Kacheong Poon wrote:
> Yunsong (Roamer) Lu wrote:
> 
>> Theoretically, it's not impossible. *Assuming* it's valuable, TCP can 
>> pass down this requirement along with other LSO information, Extended 
>> SOFT LSO can handle it easily before hardware LSO. *But*, what value do 
>> you see with such implementation?
 >
> This is the problem.  For every changes in a protocol,
> somehow the LSO may need to be changed to accommodate
I'm saying it's possible to do so if it's *really* valuable. But I don't 
see the value of supporting such MTU adjustment with LSO at all. See my 
reply below.

> that.  And how is soft LSO extended?  To have a
> construct like MDT?
Basic SOFT LSO functions as a LSO simulator that parse the over-sized 
packet to a chain of MTU-sized packets.

>> LSO is not implemented for every single transmission, but for a "stable" 
>> transmission that doesn't involve PMTU-like "trick", since such rare 
>> case won't impact performance much. :;
> 
> 
> I am not sure what is meant by "stable?"  For example,
> one common problem using VPN, such as our own punchin,
> is PMTU mismatch.  So people artificially lower the
> interface MTU hoping for the best.  When we add this
> kind of better PMTUd support to TCP, TCP can adjust the
> segment size to detect the correct size to use.  It
> is a common case.
Actually, this could be easily handled inside TCP.
Do LSO with unique MSS within each transmission, and MSS can be adjusted 
immediately for the next transmissions. Keep this in mind: LSO can 
benefit performance whenever there is >MSS payload that could be sent 
with the same MSS size.

Thanks,

Roamer


-- 

# telnet (650)-786-6759 (x86759)
Connected to Solaris.Sun.COM.
login: Lu, Yunsong
Last login: January 2, 2007 from beyond.sfbay
[EMAIL PROTECTED]    v1.04    Since Mon Dec. 22, 2003
[EMAIL PROTECTED] Networking]# cd ..
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to