George Bosilca d265981c55 Don't always retain the proc, do it only for new procs. This enforce a strict policy in the BML, it has one and only one ref on each proc. ...

This commit was SVN r30429.

2014-01-26 17:26:04 +00:00

..

allocator

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

bcol

Fixes trac:4143

2014-01-24 17:56:52 +00:00

bml

Don't always retain the proc, do it only for new procs. This enforce a strict policy in the BML, it has one and only one ref on each proc.

2014-01-26 17:26:04 +00:00

btl

btl/ugni: fix leak in new sendi function.

2014-01-22 16:32:07 +00:00

coll

Fixes trac:4143

2014-01-24 17:56:52 +00:00

common

Provide option to have synchronous copy be asynchronous with a wait. For now,

2014-01-23 15:47:20 +00:00

crcp

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

dpm

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

fbtl

add support for PLFS filesystem

2014-01-22 21:16:15 +00:00

fcoll

add proper typcasts to intptr_t to avoid warnings on 32bit systems.

2014-01-10 16:19:04 +00:00

fs

add support for PLFS filesystem

2014-01-22 21:16:15 +00:00

io

According to a report from Adam Moody, there is a compile error with

2014-01-23 14:15:35 +00:00

mpool

udreg: free huge page allocations correctly

2014-01-14 16:26:06 +00:00

mtl

Fix a recursion in mxm send flow which happens when mpi starts a new send from the context of send completion callback.

2014-01-12 17:47:03 +00:00

op

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

osc

fix communicator double-free from pt2pt component, caused by r29938.

2014-01-12 17:38:14 +00:00

pml

Cleanup a couple of abstraction breaks found by Thomas Naughton

2014-01-22 21:36:24 +00:00

pubsub

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

rcache

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

rte

For large scale systems, we would like to avoid doing a full modex during MPI_Init so that launch will scale a little better. At the moment, our options are somewhat limited as only a few BTLs don't immediately call modex_recv on all procs during startup. However, for those situations where someone can take advantage of it, add the ability to do a "modex on demand" retrieval of data from remote procs when we launch via mpirun.

2014-01-11 17:36:06 +00:00

sbgp

coll/ml: add support for blocking and non-blocking allreduce, reduce, and

2014-01-22 15:39:19 +00:00

sharedfp

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

topo

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00

vprotocol

Fix longstanding issue with our multi-project support. Rather than using

2014-01-07 22:11:15 +00:00