Dear all,

I've implemented a mechanism in the kernel removing transmission time
checksumming by prefabricating packets. However, this mechanism now seem
to be unneccessary due to on-board CPUs. However, how does this work. Is
all the protocol processing done on-board? How is IP fragmentation
handled? Is there any documentation? Which files should I look at?

Thanks.

Best regards,
Paal Halvorsen

>Folks...
>
>I have committed my changes to support in-bound and out-bound IPv4, TCP,
>and UDP checksumming to NetBSD-current.  It is currently supported on the
>NatSemi DP83820 Gigabit Ethernet (gsip), 3Com Etherlink XL (ex), and
>Alteon Tigon/Tigon2 Gigabit Ethernet (ti).  There is also support for
>caching the IPv6 TCP pseudo-header checksum, which should save some CPU
>time for IPv6 users, as well.
>
>Please report any problems you experience with this code via send-pr(1).
>
>Shar and enjoy!
>
>-- 
>        -- Jason R. Thorpe <thorpej@zembu.com>


-ph
---       . o  o   .  o  .  o ..  o ..  o .. o oo . o  . o o o
         _n_n_n____i_i _++++++_ _______ ________ _+++++++++++_
      *>(____________I I______I I_____I I______I I___________I
 __^__  /ooOOOO OOOOoo  oo ooo  oo   oo oo    oo ooo       ooo  __^__
( ___ )--------------------------------------------------------( ___ )
 | / | Paal Halvorsen   UniK - Center for technology at Kjeller | \ |
 | / |                                       University of Oslo | \ |
 | / | Phone: +47 64844731                               PB. 70 | \ |
 | / | Phone: +47 64844700 (switchboard)       N - 2027 KJELLER | \ |
 |_/_| Fax:   +47 63818146                               Norway |__|
(_____)-- E-mail: paalh@unik.no -- http://www.unik.no/~paalh --(_____)