Date: Fri, 11 Aug 2006 11:16:41 -0700 (PDT) From: Kip Macy <kmacy@fsmware.com> To: John Baldwin <jhb@freebsd.org> Cc: Divacky Roman <xdivac02@stud.fit.vutbr.cz>, Perforce Change Reviews <perforce@freebsd.org>, Robert Watson <rwatson@freebsd.org>, Roman Divacky <rdivacky@freebsd.org> Subject: Re: PERFORCE change 103633 for review Message-ID: <20060811111553.X27159@demos.bsdclusters.com> In-Reply-To: <200608111329.52886.jhb@freebsd.org> References: <200608111110.k7BBAxIO059339@repoman.freebsd.org> <200608111249.44686.jhb@freebsd.org> <20060811181817.C8215@fledge.watson.org> <200608111329.52886.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> Yes, but my assumption is that it lets the easy cases just run in userland and
> only goes into the kernel for a hard case, so you need to use atomic ops such
> as casuptr().
Correct, umtx post-dates and is meant to compete with futex.
-Kip
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060811111553.X27159>