Contemplating signal changes, but...

[Chris Faylor]

I'm contemplating changing the signal handler in some fundamental
ways.  I figure it's time since I've essentially rewritten it three
times since 1997.  I've been stabilizing things a lot in the snapshots
but I have never liked the performance hits that I"ve been introducing.

To implement my plan, I need to have some quickly available per-thread
memory.  I won't go into details, but basically what I'm looking for is
to use the base of the stack for fast per-thread storage.  I can
guarantee that the stack has space available for me down there when I
control thread creation but I can't guarantee this when I haven't
created the thread or when the cygwin DLL is being dynamically loaded.

I can probably put a front-end on CreateThread to work around the
problem with people creating threads but I was thinking of just
turning off signals entirely when the Cygwin DLL is dynamically
loaded.  I could also just revert to the old method of delivering
signals but that would be a little more work.

I wanted to see how people (i.e., Mumit) felt about the idea of
signals being inoperative in a dynamically loaded cygwin.  I don't
think that this is a big deal but I don't know for sure.

Opinions?

cgf