svn commit: r232345 - user/andre/tcp_workqueue/sys/netinet

Andre Oppermann andre at
Mon Mar 5 10:56:29 UTC 2012

On 05.03.2012 07:19, Gleb Smirnoff wrote:
> On Mon, Mar 05, 2012 at 12:15:24AM +0100, Ivan Voras wrote:
> I>  >  I'm wary on having every other Jon Doe doing his own "research" on the
> I>  >  most optimal icwnd from his point of view.
> I>
> I>  I still think that such prohibition is useless in the light of Linux
> I>  having it tunable.
> Agreed. We should have a reasonable default, but allow people to shoot
> theirselves in their foot. Some people really demand this possibility!

Thing is they don't shoot into their own foot but everybody else's foot
in a non-obvious and difficult to debug manner.  While it is user-changeable
in Linux it not through the normal easy /proc path equaling a sysctl on our

FreeBSD never had the option to adjust the ICWND except through enabling
rfc3390.  I'll hold it the same with this one and refuse to make it arbitrarily
user tuneable.  It really is not something a normal user/admin should touch.


More information about the svn-src-user mailing list