Date: Mon, 07 Feb 2005 07:33:23 +0100 From: "Poul-Henning Kamp" <phk@phk.freebsd.dk> To: Ian Dowse <iedowse@FreeBSD.org> Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/share/man/man9 Makefile timeout.9 src/sys/sys callout.h src/sys/kern kern_timeout.c Message-ID: <3212.1107758003@critter.freebsd.dk> In-Reply-To: Your message of "Mon, 07 Feb 2005 02:47:33 GMT." <200502070247.j172lXIi069498@repoman.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <200502070247.j172lXIi069498@repoman.freebsd.org>, Ian Dowse writes: >iedowse 2005-02-07 02:47:33 UTC > > FreeBSD src repository > > Modified files: > share/man/man9 Makefile timeout.9 > sys/sys callout.h > sys/kern kern_timeout.c > Log: > Add a mechanism for associating a mutex with a callout when the > callout is first initialised, using a new function callout_init_mtx(). > The callout system will acquire this mutex before calling the callout > function and release it on return. Great! But can we make softclock() use a two pass algorithm ? First pass take all the timeouts which have a mutex which we can get with mtx_lock_try() then in second pass all the rest of the callouts for this tick ? This would improve latency for well-written/locked code. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3212.1107758003>