cvs commit: src/sys/net if.h

Andrew Gallatin gallatin at cs.duke.edu
Tue Jun 12 10:03:17 UTC 2007


Andre Oppermann [andre at FreeBSD.org] wrote:
> Sam Leffler wrote:
> > Andre Oppermann wrote:
> >> andre       2007-06-11 20:08:12 UTC
> >>
> >>   FreeBSD src repository
> >>
> >>   Modified files:
> >>     sys/net              if.h   Log:
> >>   Add IFCAP_LRO flag for drivers to announce their TCP Large Receive 
> >> Offload
> >>   capabilities.
> >>     Revision  Changes    Path
> >>   1.108     +1 -0      src/sys/net/if.h
> >> http://cvsweb.FreeBSD.org/src/sys/net/if.h.diff?r1=1.107&r2=1.108
> >>
> >>
> > There are many offload capabilities defined that are not well thought 
> > out.  In particular we do not distinguish between ipv4 and ipv6 for 
> > things like cksum and tso so there's no way to disable individual 
> > features.  The ability to tweak LRO is clearly needed and clearly belong 
> > as an ifnet capability but unilateraly deciding this is the wrong approach.
> 
> We do distinguish between TSO4 and TSO6 since from the beginning.
> 
> LRO is not hardware dependent but done in software.  However it must
> be handled from inside the drivers RX function.  Our software
> implementation will be IPv4 and IPv6 aware right from the beginning.
> 
> It is not an unilateral decision if other directly affected developers
> agree (Drew, Jack).

I'm fine with it, as long as it is "OK" for the current in-driver
software implementations to not support v6.  I believe mxge and cxgb
both only supprt v4.

Drew


More information about the cvs-src mailing list