This enables IBM's software iWARP provider. With this driver you can
run iWARP/RDMA over any ethernet NIC. Useful for testing OMPI RDMA
logic without requiring an expensive RDMA adapter/infrastructure.
The Soft iWARP code is at: https://www.gitorious.org/softiwarp
At some point we added a sanity check to the btl base to ensure that
the btl flags match the available functions (this prevents user's from
specifying get or put when no function exists). This check was
disabling get for the sm btl since at the time of the check there is
no btl_get function. The simplest fix is to set a dummy value to btl_get
that will be overwritten with the proper value on btl initialization.
Closes#239.
If there are no usnic BTL modules, then just avoid sending any modex
message at all (other BTLs do this; it's safe to do).
The change is smaller than it looks: I added a "if 0 ==..." check at
the top to return immediately if there are no BTL modules. Then I
removed some now-unnecessary conditionals and un-indented as
appropriate.
Fixes#248
These two macros set the prefix for the OPAL and ORTE libraries,
respectively. Specifically, the OPAL library will be named
libPREFIXopen-pal.la and the ORTE library will be named
libPREFIXopen-rte.la.
These macros must be called, even if the prefix argument is empty.
The intent is that Open MPI will call these macros with an empty
prefix, but other projects (such as ORCM) will call these macros with
a non-empty prefix. For example, ORCM libraries can be named
liborcm-open-pal.la and liborcm-open-rte.la.
This scheme is necessary to allow running Open MPI applications under
systems that use their own versions of ORTE and OPAL. For example,
when running MPI applications under ORTE, if the ORTE and OPAL
libraries between OMPI and ORCM are not identical (which, because they
are released at different times, are likely to be different), we need
to ensure that the OMPI applications link against their ORTE and OPAL
libraries, but the ORCM executables link against their ORTE and OPAL
libraries.
This commit makes the folowing changes:
- Add support for the knem single-copy mechanism. Initially vader will only
support the synchronous copy mode. Asynchronous copy support may be added
int the future.
- Improve Linux cross memory attach (CMA) when using restrictive ptrace
settings. This will allow Open MPI to use CMA without modifying the system
settings to support ptrace attach (see /etc/sysctl.d/10-ptrace.conf).
- Allow runtime selection of the single copy mechanism. The default behavior
is to use the best available. The priority list of single-copy mehanisms is
as follows: xpmem, cma, and knem.
- Allow disabling support for kernel-assisted single copy.
- Some tuning and bug fixes.
Restore the functionality to error out (and show a helpful message) if
knem support is requested by is either not compiled in or cannot be
activated.
Thanks to Gus Correa for bringing the matter to our attention.
Somehow, this MCA param was accidentally dropped after v1.6.5. Thanks
to Gus Correa for bringing this matter to our attention.
Also moving some MCA params down from level 9 to levels 4/5.
of preregistered buffers
Before this change eager gets we retried on each progress loop. This commit
modifies the protocol to only retry eager gets when another eager get has
completed. This commit also cleans up some callback code that is no longer
needed.
The GNI_RDMAMODE_FENCE bit was a left over from
async progress work that is not needed at this point
in the gni BTL. Removing the bit also allows
for the removal of the GNI_CDM_MODE_BTE_SINGLE_CHANNEL
bit from the GNI_CdmCreate call.
This commit adds initial ugni thread safety support.
With this commit, sun thread tests (excepting MPI-2 RMA)
pass with various process counts and threads/process.
Also osu_latency_mt passes.
Properly setup the opal_process_info structure early in the initialization procedure. Define the local hostname right at the beginning of opal_init so all parts of opal can use it. Overlay that during orte_init as the user may choose to remove fqdn and strip prefixes during that time. Setup the job_session_dir and other such info immediately when it becomes available during orte_init.
This is a large update that does the following:
- Only allocate fast boxes for a peer if a send count threshold
has been reached (default: 16). This will greatly reduce the memory
usage with large numbers of local peers.
- Improve performance by limiting the number of fast boxes that can
be allocated per peer (default: 32). This will reduce the amount
of time spent polling for fast box messages.
- Provide new MCA variables to configure the size, maximum count,
and send count thresholds for fast boxes allocations.
- Updated buffer design to increase the range of message sizes that
can be sent with a fast box.
- Add thread protection around fast box allocation (locks). When
spin locks are available this should be updated to use spin locks.
- Various fixes and cleanup.
This commit was SVN r32774.
The use of restrict in the return type qualifier for mca_btl_vader_reserve_fbox
is being ignored by gnu compiler. for newer gcc, one sees this warning only
with -Wignored-qualifiers set, but for older variants of gcc it was reported
that numerous warning messages about this ignored qualifier were being
generated as vader is being compiled.
The warning reported by gcc is
btl_vader_fbox.h:53:47: warning: type qualifiers ignored on function return type [-Wignored-qualifiers]
static inline mca_btl_vader_fbox_t * restrict mca_btl_vader_reserve_fbox (struct mca_btl_base_endpoint_t *ep, const size_t size)
This commit was SVN r32714.
The question of whether or not the openib BTL supports loopback is a separate question. It may be more appropriate to make the modex be PMIX_GLOBAL for cases where openib can support loopback so someone can run without a shared memory component. I'll leave that decision to the IB vendors.
This commit was SVN r32702.
- do not have mca_btl_openib_component.default_recv_qps point to the stack
- do not reset mca_btl_openib_component.default_recv_qps in btl_openib_component_open
cmr=v1.8.3:reviewer=miked
This commit was SVN r32642.
Per #4874, code review revealed a possible race condition in the
module struct and the connectivity agent. Move the setup of the
connectivity agent listener until the module struct has been fully
setup.
This commit was SVN r32573.
When no components were able to be found, btl_base_select() was
showing the wrong help message -- one that indicated that a specific
component could not be found. And it left off a string argument, so
the end of the help message was garbage.
This commit creates a new help message for this case and updates the
show_help call to use the new message.
This commit was SVN r32572.
WHAT: Merge the PMIx branch into the devel repo, creating a new
OPAL “lmix” framework to abstract PMI support for all RTEs.
Replace the ORTE daemon-level collectives with a new PMIx
server and update the ORTE grpcomm framework to support
server-to-server collectives
WHY: We’ve had problems dealing with variations in PMI implementations,
and need to extend the existing PMI definitions to meet exascale
requirements.
WHEN: Mon, Aug 25
WHERE: https://github.com/rhc54/ompi-svn-mirror.git
Several community members have been working on a refactoring of the current PMI support within OMPI. Although the APIs are common, Slurm and Cray implement a different range of capabilities, and package them differently. For example, Cray provides an integrated PMI-1/2 library, while Slurm separates the two and requires the user to specify the one to be used at runtime. In addition, several bugs in the Slurm implementations have caused problems requiring extra coding.
All this has led to a slew of #if’s in the PMI code and bugs when the corner-case logic for one implementation accidentally traps the other. Extending this support to other implementations would have increased this complexity to an unacceptable level.
Accordingly, we have:
* created a new OPAL “pmix” framework to abstract the PMI support, with separate components for Cray, Slurm PMI-1, and Slurm PMI-2 implementations.
* Replaced the current ORTE grpcomm daemon-based collective operation with an integrated PMIx server, and updated the grpcomm APIs to provide more flexible, multi-algorithm support for collective operations. At this time, only the xcast and allgather operations are supported.
* Replaced the current global collective id with a signature based on the names of the participating procs. The allows an unlimited number of collectives to be executed by any group of processes, subject to the requirement that only one collective can be active at a time for a unique combination of procs. Note that a proc can be involved in any number of simultaneous collectives - it is the specific combination of procs that is subject to the constraint
* removed the prior OMPI/OPAL modex code
* added new macros for executing modex send/recv to simplify use of the new APIs. The send macros allow the caller to specify whether or not the BTL supports async modex operations - if so, then the non-blocking “fence” operation is used, if the active PMIx component supports it. Otherwise, the default is a full blocking modex exchange as we currently perform.
* retained the current flag that directs us to use a blocking fence operation, but only to retrieve data upon demand
This commit was SVN r32570.
Thanks to Ashley Pittman for pointing the modex should now be zero'ed
cmr=v1.8.2:ticket=trac:4871
This commit was SVN r32568.
The following Trac tickets were found above:
Ticket 4871 --> https://svn.open-mpi.org/trac/ompi/ticket/4871
PGI compilers 2013 and older do not support the following syntax :
mca_btl_scif_modex_t modex = {.port_id = mca_btl_scif_module.port_id};
so split it on two lines
cmr=v1.8.2:reviewer=hjelmn
This commit was SVN r32555.
It turns out that we ''can'' get to the endpoint destructor with the
endpoint still on the "endpoints needing ACKs" list. So if it's on
the list, remove it first, and then DESTRUCT the opal_list_item_t.
This prevents an assert() fail in debug builds. We'd like to let this
soak over the weekend.
cmr=v1.8.2:reviewer=dgoodell
This commit was SVN r32546.
Rarely -- but it happens -- the connectivity client gets ECONNREFUSED
because the connectivity agent listen() backlog is too small. Rather
than put in a loop on the client side, take the simple way out for
now: increase the backlog size to an arbitrarily-large number.
Reviewed by Dave Goodell.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32543.
Instead of waiting to destroy the connectivity agent during component
shutdown, have the module shutdown send an "unlisten" command to the
cagent that will tell it to stop listening on a given interface.
This commit was SVN r32536.
1. After we receive N abnormally-short messages (meaning: corrupted),
print a show_help message about it. N defaults to 25. N can be set
to 0 disable the message via btl_usnic_max_short_packets.
1. If we receive a completion error for something other than a
receive, display a show_help message.
Reviewed by Dave Goodell.
CMR'ing to v1.8.3, but it will require a custom patch because of the
OMPI->OPAL BTL move.
cmr=v1.8.3
This commit was SVN r32522.
There were remaining references to size of MPI_COMM_WORLD,
etc. in ugni btl which prevented building of opal library
following btl move to opal.
Eventually another mechanism will need to be found for
providing hints to BTLs about how to setup internal
resources relevant to max. number possible endpoints, etc.
Conflicts:
opal/mca/btl/ugni/btl_ugni_component.c
This commit was SVN r32507.
Show an example of using the btl_usnic_connectivity_map option. Also,
mention that another reason for the "total connectivity failure" may
be due to asymmetric / unexpected routing.
Reviewed by Dave Goodell.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32465.
The contrib/check-help-strings.pl gets confused if the topic is an
inline logic check, so separate it into two calls to show_help.
This commit was SVN r32455.
These messages were committed in the v1.8 branch in r32341, but were
never committed to the trunk (because we were waiting for the OPAL BTL
move). This commit brings the trunk and v1.8 help messages in line
with each other.
This commit was SVN r32445.
The following SVN revision numbers were found above:
r32341 --> open-mpi/ompi@5e752b4aba
Ensure that the connectivity checker agent only uses pointers from the
client that is the same process as the agent.
Not necessary for the v1.8 branch -- this is a trunk/v1.9-only problem.
This commit was SVN r32438.
Make the del_procs, module finalize, and endpoint destructors be the
same between trunk and v1.8, with one exception: the very beginning of
v1.8 module_finalize calls del_procs for each proc to simulate/pretend
the trunk/v1.9 PML behavior of calling del_procs before module_finalize.
This commit was SVN r32437.
Fixes an assertion failure in --enable-debug builds and SEGVs in normal
builds.
I'm not 100% sure I like this model, but it at least seems to be
consistent. Some variation on this scheme will need to be adapted to
the trunk, where usnic_del_procs() is called by the PML instead of
internally in usnic_finalize().
A related bug (but with different mechanics) is #4832.
This commit was SVN r32424.
Per patch from aryzhikh, initialize a copule of fields in the openib ini struct
Hard to understand why this has sat there for so long...
cmr=v1.8.2:reviewer=rhc:subject=initialize a copule of fields in the openib ini struct
This commit was SVN r32417.
The following Trac tickets were found above:
Ticket 4377 --> https://svn.open-mpi.org/trac/ompi/ticket/4377
Previously, the connectivity agent was pretty dumb: it took whatever
pings it got and ACKed them. Then we added an agent check to ensured
that the ping actually came from the source interface that it said it
came from. Now we add another check such that when a ping is received
on interface X that corresponds to usnic module Y, we ensure that the
source interface of the ping is on the all_endpoints list for module Y
(i.e., module Y expects to be able to talk to that peer interface).
This detects cases where peers have come to different conclusions
about which interfaces should be used to communicate (which is bad!).
This usually reflects a network misconfiguration.
Fixes CSCuq05389.
This commit was SVN r32383.
Ensure that incoming "ping" messages came from the IP address that
they think they came from. If they don't, drop them (because it is
probably routing error), which will likely eventually cause the
connectivity checker to timeout, and therefore cause the job to abort.
This commit was SVN r32368.
Update compat.h to only handle compatability between v1.7/v1.8 and
v1.9/2.0 (i.e., the current trunk). Remove what seems to be the last
vestiages of OMPI/ORTE pollution in the now-OPAL-ized usnic BTL.
Currently use a hard-coded constant for the MCW size (i.e.,
MPI_COMM_WORLD size) for some initialization values in the v1.9/2.0
series; still need to figure out something better there.
This commit was SVN r32365.
Adapt to moving down to OPAL: find a PML-registered error callback,
and use that when we don't have a module context and we need to abort.
Failing that, just call exit().
This commit was SVN r32362.
If --with-usnic is specified and we can't build the usnic BTL, abort.
If --without-usnic is specified, gracefully skip building the usnic
BTL. If neither is specified, do the OMPI-default behavior: try to
configure/build the usnic BTL, and if we can't, skip it.
Fixes CSCuq13889.
This commit was SVN r32349.
WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL
All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic.
This commit was SVN r32317.
