This is the mail archive of the cygwin mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Cygwin PThreads bug?


Hey everyone,

I'm writing a very thread-intensive application using Cygwin and Cygwin's PThreads implementation, and I'm running into a few problems. As far as I can tell from the spec:

http://www.opengroup.org/onlinepubs/000095399/functions/pthread_rwlock_rdlock.html

This C program:

#include <stdio.h>
#include <pthread.h>
#include <assert.h>
#include <errno.h>

int main() {
	pthread_rwlock_t rw_lock;

pthread_rwlock_init(&rw_lock,NULL);

	//Lock it once for read...
	assert(pthread_rwlock_rdlock(&rw_lock)==0);
	
	//Lock it again for read...
	int err=pthread_rwlock_rdlock(&rw_lock);

printf("Err %d\n", err);

assert (err==EAGAIN || err==0);

	if (err!=EAGAIN)
	  pthread_rwlock_unlock(&rw_lock);

pthread_rwlock_unlock(&rw_lock);

	pthread_rwlock_destroy(&rw_lock);
	
	return 0;
}

Should run correctly. That is, the second call to pthread_rwlock_rdlock should either succeed (returning zero), having acquired the read lock a second time, or it should fail and return EAGAIN, if the number of simultaneous allowed read locks has been exceeded. It should not fail and return EDEADLK (45), which it is currently doing. Am I reading the spec wrong or is Cygwin non-conforming?

Cheers,
Nicholas Sherlock


-- 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/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]