>>>>> On Mon, 2 Dec 2002 23:12:55 -0800 (PDT),
	Brian Chase <vaxzilla@jarai.org> said:

> My understanding was that Juli is suggesting a signal queue of length
> "one" for non-realtime signals using a generalized mechanism for all
> signal handling.

In that case, it's OK for me, too.
(I wrote same thing in this thread, too.)

> Admittedly, as a curious novice, I've no clue what it would take to
> implement in terms of memory or computation resources,

Using unnecessary resource is definitely bad thing, especially because 
kernel resource is wired to physical memory.
That's why I'm asking the reason why queueing multiple non-realtime
signals is needed.
--
soda