Date: Thu, 4 Aug 2016 17:33:10 +0200 From: Hans Petter Selasky <hps@selasky.org> To: Ben RUBSON <ben.rubson@gmail.com>, freebsd-net@freebsd.org Subject: Re: Unstable local network throughput Message-ID: <0b14bf39-ed71-b9fb-1998-bd9676466df6@selasky.org> In-Reply-To: <06E414D5-9CDA-46D1-A26F-0B07E76FDB34@gmail.com> References: <3C0D892F-2BE8-4650-B9FC-93C8EE0443E1@gmail.com> <bed13ae3-0b8f-b1af-7418-7bf1b9fc74bc@selasky.org> <3B164B7B-CBFB-4518-B57D-A96EABB71647@gmail.com> <5D6DF8EA-D9AA-4617-8561-2D7E22A738C3@gmail.com> <06E414D5-9CDA-46D1-A26F-0B07E76FDB34@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 08/04/16 17:24, Ben RUBSON wrote: > >> On 04 Aug 2016, at 11:40, Ben RUBSON <ben.rubson@gmail.com> wrote: >> >>> On 02 Aug 2016, at 22:11, Ben RUBSON <ben.rubson@gmail.com> wrote: >>> >>>> On 02 Aug 2016, at 21:35, Hans Petter Selasky <hps@selasky.org> wrote: >>>> >>>> The CX-3 driver doesn't bind the worker threads to specific CPU cores by default, so if your CPU has more than one so-called numa, you'll end up that the bottle-neck is the high-speed link between the CPU cores and not the card. A quick and dirty workaround is to "cpuset" iperf and the interrupt and taskqueue threads to specific CPU cores. >>> >>> My CPUs : 2x E5-2620v3 with DDR4@1866. >> >> OK, so I cpuset all Mellanox interrupts to one NUMA, as well as the iPerf processes, and I'm able to reach max bandwidth. >> Choosing the wrong NUMA (or both, or one for interrupts, the other one for iPerf, etc...) totally kills throughput. >> >> However, full-duplex throughput is still limited, I can't manage to reach 2x40Gb/s, throttle is at about 45Gb/s. >> I tried many different cpuset layouts, but I never went above 45Gb/s. >> (Linux allowed me to reach 2x40Gb/s so hardware is not a bottleneck) >> >>>> Are you using "options RSS" and "options PCBGROUP" in your kernel config? >> >> I will then give RSS a try. > > Without RSS : > A ---> B : 40Gbps (unidirectional) > A <--> B : 45Gbps (bidirectional) > > With RSS : > A ---> B : 28Gbps (unidirectional) > A <--> B : 28Gbps (bidirectional) > > Sounds like RSS does not help :/ > > Why, without RSS, do I have difficulties to reach 2x40Gbps (full-duplex) ? > Hi, Possibly because the packets are arriving at the wrong CPU compared to what RSS expects. Then RSS will invoke a taskqueue to process the packets on the correct CPU, if I'm not mistaken. The mlx4 driver does not fully support RSS. Then mlx5 does. --HPS
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?0b14bf39-ed71-b9fb-1998-bd9676466df6>