[ofa-general] Re: [PATCH 2/3][NET_BATCH] net core use batching

[Waskiewicz Jr, Peter P]

> -----Original Message-----
> From: Andi Kleen [mailto:andi at firstfloor.org] 
> Sent: Wednesday, October 10, 2007 9:02 AM
> To: Waskiewicz Jr, Peter P
> Cc: David Miller; andi at firstfloor.org; hadi at cyberus.ca; 
> shemminger at linux-foundation.org; jeff at garzik.org; 
> johnpol at 2ka.mipt.ru; herbert at gondor.apana.org.au; 
> gaagaan at gmail.com; Robert.Olsson at data.slu.se; 
> netdev at vger.kernel.org; rdreier at cisco.com; 
> mcarlson at broadcom.com; jagana at us.ibm.com; 
> general at lists.openfabrics.org; mchan at broadcom.com; 
> tgraf at suug.ch; randy.dunlap at oracle.com; sri at us.ibm.com; 
> kaber at trash.net
> Subject: Re: [ofa-general] Re: [PATCH 2/3][NET_BATCH] net 
> core use batching
> 
> > We've done similar testing with ixgbe to push maximum descriptor 
> > counts, and we lost performance very quickly in the same 
> range you're 
> > quoting on NIU.
> 
> Did you try it with WC writes to the ring or CLFLUSH?
> 
> -Andi

Hmm, I think it might be slightly different, but it still shows queue
depth vs. performance.  I was actually referring to how many descriptors
we can represent a packet with before it becomes a problem wrt
performance.  This morning I tried to actually push my ixgbe NIC hard
enough to come close to filling the ring with packets (384-byte
packets), and even on my 8-core Xeon I can't do it.  My system can't
generate enough I/O to fill the hardware queues before CPUs max out.

-PJ Waskiewicz