George Bosilca a4d99ddef6 More synchronizations for the Windows version. The problem came from ... the multiple threads accessing the OOB/registry asynchronously via the callbacks. The quickest solution (but definitively not the cleanest) is to serialize these callbacks in such a way that at any given time only one thread can execute a callbacks. This commit was SVN r15086.		2007-06-14 22:35:38 +00:00
..
configure.m4	Update the copyright notices for IU and UTK.	2005-11-05 19:57:48 +00:00
configure.params	Remove unneeded PARAM_INIT_FILE variable in configure.params files used by	2007-01-08 03:44:22 +00:00
Makefile.am	Just like r14289 on the ORTE trunk:	2007-04-12 11:19:42 +00:00
oob_tcp_addr.c	Properly handle case where user instructs the oob to not use all non-localhost	2007-05-31 02:29:44 +00:00
oob_tcp_addr.h	A couple cleanups for the IPv6 support:	2007-04-25 19:08:07 +00:00
oob_tcp_hdr.h	Rename OMPI_PROCESS_NAME_{HTON, NTOH} macros to ORTE_PROCESS_NAME_{HTON, NTOH}	2006-11-22 03:03:21 +00:00
oob_tcp_msg.c	More synchronizations for the Windows version. The problem came from	2007-06-14 22:35:38 +00:00
oob_tcp_msg.h	* use opal_free_list_item_t as the type of items stored in an opal_free_list_t,	2006-07-17 21:51:50 +00:00
oob_tcp_peer.c	Cleanup a couple warnings.	2007-06-12 14:11:06 +00:00
oob_tcp_peer.h	Merge in the ipv6 work from /tmp/ipv6-merge.	2007-04-25 01:55:40 +00:00
oob_tcp_ping.c	Merge in the ipv6 work from /tmp/ipv6-merge.	2007-04-25 01:55:40 +00:00
oob_tcp_recv.c	Fix a couple of problems with the way we were using orte_process_name_t	2007-06-11 19:12:11 +00:00
oob_tcp_send.c	Commit the orted-failed-to-start code. This correctly causes the system to detect the failure of an orted to start and allows the system to terminate all procs/orteds that *did* start.	2007-05-21 18:31:28 +00:00
oob_tcp.c	More synchronizations for the Windows version. The problem came from	2007-06-14 22:35:38 +00:00
oob_tcp.h	Properly handle case where user instructs the oob to not use all non-localhost	2007-05-31 02:29:44 +00:00