On Jan 19 08:29, SCOTT Damien wrote: > I have a multi-threaded process running under Cygwin that receives > messages from several IPC message queues (built using Cygwin's gcc, > version 4.8.3). The listener for each queue runs in its own thread, > each of which is created from the main thread. Each listener is using > a blocking call to msgrcv(). I want the main thread to be able to use > pthread_kill() to send a SIGUSR1 signal to each listening thread to > cause it to abandon its call to msgrcv() and then exit. > > My approach has been to create a signal mask containing just SIGUSR1 > and associate this with a signal handler (although the signal handler > contains no functionality, since all I need is for the blocking call > to msgrcv() to be abandoned). I initially set the signal mask to > block in the main thread, which should then be inherited by each of > the created listening threads. I only unblock the signal mask around > the call to msgrcv() in each listening thread. > > I have included a simplified version of my code below. I usually > don't see any evidence of the SIGUSR1 signal being received by either > of the listener threads. Occasionally the first thread unblocks as > expected but I have never seen both threads unblock. If I change my > code so that there is only a single listener thread, then everything > works correctly every time. Changing the msgrcv() call to a sleep > call also allows unblocking to occur correctly. Note that this > simplified code does not include removal of the created queues (since > it never manages to exit anyway) so I manually remove these after each > run using the ipcrm command. > > Example output: > > Running thread 2 with queue Id = 1441792 > Running thread 1 with queue Id = 1179649 > About to kill thread 1 > > Is there an issue with the Cygwin implementation of signal handling, > or is there a problem with my code? Any help would be much > appreciated.
It's an issue with the Cygwin code, probably. The msgrcv code ultimately hangs in a blocking ReadFile call on a named pipe, and this call is non-interruptible. Changing that requires a bigger change in Cygwin which will take time. For the time being, would it be much of a problem for you to convert your stuff to POSIX message queues(*)? These don't require running cygserver, and they are supposed not to suffer from this problem. Fixing them in case of a bug is much easier as well. Corinna (*) http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/mqueue.h.html -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat
pgpuCskVppN00.pgp
Description: PGP signature
