborg wrote:
> On Wed, Aug 09, 2006 at 10:32:36AM +0900, Magnus Damm wrote:
>  
> > Your patch seems to work as expected if I add a return 0 at the end of
> > modpost.c:secref_whitelist(). I like how you printed out the number of
> > modules being processed.
> Thanks - fixed now. My gcc (3.4.6-r1 from Gentoo did not warn)

Interesting. I have the 3.4.6-r1 ebuild installed too, but I happened to
have the 3.3.6 profile selected by default. Which explains why things
still work as expected here.

> >I have one minor comment about your patch:
> > 
> > Modpost seems to get run twice on vmlinux if the kernel is built with
> > "make all". I think it would be best to run modpost on vmlinux only when
> > vmlinux is built - never when modules are processed.
> 
> Thesecond time modpost runs vmlinux is used to pick up symbol
> information to check that all symbols are valid etc.
> The alternative was to trust the symbols being read from Module.symvers
> and that would be OK in most cases but I could imagine situations where
> Module.symvers was deleted but vmlinux kept.
> 
> So therefore the more expensive solution to run modpost twice on vmlinux
> was chosen.

I understand. I'm not that worried about build performance, more the
fact that all the warnings from vmlinux will get spit out twice.

Thanks,

/ magnus

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/