Date: Mon, 26 Sep 2022 09:49:39 -0700 From: John Fieber <jrf@ursamaris.org> To: mike tancsa <mike@sentex.net> Cc: "Pieper, Jeffrey E" <jeffrey.e.pieper@intel.com>, Jim King <jim@jimking.net>, stable@freebsd.org, kbowling@freebsd.org Subject: Re: igc problems with heavy traffic (update) Message-ID: <244B1651-9A4D-4E38-A89B-B1610992A249@ursamaris.org> In-Reply-To: <b4d1353f-f2ab-f8a0-fbb1-208c4ee39eac@sentex.net> References: <b4d1353f-f2ab-f8a0-fbb1-208c4ee39eac@sentex.net>
next in thread | previous in thread | raw e-mail | index | archive | help
> On Sep 26, 2022, at 5:57 AM, mike tancsa <mike@sentex.net> wrote: >=20 > =EF=BB=BFOn 9/24/2022 8:30 PM, John Fieber wrote: >>>> On Sep 14, 2022, at 8:03 AM, mike tancsa <mike@sentex.net> wrote: >>>=20 >>> OK, an update hence the top post. I got a new pair of boxes which use a d= ifferent Jasper Lake chipset and have i226-V vs the i225 of the previous box= . >>>=20 >>> dev.igc.0.%parent: pci2 >>> dev.igc.0.%pnpinfo: vendor=3D0x8086 device=3D0x125c subvendor=3D0x8086 s= ubdevice=3D0x0000 class=3D0x020000 >>> dev.igc.0.%location: slot=3D0 function=3D0 dbsf=3Dpci0:2:0:0 handle=3D\_= SB_.PC00.RP05.PXSX >>> dev.igc.0.%driver: igc >>> dev.igc.0.%desc: Intel(R) Ethernet Controller I226-V >>> dev.igc.%parent: >>>=20 >>> WIth a default RELENG_13, out of the box with no tweaks, I am NOT able t= o cause the transmitting nic to bounce with heave traffic. I used the same t= est script (a constant stream of iperf3 alternating in direction) maxing out= the NIC's bandwidth and all seems fine running the test for some 18hrs. Ma= ybe something different about the i225 version of this NIC that needs some d= ifferent driver defaults ? >>>=20 >>> ---Mike >>>=20 >> I also see this behavior with 13.1-RELEASE-p2 on: >>=20 >> These, however, offer unflappable performance: >>=20 >> - FreeBSD-14.0-CURRENT-amd64-20220923 >> - vyos-1.4 (for reference, what I mostly use on this hardware, via bhyve)= >>=20 > Interesting, so just to confirm, the same hardware i225, the igc under 14.= x does not see link drops under heavy load ? I wonder what the difference i= s, since the driver does not seem to be different ? Correct. NIC chips are SLNMH, B3 stepping. All the other OS versions starte= d dropping within a minute. 14-CURRENT ran for an hour at about 2.3 gigabit w= ithout a single hiccup. All the tests were done with a fresh OS install in b= hyve (with pci passthrough for the nics) followed by =E2=80=9Cpkg install ip= erf3=E2=80=9D and no other tweaks. The switch port was configured with flow c= ontrol and EEE off in all cases. I ran the tests with the test OS as the iperf client, with -R and -P4 argume= nts. -R was always the quickest failure path. The hour long run on 14-CURREN= T was with --bidi, after -R alone failed to fail after five minutes. The hardware in question is my spare one of three identical 4x2.5g ali-expre= ss celeron J4125 boxes, and as such open for more and/or longer experiments.= (My spare time to fiddle is a bigger constraint.) -john
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?244B1651-9A4D-4E38-A89B-B1610992A249>