On Wed, 20 Aug 1997, Dmitri wrote:
[A chain letter]
Neat, thanks...too bad I stopped when one line mentioned something about a
chain letter...
Circle:
Anyway, as per an idea from Jeremy, I'll be attempting to move the socket
buffering code onto my buffer system. Then we'll finally have the ability
to free large buffers with a timeout value. I'm thinking that just
allocating them on the same buffer chain as everything else would work
best. I just increase the number of persistant buffers available.
Main problem is that my (2*) patch is the only time I've every really
delved into the comm.c buffers...
Any opinions? Should we keep all the user buffers on a separate chain and
allocate say...1 small buffer for every 2 characters and 1 large buffer for
10? Or should we just keep it one chain and increase the number of general
persistant buffers? (my current favorite)
And of course, any other suggestions are great too. :)
snprintf() eat your heart out...
--
greerga@muohio.edu me@null.net | Genius may have its limitations, but stupidity
http://www.muohio.edu/~greerga | is not thus handicapped. -- Elbert Hubbard
+------------------------------------------------------------+
| Ensure that you have read the CircleMUD Mailing List FAQ: |
| http://democracy.queensu.ca/~fletcher/Circle/list-faq.html |
+------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 12/08/00 PST