Re: [gpsd-dev] What is our current bug state?

[Hal Murray]

>> It passes a few quick tests with:
>>   CLOSE_DELAY = 0.2

> I can get away with 0, 0 on my Xeon box, though the tests do run a bit
> slower than they used to.

> I guess you'll have to jack up the delays some until you find a good set.  

Please set CLOSE_DELAY for Linux back to 0.2

----------

I have a NetBSD box that needs slow mode, and the current parameters aren't 
quite slow enough.  It got through scons check, but I also run an extra batch 
of 50 alternating pairs of superstar2.log and tcp-test.log.  They were the 
ones that gave me troubles a week or two ago.

It's not the close problem.  It's dropping clumps out of the middle.

I tried:
WRITE_PAD_SLOWDOWN = 1.0
CLOSE_DELAY_SLOWDOWN = 8.0
and that still failed,

Mumble.  I'll come back to that sometime.




-- 
These are my opinions.  I hate spam.