Hey Stefan,

I'll respond inline below:

Thanks,
-Don Skidmore <donald.c.skidm...@intel.com>

> -----Original Message-----
> From: Stefan Puiu [mailto:stefan.p...@gmail.com]
> Sent: Wednesday, December 10, 2014 7:35 AM
> To: Skidmore, Donald C
> Cc: e1000-devel@lists.sourceforge.net
> Subject: Re: [E1000-devel] rx_no_dma_resources woes
> 
> Hi,
> 
> On Tue, Dec 9, 2014 at 7:22 PM, Skidmore, Donald C
> <donald.c.skidm...@intel.com> wrote:
> > Hi Stefan,
> >
> > Ubuntu is a tricky one.  I keep reminding our management that I'm
> > noticing more people in the community are using it especially LTS.
> > They (Ubuntu) make it hard though without versioning like SLES and
> > RHEL, doing kcompat can be a challenge.  Not that I think that is at
> > all related to your issue. And you are correct we do validate and
> > actively add compat support for RHEL and SLES. :)
> 
> By "without versioning" do you mean they backport stuff expected in newer
> kernel, making it hard to detect which kernel you're running on?
> I'm asking just so I know next time when I need to decide Ubuntu or their
> kernel. I guess the Linus/vanilla kernel should work (according to
> http://sourceforge.net/p/e1000/mailman/message/32788947/ at least)?

All the distros backport functionality to their releases, just Ubuntu doesn't 
have nice versioning strings we can basic our kcompat code off of.  RHEL and 
SLES do I'm sure Ubuntu will come up with something as well but it makes it 
more difficult as it is.  And yes we do support the kernel.org kernels, the 
readme that next to the tarball on source forge mentions that latest Stable 
kernel at release time.  

> 
> >
> > I think ATR may work for you with your current system setup if you map
> the interrupts in the 1-to-1 fashion I mentioned below.  Since you have your
> user space processes pinned to a given CPU all flows that transmit a SYN from
> that CPU will be redirected back to that queue/CPU pair.  Not sure you would
> be effected much by the sampling rate as it rate is our mechanism for
> identify non-SYN packet flows, it is the rate at which we check a given packet
> to see if a ATR hash bucket exist for it and if not create one.
> 
> Actually, the stuff we're receiving is strictly layer 3 - IP with a custom
> protocol, not even TCP or UDP. We would've used Layer2, but I haven't
> found a way to distribute that among cores / queues. Both RSS and ethtool -
> K/-U filters seem to require at least layer 3.
> 

That is tough.  I imagine RSS is what you will use then.

> >
> > That said I also believe your mechanism will work, possible even better
> since you don't have to wait for the ATR hash bucket to be created by the
> transmission of the SYN (meaning connections remotely initiated will most
> likely not have their initial SYN direct to the CPU you want) and won't have 
> to
> deal with the limit on ATR hash buckets (if you have LOTS of ports open from
> one IP address).  Still I would turn off ATR and thus default to RSS.  This
> however will limit you to 16 queues, which depending on the size of your
> system may or may not be an issue.
> 
> Is it possible to disable ATR from ethtool while using the older driver that
> ships with Ubuntu kernel 3.11?
> We're currently using 8 queues per NIC, so the RSS limit is OK.'

If you set the number of queues to 8 then ATR should be disabled since it needs 
to have one queue per CPU.

> 
> >
> > In truth you might want to look at perfect filter.  With it you could 
> > direct let
> the adapter to send all traffic to a given queue.  Which, from my limited
> understanding of what you're doing, looks to be just what you're looking for.
> Basically map an IP address to a give queue.
> 
> I'm using:
> 
> ethtool -K $DEV ntuple on
> [... then for each src IP]
> ethtool -U $DEV flow-type ip4 src-ip 10.10.10.$ip_digit action $i
> 
> Not sure if I'm using perfect filter or not... Or what I would need to change 
> to
> use perfect filter.

Yeah if you are turning of perfect filter, which by the way also disables ATR.  
Which sounds like something you wanted to do anyway.

> 
> Thanks for your help,
> Stefan.
> 
> >
> > Thanks,
> > -Don Skidmore <donald.c.skidm...@intel.com>
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel&#174; Ethernet, visit 
http://communities.intel.com/community/wired

Reply via email to