Subj : Re: pthreads-win32 question...
To   : comp.programming.threads
From : Randy Howard
Date : Fri Feb 11 2005 12:03 am

In article <cugruc$87i$1@nntp.webmaster.com>, davids@webmaster.com says...
> 
> "Randy Howard" <randyhoward@FOOverizonBAR.net> wrote in message 
> news:MPG.1c75abaf13494ec1989f8b@news.verizon.net...
> 
> > Per Posix pthreads, is it "legal" wrt the standard for an implementation
> > (such as pthreads-win32) to return 0 for PTHREAD_STACK_MIN, or 0 for the
> > default stack size obtained from pthread_attr_getstacksize(), after
> > pthread_attr_init() ?
> 
>     Stack operations are inherently non-portable. One could even concieve of 
> a conforming implementation that didn't have a stack.

I don't disagree, however, my understanding is if _POSIX_THREAD_ATTR_STACKSIZE 
is defined by the implementation, then not only does it have a stack, but you 
have some degree of control over it with the appropriate pthread functions.

It seems to me that if it does have it, then having either the minimum stack 
size, or the current stack size be zero on that platform is counterintuitive.

-- 
Randy Howard (2reply remove FOOBAR)
"Making it hard to do stupid things often makes it hard
 to do smart ones too." -- Andrew Koenig

.