POSIX semaphore abandonment problem

[Tristan Schmelcher]

Erm, I didn't. :/ Thanks. I had written off pthread mutexes because I
thought there wouldn't be a way to atomically create and initialize
them into a file on disk, but I suppose the app could do the same
thing as the sem_open implementation does and just create a temporary
file first and then try to move it without overwriting another, via
link(2). So yeah, I suppose there is no need for a new API.

Thanks.

On Tue, Nov 10, 2009 at 5:10 PM, Samuel Thibault
<samuel.thibault@ens-lyon.org> wrote:
> Hello,
>
> Do you know the robust mutexes, EOWNERDEAD and the
> pthread_mutex_consistent() call?
>
> Samuel
>