[rc5] v2 Mac client question

Jeff Leigh robocop at u.washington.edu
Fri Jul 11 17:23:14 EDT 1997


That is what I was saying. BTW, how did you have an input buffer of 56,
but an output buffer of only 50?? 

I think the input and output logic is separate. The input buffer gets new
keys when it is empty and the output buffer sends out completed keys when
it is full. Neither one knows what the other one is doing. When the block
of 50 is sent out, no new ones are retrieved because there are still keys
waiting in the buff-in file.


On Fri, 11 Jul 1997, Ken Hildabolt wrote:

> In <Pine.OSF.3.96.970711154449.14788A-100000 at saul3.u.washington.edu>, on
> 07/11/97 
>    at 03:46 PM, Jeff Leigh <robocop at u.washington.edu> said:
> 
> I don't agree with the following at all. Take my case where I had 56 keys
> running at one time. It would run until the buff-out had a count of 50 in
> it. Then it would connect and send 49, but would not get any new. Then it
> would do the remaining ones in the in-buff (6 or 7 or whatever), then
> connect, not send anything but would get 50 more and then repeat this
> process.
> 
> That's why I wrote a message questioning the logic that the program uses.
> 
> Ken
> 
> }If you are just buffering one block, it should send it out as soon as it
> }is done. If not, then it should write it to a file called buff-out and
> }then send it later when the output buffer is full. Either way it should
> }still say 'Sent x block(s) to server' somewhere in the logfile.
> 
> -----------------------------------------------------------
>   "Ken Hildabolt" <hildajk1 at aplcomm.jhuapl.edu>
> Applied Physics Laboratory / The Johns Hopkins University
> -----------------------------------------------------------
> 
> ----
> To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.
> 

----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list