linux-kernel - Re: Faster getcpu() and sched

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>] [day] [month] [year] [list]

Message-ID: <20080929220717.GU25711@one.firstfloor.org>
Date:	Tue, 30 Sep 2008 00:07:17 +0200
From:	Andi Kleen <andi@...stfloor.org>
To:	dean gaudet <odo@...gle.com>
Cc:	Andi Kleen <andi@...stfloor.org>, Pardo <pardo@...gle.com>,
	linux-kernel@...r.kernel.org, mbligh@...gle.com,
	briangrant@...gle.com, nil@...gle.com, jyasskin@...gle.com
Subject: Re: Faster getcpu() and sched_getcpu()

On Mon, Sep 29, 2008 at 02:55:26PM -0700, dean gaudet wrote:
> Andi Kleen wrote:
> > On Mon, Sep 29, 2008 at 11:01:26AM -0700, Pardo wrote:
> >   
> >>> [Maybe disable frame pointers for vsyscall.c and the vdso?]
> >>>       
> >> IIRC, some vsyscall.c code needs them enabled, so Dean's earlier patch split
> >>     
> >
> > I don't think it really needs it.
> >
> >   
> >> vsyscall.c, creating a vsyscall_user.c for code which can run without them.
> >> Seem reasonable?
> >>     
> >
> > Seems unnecessarily complicated.
> >   
> 
> i disagree that it's complicated to have two files, and disagree that
> it's unnecessary to have two files.

It's unnecessary to have frame pointer in the kernel functions I meant.
I agree with you that disabling redzone is needed for kernel code,
but without frame pointers (which are generally a bad idea for 
performance and should not have been added to the 64bit port ever) 
redzone is also not particularly expensive and it shouldn't be needed
to do anything complicated (like splitting files) just for the few
cycles.

-Andi

-- 
ak@...ux.intel.com
--
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/