Date: Sat, 28 Nov 1998 21:31:16 -0500 (EST) From: Daniel Eischen <eischen@vigrid.com> To: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, eivind@FreeBSD.ORG, lists@tar.com Subject: Re: cvs commit: src/include pthread.h src/lib/libc_r/uthread uthread_mattr_kind_np.c uthread_mutex.c Message-ID: <199811290231.VAA28198@pcnet1.pcnet.com>
next in thread | raw e-mail | index | archive | help
> >eivind 1998/11/28 15:52:59 PST > > > > Modified files: > > include pthread.h > > lib/libc_r/uthread uthread_mattr_kind_np.c uthread_mutex.c > > Log: > > Add support for pthread_mutexattr_settype(). As a side effect of > > testing this, fix MUTEX_TYPE_COUNTING_FAST. Recursive locks now work. > ^^^^^^^^^^^^^^^^^^^^^^^^ > You might want to check localtime.c in libc. I seem to recall from > looking at this a couple of weeks ago that the "thread safe", ie. > libc_r, implementation makes recursive calls to pthread_mutex_lock. > Therefore, its possible this change will break some of the localtime > related functions. If so, it could be fixed by having localtime.c > use recursive mutexes, or better yet, by rewriting it so it doesn't > make recursive calls to pthread_mutex_lock. And can we also make a default mutex an error checking mutex? Both POSIX and Single UNIX Spec 2 allow this. SS2 allows that you can map PTHREAD_MUTEX_DEFAULT to any one of the other mutex types - it should be mapped to PTHREAD_MUTEX_ERRORCHECK. If at all possible, we should let the programmer know that he's hit a deadlock condition. Dan Eischen eischen@vigrid.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199811290231.VAA28198>