Date: Tue, 1 May 2018 19:00:54 -0400 From: Jonathan Looney <jonlooney@gmail.com> To: hiren panchasara <hiren@strugglingcoder.info> Cc: Randall Stewart <rrs@freebsd.org>, src-committers <src-committers@freebsd.org>, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r332770 - in head/sys: conf netinet netinet/tcp_stacks sys Message-ID: <CADrOrmuVuSV2y3DRVxDfD7xT46rMkcEj19X1MQhSwVa%2BSZs6ng@mail.gmail.com> In-Reply-To: <20180430071646.GE3653@strugglingcoder.info> References: <201804191338.w3JDc0to002106@repo.freebsd.org> <20180430071646.GE3653@strugglingcoder.info>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Apr 30, 2018 at 3:16 AM, hiren panchasara < hiren@strugglingcoder.info> wrote: > > In my understanding, default stack currently cannot use this mechanism. When do > you think that'll be possible? I think I can speak to Randall's plans for this. Randall chose not to include in this commit the hooks for the default stack to use the high-precision timers. I believe his immediate priorities are upstreaming RACK and BBR. After that, if there is demand, he may upstream the (relatively untested) code that allows the default stack to use the high-precision timers (protected by a non-default kernel option) so others can choose to experiment with it. (By the way, we're hoping to change the terminology away from describing the traditional FreeBSD stack as the "default" stack. In theory, someone can make any stack be their local default. We'll need to figure out what to actually call it at some point. My suggestion was the "FreeBSD" stack, although that is lacking in some imagination. In any case, we should have that discussion at some point in the future.) Jonathan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CADrOrmuVuSV2y3DRVxDfD7xT46rMkcEj19X1MQhSwVa%2BSZs6ng>