Mail Archives: cygwin/2007/11/06/10:30:54
On Nov 6 15:12, Bruno Haible wrote:
> Hi,
>
> In the test program below, pthread_key_create is failing with return value
> EBUSY with no good reason.
> [...]
> The Cygwin version that I'm using is:
> $ grep VERSION /usr/include/cygwin/version.h | head -2
> #define CYGWIN_VERSION_DLL_MAJOR 1005
> #define CYGWIN_VERSION_DLL_MINOR 19
Time to update.
However, regardless of the version you're using, it's a bug in Cygwin.
The pthread_key_create function checks the incoming pthread_key_t pointer
for being a valid object already and,if so, bails out with EBUSY. This
is arguably wrong, as your example shows, which reuses pthread_key_t
slots already used before. Besides, EBUSY isn't defined as a valid
error code returned by pthread_key_create.
I'll fix that in CVS. For the time being, a working workaround would
be:
for (i = KEYS_COUNT - 1; i >= 0; i--) {
fprintf (stderr, "trying to get key #%d\n", i); fflush (stderr);
mykeys[i] = (pthread_key_t) NULL;
int err = pthread_key_create (&mykeys[i], free);
if (err != 0) { fprintf (stderr, "pthread_key_create -> %d\n", err); abort (); }
}
Thanks for the testcase!
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -