Date: Tue, 25 Apr 2006 09:34:16 +0400 (MSD) From: Dmitry Morozovsky <marck@rinet.ru> To: Colin Percival <cperciva@freebsd.org> Cc: cvs-src@freebsd.org, src-committers@freebsd.org, cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/amd64/amd64 mp_machdep.c src/sys/i386/i386 mp_machdep.c Message-ID: <20060425093350.Y57625@woozle.rinet.ru> In-Reply-To: <200604242117.k3OLH2RG032117@repoman.freebsd.org> References: <200604242117.k3OLH2RG032117@repoman.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 24 Apr 2006, Colin Percival wrote: CP> Adjust dangerous-shared-cache-detection logic from "all shared data CP> caches are dangerous" to "a shared L1 data cache is dangerous". This CP> is a compromise between paranoia and performance: Unlike the L1 cache, CP> nobody has publicly demonstrated a cryptographic side channel which CP> exploits the L2 cache -- this is harder due to the larger size, lower CP> bandwidth, and greater associativity -- and prohibiting shared L2 CP> caches turns Intel Core Duo processors into Intel Core Solo processors. CP> CP> As before, the 'machdep.hyperthreading_allowed' sysctl will allow even CP> the L1 data cache to be shared. Any chance to MFC this to upcoming releases? Sincerely, D.Marck [DM5020, MCK-RIPE, DM3-RIPN] ------------------------------------------------------------------------ *** Dmitry Morozovsky --- D.Marck --- Wild Woozle --- marck@rinet.ru *** ------------------------------------------------------------------------
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060425093350.Y57625>