On Wednesday 07 December 2005 21:30, Christiaan Simons wrote:
> IMHO we need to change our reference implementation
> to be correct for any architecture.
>
> I've got some ideas:
>
> * Add octets (byte-by-byte) instead of 16 bit words as shown in RFC1071 p5.
> This will lower the performance for >8 bit architectures (a lot),
> but will be alignment independant.

I would opt for this solution as it will provide an algorithm (although a 
really slow one) that works for everybody. It can also be used as a reference 
for implementing an optimized solution for different architectures. There 
should be a big warning text accompanying it that it is grossly suboptimal 
and should not be used in production environments, though.

I think that it would be beneficial to collect all the optimized 
implementations of the checksum routine under the arch/ directory so that 
there would be a good chance of having one available when starting to use 
lwip. And if not, they'd be good references for someone wanting to provide 
their own.

-- 
Atte Kojo
Software Development Engineer
Elekta Neuromag Oy
Street address: Elimäenkatu 22, Helsinki, Finland
Mailing address: P.O. Box 68, FIN-00511 HELSINKI, Finland
Tel: +358 9 756 24084 (direct), +358 9 756 2400 (operator)
Fax: +358 9 756 24011
E-mail: [EMAIL PROTECTED]
www.elekta.com


_______________________________________________
lwip-users mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/lwip-users

Reply via email to