Re: SWS for rcvbuf < MTU

[Alex Sidorenko <alexandre.sidorenko@hp.com>]

On March 13, 2007 03:01:50 pm John Heffner wrote:
> Sorry for the long delay in response, I've been on vacation.  I'm okay
> with your patch, and I can't think of any real problem with it, except
> that the behavior is non-standard.  Then again, Linux acking in general
> is non-standard, which has created the bug in the first place. :)  The
> only thing I can think where it might still ack too often is if
> free_space frequently drops just below full_space/2 for a bit then rises
> above full_space/2.
>
> I've also attached a corrected version of my earlier patch that I think
> solves the problem you noted.

Hello John,

I think your corrected version should avoid the problem I mentioned, but I 
will still ask the customer to test the new version. It will be interesting 
to compare traffic patterns with those observed with stock kernel and my 
earlier patch.

It might take some time to do proper testing, I'll inform you as soon as there 
are any results.

Thanks,
Alex


-- 
------------------------------------------------------------------
Alexandre Sidorenko             email: alexs@hplinux.canada.hp.com
Global Solutions Engineering:   Unix Networking
Hewlett-Packard (Canada)
------------------------------------------------------------------