Re: [g-a-devel]Re: Gail Range strangeness ...



Hi Bill,

On Fri, 2002-01-04 at 16:36, Bill Haneman wrote:
> Let me know if there are any other than the listener list stuff we've 
> previously discussed.

	It is indeed the listener list stuff; in debugging it I'm seeing some
extraordinarily deep stack traces in the registry - lots and lots of
re-enterancy happening - perhaps because we're too slow to process and
forward requests :-) [ or possibly we need to push the requests
asynchronously or somesuch - would stop blocking, and help us mightily
in many ways ].

	So - if you could add to the TODO:

	* Registry:
		* async event pushing
		* max buffering handling on clients - to kill
 		  non-responding clients.

	If we don't implement the 2nd we'll end up with the 'Gconf issue'
nailing us - so that's quite important.

	Also - we should really use the re-enterant list code in cspi/ what is
your preferred way to expose it ? we could have a libspi/spi-private.h
that cspi used internally - there seems no shame there, and I could
split them out into libspi/spi-utils.c or somesuch - howabout that ?

	Regards,

		Michael.


-- 
 mmeeks gnu org  <><, Pseudo Engineer, itinerant idiot




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]