Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 10 Dec 2015 10:40:02 -0800
From:      Adrian Chadd <adrian.chadd@gmail.com>
To:        Denis Pearson <dennix.pearson@gmail.com>
Cc:        "Eggert, Lars" <lars@netapp.com>, "Pieper, Jeffrey E" <jeffrey.e.pieper@intel.com>,  Kevin Oberman <rkoberman@gmail.com>, Daniel Engberg <daniel.engberg.lists@pyret.net>,  "freebsd-net@freebsd.org" <freebsd-net@freebsd.org>
Subject:   Re: ixl 40G bad performance?
Message-ID:  <CAJ-Vmo=cQQjfKhKB_D6Uc2e9TPnQwK2EXBka_sACwCkOCdYvgA@mail.gmail.com>
In-Reply-To: <CAFC3-mQ2d1GKCRCEF8%2BC-u1-EVy094w0zCDv7QHeYnaHK-NHrw@mail.gmail.com>
References:  <5aae0ee63c44627223d5d179f1901d00@pyret.net> <CAN6yY1t9Tw0j=uwaw1GK47r5=F-zeuz2hps_Ez3Y_QC-QSAGKA@mail.gmail.com> <0E4C2D93-FBAF-48CB-A704-499ABFC892B9@netapp.com> <2A35EA60C3C77D438915767F458D6568807F2A8A@ORSMSX111.amr.corp.intel.com> <99E53825-99F8-4E82-A710-6BC07B123F77@netapp.com> <2A35EA60C3C77D438915767F458D6568807F2D52@ORSMSX111.amr.corp.intel.com> <A546ABEA-D495-461F-9441-31F70AACC146@netapp.com> <E21A5504-7780-4D84-AA5B-7A3F6F968FC7@netapp.com> <CAFC3-mQ2d1GKCRCEF8%2BC-u1-EVy094w0zCDv7QHeYnaHK-NHrw@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On 10 December 2015 at 10:29, Denis Pearson <dennix.pearson@gmail.com> wrote:
> On Thu, Dec 10, 2015 at 2:18 PM, Eggert, Lars <lars@netapp.com> wrote:
>
>> On 2015-10-26, at 18:40, Eggert, Lars <lars@netapp.com> wrote:
>> > On 2015-10-26, at 17:08, Pieper, Jeffrey E <jeffrey.e.pieper@intel.com>
>> wrote:
>> >> As a caveat, this was using default netperf message sizes.
>> >
>> > I get the same ~3 Gb/s with the default netperf sizes and driver 1.4.5.
>>
>> Now there is version 1.4.8 on the Intel website, but it doesn't change
>> things for me.
>>
>
> I had the opportunity to see similar numbers and behavior while using XL710
> 1.4.3 as of FreeBSD r291085 while in DPDK poll mode, but driver 1.2.8 as of
> r292035 was providing expected numbers. While removing rxcsum/txcsum did
> not provide differences, fully removing RSS + disabling rx/cxsum support
> provided better numbers.

Can someone debug this a bit more? (My kit with ixl NICs in it is
still not up and available. :( )

Device RSS, even without kernel RSS enabled, shouldn't cause a massive
performance drop. If it is then something else odd is going on.

Do you have a diff where you removed things?


-adrian

> However now with driver 1.4.8 and the same set of hardware setup, except
> for a different transceiver, I can get 36Gbps/24Mpps with no further
> tweaks, so if you can replace your transceiver, shall be a different test
> as a starting point.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-Vmo=cQQjfKhKB_D6Uc2e9TPnQwK2EXBka_sACwCkOCdYvgA>