Adding thread support for BeOS by Chris Herborth.
diff --git a/Python/thread_beos.h b/Python/thread_beos.h
new file mode 100644
index 0000000..f2024f3
--- /dev/null
+++ b/Python/thread_beos.h
@@ -0,0 +1,388 @@
+/***********************************************************
+Copyright 1991-1995 by Stichting Mathematisch Centrum, Amsterdam,
+The Netherlands.
+
+                        All Rights Reserved
+
+Permission to use, copy, modify, and distribute this software and its
+documentation for any purpose and without fee is hereby granted,
+provided that the above copyright notice appear in all copies and that
+both that copyright notice and this permission notice appear in
+supporting documentation, and that the names of Stichting Mathematisch
+Centrum or CWI or Corporation for National Research Initiatives or
+CNRI not be used in advertising or publicity pertaining to
+distribution of the software without specific, written prior
+permission.
+
+While CWI is the initial source for this software, a modified version
+is made available by the Corporation for National Research Initiatives
+(CNRI) at the Internet address ftp://ftp.python.org.
+
+STICHTING MATHEMATISCH CENTRUM AND CNRI DISCLAIM ALL WARRANTIES WITH
+REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL STICHTING MATHEMATISCH
+CENTRUM OR CNRI BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL
+DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
+PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
+TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
+PERFORMANCE OF THIS SOFTWARE.
+
+BeOS thread support by Chris Herborth (chrish@qnx.com)
+******************************************************************/
+
+#include <kernel/OS.h>
+#include <support/SupportDefs.h>
+#include <errno.h>
+
+/* ----------------------------------------------------------------------
+ * Fast locking mechanism described by Benoit Schillings (benoit@be.com)
+ * in the Be Developer's Newsletter, Issue #26 (http://www.be.com/).
+ */
+typedef struct benaphore {
+	sem_id _sem;
+	int32  _atom;
+} benaphore_t;
+
+static status_t benaphore_create( const char *name, benaphore_t *ben );
+static status_t benaphore_destroy( benaphore_t *ben );
+static status_t benaphore_lock( benaphore_t *ben );
+static status_t benaphore_timedlock( benaphore_t *ben, bigtime_t micros );
+static status_t benaphore_unlock( benaphore_t *ben );
+
+static status_t benaphore_create( const char *name, benaphore_t *ben )
+{
+	if( ben != NULL ) {
+		ben->_atom = 0;
+		ben->_sem = create_sem( 0, name );
+		
+		if( ben->_sem < B_NO_ERROR ) {
+			return B_BAD_SEM_ID;
+		}
+	} else {
+		return EFAULT;
+	}
+	
+	return EOK;
+}
+
+static status_t benaphore_destroy( benaphore_t *ben )
+{
+	if( ben->_sem >= B_NO_ERROR ) {
+		status_t retval = benaphore_timedlock( ben, 0 );
+		
+		if( retval == EOK || retval == EWOULDBLOCK ) {
+			status_t del_retval = delete_sem( ben->_sem );
+			
+			return del_retval;
+		}
+	}
+
+	return B_BAD_SEM_ID;
+}
+
+static status_t benaphore_lock( benaphore_t *ben )
+{
+	int32 prev = atomic_add( &(ben->_atom), 1 );
+	
+	if( prev > 0 ) {
+		return acquire_sem( ben->_sem );
+	}
+	
+	return EOK;
+}
+
+static status_t benaphore_timedlock( benaphore_t *ben, bigtime_t micros )
+{
+	int32 prev = atomic_add( &(ben->_atom), 1 );
+	
+	if( prev > 0 ) {
+		status_t retval = acquire_sem_etc( ben->_sem, 1, B_TIMEOUT, micros );
+		
+		switch( retval ) {
+		case B_WOULD_BLOCK:	/* Fall through... */
+		case B_TIMED_OUT:
+			return EWOULDBLOCK;
+			break;
+		case B_OK:
+			return EOK;
+			break;
+		default:
+			return retval;
+			break;
+		}
+	}
+	
+	return EOK;
+}
+
+static status_t benaphore_unlock( benaphore_t *ben )
+{
+	int32 prev = atomic_add( &(ben->_atom), -1 );
+	
+	if( prev > 1 ) {
+		return release_sem( ben->_sem );
+	}
+	
+	return EOK;
+}
+
+/* ----------------------------------------------------------------------
+ * Initialization.
+ */
+static void _init_thread( void )
+{
+	/* Do nothing. */
+	return;
+}
+
+/* ----------------------------------------------------------------------
+ * Thread support.
+ *
+ * Only ANSI C, renamed functions here; you can't use K&R on BeOS,
+ * and there's no legacy thread module to support.
+ */
+
+static int32 thread_count = 0;
+
+int PyThread_start_new_thread( void (*func)(void *), void *arg )
+{
+	status_t success = 0;
+	thread_id tid;
+	char name[B_OS_NAME_LENGTH];
+	int32 this_thread;
+
+	dprintf(("start_new_thread called\n"));
+
+	/* We are so very thread-safe... */
+	this_thread = atomic_add( &thread_count, 1 );
+	sprintf( name, "python thread (%d)", this_thread );
+
+	tid = spawn_thread( (thread_func)func, name,
+	                    B_NORMAL_PRIORITY, arg );
+	if( tid > B_NO_ERROR ) {
+		success = resume_thread( tid );
+	}
+
+	return ( success == B_NO_ERROR ? 1 : 0 );
+}
+
+long PyThread_get_thread_ident( void )
+{
+	/* Presumed to return the current thread's ID... */
+	thread_id tid;
+	tid = find_thread( NULL );
+	
+	return ( tid != B_NAME_NOT_FOUND ? tid : -1 );
+}
+
+static void do_exit_thread( int no_cleanup )
+{
+	int32 threads;
+
+	dprintf(("exit_thread called\n"));
+
+	/* Thread-safe way to read a variable without a mutex: */
+	threads = atomic_add( &thread_count, 0 );
+
+	if( threads == 0 ) {
+		/* No threads around, so exit main(). */
+		if( no_cleanup ) {
+			_exit(0);
+		} else {
+			exit(0);
+		}
+	} else {
+		/* Oh, we're a thread, let's try to exit gracefully... */
+		exit_thread( B_NO_ERROR );
+	}
+}
+
+void PyThread_exit_thread( void )
+{
+	do_exit_thread(0);
+}
+
+void PyThread__exit_thread( void )
+{
+	do_exit_thread(1);
+}
+
+#ifndef NO_EXIT_PROG
+static void do_exit_prog( int status, int no_cleanup )
+{
+	dprintf(("exit_prog(%d) called\n", status));
+
+	/* No need to do anything, the threads get torn down if main() exits. */
+
+	if (no_cleanup) {
+		_exit(status);
+	} else {
+		exit(status);
+	}
+}
+
+void PyThread_exit_prog( int status )
+{
+	do_exit_prog(status, 0);
+}
+
+void PyThread__exit_prog( int status )
+{
+	do_exit_prog(status, 1);
+}
+#endif /* NO_EXIT_PROG */
+
+/* ----------------------------------------------------------------------
+ * Lock support.
+ */
+
+static int32 lock_count = 0;
+
+type_lock PyThread_allocate_lock( void )
+{
+	benaphore_t *lock;
+	status_t retval;
+	char name[B_OS_NAME_LENGTH];
+	int32 this_lock;
+	
+	dprintf(("allocate_lock called\n"));
+
+	lock = (benaphore_t *)malloc( sizeof( benaphore_t ) );
+	if( lock == NULL ) {
+		/* TODO: that's bad, raise MemoryError */
+		return (type_lock)NULL;
+	}
+
+	this_lock = atomic_add( &lock_count, 1 );
+	sprintf( name, "python lock (%d)", this_lock );
+
+	retval = benaphore_create( name, lock );
+	if( retval != EOK ) {
+		/* TODO: that's bad, raise an exception */
+		return (type_lock)NULL;
+	}
+
+	dprintf(("allocate_lock() -> %lx\n", (long)lock));
+	return (type_lock) lock;
+}
+
+void PyThread_free_lock( type_lock lock )
+{
+	status_t retval;
+
+	dprintf(("free_lock(%lx) called\n", (long)lock));
+	
+	retval = benaphore_destroy( (benaphore_t *)lock );
+	if( retval != EOK ) {
+		/* TODO: that's bad, raise an exception */
+		return;
+	}
+}
+
+int PyThread_acquire_lock( type_lock lock, int waitflag )
+{
+	int success;
+	status_t retval;
+
+	dprintf(("acquire_lock(%lx, %d) called\n", (long)lock, waitflag));
+
+	if( waitflag ) {
+		retval = benaphore_lock( (benaphore_t *)lock );
+	} else {
+		retval = benaphore_timedlock( (benaphore_t *)lock, 0 );
+	}
+	
+	if( retval == EOK ) {
+		success = 1;
+	} else {
+		success = 0;
+		
+		/* TODO: that's bad, raise an exception */
+	}
+
+	dprintf(("acquire_lock(%lx, %d) -> %d\n", (long)lock, waitflag, success));
+	return success;
+}
+
+void PyThread_release_lock( type_lock lock )
+{
+	status_t retval;
+	
+	dprintf(("release_lock(%lx) called\n", (long)lock));
+	
+	retval = benaphore_unlock( (benaphore_t *)lock );
+	if( retval != EOK ) {
+		/* TODO: that's bad, raise an exception */
+		return;
+	}
+}
+
+/* ----------------------------------------------------------------------
+ * Semaphore support.
+ *
+ * Guido says not to implement this because it's not used anywhere;
+ * I'll do it anyway, you never know when it might be handy, and it's
+ * easy...
+ */
+type_sema PyThread_allocate_sema( int value )
+{
+	sem_id sema;
+	
+	dprintf(("allocate_sema called\n"));
+
+	sema = create_sem( value, "python semaphore" );
+	if( sema < B_NO_ERROR ) {
+		/* TODO: that's bad, raise an exception */
+		return 0;
+	}
+
+	dprintf(("allocate_sema() -> %lx\n", (long) sema));
+	return (type_sema) sema;
+}
+
+void PyThread_free_sema( type_sema sema )
+{
+	status_t retval;
+	
+	dprintf(("free_sema(%lx) called\n", (long) sema));
+	
+	retval = delete_sem( (sem_id)sema );
+	if( retval != B_NO_ERROR ) {
+		/* TODO: that's bad, raise an exception */
+		return;
+	}
+}
+
+int PyThread_down_sema( type_sema sema, int waitflag )
+{
+	status_t retval;
+
+	dprintf(("down_sema(%lx, %d) called\n", (long) sema, waitflag));
+
+	if( waitflag ) {
+		retval = acquire_sem( (sem_id)sema );
+	} else {
+		retval = acquire_sem_etc( (sem_id)sema, 1, B_TIMEOUT, 0 );
+	}
+	
+	if( retval != B_NO_ERROR ) {
+		/* TODO: that's bad, raise an exception */
+		return 0;
+	}
+
+	dprintf(("down_sema(%lx) return\n", (long) sema));
+	return -1;
+}
+
+void PyThread_up_sema( type_sema sema )
+{
+	status_t retval;
+	
+	dprintf(("up_sema(%lx)\n", (long) sema));
+	
+	retval = release_sem( (sem_id)sema );
+	if( retval != B_NO_ERROR ) {
+		/* TODO: that's bad, raise an exception */
+		return;
+	}
+}