Re: XSK + ZC, shared UMEM and multiple Fill/Completion queues - broken?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Aug 9, 2022 at 5:12 PM Alasdair McWilliam
<alasdair.mcwilliam@xxxxxxxxxxx> wrote:
>
>
>
> > On 9 Aug 2022, at 15:58, Magnus Karlsson <magnus.karlsson@xxxxxxxxx> wrote:
> >
> >
> > I can reach it thanks. Could you please send me the command line you
> > use to trigger the problem so I can try out exactly that on my system?
> > And I guess it is the "multi FCQ" build that breaks.
>
> Certainly:
>
> ./xdpsock_multi --l2fwd --interface ice0 --queue 0 --channels 4 --poll --busy-poll --zero-copy
>
> Hardware info:
>
> # ethtool -i ice0
> driver: ice
> version: 5.18.10-1.el8.elrepo.x86_64
> firmware-version: 2.50 0x800077a8 1.2960.0
> expansion-rom-version:
> bus-info: 0000:03:00.0
> supports-statistics: yes
> supports-test: yes
> supports-eeprom-access: yes
> supports-register-dump: yes
> supports-priv-flags: yes
>
> # lspci -s 03:00.0 | grep -i E810
> 03:00.0 Ethernet controller: Intel Corporation Ethernet Controller E810-XXV for SFP (rev 02)
>
> # ethtool -g ice0
> Ring parameters for ice0:
> Pre-set maximums:
> RX:             8160
> RX Mini:        n/a
> RX Jumbo:       n/a
> TX:             8160
> Current hardware settings:
> RX:             4096
> RX Mini:        n/a
> RX Jumbo:       n/a
> TX:             4096
>
> # ethtool -l ice0
> Channel parameters for ice0:
> Pre-set maximums:
> RX:             16
> TX:             16
> Other:          1
> Combined:       16
> Current hardware settings:
> RX:             0
> TX:             0
> Other:          1
> Combined:       4

OK, I believe I have found the problem and have produced a fix for it.
As usual, it is something simple and stupid, sigh. I will post a patch
here tomorrow. Would you be able to test it and see if it fixes the
problem with corrupted packets in your app? If it does, then I will
post the patch on the netdev mailing list. Just note that even if this
fixes this, there is still your RSS problem. Maciej is looking into
that. You were also reporting stalls, which we should examine if they
still occur after fixing the corrupted packets and the RSS problem.

/Magnus



[Index of Archives]     [Linux Networking Development]     [Fedora Linux Users]     [Linux SCTP]     [DCCP]     [Gimp]     [Yosemite Campsites]

  Powered by Linux