>>>>> On Wed, 30 Jun 2004 13:08:25 -0400, Robert Story (Coders) <[EMAIL PROTECTED]> 
>>>>> said:

DS> Robert> it stands at 2-1.
DS> 
DS> Yes - I'm getting used to that :-(

Robert> Well, we've all been there. You just need to come work for us,
Robert> and then together we can rule the world! Mwuuuhaaaaahaaaaa!

I've scanned the thread (I'm trying to catch up on mail again after
traveling and deadline pushing).  My summary, bluntly put:

1) The auto_next flag is good because it lets people reduce the stack
   overhead for those with weak stacks.

2) However, Robert is wrong that it helps APIs.  Since we can not and
   should not delete the ability to have handlers do things
   themselves, it is unlikely it'll save us a huge amount of work if
   we change APIs in the future.  It's much more likely that since
   there will be so many places that don't use it because they have to
   call down themselves that we'll never change the APIs but just
   augment them or add new ones in a backwards-compatible fashion.

3) The auto_next flag does make tracing stuff much more of a pain.
   I'd be tempted to make the flag dynamic in some way, and have a
   macro or null function called based on the current flag setting...
   writing code that will work optimized in some cases, but allow for
   better debugging in others.  sigh...  but it would be a mess, so
   never mind.

   <aside> I've often thought we need to introduce a whole slew of
   macros which would end up producing a Net-SNMP specific coding
   style that would transcribe into real C in order to facility
   short-code writing.  Realistically, though, I don't think this is
   wise every time I think it through because as much as it helps code
   size, it increases the learning curve as well.  Try reading the
   linux kernel and understanding it.  I'm fairly robust at C, but I
   hate reading the linux kernel.  And I've even read a book about it.

4) I've actually always thought that caching should be done
   automatically when possible instead.  Like the way we have the
   injectable handler that wraps around any iterator (and should do
   other) helper implemented modules.  Less code, free caching, just
   not *quite* as efficient at collecting data.  However, it turns out
   that the caching itself is so fast in the tree structure that even
   when applied to something that already caches underneath (like the
   tcp table) you *still* got a speed improvement.

5) There was a 5.  I forgot it.

-- 
Wes Hardaker
Sparta


-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com
_______________________________________________
Net-snmp-coders mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to