[mpich2-commits] r5435 - mpich2/trunk

[balaji at mcs.anl.gov]

Author: balaji
Date: 2009-10-08 12:35:22 -0500 (Thu, 08 Oct 2009)
New Revision: 5435

Modified:
   mpich2/trunk/README.vin
   mpich2/trunk/RELEASE_NOTES
Log:
Updated the README to mention the 8 EP restriction in MX and removed
it from the RELEASE_NOTES.

Reviewed by thakur at mcs.


Modified: mpich2/trunk/README.vin
===================================================================
--- mpich2/trunk/README.vin	2009-10-08 14:06:24 UTC (rev 5434)
+++ mpich2/trunk/README.vin	2009-10-08 17:35:22 UTC (rev 5435)
@@ -547,7 +547,8 @@
 shared library search path. This can be done by adding the path to
 /etc/ld.so.conf, or by setting the LD_LIBRARY_PATH variable in your
 .bashrc (or .tcshrc) file.  It's also possible to set the shared
-library search path in the binary. If you're using gcc, you can do this by adding
+library search path in the binary. If you're using gcc, you can do
+this by adding
 
   LD_LIBRARY_PATH=/path/to/lib
 
@@ -557,6 +558,12 @@
 
 ... as arguments to configure.
 
+By default, MX allows for only eight endpoints per node causing
+ch3:nemesis:mx to give initialization errors with greater than 8
+processes on the same node (this is an MX error and not an inherent
+limitation in the MPICH2/Nemesis design). If needed, this can be set
+to a higher number when MX is loaded. We recommend the user to contact
+help at myri.com for details on how to do this.
 
 Shared-memory optimizations are enabled by default to improve
 performance for multi-processor/multi-core platforms. They can be

Modified: mpich2/trunk/RELEASE_NOTES
===================================================================
--- mpich2/trunk/RELEASE_NOTES	2009-10-08 14:06:24 UTC (rev 5434)
+++ mpich2/trunk/RELEASE_NOTES	2009-10-08 17:35:22 UTC (rev 5435)
@@ -11,11 +11,6 @@
 
  * ch3:nemesis:mx does not support dynamic processes at this time.
 
- * ch3:nemesis:mx does not typically work with greater than 8 processes
-   on a node (initialization failures will occur).  This is an MX
-   limitation due to limited SRAM on the Myrinet NICs, not an inherent
-   limitation of MPICH2/Nemesis.
-
  * ch3:ssm and ch3:shm do not support thread safety.
 
  * ch3:shm does not support dynamic processes (e.g., MPI_Comm_spawn).