1
1
Граф коммитов

190 Коммитов

Автор SHA1 Сообщение Дата
Nathan Hjelm
5334d22a37 ompi/group: release ompi_proc_t's at group destruction
This commit changes the way ompi_proc_t's are retained/released by
ompi_group_t's. Before this change ompi_proc_t's were retained once
for the group and then once for each retain of a group. This method
adds unnecessary overhead (need to traverse the group list each time
the group is retained) and causes problems when using an async
add_procs.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-11-30 23:03:47 -07:00
Nathan Hjelm
9ef0821856 osc/rdma: fix some threading bugs
There were two bugs in osc/rdma when using threads:

 - Deadlock is ompi_osc_rdma_start_atomic. This occurs because
   ompi_osc_rdma_frag_alloc is called with the module lock. To fix the
   issue the module lock is now recursive. In the future I will add a
   new lock to protect just the current rdma fragment.

 - Do not drop the lock in ompi_osc_rdma_frag_alloc when calling
   ompi_osc_rdma_frag_complete. Not only is it not needed but dropping
   the lock at this point can cause a competing thread to mess up the
   state.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-11-12 20:25:57 -07:00
Rolf vandeVaart
87a4cc6118 Disable the use of osc rdma when we detect a GPU buffer as it is not supported in that component.
This forces a failover to the osc pt2pt component. Fixes #1012
2015-10-28 14:47:45 -04:00
Jeff Squyres
140cf90e3e osc_rdma: minor compiler warning stomp 2015-10-23 06:21:56 -07:00
Nathan Hjelm
63e744ffc6 osc/rdma: use only a single btl registration for local state
This commit fixes a bug that can occur on Cray Gemini networks. If
multiple registrations are used for the local state then we looks the
atomicity guarantees. To avoid issues like this use only a single
registration handle for all local state on a node.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-10-22 15:51:19 -06:00
Nathan Hjelm
97c9732bad osc/rdma: bug fixes
This commit fixes the following:

 - CIDs 1328491, 1328492: Dead code caused by typos in a prior
   commit.

 - Fix the calculation of dynamic memory regions. This was causes
   incorrect RMA range errors when accessing the last partial page of
   an attachment.

 - Fix a SEGV when using dynamic memory windows with local state (all
   processes on the same node).

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-10-22 09:49:38 -06:00
Nathan Hjelm
9476c7bbca osc/rdma: use standard verbosity levels
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-10-21 12:31:41 -06:00
Nathan Hjelm
b8ee05d352 osc/rdma: bug fixes
This commit fixes several bugs in the osc/rdma component:

 - Complete aggregated requests immediately. Completion of RMA
   requests indicates local completion anyway. This fixes a hang in
   the c_reqops test.

 - Correctly mark Rget_accumulate requests.

 - Set the local base flag correctly on the local peer.

 - Clear or set the no locks flag on the window if the value is
   changed by MPI_Win_set_info.

 - Actually update the target when using MPI_OP_REPLACE.

Fixes open-mpi/ompi#1010

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-10-20 15:27:15 -06:00
Nathan Hjelm
e11f014c6e osc/rdma: fix segmentation fault when running 1 ppn
This commit fixes an issue identified by @rolfv. The local peer was
not being correctly initialized when running with a single process on
a node.

This fixes open-mpi/ompi#1010

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-10-14 12:40:52 -06:00
Nathan Hjelm
5fd9c35957 osc/rdma: fix incorrect assert
This commit fixes MTT failures in debug builds.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-09-29 15:37:40 -06:00
Nathan Hjelm
7b8ec48c68 osc/rdma: fix typos inarguments to btl_atomic_[f]op
Signed-off-by: Nathan Hjelm <hjelmn@me.com>
2015-09-29 08:09:00 -06:00
Nathan Hjelm
552e1b59a5 osc/rdma: fix coverity issues
Fixes CID 1324730, 1327429, 1324728, 1196633, 1324731, 1324727, and
1196632: Logically dead code

OMPI_OSC_RDMA_REQUEST_ALLOC can never return a NULL request. Removed
unnecessary NULL checks.

Signed-off-by: Nathan Hjelm <hjelmn@me.com>
2015-09-26 12:45:14 -06:00
Nathan Hjelm
f6920aa916 osc/rdma: check for usable btls during query
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-09-22 17:08:28 -06:00
Nathan Hjelm
974061c38f osc: fixed issues identified by coverity
Fix CID 1324733: Null pointer dereferences  (FORWARD_NULL)
Fix CID 1324734: Null pointer dereferences  (FORWARD_NULL)
Fix CID 1324735: Null pointer dereferences  (FORWARD_NULL)
Fix CID 1324736: Null pointer dereferences  (FORWARD_NULL)
Fix CID 1324737: Null pointer dereferences  (FORWARD_NULL)
Fix CID 1324751: Memory - illegal accesses  (USE_AFTER_FREE)
Fix CID 1324750: (USE_AFTER_FREE)
Fix CID 1324749: Memory - corruptions  (USE_AFTER_FREE)
Fix CID 1324748: Memory - illegal accesses  (USE_AFTER_FREE)
Fix CID 1324747: (USE_AFTER_FREE)
Fix CID 1324746: Memory - corruptions  (USE_AFTER_FREE)

Add missing return on an error path.

Fix CID 1324745: Code maintainability issues  (UNUSED_VALUE)

Ignore return code from barrier. It was not being used anyway.

Fix CID 1324738: Null pointer dereferences  (FORWARD_NULL)
Fix CID 1324741: Null pointer dereferences  (REVERSE_INULL)

module->selected_btl can not be NULL in osc/rdma during normal
operation. Removed the unnecessary NULL check.

Fix CID 1324752: Memory - illegal accesses  (USE_AFTER_FREE)

Move ompi_osc_pt2pt_module_lock_remove to before the lock is freed.

Fix CID 1324744: Uninitialized variables  (UNINIT)
Fix CID 1324743: Uninitialized variables  (UNINIT)

This array is not used unitialized but there is no reason not to use
calloc here to silence the warning.

The following CID is a false positive: 1324742. I will mark it such in
coverity.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-09-22 09:23:39 -06:00
Nathan Hjelm
d8df9d414d osc/rdma: add true RDMA one-sided component
This commit adds support for performing one-sided operations over
supported hardware (currently Infiniband and Cray Gemini/Aries). This
component is still undergoing active development.

Current features:

 - Use network atomic operations (fadd, cswap) for implementing
   locking and PSCW synchronization.

 - Aggregate small contiguous puts.

 - Reduced memory footprint by storing window data (pointer, keys,
   etc) at the lowest rank on each node. The data is fetched as each
   process needs to communicate with a new peer. This is a trade-off
   between the performance of the first operation on a peer and the
   memory utilization of a window.

TODO:

 - Add support for the accumulate_ops info key. If it is known that
   the same op or same op/no op is used it may be possible to use
   hardware atomics for fetch-and-op and compare-and-swap.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2015-09-16 15:01:33 -06:00
Howard Pritchard
e49a37c034 ownership: update ownership files
per discussions at OMPI devel workshop

Signed-off-by: Howard Pritchard <howardp@lanl.gov>
2015-06-25 10:04:42 -06:00
Howard Pritchard
bf89131f9e add owner files to opa/ompi/orte mca directories
This commit adds an owner file in each of the component directories
for each framework.  This allows for a simple script to parse
the contents of the files and generate, among other things, tables
to be used on the project's wiki page.  Currently there are two
"fields" in the file, an owner and a status.  A tool to parse
the files and generate tables for the wiki page will be added
in a subsequent commit.
2015-02-22 15:10:23 -07:00
Nathan Hjelm
9eba7b9d35 Rename the OSC "rdma" component to pt2p to better reflect that it does not actually use btl rdma 2015-01-06 13:38:55 -07:00
Nathan Hjelm
1b564f62bd Revert "Merge pull request #275 from hjelmn/btlmod"
This reverts commit ccaecf0fd6, reversing
changes made to 6a19bf85dd.
2014-11-19 23:22:43 -07:00
Nathan Hjelm
5a0a48c3c4 osc: remove lingering rdma component files 2014-11-19 12:11:54 -07:00
Nathan Hjelm
45d1fac8af ugni thread safety fixes 2014-11-19 11:33:03 -07:00
Nathan Hjelm
29e4e1c90a Rename the OSC "rdma" component to pt2p to better reflect that it does not actually use btl rdma 2014-11-19 11:33:03 -07:00
Ralph Castain
616f0894ce Add missing parens on values being passed to OPAL_THREAD_ADD32 2014-10-31 19:11:48 -07:00
Nathan Hjelm
672d96704c osc/rdma: fix regression introduced by eed7b45db5
The ompi_osc_signal_outgoing was moved from ompi_osc_rdma_frag_start to frag_send
which gave correct results for the bug reproducer but hangs with simple OSC
tests. Moved the ompi_osc_signal_outgoing back and it now passes all tests.

Closes #256
2014-10-30 23:16:11 -06:00
Nathan Hjelm
23dd3af946 osc/rdma: use unsigned types for all counters
Some of the counters used by the "rdma" one-sided component are intended
to overflow. Since overflow behavior is undefined for signed integers in
C it is safer to use unsigned integers here.
2014-10-22 15:36:15 -06:00
Nathan Hjelm
eed7b45db5 osc/rdma: fix issue identified by Berk Hess
osc/rdma uses counters to determine if all messages have been received
before exiting synchronization calls. The problem is that the active
target counter is always increasing (never zeroed). If over 2^31-1
messages are sent this causes the counter to overflow (in itself this
isn't an error). This causes test/wait to return before the communication
is complete. There is an additional error in the use of the fragment
flush function. If PSCW synchronization is in use this function CAN NOT
be called unless a post message has arrived.

Relevant mailing list thread: http://www.open-mpi.org/community/lists/devel/2014/10/16016.php

This commit fixes both issues. Tested against MTT and issue reproducer.

Closes #224.
2014-10-07 11:45:22 -06:00
Gilles Gouaillardet
5f1e0f284a Fix compilation when --enable-hetorogeneous
This commit was SVN r32410.
2014-08-04 10:35:08 +00:00
Ralph Castain
552c9ca5a0 George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-)
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.
2014-07-26 00:47:28 +00:00
Ralph Castain
3d1b32a2c6 Silence warning
cmr=v1.8.2:reviewer=hjelmn

This commit was SVN r32231.
2014-07-14 19:27:30 +00:00
Nathan Hjelm
32ab6f850e osc/rdma: fix warning
cmr=v1.8.2:reviewer=rhc

This commit was SVN r32201.
2014-07-10 18:42:55 +00:00
Nathan Hjelm
b6abe68972 osc/rdma: check for more types of window access violations
This commit adds a check to see if the target is in an access epoch. If
not we return OMPI_ERR_RMA_SYNC. This fixes test_start3 in the onesided
test suite. The cost of this extra check is 1 byte/peer for the boolean
flag indicating that the peer is in an access epoch.

I also fixed a problem where mupliple unexpected post messages are not
correctly handled.

cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r32160.
2014-07-08 21:11:12 +00:00
Ralph Castain
f3cb124e50 Revert r32082 and r32070 - the developer's conference has decided to go a different direction on the threaded progress effort. This will involve some degree of prototyping to understand the tradeoffs prior to making a final design decision, and so we'll hold off on the final change until that is completed.
This commit was SVN r32089.

The following SVN revision numbers were found above:
  r32070 --> open-mpi/ompi@12d92d0c22
  r32082 --> open-mpi/ompi@aa6438ef7a
2014-06-25 20:43:28 +00:00
Ralph Castain
12d92d0c22 Per the OMPI developer conference, remove the last vestiges of OMPI_USE_PROGRESS_THREADS
This commit was SVN r32070.
2014-06-24 17:05:11 +00:00
Nathan Hjelm
fd21b244ce osc/rdma: better name for lookup function
cmr=v1.8.2:ticket=trac:4732:reviewer=ompi-rm1.8

This commit was SVN r32021.

The following Trac tickets were found above:
  Ticket 4732 --> https://svn.open-mpi.org/trac/ompi/ticket/4732
2014-06-17 19:49:17 +00:00
Nathan Hjelm
390f8f52b4 osc/rdma: clean up group process matching a bit
cmr=v1.8.2:ticket=trac:4732:reviewer=dgoodell

This commit was SVN r32018.

The following Trac tickets were found above:
  Ticket 4732 --> https://svn.open-mpi.org/trac/ompi/ticket/4732
2014-06-17 17:48:30 +00:00
Nathan Hjelm
7f20868179 osc/rdma: ensure matching of post/start calls
The post and start window calls are supposed to be matching. The code
did not check to see that an incoming post matched with the start call.
This commit fixes the bug by placing the post on a pending list that
will be checked by the next call to start.

cmr=v1.8.2:reviewer=dgoodell

This commit was SVN r32017.
2014-06-17 15:23:06 +00:00
Nathan Hjelm
927098d567 osc/rdma: fix hang when accumulating with MPI_REPLACE
The replace callback did not increment the incoming frag counter. This
leads to a hang during synchronization. This commit adds the increment
and also puts the request on the garbage collection list to fix a leak.

This fixes a hang found when running the mpich test suite.

cmr=v1.8.2:reviewer=bbenton

This commit was SVN r32016.
2014-06-17 14:53:29 +00:00
Nathan Hjelm
7f6de57653 osc/rdma: fix accumulate fragment size calculation
The wrong type was used when calculating the amount of space needed
for an accumulate fragment. Fixed the calculation and took the
opportunity to eliminate the get_acc header as it is identical to the
acc header.

This fixes trac:4719 and #4718

Tracking these fixes for 1.8.2 in this CMR.

Throwing this to Brad for review as he is the one who ran into the issue.

cmr=v1.8.2:reviewer=bbenton

This commit was SVN r32015.

The following Trac tickets were found above:
  Ticket 4719 --> https://svn.open-mpi.org/trac/ompi/ticket/4719
2014-06-17 14:53:24 +00:00
Nathan Hjelm
2f96f16416 osc/rdma: ensure eager sends are active before checking for sync errors
in self optimization

This addresses an issue found with the MPICH pscw_ordering test. Eager sends
were not yet active (which is ok for the standard path) but not ok for the
self optimization. Fixed by waiting for all post messages before checking
the sync state.

Fixes trac:4724

Tracking the 1.8.2 issue in this CMR.

cmr=v1.8.2:reviewer=bbenton

This commit was SVN r32012.

The following Trac tickets were found above:
  Ticket 4724 --> https://svn.open-mpi.org/trac/ompi/ticket/4724
2014-06-17 04:53:47 +00:00
Nathan Hjelm
2f5b1ca4cf osc/rdma: do not leak the receive request
This commit fixes a bug that can cause request and communicator leaks
when cleaning up an OSC window. The should prevent a hang seen with
IMB-EXT.

cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r31539.
2014-04-28 19:55:18 +00:00
Yossi Etigin
7efb724d7b osc/rdma: fix deadlock with put_long protocol.
When sending PUT_LONG, the data is sent before headers, and sometimes 
the header is not flushed immediately. This creates a lot of unexpected 
receives in the peer, since it would posts a receive only when gets the 
header, which makes it run out of receive buffers. When the sender 
eventually flushes the window, the receiver already has no buffers to 
receive the header, which causes a deadlock.

The fix is to always flush the headers when doing put_long.

cmr=v1.8.1:reviewer=hjelmn

This commit was SVN r31378.
2014-04-13 16:24:56 +00:00
Nathan Hjelm
a31bfbeb2c osc/rdma: fix typo in get accumulate path
There was a typo in the ompi_osc_gacc_long_start that was causing a
segmentation fault when executing long get accumulate operations.

cmr=v1.8.1:reviewer=jsquyres

This commit was SVN r31353.
2014-04-08 21:54:52 +00:00
Nathan Hjelm
fdf4c3b900 osc/rdma: really fix active message support
The last fix prevented a hang but had some cases where the results were
wrong. Fixed. Tested with armci, openmpi/ibm, openmpi/onesided.

cmr=v1.8:reviewer=jsquyres

This commit was SVN r31284.
2014-03-28 22:06:16 +00:00
Nathan Hjelm
ee7a1478ee osc/rdma: fix test/wait hang
There are differences between how active and passive messages are
accounted for in this component. Active message counts on the sender
side are set to zero before the control message is sent so we do not
have to add one to the expected number of messages or we end up
double counting the control message. This commit should fix that error.

Fixes regression in one-sided/test_rma1

cmr=v1.8:reviewer=jsquyres

This commit was SVN r31281.
2014-03-28 20:49:20 +00:00
Nathan Hjelm
020f011552 osc/rdma: fix bugs in lock_all and flush_all
This commit fixes two bugs:

 - We were not correctly setting the lock type in the outstanding lock
   for lock_all. This caused undefined behavior.

 - flush_all was incorrectly checking for comm size - 1 lock acks but
   comm size flush acks. This is the reverse of what was intended.

cmr=v1.8:reviewer=jsquyres

This commit was SVN r31226.
2014-03-25 23:39:43 +00:00
Nathan Hjelm
0d703759f6 osc/rdma: fix possible error when encountering accumulate lock contention
It is possible to get into a situation where a small accumulate operation
can not be completed because a large accumulate operation holds the lock.
In this case we may return from wait/flush/etc before the operation is
complete. To handle this case increment the expected incoming fragment
count when queuing an accumulate operation and increment the incoming
fragment count after processing the accumulate operation.

cmr=v1.8:reviewer=jsquyres

This commit was SVN r31224.
2014-03-25 21:00:43 +00:00
Nathan Hjelm
3df85b47e9 osc/rdma: quiet warning in r31197
cmr=v1.8:ticket=trac:4441

This commit was SVN r31223.

The following SVN revision numbers were found above:
  r31197 --> open-mpi/ompi@0ed44f2fdb

The following Trac tickets were found above:
  Ticket 4441 --> https://svn.open-mpi.org/trac/ompi/ticket/4441
2014-03-25 21:00:36 +00:00
Nathan Hjelm
d681eb4655 osc/rdma: fix warnings introduced by r31204
cmr=v1.8:ticket=trac:4449

This commit was SVN r31221.

The following SVN revision numbers were found above:
  r31204 --> open-mpi/ompi@949abe45cd

The following Trac tickets were found above:
  Ticket 4449 --> https://svn.open-mpi.org/trac/ompi/ticket/4449
2014-03-25 21:00:19 +00:00
Nathan Hjelm
949abe45cd osc: fix datatype related issues in the one-sided code
This commit fixes two issues:

 - osc/rdma: The target side of an accumulate was using the target datatype
   in the receive to the packed buffer. This was conflicting with the way
   the reduction is done into the target buffer. Changed the receive to use
   the primitive datatype.

 - osc/base: The copy table was completely wrong. Fixed the table to match
   the underlying datatypes (which are opal not ompi datatypes).

 - osc/base: There is a problem using the optimized description. Fall back
   on using the non-optimized description until we can understand what is
   going wrong.

cmr=v1.8:reviewer=jsquyres

This commit was SVN r31204.
2014-03-25 15:28:48 +00:00
Nathan Hjelm
bc55276844 osc/rdma: fix bug in the active message code that could cause erroneous
results

The code to handle completion messages did not correctly increment the
number of expected messages. This could cause wait to return before all
incoming messages are complete.

I also added a check to ensure that start returns an error if we are in
a passive access epoch.

cmr=v1.8:reviewer=jsquyres

This commit was SVN r31203.
2014-03-25 15:28:36 +00:00