Commit by rhc :: r22848 /trunk/ (10 files in 4 dirs):

With Jeff's help, get the libraries to link as required.

Update ompi_info and orte-info to include the new framework.

Fix some selection logic and a typo'd variable name

Still remains ompi_ignored until we complete testing

Commit by jurenz :: r22847 /trunk/ompi/contrib/vt/vt/ (4 files in 2 dirs):

Fixed segfault which might occur if the application uses fork's

Commit by jurenz :: r22846 /vendor/vampirtrace/VampirTrace-5.8.1.10/:

tagging VampirTrace-5.8.1.10

Commit by jurenz :: r22845 /vendor/vampirtrace/current/vtlib/ (vt_comp_gnu.c vt_otf_gen.c vt_otf_trc.c):

Fixed segfault which might occur if the application uses fork's

Commit by bosilca :: r22844 /trunk/ompi/mca/mpool/pcie/:

The mpool should take the same route as the rest of the pcie modules.

Commit by rhc :: r22843 /trunk/orte/mca/state/ (7 files in 2 dirs):

Use the OMPI_CHECK_PACKAGE macro to check both header file and library existence before building the component.

Still haven't gotten the right libraries linked in...so add ompi_ignore/unignore until we get it all fully integrated.

Commit by rhc :: r22842 /trunk/README:

Update the README file to reflect the new thread configure options

Commit by rhc :: r22841 /trunk/ (19 files in 13 dirs):

Merge in the modified thread configure option branch per today's telecon.

Remove the --enable-progress-threads option as this is no longer functional, and hardcode OPAL_ENABLE_PROGRESS_THREADS to 0.

Replace the --enable-mpi-threads option with --enable-mpi-thread-multiple as this is clearer as to meaning. This option automatically turns "on" opal thread support if it wasn't already so specified. If the user specifies --disable-opal-multi-threads --enable-mpi-thread-multiple, we will error out with a message

Add a new --enable-opal-multi-threads option that turns "on" opal thread support without doing anything wrt mpi-thread-multiple

Commit by rhc :: r22840 /trunk/ompi/config/ompi_check_pcie.m4:

Remove stale config file - the pcie btl has already been removed

Commit by rhc :: r22839 /trunk/ (26 files in 8 dirs):

Add a new framework to ORTE for saving and recovering state information. Two components are included that use the db or dbm library for storing the data, with a distributed hash table component coming later.

Note that each of these components will only be selected if specifically requested - otherwise, a "NULL" component will be used. The framework is only opened by the HNP and orteds, though neither is currently coded to save/restore state

Commit by rhc :: r22838 /branches/v1.4/ (4 files in 4 dirs):

Closes #2309: update openib INI file

Submitted by jsquyres, no review required, RM-approved by bosilca

r22708 and r22723
Update the openib INI file

Commit by rhc :: r22837 /branches/v1.4/ (4 files in 4 dirs):

Closes #2304: fix unlocked access in osc_rdma

Submitted by cyeoh, reviewed by bosilca, RM-approved by bosilca

r22700
fixes a problem where there is an unlocked access to the mca_osc_rdma_component.c_modules in ompi_osc_rdma_windx_to_module.

Commit by rhc :: r22836 /branches/v1.4/ (4 files in 4 dirs):

 Closes #2298: remove race condition in openib endpoint

Submitted by cyeoh, reviewed by bosilca, RM-approved by bosilca

r22684
This fixes #2265 removing a race in the openib btl endpoint when increasing sequence numbers.

Commit by rhc :: r22835 /branches/v1.4/ompi/mca/ (6 files in 3 dirs):

Closes #2297: delay release of registered memory

Submitted by cyeoh, reviewed by bosilca, RM-approved by bosilca

r22683 (patch)
The release of memory used by registration lists in rcaches must be delayed until the rcache lock is not held or deadlock can occur ( fixes #2111 ). Should not deregister memory with the rcache lock held otherwise a deadlock can occur as the lower level infiniband libraries can free memory ( fixes #2110 )

Commit by rhc :: r22834 /branches/v1.4/ (6 files in 4 dirs):

Closes #2296: protect ip_procs list

Submitted by cyeoh, reviewed by bosilca, RM-approved by bosilca

r22682
The ib_procs list in the openib btl is accessed without the ib lock in some cases. This causes races when running multithreaded. This patch adds protection of the ib_procs list with the ib_lock.

Commit by rhc :: r22833 /branches/v1.4/ (9 files in 6 dirs):

Closes #2291: fix rcache vma lock

Submitted by cyeoh, reviewed by bosilca, RM-approved by bosilca

r22679
Destroying an rcache vma while the rcache lock is held as this can result in a low level free of memory which can require the rcache lock resulting in a deadlock

Commit by rhc :: r22832 /branches/v1.4/ (4 files in 4 dirs):

Closes #2288: Fix lock in mca_btl_openib_alloc

Submitted by cyeoh, reviewed by bosilca, RM-approved by bosilca

r22678
Fixes bug where the wrong lock is taken in mca_btl_openib_alloc when protecting the no_wqe_pending_frags list.

Commit by jsquyres :: r22831 /trunk/ompi/ (8 files in 2 dirs):

Refs #2273

After talking to both Brian and George, the conensus was to just
remove the flag and the test function. Begone, evil spirits, BEGONE!

Commit by rhc :: r22830 /branches/v1.4/orte/mca/plm/process/plm_process_module.c:

Cleanup a variable name per Shiqing's request

Commit by rusraink :: r22829 /trunk/ (5 files in 5 dirs):

• Further patches for support on NetBSD (and DragonFly) by Aleksej Saushev. Dont use bash or bashism in shell scripts We should use Posix' setpgid(0,0), which is equivalent to setpgrp().