Date: Thu, 9 Aug 2018 11:22:43 -0600 From: Warner Losh <imp@bsdimp.com> To: Mark Johnston <markj@freebsd.org> Cc: Mark Millard <marklmi@yahoo.com>, "freebsd-arm@freebsd.org" <freebsd-arm@freebsd.org> Subject: Re: RPI3 swap experiments ["was killed: out of swap space" with: "v_free_count: 5439, v_inactive_count: 1"] Message-ID: <CANCZdfq42nWEDTShA1zyxLbTY07R4_xMDwCrdhXgE168jA=79g@mail.gmail.com> In-Reply-To: <20180809170900.GD68459@raichu> References: <6BFE7B77-A0E2-4FAF-9C68-81951D2F6627@yahoo.com> <20180802002841.GB99523@www.zefox.net> <20180802015135.GC99523@www.zefox.net> <EC74A5A6-0DF4-48EB-88DA-543FD70FEA07@yahoo.com> <20180806155837.GA6277@raichu> <20180808153800.GF26133@www.zefox.net> <20180808204841.GA19379@raichu> <20180809065648.GB30347@www.zefox.net> <20180809152152.GC68459@raichu> <915DFC7F-7AC9-484F-8619-C386FF077769@yahoo.com> <20180809170900.GD68459@raichu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Aug 9, 2018 at 11:09 AM, Mark Johnston <markj@freebsd.org> wrote: > On Thu, Aug 09, 2018 at 09:21:38AM -0700, Mark Millard wrote: > > Would something like the patch that produced the messages > > like: > > > > waited 3s for async swap write > > waited 3s for swap buffer > > > > be appropriate as able to be enabled via a sysctl or in > > some other way? In other words: in the source code by standard, > > off by default, but able to be enabled without patching, > > possibly without rebuilding? > > Yes, I think so. We don't have a lot of visibility into the behaviour > of the pageout threads, and I have some ideas on how to improve that. > Swap I/O latency measurements are something I've wanted in the past. > There's two ways to measure this: total latency (which is most likely the best) and latency in the hardware (which is a better diagnostic for hardware failure / behavior changes). There's some code to do power-of-two bucket latency measurements in the CAM I/O scheduler that could be abstracted out. It's rock simple. There's some code coming into the tree that may give more nuanced results, but it's not clear when that will land. Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANCZdfq42nWEDTShA1zyxLbTY07R4_xMDwCrdhXgE168jA=79g>