Author: bugman Date: Thu Mar 8 12:21:13 2012 New Revision: 15464 URL: http://svn.gna.org/viewcvs/relax?rev=15464&view=rev Log: Copied the OpenMPI + Linux 2.6 100% master problem to the API documentation. The text has also been updated a little. Modified: 1.3/multi/__init__.py 1.3/multi/test_implementation.py Modified: 1.3/multi/__init__.py URL: http://svn.gna.org/viewcvs/relax/1.3/multi/__init__.py?rev=15464&r1=15463&r2=15464&view=diff ============================================================================== --- 1.3/multi/__init__.py (original) +++ 1.3/multi/__init__.py Thu Mar 8 12:21:13 2012 @@ -106,6 +106,16 @@ ========= See the script 'test_implementation.py' for a basic example of a reference, and full, implementation of the multi-processor package. + + +5 Issues +======== + +For multi-core systems and Linux 2.6, the following might be required to prevent the master processor from taking 100% of one CPU core while waiting for the slaves: + +# echo "1" > /proc/sys/kernel/sched_compat_yield + +This appears to be an OpenMPI problem with late 2.6 Linux kernels. """ Modified: 1.3/multi/test_implementation.py URL: http://svn.gna.org/viewcvs/relax/1.3/multi/test_implementation.py?rev=15464&r1=15463&r2=15464&view=diff ============================================================================== --- 1.3/multi/test_implementation.py (original) +++ 1.3/multi/test_implementation.py Thu Mar 8 12:21:13 2012 @@ -27,9 +27,11 @@ Linux ----- -For multi-core systems and late 2.6 Linux kernels, the following as root might be required to prevent the master processor from taking 100% of one CPU core while waiting for the slaves. +For multi-core systems and Linux 2.6, the following might be required to prevent the master processor from taking 100% of one CPU core while waiting for the slaves: # echo "1" > /proc/sys/kernel/sched_compat_yield + +This appears to be an OpenMPI problem with late 2.6 Linux kernels. """ # Python module imports.