This is weird - I'm not a pth guru (by any means), but I've used it a
fair amount, and am pretty familiar with it. A strange error has me
confused - right now, I've got a bunch of really cpu-intensive,
stack-heavy threads. They work fine. I added another one in, probably
the simplest of all, that basically opens a file, and does a pth_select
call on it. The select wakes up, but if i try to read from the file
descriptor, I get a stack overflow on that thread. Raising it's stack
size to 128000 solves it, but I'm wondering why a simple select/read
call can overflow the stack while the other threads are dozens of levels
deep, with a lot more stack-based data structures. Its more of a
curiosity than anything - there's no global stack parameters or anything
that would come into play by adding another thread, would there?
(Actually, I shut down several of the others, and the problem is
consistent, so it's probably not a matter of adding 1 to n and crossing
some threshold.)
Any conjectures are welcome.
--Matt Sherer
______________________________________________________________________
GNU Portable Threads (Pth) http://www.gnu.org/software/pth/
User Support Mailing List [EMAIL PROTECTED]
Automated List Manager (Majordomo) [EMAIL PROTECTED]
