linux-kernel - Re: Choosing a HyperThreading/SMP/MultiCore kernel ?

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: <200701140221.l0E2LOYS010969@turing-police.cc.vt.edu>
Date:	Sat, 13 Jan 2007 21:21:24 -0500
From:	Valdis.Kletnieks@...edu
To:	lsorense@...lub.uwaterloo.ca (Lennart Sorensen)
Cc:	Sunil Naidu <akula2.shark@...il.com>, linux-kernel@...r.kernel.org
Subject: Re: Choosing a HyperThreading/SMP/MultiCore kernel ?

On Sat, 13 Jan 2007 12:54:43 EST, Lennart Sorensen said:
> On Fri, Jan 12, 2007 at 10:38:43PM -0500, Valdis.Kletnieks@...edu wrote:
> > CONFIG_MCORE2=y
> 
> Oh good.  Makes life much simpler for users.

After writing that, I actually went back and *checked* the fine print.
It turns out that unless you have installed a not-yet-escaped release of
gcc, -mtune=core2 doesn't work, so it punts to -mtune=generic.  Wandering
over to http://gcc.gnu.org and searching the mailing lists, it seems that
on most of the benchmarks, -mtune=core2 was only a 0.5% or so win on most
stuff in its current form.

Content of type "application/pgp-signature" skipped