[Barrelfish-users] 82599EB NICs not work

Reto Achermann reto.achermann at inf.ethz.ch
Thu Feb 5 09:28:03 CET 2015


Hi Tom,

We also had some troubles with the e10k. However, we received a patch 
that fixes the initialization of the card. However, the patch has not 
made its way to the release yet.

@Simon: Antoine sent it to Stefan, however w/o sign off.

-- Reto


On 02/05/2015 05:12 AM, tomsun.0.7 wrote:
> Yes, I've already tried this before asking this question (commented out
> the lines both in /e10k_cdriver.c/ and /Hakefile/), but it didn't work
> either.
>
> On Thu, Feb 5, 2015 at 2:20 AM, Simon Peter <speter at inf.ethz.ch
> <mailto:speter at inf.ethz.ch>> wrote:
>
>     The e10k driver should figure out automatically whether the device
>     is in SR-IOV mode or not. However, you can try commenting out the line:
>
>     #define VTON_DCBOFF
>
>     in usr/drivers/e10k/e10k_cdriver.__c and see if it helps (no guarantee).
>
>     On 02/04/2015 06:47 PM, tomsun.0.7 wrote:
>
>         Hi Simon,
>
>         It didn't work... I even /break/ the initialization part of SR-IOV
>         directly for all devices like you did to e1000 NIC, still I got this
>         panic...
>
>         How does e10k driver check the existence of SR-IOV capability
>         actually?
>
>         Tom
>
>         On Thu, Feb 5, 2015 at 12:16 AM, Simon Peter <speter at inf.ethz.ch
>         <mailto:speter at inf.ethz.ch>
>         <mailto:speter at inf.ethz.ch <mailto:speter at inf.ethz.ch>>> wrote:
>
>              Hi Tom,
>
>              You might want to try disabling SR-IOV support, which you
>         can do by
>              passing numvfs=0 to the PCI domain.
>
>              Simon
>
>              On 02/04/2015 06:41 AM, tomsun.0.7 wrote:
>
>                  Hi,
>
>                  I tried to work on bare-metal then, but still I came
>         across some
>                  problems.
>
>                  I started Barrelfish as your README guided and then
>         modified the
>                  GRUB
>                  entries to start e10ktest. The system started with a
>         mess (maybe
>                  result
>                  from ANSI colors), but worked well without fish and angler.
>                  However, if
>                  I started the e10ktest by writing this line into grub:
>
>                       legacy_initrd '/barrelfish/x86_64/sbin/____e10ktest'
>                       '/barrelfish/x86_64/sbin/____e10ktest' 'core=1'
>         'queue=0'
>
>                  then I got a panic:
>
>                       RSC not supported in SR-IOV mode!
>
>
>                  I tried to add a parameter /rsc=0 /as driver code guided,
>                  however, it
>                  didn't work at all.
>
>                  So, what should I do to solve this?
>
>                  Looking forward to your kind reply, thanks.
>
>                  Tom
>
>
>
>
>
>                  ___________________________________________________
>                  Barrelfish-users mailing list
>                  Barrelfish-users at lists.inf.__e__thz.ch <http://ethz.ch>
>                  <mailto:Barrelfish-users at __lists.inf.ethz.ch
>         <mailto:Barrelfish-users at lists.inf.ethz.ch>>
>         https://lists.inf.ethz.ch/____mailman/listinfo/barrelfish-____users
>         <https://lists.inf.ethz.ch/__mailman/listinfo/barrelfish-__users>
>
>         <https://lists.inf.ethz.ch/__mailman/listinfo/barrelfish-__users
>         <https://lists.inf.ethz.ch/mailman/listinfo/barrelfish-users>>
>
>
>
>
>
>
>
> _______________________________________________
> Barrelfish-users mailing list
> Barrelfish-users at lists.inf.ethz.ch
> https://lists.inf.ethz.ch/mailman/listinfo/barrelfish-users
>



More information about the Barrelfish-users mailing list