[ewg] Re: [ofa-general] [PATCH] IB/ipoib: fix net queue lockup

[Eli Cohen]

On Wed, 2008-04-30 at 20:05 -0700, Roland Dreier wrote:
> > we haves seen a few other cases where a large tx queue is needed. I
>  > think we should choose a larger default value than the current 64.
> 
> maybe yes, maybe no... what are the cases where it is needed?
> 
> The send queue is basically acting as a "shock absorber" for bursty
> traffic.  If the queue is filling up because of a steady traffic rate,
> then making the queue bigger means it will just take a little longer to
> fill.  The way a longer send queue helps I guess is if the send queue is
> emptying out before the transmit queue is woken up...
I agree, but I want to have a larger buffer to absorb larger picks. For
example, after applying this patch I tested how many times the net queue
is stopped and woken up when running four streams of netperf, udp, small
packets. When using the default 64 tx queue size it happened 500 times.
When I used a 256 tx queue size it happened only 37 times. This makes me
think that we have larger picks that a larger queue size can help
handle. 

Also looking for example on Broadcom bnx2 driver on my machine, it uses
a 1000 tx queue len.