Date: Sun, 17 Feb 2002 23:42:18 -0800 From: Terry Lambert <tlambert2@mindspring.com> To: Bruce Evans <bde@zeta.org.au> Cc: Matthew Dillon <dillon@apollo.backplane.com>, Poul-Henning Kamp <phk@critter.freebsd.dk>, Julian Elischer <julian@elischer.org>, Alfred Perlstein <bright@mu.org>, arch@FreeBSD.ORG, jhb@FreeBSD.ORG, peter@wemm.org, jake@locore.ca Subject: Re: gettimeofday() and crhold()/crfree() (was Re: gettimeofday()andcopyout().Is copyout() MPSAFE on non-i386 archs? ) Message-ID: <3C70B05A.B41BCF9E@mindspring.com> References: <20020218182727.C5246-100000@gamplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Bruce Evans wrote: > > .4nsec * 22,000 transactions/sec * 5 timestamps/transaction > > = 440us/sec spent on time stamps. > > This is not long :-), but 440 msec is. > > > Up that to 2uS and... = 220ms/sec on time stamps = ~1/4 of > > all available time spent on time stamps. In practice, the number was closer to 8-10% of time burnt on the extra protection domain crossings. YMMV, depending on the hardware you end up using, but 10% was low hanging fruit, and so worth winning back. The 25% cost for the 2uS number is probably important to people running blatantly low end hardware, which is is why I calculated it from there (well, that, and the overhead I actually had was not in the set of numbers presented by other people for use in calculations ;^)). I guess if we could all afford to have $800 ServerWorks motherboards sontributing to our COGS for our products, this would be a better world... 8-) 8-). -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3C70B05A.B41BCF9E>