linux-kernel - Re: [PATCH] Uninline kcalloc()

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

Message-ID: <5202.1190639853@turing-police.cc.vt.edu>
Date:	Mon, 24 Sep 2007 09:17:33 -0400
From:	Valdis.Kletnieks@...edu
To:	Kyle Moffett <mrmacman_g4@....com>
Cc:	Alexey Dobriyan <adobriyan@...il.com>, akpm@...l.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] Uninline kcalloc()

On Mon, 24 Sep 2007 08:59:56 EDT, Kyle Moffett said:

> Proper fix is to give __kmalloc a "void *caller" parameter and have  
> all of the various wrapper functions pass in the value of  
> __builtin_return_address() appropriately.  I believe that even works  
> properly for inline functions which may or may not be inlined.

I think I looked at that, and it wasn't as easy as it looked, because there
were ugly corner cases for what __builtin_return_address() returned depending
on exactly what did or didn't get inlined. Basically, it's ugly stuff - if you
try to use __builtin_retur_address inside kcalloc to pass to __kmalloc, and
then kcalloc gets inlined, you end up passing not the routine you wanted (the
caller of kcalloc), but the *parent* of that...

Content of type "application/pgp-signature" skipped