Date: Fri, 27 Oct 2006 23:04:56 +0200 From: "Simon L. Nielsen" <simon@FreeBSD.org> To: gnn@freebsd.org Cc: Lev Serebryakov <lev@freebsd.org>, Robert Watson <rwatson@freebsd.org>, current@freebsd.org Subject: Re: KSE, libpthread & libthr: almost newbie question Message-ID: <20061027210455.GA1073@zaphod.nitro.dk> In-Reply-To: <m2r6wtsqhj.wl%gnn@neville-neil.com> References: <917908193.20061027102647@serebryakov.spb.ru> <20061027103924.F79313@fledge.watson.org> <m2r6wtsqhj.wl%gnn@neville-neil.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2006.10.27 17:55:20 +0200, gnn@freebsd.org wrote: > At Fri, 27 Oct 2006 14:02:59 +0100 (BST), > rwatson wrote: > > (3). One of the current theories bouncing around the kernel > > developer community is that the complexity and overhead of (2) > > outweighs many of the benefits of KSE, and that by making it an > > option, we can better evaluate the impact. Notice that this isn't > > just about code complexity, but also about scheduler overhead. > > David Xu has reported a non-trivial performance change from the > > reduced overhead of the scheduler paths. So now we're at a point > > where we can more fully evaluate the impact of KSE (since we can > > actually compile it out of the scheduler). Before anything further > > can be done, we now need to do that evaluation. > > > > And speaking of evaluation if people can follow the advice here: > > http://wikitest.freebsd.org/BenchmarkAdvice > > It would be greatly appreciated. Note that the text copy/pasted here is actually already in our developmers handbook (and has been since shortly after phk's mail): http://www.freebsd.org/doc/en_US.ISO8859-1/books/developers-handbook/testing.html -- Simon L. Nielsen
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20061027210455.GA1073>