Date: Mon, 13 Jan 2014 21:08:43 -0800 From: Adrian Chadd <adrian@freebsd.org> To: John Baldwin <jhb@freebsd.org> Cc: "freebsd-arch@freebsd.org" <freebsd-arch@freebsd.org> Subject: Re: Acquiring a lock on the same CPU that holds it - what can be done? Message-ID: <CAJ-VmomhbDcwL-eGK6Doh%2BXjUkcaFZYbZxZaNKGgNgRREccesw@mail.gmail.com> In-Reply-To: <201401131443.52550.jhb@freebsd.org> References: <CAJ-Vmok-AJkz0THu72ThTdRhO2h1CnHwffq=cFZGZkbC=cWJZA@mail.gmail.com> <9508909.MMfryVDtI5@ralph.baldwin.cx> <CAJ-Vmo=rayYvUYsNLs2A-T=a7WbrSA%2BTUPgDoGCHdbQjeJ9ynw@mail.gmail.com> <201401131443.52550.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 13 January 2014 11:43, John Baldwin <jhb@freebsd.org> wrote: > On Thursday, January 09, 2014 1:44:51 pm Adrian Chadd wrote: >> On 9 January 2014 10:31, John Baldwin <jhb@freebsd.org> wrote: >> > On Friday, January 03, 2014 04:55:48 PM Adrian Chadd wrote: >> >> Hi, >> >> >> >> So here's a fun one. >> >> >> >> When doing TCP traffic + socket affinity + thread pinning experiments, >> >> I seem to hit this very annoying scenario that caps my performance and >> >> scalability. >> >> >> >> Assume I've lined up everything relating to a socket to run on the >> >> same CPU (ie, TX, RX, TCP timers, userland thread): >> > >> > Are you sure this is really the best setup? Especially if you have free CPUs >> > in the system the time you lose in context switches fighting over the one >> > assigned CPU for a flow when you have idle CPUs is quite wasteful. I know >> > that tying all of the work for a given flow to a single CPU is all the rage >> > right now, but I wonder if you had considered assigning a pair of CPUs to a >> > flow, one CPU to do the top-half (TX and userland thread) and one CPU to >> > do the bottom-half (RX and timers). This would remove the context switches >> > you see and replace it with spinning in the times when the two cores actually >> > contend. It may also be fairly well suited to SMT (which I suspect you might >> > have turned off currently). If you do have SMT turned off, then you can get >> > a pair of CPUs for each queue without having to reduce the number of queues >> > you are using. I'm not sure this would work better than creating one queue >> > for every CPU, but I think it is probably something worth trying for your use >> > case at least. >> > >> > BTW, the problem with just slapping critical enter into mutexes is you will >> > run afoul of assertions the first time you contend on a mutex and have to >> > block. It may be that only the assertions would break and nothing else, but >> > I'm not certain there aren't other assumptions about critical sections and >> > not ever context switching for any reason, voluntary or otherwise. >> >> It's the rage because it turns out it bounds the system behaviour rather nicely. > > Yes, but are you willing to try the suggestion? This doesn't restrict to you > a single queue-pair. It might net you 1 per core (instead of 1 per thread), > but that's still more than 1. Sure. I can also try your suggestion of binding them to SMT pairs and see if that has any effect. But I'm specifically looking to _avoid_ contention at all in the main data path, not try to occasionally have the cores spin. -a
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-VmomhbDcwL-eGK6Doh%2BXjUkcaFZYbZxZaNKGgNgRREccesw>