linux-kernel - Re: Unnecessary overhead with stack protector.

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 for Android: free password hash cracker in your pocket

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

Message-ID: <4ADF5FE2.7070604@redhat.com>
Date:	Wed, 21 Oct 2009 14:24:18 -0500
From:	Eric Sandeen <sandeen@...hat.com>
To:	Arjan van de Ven <arjan@...radead.org>
CC:	Ingo Molnar <mingo@...e.hu>, Dave Jones <davej@...hat.com>,
	Linux Kernel <linux-kernel@...r.kernel.org>,
	Thomas Gleixner <tglx@...utronix.de>, esandeen@...hat.com,
	cebbert@...hat.com, "H. Peter Anvin" <hpa@...or.com>
Subject: Re: Unnecessary overhead with stack protector.

Eric Sandeen wrote:
> Eric Sandeen wrote:

..

> but maybe more to Dave's original point, xfs on x86_64 in my tree had
> 243 functions with minimal stack usage of 8 bytes.  w/
> CC_STACKPROTECTOR_ALL in force, I end up with these sizes for those
> functions:
> 
>   count bytes
>       3 16
>     236 24
>       1 32
>       5 40
> 
> 8->24 bytes is pretty significant too, w/ a 200% increase, if you add a
> few up...

And on top of that (sorry for the self-replies) there are 600+ functions
in xfs that didn't even register a stack footprint at all (i.e. no sub
%rsp in disassembly) w/o stackprotector, which now have 16 or 24 bytes
with it on.

Forgive me if I'm not using the right tools to look, but it seems to me
that in the aggregate, CC_STACKPROTECTOR_ALL has a pretty big impact.

-Eric
--
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/