ibv_create_ah() can also return EHOSTUNREACH, which means that there
is no route to the peer. Treat that as a non-fatal warning.
Reviewed by Reese Faucette.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32135.
There's no need for the port number (since usNIC has no port numbers),
and make the wording the same as other help messages.
Reviewed by Reese Faucette.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32134.
no need to #include <math.h> ...
cmr=v1.8.2:reviewer=miked:ticket=4759
This commit was SVN r32121.
The following Trac tickets were found above:
Ticket 4759 --> https://svn.open-mpi.org/trac/ompi/ticket/4759
The distances as returned by hwloc_get_whole_distance_matrix_by_type are typ float.
This patch handle all distances as float.
cmr=v1.8.2:reviewer=miked
This commit was SVN r32120.
cmr=v1.8.2:reviewer=tkordenbrock:subject=Portals4/MTL hanging fix
This commit was SVN r32113.
The following Trac tickets were found above:
Ticket 4681 --> https://svn.open-mpi.org/trac/ompi/ticket/4681
cmr=v1.8.2:reviewer=tkordenbrock:subject=Move r32112 to v1.8.2 branch
This commit was SVN r32112.
The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
r32112
The following Trac tickets were found above:
Ticket 4682 --> https://svn.open-mpi.org/trac/ompi/ticket/4682
RHEL 7 has shipped with kernel support for the RDMA_TRANSPORT_USNIC
enum, but ''not'' the RDMA_TRANSPORT_USNIC_UDP enum. This means that
when you install usNIC drivers from cisco.com, the kernel will report
IBV_TRANSPORT_USNIC, even though the transport is actually using UDP.
Therefore, we have to modify the logic in common/verbs to do the
additional magic probe if the device reports either an
IBV_TRANSPORT_IWARP or IBV_TRANSPORT_USNIC (because both of those might
be lies -- do the probe to figure out the real transport).
The code changed by this patch is fairly trivial; it simply moves the
logic of the magic probe to its own short function, and then calls that
short function in both the IBV_TRANSPORT_(IWARP|USNIC) cases. It looks
longer because several lengthy comments were also updated.
Authored-by: Jeff Squyres <jsquyres@cisco.com>
Reviewed-by: Dave Goodell <dgoodell@cisco.com>
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32098.
the other collective modules. If we endup without some of the
collective the code will raise an error anyway.
cmr=v1.8.2:reviewer=hjelmn
This commit was SVN r32096.
Based on extensive discussions before/at the June 2014 developer's
meeting, put a lengthy comment explaining a second reason why we
''must'' use an RTE barrier during MPI_FINALIZE and
MPI_COMM_DISCONNECT (i.e., unreliable transports). Slightly explain
more the original reason why we do this, too (BTLs can lie/buffer a
message without actually injecting it on the network).
This commit was SVN r32095.
rtnetlink doesn't check the source address when determining whether to
return route info for a query. So we need to check that the OIF matches
the OIF of the source interface name. Without this check, OMPI might
pair a local interface which does not have a route to a particular
remote interface.
Fixes Cisco bug CSCup55797.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r32090.
At the developer meeting today, the question was raised as to whether
the SCTP BTL was maintained any more. I emailed Alan Wagner to see if
he had any interest/resources to continue to maintain the SCTP BTL.
He indicated that he unfortunately had any resources to maintain it;
it would be fine to remove the SCTP BTL from the trunk.
So long, SCTP BTL... fare thee well...
This commit was SVN r32075.
1: find/create procs, and create associated endpoint for each
2: resolve peer addresses
The 2nd part is done as a separate loop so that the address lookups
can be parallelized.
The overall result is to split usnic_add_procs() into two smaller,
simpler parts.
cmr=v1.8.2:ticket=trac:4734
This commit was SVN r32062.
The following Trac tickets were found above:
Ticket 4734 --> https://svn.open-mpi.org/trac/ompi/ticket/4734
When ibv_create_ah() fails due to an address resolution failure, it
really only means that we can't reach that one peer -- so we should
just ignore that one peer. If ibv_create_ah() fails for some other
reason, then give up on the entire usnic_X device.
Change the show_help() message that is displayed when ibv_create_ah()
fails due to address resolution failure; indicate that it's likely a
routing problem. Also opal_output_verbose() the same info, since
show_help() is de-duplicated (and this particular show_help() message
can be squelched).
Fixes Cisco bugs CSCup35851 and CSCup35872.
cmr=v1.8.2:ticket=trac:4734
This commit was SVN r32061.
The following Trac tickets were found above:
Ticket 4734 --> https://svn.open-mpi.org/trac/ompi/ticket/4734
Move away from verbs-specific terms "device" and "port" in the usnic
BTL help messages. Replace them with "usNIC interface" (since usNIC
has no concept of a port).
cmr=v1.8.2:ticket=trac:4734
This commit was SVN r32029.
The following Trac tickets were found above:
Ticket 4734 --> https://svn.open-mpi.org/trac/ompi/ticket/4734
Move MACLEN and IPV4LEN into _util.h and rename them to be MACSTRLEN
and IPV4STRLEN, respectively.
cmr=v1.8.2:ticket=trac:4734
This commit was SVN r32028.
The following Trac tickets were found above:
Ticket 4734 --> https://svn.open-mpi.org/trac/ompi/ticket/4734
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.
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.
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
This changeset :
- always call the low/level implementation for :
* MPI_Alltoallv
* MPI_Neighbor_alltoallv
* MPI_Alltoallw
* MPI_Neighbor_alltoallv
- fix mca_coll_tuned_alltoallv_intra_basic_inplace
so zero size types are correctly handled
cmr=v1.8.2:reviewer=bosilca:ticket=4715
This commit was SVN r32013.
The following Trac tickets were found above:
Ticket 4715 --> https://svn.open-mpi.org/trac/ompi/ticket/4715
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
Brad correctly pointed out that the total window size should not be an
int. Changed it to an unsigned long.
cmr=v1.8.2:reviewer=bbenton
This commit was SVN r32010.
Only one field is valid for RMA requests: MPI_ERROR. This field is set
to the correct value in ompi_request_empty so there is no reason to
allocate and keep track of osc/sm requests because they are always
complete on return. Since we are no longer using the osc/sm request
structure or free list they are now removed.
Closes trac:4723
Tracking this issue with the CMR. Brad, can you verify the issue is indeed fixed.
cmr=v1.8.2:reviewer=bbenton
This commit was SVN r32009.
The following Trac tickets were found above:
Ticket 4723 --> https://svn.open-mpi.org/trac/ompi/ticket/4723
Correctly handle the corner case in MPI_Alltoallv when
some tasks have no data to transfer and some other tasks
do have data to transfer.
This test case is covered in ibm/collective/alltoallv_somezeros
from the ompi-tests repo.
cmr=v1.8.2:reviewer=bosilca
This commit was SVN r31985.
This would be a really, really weird case if it ever happens (i.e.,
you have usnics but the agent process failed somewhere in MPI_INIT
such that the agent never appears), but having an infinite loop
doesn't seem like a good idea.
(does not need to go to v1.8 because v1.8 still uses RML for
communication for the connectivity checker)
This commit was SVN r31932.
if eager rdma is used, endpoint reference_count is greater than one.
this commit is a temporary fix that OBJ_RELEASE the endpoint as much as needed.
thought this is likely correct, it can be suboptimal and hence needs to be reviewed
cmr=v1.8.2:reviewer=hjelmn
This commit was SVN r31922.
http://www.open-mpi.org/community/lists/devel/2014/05/14822.php
Revamp the ORTE global data structures to reduce memory footprint and add new features. Add ability to control/set cpu frequency, though this can only be done if the sys admin has setup the system to support it (or you run as root).
This commit was SVN r31916.
We were still leaking 1) file descriptors for data files, and 2) some
control files. I fixed both of these leaks and everything is looking
good. This should fix the bug where we are running out of file
descriptors when running the loop_spawn test. I also too the
opportunity to refactor the code a bit to make the mapping/unmapping
simpler. This should help avoid these sorts of issues in the future.
Depends on #4678
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31893.
if in_ptr is NULL, the MAP_FIXED flag cannot be passed to mmap
this caused a hang in topology/cart and topology/sub from ibm
test suite on trunk.
cmr=v1.8.2:reviewer=hjelmn
This commit was SVN r31890.
Per Ralph :
"I noticed that we are incrementing and decrementing the opal_progress_event state.
However, this no longer has any impact whatsoever on the RML as that is running in
the independent ORTE event thread. So all this actually does is impact the MPI layer
by adding an unnecessary overhead."
Thanks Ralph for pointing this :-)
cmr=v1.8.2:reviewer=rhc:ticket=4671
This commit was SVN r31887.
The following Trac tickets were found above:
Ticket 4671 --> https://svn.open-mpi.org/trac/ompi/ticket/4671
since r31716 mca_topo_base_comm_cart_2_2_0_t is an object
and must be allocated/freed with OBJ_NEW/OBJ_RELEASE.
this fixes topology/cart_sub_zero from the ibm test suite.
v1.8 does not use objects, so no cmr for this branch
This commit was SVN r31883.
The following SVN revision numbers were found above:
r31716 --> open-mpi/ompi@e3df77548d
operation. Ralph will fix shortly.
For the time being, put back the original code...
Refs trac:4669
This commit was SVN r31872.
The following Trac tickets were found above:
Ticket 4669 --> https://svn.open-mpi.org/trac/ompi/ticket/4669
Now that the infrastructure is calling BTL del_procs() before the BTL
finalize(), the usnic BTL had to re-order some of its teardown
sequence to avoid assert() failing.
This is part of a larger conversation involving #4669. Since
MPI_FINALIZE and MPI_COMM_DISCONNECT currently use an
oob/grpcomm-based barrier, the usnic BTL can ''absolutely know'' that
these endpoints and procs will no longer be used. If the ORTE DPM
goes back to a PML-based barrier, the usnic BTL will need to grow more
complex teardown semantics (a la TCP socket FIN/ACK/FIN_WAIT states).
Refs trac:4669
This commit was SVN r31871.
The following Trac tickets were found above:
Ticket 4669 --> https://svn.open-mpi.org/trac/ompi/ticket/4669
to be based on grpcomm (i.e., an out-of-band based barrier) rather
than the simplistic PML-based barrier that it currently uses.
This is pending a larger discussion with Nathan and George, but it
will allow the usnic BTL to stop assert()-failing in light of the
recent del_procs() change.
This commit was SVN r31870.
Based on a suggestion from George on #31806, use the pre-computed
sizes rather than duplicating the computation math (which may change
someday in the future).
cmr=v1.8.2:ticket=trac:4647
This commit was SVN r31841.
The following Trac tickets were found above:
Ticket 4647 --> https://svn.open-mpi.org/trac/ompi/ticket/4647
a memory leak. Similar changes shuld be applied to all the
other PML that are copies of OB1. This patch is related to
#4653.
This commit was SVN r31838.
Avoid sending/receiving zero size messages in order to be compliant
with the top-level modification
cmr=v1.8.2:ticket=4651:reviewer=bosilca
This commit was SVN r31836.
The following Trac tickets were found above:
Ticket 4651 --> https://svn.open-mpi.org/trac/ompi/ticket/4651
of existing BTL fo an endpoint, all the others are just partial list.
Thus, all the cleaning should first be done in the btl_send array,
and them in the other arrays (btl_eager and btl_rdma).
This commit was SVN r31834.
This commit fixes two issues:
- This intent of the code @ bml_r2.c:486 is to prevent calling the
btl_del_procs more than once for a given proc. Gilles correctly
identified there was a problem in this code but r31786 we not the
correct fix.
- Fix a segmentation fault in r2 finalize revealed by the fact we
actually call del_procs now.
cmr=v1.8.2:reviewer=ggouaillardet:ticket=trac:4645
This commit was SVN r31829.
The following SVN revision numbers were found above:
r2 --> open-mpi/ompi@58fdc18855
r31786 --> open-mpi/ompi@fc96b0a7b8
The following Trac tickets were found above:
Ticket 4645 --> https://svn.open-mpi.org/trac/ompi/ticket/4645
MPI_Isend.
I filed an RFC for this optimization some time back. It is a
relatively simple optimization. If the data associated with an
MPI_Isend can be put on the wire without allocating an MPI_Request
then do so. In this case we can legally return omp_request_empty
which will correctly indicate that the request is complete and that is
was not cancelled (these are the only requirements on send requests).
cmr=v1.8.3:reviewer=bosilca
This commit was SVN r31828.
Basesmuma was vallocing space for control data then mmapping over that
data. Nothing in the code suggests any need for mmapping a specific
address so I did the following to remove the leak:
- Removed the valloc of the buffer space
- ftruncate the mmaped file to ensure there is sufficient memory to
allocate space for the control data.
Ideally this code should be using opal/shmem but that is a larger
change. Keeping it simple for now.
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31822.
Make sure to cleanup the registation cache when removing an
endpoint. This leak only affects systems with XPMEM installed.
Since this is in code specific to XPMEM not sure who could review so
sending it directly to the RM.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31821.
This commit :
- Correctly retrieve the communicator size when
checking memory and parameters
- Ensure (sendtype,sendcount) and (recvtype,recvcount)
matches and return with MPI_ERR_TRUNCATE otherwise
- Return with MPI_SUCCESS without invoking the low level
if no data is going to be transferred
- Fixes trac:4506
cmr=v1.8.2:reviewer=bosilca
This commit was SVN r31815.
The following Trac tickets were found above:
Ticket 4506 --> https://svn.open-mpi.org/trac/ompi/ticket/4506
r30343 introduced the optimization of putting the OB1 sendreq and
recvreq on the stack for blocking sends and receives. However, the
requests did not contain enough storage for the data that is normally
immediately ''after'' the request (e.g., BTL data).
This commit changes these requests to be pointers and to use alloca()
to get enough total space for the OB1 request and all the associated
data.
The change is smaller than it looks; most of it is just changing from
"foo.bar" to "foo->bar" notation (etc.).
Submitted by Jeff, reviewed by Nathan. But we want George to look at
this (and get a little soak time on the trunk) before moving to v1.8.
cmr=v1.8.2:reviewer=bosilca
This commit was SVN r31806.
The following SVN revision numbers were found above:
r30343 --> open-mpi/ompi@2b57f4227e
This commit fixes leaks of bml endpoints in finalize. A summary of the
bugs/fixes is below.
1) ompi_mpi_finalize used ompi_proc_all to get the list of procs but
never released the reference to them (ompi_proc_all called
OBJ_RETAIN on all the procs returned). When calling del_procs at
finalize it should suffice to call ompi_proc_world which does not
increment the reference count.
2) del_procs is called BEFORE ompi_comm_finalize. This leaves the
references to the procs from calling the pml_add_comm
function. The fix is to reorder the calls to do omp_comm_finalize,
del_procs, pml_finalize instead of del_procs, pml_finalize,
ompi_comm_finalize.
3) The check in del_procs in r2 checked for a reference count of
1. This is incorrect. At this point there should be 2 references:
1 from ompi_proc, and another from the add_procs. The fix is to
change this check to look for a reference count of 22. This check
makes me extremely uncomforable as nothing will call del_procs if
the reference count of a procs is not 2 when del_procs is
called. Maybe there should be an assert since this is a developer
error IMHO.
cmr=v1.8.2:reviewer=bosilca
This commit was SVN r31782.
The following SVN revision numbers were found above:
r2 --> open-mpi/ompi@58fdc18855
netpatterns_setup_narray_knomial_tree.
Fix a bug in ptpcoll that caused memory allocated by
netpatterns_setup_narray_knomial_tree to leak.
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31781.
vagrind correctly indicated that the mpool is needed when tearing down
the free lists. Reordered the teardown code to correct this.
My code so no review required.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31780.
Two leaks are fixed in this commit:
- Do not leak btl component list items.
- Do not leak the nodename when decoding the pidmap.
cmr=v1.8.2:reviewer=rhc
This commit was SVN r31779.
This commit fixes three leaks:
- bml/r2: fix leak of del_procs in mca_bml_r2_del_procs
- Release the modex data in btl/scif, btl/ugni, and btl/vader
- ompi_mpi_finalize: close the allocator framework
cmr=v1.8.2:reviewer=jsquyres
This commit was SVN r31778.
The following SVN revision numbers were found above:
r2 --> open-mpi/ompi@58fdc18855
a brand new dummy connection has to be used to properly
trigger the main thread termination and avoid timeout in
the main thread
cmr=v1.8.2:reviewer=hjelmn
This commit was SVN r31772.
The local buffer object is only used by the root of the group,
so move down the buffer declaration and initialization.
cmr=v1.8.2:reviewer=rhc
This commit was SVN r31771.
An OBJ_DESTRUCT was missing for mca_pml_ob1.send_ranges causing a
memory leak. Identified by valgrind.
cmr=v1.8.2:reviewer=jsquyres
This commit was SVN r31768.
scif_close is not causing scif_poll in the listening thread to return
as expected. To ensure the thread exits attempt to make a local
connection to wake up the thread before calling pthread_join.
cmr=v1.8.2:reviewer=ggouaillardet
This commit was SVN r31756.
closed
We were leaving the selected component open. This commit should
eliminate a leak detected by valgrind.
cmr=v1.8.2:reviewer=jsquyres
This commit was SVN r31749.
The smsg_mboxes free list was not getting destructed. The construct
has been moved to module initialization and a matching destruct is now
in the module destruct.
This commit was SVN r31746.
This commit fixes memory leaks discovered in the sbgp setup code. We
were leaking an opal_argv as well as some list items. I took the
opportunity to clean up the code a little which includes making use of
the opal_argv_free function.
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31745.
The items in the available bcol list were getting leaked. This commit
fixes this leak. I also cleaned up the code a bit. This includes
making use of the opal_argv_free function.
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31744.
It is essential to call mca_base_framework_close for every framework
that is opened. coll/ml was not doing this so neither bcol nor sbgp
were getting cleaned up. This commit fixes this omission.
Also fixed a leak caused by calling OBJ_DESTRUCT for something created
with OBJ_NEW. With these changes coll/ml appears to be valgrind clean.
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31743.
Simple bug. The dist_graph pointer must be a constructed object. The
change from malloc to OBJ_NEW was missing from r31716. Tested with MTT
and everything looks ok now.
This commit was SVN r31739.
The following SVN revision numbers were found above:
r31716 --> open-mpi/ompi@e3df77548d
There is no reason to cancel the listening thread. It should die
automatically when the file descriptor is closed. It is sufficient
to just wait for the thread to exit with pthread join.
cmr=v1.8.2:ticket=trac:4616:reviewer=jsquyres
This commit was SVN r31738.
The following Trac tickets were found above:
Ticket 4616 --> https://svn.open-mpi.org/trac/ompi/ticket/4616
We were leaking file descriptors when coll/ml was in use. It turn out
this was because basesmuma was failing to unmap files it had previously
mapped. This commit cleans up the setup code to ensure that we only
attempt to map the control files once per module and then ensures the
files are unmapped when the module is released.
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31737.
algorithm
Per suggestion from Manju make sure there isn't a gap in the size ranges
for the available algorithms.
cmr=v1.8.2:ticket=trac:4437:reviewer=ompi-rm1.8
This commit was SVN r31728.
The following Trac tickets were found above:
Ticket 4437 --> https://svn.open-mpi.org/trac/ompi/ticket/4437
MPI_Cart_Create/MPI_Graph_create/MPI_Dist_Graph
Fixes trac:4581
This commit was SVN r31716.
The following Trac tickets were found above:
Ticket 4581 --> https://svn.open-mpi.org/trac/ompi/ticket/4581
In preparation for moving the BTLs down to OPAL, discontinue the use
of the RML for connectivity client/agent communication. Instead, use
local unix domain sockets in the job session directory (all
communication is between processes on the same server, so unix domain
sockets are fine).
This commit was SVN r31710.
Add the component use_udp value into the modex. If my component's
use_udp value doesn't agree with the use_udp value from a peer's modex
data, print a helpful message and disqualify the usnic BTL (the usnic
BTL will not be used). This prevents accidental customer
misconfigurations.
Reviewed by Dave Goodell
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31689.
Trivial struct re-ordering to eliminate holes in the middle of the
struct (although there's still a hole at the end) and reduce the
overall size of the struct from 64 to 56 bytes. Also change mtu from
int to uint16_t; there was no need for it to be that large.
Reviewed by Dave Goodell
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31688.
Fix mismatch between the MCA param (which expresses the timeout in
*mili*seconds) and the struct timeval timeout (which expresses the
timeout in *micro*seconds).
Reviewed by Dave Goodell
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31687.
top_ompi_srcdir -> OMPI_TOP_SRCDIR
top_ompi_builddir -> OMPI_TOP_BUILDDIR
We also split the srcdir/builddir flags according to their local tree (e.g., OPAL_TOP_SRCDIR), and tied them all together in configure.ac. Renamed ompi_ignore and ompi_unignore to be opal_<foo> as these are agnostic markers.
Only thing left is ompilibdir being treated similar to what we dif for srcdir/builddir. Coming soon.
This commit was SVN r31678.
In abusive MPI communication patterns, sending a UDP ping only once a
second may not be sufficient -- all the UDP pings may be dropped. So
increase the frequency of the pings to every quarter second, and allow
more total pings to be sent.
Total timeout time is still the same (10 seconds) -- we'll just now
try 40 times (i.e., once every quarter second) as opposed to 10 times
(i.e., once a second). Testing has shown that this frequency allows
the connectivity checker to always succeed even in the many-to-one
abusive communication patterns.
cmr=v1.8.2:reviewer=dgoodell
This commit was SVN r31602.
We're passing a char foo[x] into PACK_BYTES, so we don't need to take
its address in the macro. This is parallel to the UNPACK_BYTES macro
(where we pass a char bar[x] into it, and don't take its address in
the macro).
The value we're packing is only used to output in a show_help message,
which is why this wasn't noticed before (i.e., it's not used in
network or addressing that would have caused a failure).
cmr=v1.8.2:reviewer=dgoodell
This commit was SVN r31594.
Use the new opal dstore API (vs. the old RTE DB API).
(dstore is not going to the v1.8 series, so there's no need to CMR
this to v1.8)
This commit was SVN r31580.
Not closing this file descriptor will cause us to leak file
descriptors. It is safe to close the file after it has been mmapped.
cmr=v1.8.2:reviewer=manjugv
This commit was SVN r31579.
There's a requirement in several places (e.g., opal dstore) that
sizeof(ompi_process_name_t) -- which comes from the compile-time
selected ompi/mca/rte component -- is equal to sizeof(uint64_t). If
it's not, Bad Things will happen.
So put an assert here to catch that case.
This commit was SVN r31577.
The old default context id maximum was committed to the trunk in
2006. After some discussion with Intel it appears this is restricting
the mtl to an arbirarly small number of communicators. Increasing the
default to allow up to 2^16 - 1 context ids.
Refs trac:4574
cmr=v1.8.2
This commit was SVN r31574.
The following Trac tickets were found above:
Ticket 4574 --> https://svn.open-mpi.org/trac/ompi/ticket/4574
http://www.open-mpi.org/community/lists/devel/2014/04/14496.php
Revamp the opal database framework, including renaming it to "dstore" to reflect that it isn't a "database". Move the "db" framework to ORTE for now, soon to move to ORCM
This commit was SVN r31557.
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.
We will track #4568 from the 1.8 CMR.
Closes trac:4568
cmr=v1.8.2:reviewer=jsquyres
This commit was SVN r31535.
The following Trac tickets were found above:
Ticket 4568 --> https://svn.open-mpi.org/trac/ompi/ticket/4568
This commit will improve the message rate when using the sendi function
by not waiting for the send to get to the remote process.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31526.
feature
This commit should fix a hang seen when running some of the one-sided
tests. The downside of this fix is it reduces the maximum size of the
messages that use the fast boxes. I will fix this in a later commit.
To improve performance under a heavy load I introduced sequencing to
ensure messages are given to the pml in order. I have seen little-no
impact on the message rate or latency with this change and there is a
clear improvement to the heavy message rate case.
Lets let this sit in the trunk for a couple of days to ensure that
everything is working correctly.
cmr=v1.8.2:reviewer=jsquyres
This commit was SVN r31522.
Patch from Gilles Gouaillardet on #4517 to fix handling 0-sized
messages in coll tuned with MPI_ALLTOALLV and MPI_IN_PLACE.
Reviewed by Jeff Squyres.
Fixes trac:4517
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31521.
The following Trac tickets were found above:
Ticket 4517 --> https://svn.open-mpi.org/trac/ompi/ticket/4517
Patch from Gilles Gouaillardet on #4506 to correctly handle 0-sized
messages in coll/basic MPI_Alltoallv and MPI_Alltoallw.
Reviewed by Jeff Squyres.
Fixes trac:4506.
cmr=v1.8.2:reviewer=ompi-rm1.8
This commit was SVN r31519.
The following Trac tickets were found above:
Ticket 4506 --> https://svn.open-mpi.org/trac/ompi/ticket/4506
Ensure to also OBJ_RELEASE the neightbor and ineighbor modules.
Fixes trac:4444 (this patch is from that ticket).
This commit was SVN r31516.
The following Trac tickets were found above:
Ticket 4444 --> https://svn.open-mpi.org/trac/ompi/ticket/4444
The algorithm was failing ibm/collective/allgather and iallgather. I
cleaned up the code to eliminate duplicate code paths and tracked the
issue down to an error in the way extra nodes in the knomial exchange
are handled. The new code is more compact and has been tested with up
to 64 ranks with the ibm test suite.
cmr=v1.8.1:reviewer=manjugv
This commit was SVN r31419.
The file coll_ml_ibarrier.c wasn't included in coll/ml's Makefile.am
and the setup code from coll_ml_hier_algorithms_ibarrier.c was not
being called. It looks like this code is stale and has long since been
replaced by the code in coll_ml_barrier.c
Once all these little CMRs are approved I may make it into one roll-up
CMR to make it easier on the RM.
cmr=v1.8.1:reviewer=manjugv
This commit was SVN r31418.
a segmentation fault in the reduce cleanup
Some of the changes address false warnings produced by scan-build. I
added asserts and changed some malloc calls to calloc to silence these
warnings.
The was one issue in cleanup for reduce since the component_functions
member is changed by the allreduce call. There may be other issues
with how this code works but releasing the allocated
component_functions after setting up the static functions addresses
the primary issue (SIGSEGV).
cmr=v1.8.1:reviewer=manjugv
This commit was SVN r31417.
Two things to note:
- This change will allow us to expand the BTL interface without
having to worry about modifying BTLs that will not support the new
interfaces. More on this will come later this year as part of the
1.9 series.
- C99 guarantees that uninitialed members of structs declared outside
of functions (DATA binary section) will be initialized with
0's. This allows us to drop stuff like .btl_flags = 0, or .btl_get
= NULL.
This commit was SVN r31388.
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.
While testing one-sided on LANL systems I found a couple more OSC
bugs that were not caught during the initial testing:
- In the passive target code we read the read lock count as a
char instead of the intended uint32_t. This causes lock to
lockup when using shared locks after 127 iterations.
- The post code used the wrong group when trying to increment post
counters. This causes a segmentation fault.
- Both the post and wait code used the wrong check in the inner
loop leading to an infinite loop.
cmr=v1.8.1:reviewer=jsquyres
This commit was SVN r31354.
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.
some of the collective modules, the shared memory and the profiling
interface. I left out VT, dynamic fcoll and seq rmaps.
cmr=v1.8.1:reviewer=jsquyres:subject=silence Coverity reported warnings
This commit was SVN r31309.
This commit fixes two nasty races:
- One can occur if the connection request message and connection completion
message arrive out of order. This can happen normally when adaptive routing
is used and also in a timeout situation where a UD message is lost.
- One occurs when handling an ack at the same time as we are handling the
message timeout. In this case we can not free the message or the timeout
will be operating on invalid data. This fix is a band-aid until I can come
up with a better approach. Instead of freeing the message it is marked
as inactive and the event callback is triggered immediately (this has no
affect if the callback is already active). The callback then frees the
message if it is inactive.
cmr=v1.8.1:reviewer=pasha
This commit was SVN r31305.
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.
It might be possible (don't know) for a datatype to made of a contiguous block
of a primitive datatype and have an lb. If this is ever the case the code
would have done the wrong thing. Add the lb in to be safe.
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31283.
Discussed this with Manju and we decided to back this one out until a later time.
This reverts commit r31188 and closes trac:4435
This commit was SVN r31282.
The following SVN revision numbers were found above:
r31188 --> open-mpi/ompi@f1dd589092
The following Trac tickets were found above:
Ticket 4435 --> https://svn.open-mpi.org/trac/ompi/ticket/4435
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.
There were a couple of issues with the memory leak fixes and several more verbose
issues. This fixes those issues.
cmr=v1.8.1:ticket=trac:4473
This commit was SVN r31273.
The following Trac tickets were found above:
Ticket 4473 --> https://svn.open-mpi.org/trac/ompi/ticket/4473
This fixes more issues identified by armci. More issues still remain and fixes are
coming for those as well.
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31272.
Also, since I put some of the macros for these silent/verbose rules up
in the top-level Makefile.man-page-rules file, I renamed it to
Makefile.ompi-rules.
I've had this sitting around for a while; now seems like as good a
time as any to commit it.
This commit was SVN r31271.
directory not the job's
This bug didn't affect the correctness of the vader results just the
cleanup. This commit removes an error message about removing a non-existent
file.
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31265.
Thanks to ggouaillardet for finding and fixing these issues.
Closes trac:4460
cmr=v1.8.1:reviewer=manjugv
This commit was SVN r31264.
The following Trac tickets were found above:
Ticket 4460 --> https://svn.open-mpi.org/trac/ompi/ticket/4460
If ompi_modex_recv() fails with OPAL_ERR_DATA_VALUE_NOT_FOUND, it
simply means that the peer process did not put any usnic BTL modex
info -- it is not an error. So have the usnic BTL simply ignore that
peer (vs. disqualifying itself / treating this like a real error).
Refs trac:4442.
This commit was SVN r31258.
The following Trac tickets were found above:
Ticket 4442 --> https://svn.open-mpi.org/trac/ompi/ticket/4442
This commit should finish the work started for #869. Closing that ticket
with this commit.
Closes trac:869
cmr=v1.8.1:reviewer=jsquyres
This commit was SVN r31257.
The following Trac tickets were found above:
Ticket 869 --> https://svn.open-mpi.org/trac/ompi/ticket/869
in ompi_mtl_mxm_add_procs, define the ep_index variable only
for an older version of mxm.
submitted by Alina, reviewed by Mike.
cmr=v1.8:reviewer=ompi-rm1.8
This commit was SVN r31245.
The error doesn't prevent the user from running so there is no reason
to display it unless the user requested it (through coll_ml_verbose).
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31242.
Fix a one line bug when dealing with non-contiguous sends in prepare_src. Bug was
identified by the intel test suite.
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31232.
the case fix in ompi_osc_base_process_op in r31204.
There are two cases that needed to be handled:
- The target is a simple datatype (contiguous block of a primitive
type) but the origin is not. In this case we still need to pack
the origin data but we can not rely on the convertor to do the
unpack (see r31204).
- Both the origin and target datatypes are simple datatypes. In this
case we can use ompi_op_reduce to do the accumulation without having
to pack the origin data.
cmr=v1.8:ticket=trac:4449
This commit was SVN r31231.
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
Fixed two bugs:
- Use module->comm NOT comm to get the CID for the shared memory backing
file. This fixes the case where there are multiple shared memory windows
at the same time.
- Remember to unlink the shared memory backing file.
Refs trac:4438
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31227.
The following Trac tickets were found above:
Ticket 4438 --> https://svn.open-mpi.org/trac/ompi/ticket/4438
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.
In most cases, bad messages received by the connectivty checker are
just dropped. However, in one specific code path, a bad packet caused
an abort. Doh!
This commit does two things:
1. Improve verbose messages for all these cases
1. Simply drop incoming messages that cannot be identified as ACKs or PINGs
Submitted by Jeff Squyres, reviewed by Dave Goodell.
cmr=v1.8:reviewer=ompi-rm1.8
This commit was SVN r31225.
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.
of the primitive datatype
In this case we can not use the convertor to run the accumulate operation
since the datatype is a more or less a primitive type.
cmr=v1.8:ticket=trac:4449
This commit was SVN r31222.
The following Trac tickets were found above:
Ticket 4449 --> https://svn.open-mpi.org/trac/ompi/ticket/4449
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.
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.
ROMIO code assumes all processes will use the same ROMIO driver. we
were not reaching the "find a common file system" logic when NFS was
enabled, everyone stat-ed the file system without errors, but some
processees found a different file system (like if some processes are
writing to NFS and others to UFS)
See discussion beginning here:
http://lists.mpich.org/pipermail/discuss/2014-March/002403.html
Tested-by: Jeff Squyres <jsquyres@cisco.com>
Submitted by Rob Lathan, reviewed by Jeff Squyres
cmr=v1.8:reviewer=ompi-rm1.8
This commit was SVN r31201.
the ompi_common_verbs_find_ports function had a call to
ompi_ibv_get_device_list, but not to ompi_ibv_free_device_list.
fixed by Alina, reviewed by Vasily/Mike.
cmr=v1.8:reviewer=ompi-rm1.8
This commit was SVN r31200.
This commit adds large datatype description support to the osc/rdma
component. Support is provided by an additional send/recv of the datatype
description if the description does not fit in an eager buffer. The
code is designed to require minimal new code and not for speed. We
consider this code path to be a slow path.
Refs trac:1905
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31197.
The following Trac tickets were found above:
Ticket 1905 --> https://svn.open-mpi.org/trac/ompi/ticket/1905
When we are only using local ranks basesmuma needs to provide an allreduce
function for both large and small message or else the coll/ml selection
logic will fail. In the future this logic should probably be updated to
just disable allreduce in coll/ml instead of disabling coll/ml. For now
it should be correct to say the basesmuma allgather works for larger
messages.
cmr=v1.8:reviewer=manjugv
This commit was SVN r31190.
a hierarchy actually matches a bcol that is in use.
There was a bug in one of the paths to calculate the ml buffer size. I fixed
the bug and squashed all the paths together to avoid further issues (the
result was correct in another path that calculated the same value).
Additionally, the i_hier was being used as the bcol_index. This is not
correct in a couple of cases so I added a variable to keep track of the
real bcol_index.
cmr=v1.8:reviewer=pasha
This commit was SVN r31189.
bound.
This case is correctly handled by coll/ml so remove the check that diables
coll/ml in the not bound case.
cmr=v1.8:reviewer=manjugv
This commit was SVN r31188.
This patch fixes two leaks:
- Fix typo in fallback collective code that caused coll/ml to retain
the ibcast module twice but only release it once. One of those ibcast
saves was supposed to be bcast.
- Do not check for module initialization in the module destructor. It
is possible to destruct a module that is partially setup.
cmr=v1.8:reviewer=manjugv
This commit was SVN r31187.
When you call MPI_Graph_create with a old_comm of size N, and pass
nnodes=(N=1), then the Nth proc is supposed to get MPI_COMM_NULL out.
The code in this base function didn't properly handle the proc(s) that
are supposed to get MPI_COMM_NULL out.
cmr=v1.7.5:reviewer=hjelmn
This commit was SVN r31145.
This isn't causing any errors that I know about but it does fix an
annoying valgrind warning. Simple fix, no review required.
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r31130.
There are situations where coll/ml does not initialize properly. These will
eventually need to be fixed but in the meantime it is better to not always
print an error message because the collective framework can still fall back
on another collective module. This commit reduces the verbose output.
cmr=v1.7.5:reviewer=manjugv
This commit was SVN r31129.
It is usually not a good idea to assert when something is not implemented
or something goes wrong. Replace asserts with debug output and return.
cmr=v1.7.5:reviewer=manjugv
This commit was SVN r31128.
The necessary information is stored in the proc object. There is no need
to allgather the local process data to determine if another rank is on
the same socket.
cmr=v1.7.5:reviewer=manjugv
This commit was SVN r31127.
After discussion with Manju we decided to update these the process count
limits of the shared memory collectives to an arbitrarily large number.
cmr=v1.7.5:ticket=trac:4405
This commit was SVN r31126.
The following SVN revision numbers were found above:
r31096 --> open-mpi/ompi@3f469d08e7
The following Trac tickets were found above:
Ticket 4405 --> https://svn.open-mpi.org/trac/ompi/ticket/4405
* Ensure that all endpoints[x] values are initialized to NULL
* If ibv_create_ah fails, remove each endpoint from the
module->all_endpoints list so that the endpoint can be destructed
properly.
Submitted by Jeff Squyres, reviewed by Dave Goodell.
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r31111.
This provides full locality - i.e., not just node-level, but all the way down to whatever common binding level exists between the procs.
cmr=v1.7.5:reviewer=jsquyres
This commit was SVN r31106.
Also fixed spelling: IS_NOT_RECHABLE -> IS_NOT_REACHABLE.
Also mark a few places where opal_show_help() should have been used;
Manju will take care of these.
This commit was SVN r31104.
In r31071 I modified the logic to not increment the hierarchy level if
no processes were selected by that sbgp. That fixed a problem seen on
systems where we don't support process binding. The problem is there
is a case where we actually did select processes yet the number of
selected processes is 0. We need to increment the hierarchy in this case
as well.
This should fix the segmentation fault found by recent MTT runs. Once
this is committed to 1.7.5 remove the .ompi_ignore's from coll/ml and
bcol/ptpcoll. Tested with ompi-tests/ibm.
cmr=v1.7.5:reviewer=rhc
This commit was SVN r31081.
The following SVN revision numbers were found above:
r31071 --> open-mpi/ompi@1911d97044
This was causing JVMs to run out of stack space, and all manner of
badness ensued.
Instead, use the heap -- that's what it's there for.
cmr=v1.7.5:reviewer=rhc:subject=make coll/ml use the heap for large debug array
This commit was SVN r31073.
fails to select any processes on any nodes.
Also modified basesmsocket to only print debugging info to the framework
output.
cmr=v1.7.5:reviewer=jsquyres
This commit was SVN r31071.
It is not valid to call flush outside a passive target epoch nor is
it valid to call lock/lock_all when no_locks is set. In the former
we were just semantically incorrect and the later would crash and
burn.
cmr=v1.7.5:ticket=trac:4382
This commit was SVN r31046.
The following Trac tickets were found above:
Ticket 4382 --> https://svn.open-mpi.org/trac/ompi/ticket/4382
This fixes a bug in r31029 which removes the use of the pml base request
(also not a good way since cm doesn't use the base request). We now allocate
a data structure (ugh) to determine the needed information. Tested with
mtt/onesided.
cmr=v1.7.5:ticket=trac:4379
This commit was SVN r31044.
The following SVN revision numbers were found above:
r31029 --> open-mpi/ompi@29e00f9161
The following Trac tickets were found above:
Ticket 4379 --> https://svn.open-mpi.org/trac/ompi/ticket/4379
- Return an error if the caller specified both MPI_MODE_NOPRECEDE and
MPI_MODE_NOSUCCEED to MPI_Win_fence.
- Return an error if the caller attempts to enter an active target
epoch while already in a passive target epoch.
- End an active target epoch if MPI_Win_fence is called with
MPI_MODE_NOSUCCEED.
cmr=v1.7.5:ticket=trac:4382
This commit was SVN r31043.
The following Trac tickets were found above:
Ticket 4382 --> https://svn.open-mpi.org/trac/ompi/ticket/4382
need to enable the access epoch in MPI_Win_fence.
I missed this change when I fixed the semantics of MPI_Win_create. With
this commit our one-sided MTT runs are now running clean.
cmr=v1.7.5:reviewer=dgoodell
This commit was SVN r31041.
It seems we can't release accumulate buffers in completion callbacks
because the btls don't release registration resources until after the
callback has fired. The fix is to keep track of the unused buffers and
free them later. This should resolve issues when running IMB-EXT and
IMB-RMA.
cmr=v1.7.5:reviewer=jsquyres
This commit was SVN r31029.
Dave Goodell correctly pointed out that it is unusual to return MPI
error classes from internal ompi functions. Correct this in the RMA
case by adding an internal error code to match MPI_ERR_RMA_SYNC.
This does change OMPI_ERR_MAX. I don't think this will cause any
problems with ABI.
cmr=v1.7.5:reviewer=jsquyres
This commit was SVN r31012.
This commit resolves a number of crashed discovered my the onesided
tests in MTT. The functions in question were operating on the assumption
the user was calling RMA functions correctly.
cmr=v1.7.5:reviewer=jsquyres
This commit was SVN r31008.
NOTE: I transferred the oshmem-disabled-by-default from the 1.7 branch to the trunk to minimize future disruption if/when we change that option.
cmr=v1.8:reviewer=jsquyres
This commit was SVN r31006.
- -check-shmem-params is OFF by default. It checks OSHMEM API params and will abort on bad input
- hcoll do not save fallback coll pointers for unsupported collectives.
fixed by Val, Roman, reviewed by Miked/Igor
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30995.
The datatype unpacking code assumes that the packed datatype buffer has the
same alignment as an OPAL_PTRDIFF_TYPE. This was not enforced by the rdma
one-sided component. I changed the ordering and sized of various osc/rdma
headers to ensure their sizes are a multiple of 8-bytes and modified the
fragment allocation call to ensure all headers are 8-byte aligned. While
not the cleanest way to handle this situation it should resolve the issue.
Fixes trac:4315
cmr=v1.7.5:reviewer=jsquyres
This commit was SVN r30974.
The following Trac tickets were found above:
Ticket 4315 --> https://svn.open-mpi.org/trac/ompi/ticket/4315
assumes the send request is derived from mca_pml_base_send_request_t,
but this is not true for pml cm, so we end up freeing invalid pointer.
We cannot take the data pointer from the pml send request, so we pass
the allocated buffer pointer in req_complete_cb_data, and put the
osc_rdma_module pointer in that buffer as well.
Previously, osc_pt2pt was used with pml_cm which didn't have this
problem.
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30967.
* clang warning stomp
* memory barrier for volatile variable use
These can go to 1.7.5 or can slip to v1.8 -- RM decision.
Submitted by Jeff Squyres, reviewed by Dave Goodell
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30944.
* Older versions of libusnic_verbs actually return 0 when querying for
an unknown port. So also check for a magic ID in the returned data
to *really* know if the usnic extensions are supported.
* Use a union (in the common_verbs area) and memcpy (in the btl) to
avoid undefined C type aliasing behavior.
* Ensure to memset the function table to 0 if the usnic extensions
are not supported.
Submitted by Jeff Squyres, reviewed by Dave Goodell.
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30935.
When compiling --with-ft there are a few compiler warnings about
unused variables. This patch fixes those compiler warnings.
This commit was SVN r30927.
Realistically, the usnic BTL doesn't need to know anything about the
underlying transport except for its header length (so that it knows
where the payload begins in a received buffer). So remove the use of
the specific transport prefix union and just rely on the usnic verbs
extension to tell us what the header length is if we're using the
usNIC/UDP transport, or sizeof(struct ibv_grh) if we're using usNIC/L2
transport.
This commit was SVN r30914.
Check the IBV_TRANSPORT_* values. In the case of IBV_TRANSPORT_IWARP,
there's an ambiguity and we need to also check to see whether the
usnic verbs externsion probe exists.
This commit was SVN r30913.
If they exist, call the usnic verbs extensions to both enable UDP
support and get the UD receiver header length that should be used
(rather than assume 40/struct GRH).
This commit was SVN r30912.
- now add_procs can be called more than once (during MPI_INIT and Inter_Comm_Create)
- adjust MXM to this reality
fixed by Alina, reviewed by Yossi/Mike
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30907.
This allows compilers to know that the code path(s) where
ompi_rte_abort() is invoked won't return (and therefore won't warn in
certain cases).
cmr=v1.8:reviewer=rhc
This commit was SVN r30891.
We don't use this functionality any more; we use the transport_type
and device name to identify usnic devices. It's slightly easier
because we can transport_type+name from ibv_device_open() and don't
have to do an additional ibv_query_device() to get its attributes.
Reviewed by Dave Goodell.
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30882.
Follow on to SVN trunk r30850: consolidate the ibv_create_ah() calls
into a single loop, MPI_WAITALL-style. That is, call the (effectively
non-blocking) ibv_create_ah() for each endpoint. If we get
NULL+EAGAIN, it means that the UDP ARP is still ongoing down in the
kernel, so just try again later. We put these all into a single loop
because it allows us to parallelize the ARP progress in the kernel.
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30879.
The following SVN revision numbers were found above:
r30850 --> open-mpi/ompi@3641500442
r30852 --> open-mpi/ompi@4e282a3295
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Basically: since usnic is a connectionless transport, we do not get
OS-provided services "for free" that connection-oriented transports
get, namely: "hey, I wasn't able to make a connection to peer X", and
"hey, your connection to peer X has died."
This connectivity-checker runs in a separate progress thread in the
usnic BTL in local rank 0 on each server. Upon first send in any
process, the connectivty-checker agent will send some UDP pings to the
peer to ensure that we can reach it. If we can't, we'll abort the job
with a nice show_help message.
There's a lengthy comment in btl_usnic_connectivity.h explains the
scheme and how it works.
Reviewed by Dave Goodell.
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30860.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
finalize.
Closes trac:4290
cmr=v1.7.5:reviewer=miked
This commit was SVN r30854.
The following Trac tickets were found above:
Ticket 4290 --> https://svn.open-mpi.org/trac/ompi/ticket/4290
- Fix several typos is osc/rdma.
- Fix a locking issue in osc/sm that was caused by an incorrect
assumption about the semantics of opal_atomic_add_32.
- Always unlock the accumulation lock in osc/sm.
- The base of a processes shared memory window should be NULL if
the size is zero. Fixed.
cmr=v1.7.5:ticket=trac:4304
This commit was SVN r30853.
The following Trac tickets were found above:
Ticket 4304 --> https://svn.open-mpi.org/trac/ompi/ticket/4304
Follow on to SVN trunk r30850: consolidate the ibv_create_ah() calls
into a single loop, MPI_WAITALL-style. That is, call the (effectively
non-blocking) ibv_create_ah() for each endpoint. If we get
NULL+EAGAIN, it means that the UDP ARP is still ongoing down in the
kernel, so just try again later. We put these all into a single loop
because it allows us to parallelize the ARP progress in the kernel.
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30852.
The following SVN revision numbers were found above:
r30850 --> open-mpi/ompi@3641500442
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
ibv_create_ah() may need to effect an ARP resolution, which may take
some time. Rather than blocking in ibv_create_ah(), the usNIC driver
may return NULL and set errno to EAGAIN indicating that we should try
again (i.e., the ARP resolution is proceeding under the covers).
So add a simple loop here to loop over ibv_create_ah() until it
returns non-(NULL+EAGAIN). A future commit will make this a bit more
efficient.
Authored-by: Jeff Squyres <jsquyres@cisco.com>
Reviewed-by: Dave Goodell <dgoodell@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30850.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Prior to this commit we matched local interfaces to remote interfaces in
order to create endpoints in a simplistic way. If any remote interfaces
were on the same subnet as any of our local interfaces then only local
interfaces would be paired (IP-routed remote interfaces would be
ignored).
This commit introduces a more general scheme which attempts to make the
"best" pairing of local interfaces to remote interfaces. We now cast
the problem as a graph theory problem known as the "Assignment Problem",
or finding a maximum-cardinality, minimum-weight bipartite matching. We
solve this problem by reducing the bipartite graph of interface
connectivity to a flow network and then solving for a minimum cost flow.
This is then easily converted into back into a matching on the original
bipartite graph.
In the new scheme, interfaces on the same subnet are preferred over
interfaces requiring intermediate routing hops and higher bandwidth
links are preferred over lower bandwidth links.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30849.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Querying the OS routing table is important for making decisions about
which local and remote interfaces should be paired into reliable
communication channels.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30848.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
This code is intended to support usNIC interface matching functionality.
We currently view that problem as essentially the "Assignment Problem"
(http://en.wikipedia.org/wiki/Assignment_problem), for which there are
many possible solution approaches, including flow-network analysis. In
the future, we might transition to a more nuanced view of the problem
which would likely also be flow-network based.
To this end, the current code focuses on providing one major algorithm
to the core usnic BTL: `ompi_btl_usnic_solve_bipartite_assignment`. It
also exposes several typical and necessary functions for constructing,
manipulating, and querying weighted, directed graphs.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30847.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30846.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
This commit adds mechanisms for writing and running unit tests in the
usnic BTL. The short version of how to run the tests is:
1. Configure with `--enable-ompi-btl-usnic-unit-tests`. This will cause
the unit testing code and test runner utility to be built.
2. Run the tests by running `ompi_btl_usnic_run_tests`.
See `README.test` for full details.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30845.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
These includes only exist in the Cisco-internal usnic-v1.6 code base,
but they should not exist anywhere except btl_usnic_compat.h in order to
minimize source differences between usnic-v1.6 and v1.7/trunk.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30844.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Lower layer (hardware or software) bugs can result in a mismatch between
our BTL-layer payload size and the actual packet length. We now check
that in order to catch these cases, which otherwise can result in
MPI-layer message corruption.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30843.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
We were missing a debug message for a very common recv case, making it a
bit harder to follow a debug log.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30842.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
There was a duplicated subnet check in the sender hash lookup routine.
This caused receivers to always fail the sender hash lookup if the
sender was in a different subnet, so the receiver would discard the
packet as though it were coming from a different job.
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30841.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
If ibv_create_ah fails, we will not initialize the `endpoint->proc`.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30840.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
This functionality is required for routable UDP/IP usnic traffic.
Previously we would only setup endpoints for remote interfaces on the
same subnet as the current module's local interface. This behavior
still holds if two processes share any common subnets. However, if the
two processes only have no subnets in common then we assume that all
interfaces are reachable from all other interfaces and wire them up in a
1-1, randomly-matched order somewhat similarly to the "tcp" BTL's
behavior.
Only match in different subnets if we detect UDP support in the lower
layer.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30839.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
This commit decouples OMPI deployment from the version(s) of the lower
layers of the stack by probing for UDP support.
Verbs applications assume a 40-byte header (there is no current
mechanism for querying payload offset). So to support a 42-byte UDP
header without causing existing applications like ibv_ud_pingpong or
older versions of OMPI to crash, we must inform libusnic_verbs that we
are aware of the nonstandard payload offset. We do this by overriding
the `transport_type` field of the device to be 42 before calling
`ibv_open_device`. If the library resets it to something else, then we
know the lower layers are UDP capable. Otherwise we use the older
custom-L2 format.
This necessitated some minor ugliness in common_verbs, but it's as tidy
as Jeff and I know how to make it right now.
This commit only adds support for UDP headers and connectivity over the
same L2 network, it does not touch routing or interface pairing.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30838.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Just trying to be deliberate about keeping fastpath-accessed fields
grouped together to fit into the same 64-byte cache line.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30837.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Authored-by: Jeff Squyres <jsquyres@cisco.com>
Reviewed-by: Reese Faucette <rfaucett@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30836.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Authored-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30835.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Authored-by: Reese Faucette <rfaucett@cisco.com>
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30834.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Authored-by: Reese Faucette <rfaucett@cisco.com>
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30833.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
Valgrind showed this one, just a bit of sloppiness with the reference
counting.
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30832.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
The logic did not correctly perform the OR behavior that is described
in the doxy docs for this function. This commit fixes the logic so
that a port will be included if it has supports any of the
capabilities indicated by the passed-in flags.
Authored-by: Jeff Squyres <jsquyres@cisco.com>
Reviewed-by: Dave Goodell <dgoodell@cisco.com>
cmr=v1.7.5:ticket=trac:4253
This commit was SVN r30831.
The following Trac tickets were found above:
Ticket 4253 --> https://svn.open-mpi.org/trac/ompi/ticket/4253
1. Changed rng_buff_t --> opal_rng_buff_t
2. All global variables obey the prefix rule
3. Old code has been removed
4. Found a couple of unnecessary includes
Refs trac:4298
This commit was SVN r30807.
The following Trac tickets were found above:
Ticket 4298 --> https://svn.open-mpi.org/trac/ompi/ticket/4298
We're going to be bringing a bunch of usnic code to the SVN trunk
soon, and I basically brought this commit over out of order. So I'm
reverting it for now; the same functionality will come back shortly.
This commit was SVN r30805.
The following SVN revision numbers were found above:
r30804 --> open-mpi/ompi@5bedcc15bf
These constants are now upstream (see
https://git.kernel.org/cgit/libs/infiniband/libibverbs.git/commit/?id=f57a9c67eabb9e7f19c624ac3c8c27b7be55796c),
so let's support them properly in Open MPI.
Added bonus: consolidating these checks up in
ompi_check_openfabrics.m4 allowed removing some custom checks and
AC_DEFINE's from the usnic configure.m4 script.
Also change the usnic/configure.m4 check for IBV_EVENT_GID_CHANGE to
use AC_CHECK_DECLS (vs. AC_CHECK_DECL).
cmr=v1.7.5:reviewer=dgoodell
This commit was SVN r30804.
* Use the prefix rule for global variables
* Elimiante seed_prng() since it isn't necessary any more
These files will need to get edited again then the RNG type obeys the
prefix rule.
Refs trac:4298
This commit was SVN r30803.
The following SVN revision numbers were found above:
r30801 --> open-mpi/ompi@e39d9f4080
The following Trac tickets were found above:
Ticket 4298 --> https://svn.open-mpi.org/trac/ompi/ticket/4298
- MXM uses libtool versioning scheme which is enough, no need additional in OMPI
reviewed by yossi
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30768.
Some older versions of libibverbs do not have `ibv_event_type_str`,
leading to compilation failures on older machines, irrespective of
whether they could ever support usNIC anyway. If we encounter any other
build issues related to "old verbs" then we should just cause the usnic
BTL to disqualify itself when it encounters "old" traits.
Thanks to Paul Hargrove for reporting the issue:
http://www.open-mpi.org/community/lists/devel/2014/02/14056.php
Reviewed-by: Jeff Squyres <jsquyres@cisco.com>
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30674.
only goes up to VADER_MAX_ADDRESS instead of 0xfffffffffffffffful.
cmr=v1.7.5:ticket=trac:4216
This commit was SVN r30669.
The following Trac tickets were found above:
Ticket 4216 --> https://svn.open-mpi.org/trac/ompi/ticket/4216
The error was caused by leaving the pipe to the async thread uninitialized, then writing to it regardless of this.
Fix is to check the existance of the async thread and the pipe to it.
reviewd by miked
cmr=v1.7.5:reviewer=ompi-rm1.7
This commit was SVN r30644.
The initialization code did several allgathers on void *'s using
MPI_LONG_LONG_INT. This will produce the wrong result on 32-bit
platforms. Instead use MPI_BYTE with count = sizeof (void *).
cmr=v1.7.5:ticket=trac:4158
This commit was SVN r30627.
The following Trac tickets were found above:
Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
Found two bugs in basesmuma:
- Release all resources when tearing down the bcol module.
- Allways call the allreduce in the smcm code. We do not know
beforehand whether all procs have all the files mapped.
cmr=v1.7.5:ticket=trac:4158
This commit was SVN r30623.
The following Trac tickets were found above:
Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
This is hot-fix patch for the issue reported by Ralph.
In future we plan to restructure ml data structure layout.
Tested by Nathan.
cmr=v1.7.5:ticket=trac:4158
This commit was SVN r30619.
The following Trac tickets were found above:
Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
This commit was SVN r30605.
The following SVN revision numbers were found above:
r30600 --> open-mpi/ompi@7d2c4cb468
r30602 --> open-mpi/ompi@9e751a0302
r30604 --> open-mpi/ompi@3012c280cf
Revision number ranges (suitable for "git log"):
r30602-30604 --> open-mpi/ompi@9e751a03^..3012c280
The problem was caused by the static request optimization. The buffered send case
is much like the isend case in that the request structure may be needed after
MPI_Bsend completes. Fix this case by calling isend and freeing the resulting
request.
cmr=v1.7.5:ticket=trac:4149
This commit was SVN r30601.
The following Trac tickets were found above:
Ticket 4149 --> https://svn.open-mpi.org/trac/ompi/ticket/4149
them, but it's going to take a little time (at least one day). So
Nathan says it's ok to .ompi_ignore coll ml until he's able to fix it.
This commit was SVN r30600.
This change does not appear to increase the small message latency of ping-pong
benchmarks and fixes an issue found by our ibm datatype tests.
Fixes trac:4232
cmr=v1.7.5:ticket=trac:4149
This commit was SVN r30598.
The following Trac tickets were found above:
Ticket 4149 --> https://svn.open-mpi.org/trac/ompi/ticket/4149
Ticket 4232 --> https://svn.open-mpi.org/trac/ompi/ticket/4232
opal does not always define MB. It is recommended that opal_atomic_[rw]mb is
called instead. We will need to address the cases where these functions are
no-ops on weak-memory ordered cpus.
cmr=v1.7.5:ticket=trac:4158
This commit was SVN r30534.
The following Trac tickets were found above:
Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
Several changes are contained in this commit:
- Clean up tabs and trailing whitespaces
- Use consistent indentation in changed files
- Remove unused code. None of the removed code will ever have been
used in a trunk build.
- Clean up the smcm code quite a bit
- Do not fflush stderr and use opal_output instead of fprintf.
These changes have been tested on Cray XE-6 and PSM systems.
cmr=v1.7.5:ticket=trac:4158
This commit was SVN r30533.
The following Trac tickets were found above:
Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
During the commits to make the C/R code compile again the
blocking receive calls were replaced by non-blocking
which broke the code. This patch uses ORTE_WAIT_FOR_COMPLETION()
to wait until the non-blocking calls have finished.
This commit was SVN r30486.
This commit fixes one warning that should have caused coll/ml to segfault
on reduce. The fix should be correct but we will continue to investigate.
cmr=v1.7.5:ticket=trac:4158
This commit was SVN r30477.
The following Trac tickets were found above:
Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
for 32-bit architectures.
This commit also modifies _OMPI_CHECK_HEADER to use AC_CHECK_HEADERS instead
of AC_CHECK_HEADER. This allows components to check for multiple headers
instead of just one. The new semantics of the header check in OMPI_CHECK_PACKAGE
are to return success if at least one of the specified headers exists. The new
semantics will not break current usage.
cmr=v1.7.5:ticket=trac:4053
This commit was SVN r30476.
The following Trac tickets were found above:
Ticket 4053 --> https://svn.open-mpi.org/trac/ompi/ticket/4053
After IM with Nathan, apply patch from ticket after verification by Paul Hargrove that it fixes the problem on non-x86 32-bit platforms
Verified by Paul, RM-approved
cmr=v1.7.4:reviewer=ompi-gk1.7
This commit was SVN r30411.
The following Trac tickets were found above:
Ticket 4143 --> https://svn.open-mpi.org/trac/ompi/ticket/4143
ROMIO and Lustre 2.4.0. It has been solved upstream already; here's
the ticket:
http://trac.mpich.org/projects/mpich/ticket/1973
And here's the commit that fixed it:
a0c4278f14
OMPI does not have the other code referred to in that git commit (in
ad_lustre_hints.c).
Thanks to Adam Moody for reporting the issue.
cmr=v1.7.4:reviewer=hjelmn:subject=Fix ROMIO compile error w/ Lustre 2.4
This commit was SVN r30393.
btl sendi functions currently can not handle the descriptor being NULL. The
send inline optimization was assuming (incorrectly) that NULL was ok.
cmr=v1.7.5:ticket=trac:4149
This commit was SVN r30364.
The following Trac tickets were found above:
Ticket 4149 --> https://svn.open-mpi.org/trac/ompi/ticket/4149
allgather.
The new collectives provide a signifigant performance increase over tuned for
small and medium messages. We are initially setting the priority lower than
tuned until this has had some time to soak in the trunk. Please set
coll_ml_priority to 90 for MTT runs.
Credit for this work goes to Manjunath Gorentla Venkata (ORNL), Pavel Shamis (ORNL),
and Nathan Hjelm (LANL).
Commit details (for reference):
Import ORNL's collectives for MPI_Allreduce, MPI_Reduce, and MPI_Allgather.
We need to take the basesmuma header into account when calculating the
ptpcoll small message thresholds. Add a define to bcol.h indicating the
maximum header size so we can take the header into account while not
making ptpcoll dependent on information from basesmuma.
This resolves an issue with allreduce where ptpcoll overwrites the
header of the next buffer in the basesmuma bank.
Fix reduce and make a sequential collective launcher in coll_ml_inlines.h
The root calculation for reduce was wrong for any root != 0. There are
four possibilities for the root:
- The root is not the current process but is in the current hierarchy. In
this case the root is the index of the global root as specified in the
root vector.
- The root is not the current process and is not in the next level of the
hierarchy. In this case 0 must be the local root since this process will
never communicate with the real root.
- The root is not the current process but will be in next level of the
hierarchy. In this case the current process must be the root.
- I am the root. The root is my index.
Tested with IMB which rotates the root on every call to MPI_Reduce. Consider
IMB the reproducer for the issue this commit solves.
Make the bcast algorithm decision an enumerated variable
Resolve various asset failures when destructing coll ml requests.
Two issues:
- Always reset the request to be invalid before returning it to the
free list. This will avoid an asset in ompi_request_t's destructor.
OMPI_REQUEST_FINI does this (and also releases the fortran handle
index).
- Never explicitly construct or destruct the superclass of an opal
object. This screws up the class function tables and will cause
either an assert failure or a segmentation fault when destructing
coll ml requests.
Cleanup allgather.
I removed the duplicate non-blocking and blocking functions and modeled
the cleanup after what I found in allreduce. Also cleaned up the code
somewhat.
Don't bother copying from the send to the recieve buffer in
bcol_basesmuma_allreduce_intra_fanin_fanout if the pointers are the
same.
The eliminates a warning about memcpy and aliasing and avoids an
unnecessary call to memcpy.
Alwasy call CHECK_AND_RELEASE on memsync collectives.
There was a call to OBJ_RELEASE on the collective communicator but
because CHECK_AND_RECYLCE was never called there was not matching call
to OBJ_RELEASE. This caused coll ml to leak communicators.
Make allreduce use the sequential collective launcher in coll_ml_inlines.h
Just launch the next collective in the component progress.
I am a little unsure about this patch. There appears to be some sort
of race between collectives that causes buffer exhaustion in some cases
(IMB Allreduce is a reproducer). Changing progress to only launch the
next bcol seems to resolve the issue but might not be the best fix.
Note that I see little-no performance penalty for this change.
Fix allreduce when there are extra sources.
There was an issue with the buffer offset calculation when there are
extra sources. In the case of extra sources == 1 the offset was set
to buffer_size (just past the header of the next buffer). I adjusted
the buffer size to take into accoun the maximum header size (see the
earlier commit that added this) and simplified the offset calculation.
Make reduce/allreduce non-blocking. This is required for MPI_Comm_idup
to work correctly.
This has been tested with various layouts using the ibm testsuite and
imb and appears to have the same performance as the old blocking version.
Fix allgather for non-contiguous layouts and simplify parsing the
topology.
Some things in this patch:
- There were several comments to the effect that level 0 of the
hierarchy MUST contain all of the ranks. At least one function
made this assumption but it was not true. I changed the sbgp
components and the coll ml initization code to enforce this
requirement.
- Ensure that hierarchy level 0 has the ranks in the correct
scatter gather order. This removes the need for a separate
sort list and fixes the offset calculation for allgather.
- There were several passes over the hierarchy to determine
properties of the hierarchy. I eliminated these extra passes
and the memory allocation associated with them and calculate the
tree properties on the fly. The same DFS recursion also handles
the re-order of level 0.
All these changes have been verified with MPI_Allreduce, MPI_Reduce, and
MPI_Allgather. All functions now pass all IBM/Open MPI, and IMB tests.
coll/ml: correct pointer usage for MPI_BOTTOM
Since contiguous datatypes are copied via memcpy (bypassing the convertor) we
need to adjust for the lb of the datatype. This corrects problems found testing
code that uses MPI_BOTTOM (NULL) as the send pointer.
Add fallback collectives for allreduce and reduce.
cmr=v1.7.5:reviewer=pasha
This commit was SVN r30363.
Per RFC. There are two optimizations in this commit:
- Allocate requests for blocking sends and receives on the stack. This
bypasses the request free list and saves two atomics on the critical path.
This change improves the small message ping-pong by 50-200ns on both AMD
and Intel CPUs.
- For small messages try to use the btl sendi function before intializing a
send request. If the sendi fails or the btl does not have a sendi function
silently fallback on the standard send path.
cmr=v1.7.5:reviewer=brbarret
This commit was SVN r30343.
Adds coll_hcoll_np mca parameter similar to that of fca component (defaults to 32). Those who use hcoll be aware that from now on the communicators less than 32 procs will run w/o hcoll by default. - Resolves fallback issue in case libhcoll runs out of allowed contexts. The solution is moving hcoll_context_create from comm_enable to comm_query. Shortly, comm_enable should never return OMPI_ERROR in the coll component with highest priority (hcoll). Otherwise the ompi coll_base_select will unselect the coll funtion pointers and module references leaving the communicator w/o coll pointer. This will cause the fail. Same behavior can be reproduced even with tuned if one would hardcore some "return OMPI_ERROR" into it's module_enable funtion. - Additionally, removed all the dead code under #if 0; removed unused variables (path for library, active_modules list) and classes (module list wrapper)
Fixed by Val, Reviewed by Devendar/Josh/Miked
cmr=v1.7.4:reviewer=ompi-rm1.7
This commit was SVN r30341.
This commit fixes an error path that occurs when huge page allocations are
enabled. In this case we allocate a huge page and try to register it but fail.
We then were calling free on the opal object. Fix this by calling the proper free
function.
cmr=v1.7.4:reviewer=rhc
This commit was SVN r30289.
Also add a verbose flag so one can see what devices are selected as well as another flag to override
locality information and use all devices on the node.
This commit was SVN r30287.
NetBSD puts the AIO functions in -lrt, vs. the usual libc. So we
need the fbtl/posix configure.m4 to test for -lrt properly.
Reviewed by Jeff Squyres.
cmr=v1.7.4:reviewer=ompi-rm1.7:subject=Fix NetBSD use of -laio
This commit was SVN r30274.
NOTE: launch performance will be absolutely awful if you do this with BTLs that aren't configured to modex_recv on first message!
Even with "modex on demand", we still have to do a barrier in place of the modex - we simply don't move any data around, which does reduce the time impact. The barrier is required to ensure that the other proc has in fact registered all its BTL info and therefore is prepared to hand over a complete data package. Otherwise, you may not get the info you need. In addition, the shared memory BTL can fail to properly rendezvous as it expects the barrier to be in place.
This behavior will *only* take effect under the following conditions:
1. launched via mpirun
2. #procs is greater than ompi_hostname_cutoff, which defaults to UINT32_MAX
3. mca param rte_orte_direct_modex is set to 1. At the moment, we are having problems getting this param to register properly, so only the first two conditions are in effect. Still, the bottom line is you have to *want* this behavior to get it.
The planned next evolution of this will be to make the direct modex be non-blocking - this will require two fixes:
1. if the remote proc doesn't have the required info, then let it delay its response until it does. This means we need a way for the MPI layer to tell the RTE "I am done entering modex data".
2. adjust the SM rendezvous logic to loop until the required file has been created
Creating a placeholder to bring this over to 1.7.5 when ready.
cmr=v1.7.5:reviewer=hjelmn:subject=Enable direct modex at scale
This commit was SVN r30259.
intended to be used and emits a compile-time warning.
Thanks to Paul Hargrove for identifying the issue.
cmr=v1.7.4:reviewer=hjelmn:subject=remove/replace malloc.h
This commit was SVN r30231.
Avoid compiler warning about (unnecessarily) initializing 2 variables
during instantiation at the top of a switch block (but outside of any
case statements): just declare the variables at the top of the outter
block. They're already safely initialized, so don't worry about
initializing them in the instantiation.
Reviewed by Dave Goodell.
cmr=v1.7.4:reviewer=ompi-rm1.7:subject=Don't instantiate+init variables in a switch block
This commit was SVN r30228.
ob1 dummy registration to actually be used when using udreg. Fix this by
always setting reg to NULL when mpool/udreg's register function fails.
cmr=v1.7.4:reviewer=rhc
This commit was SVN r30214.
Set comm attribute with keyval.
Wait for pending hcoll module tasks in comm delete callback where PML
still valid on the communicator. safely destroy hcoll context during
hcoll module destructor.
Author: Devendar Bureddy
reviewed by miked
cmr=v1.7.4:reviewer=ompi-rm1.7
This commit was SVN r30175.
needed for correctness. The if_include/if_exclude are level 1, and
the TCP port range params are level 2; this parameter seems to be on
par with the TCP port range params.
Refs trac:4019
This commit was SVN r30161.
The following Trac tickets were found above:
Ticket 4019 --> https://svn.open-mpi.org/trac/ompi/ticket/4019
* Remove some set-but-not-used variables
* Make a convenience function return void (we weren't using the
return code, anyway)
* Mark a function as inline (it was supposed to be inline anyway)
Reviewed by Dave Goodell.
cmr=v1.7.5:reviewer=ompi-rm1.7:subject=Fix usnic BTL compiler warnings
This commit was SVN r30160.
Thanks to Tetsuya Mishima for detecting it!
cmr=v1.7.4:reviewer=jsquyres:subject=Correct tcp_not_use_nodelay option processing
This commit was SVN r30157.
- HCOLL close without init
- Call hcoll progress after comm finalize
- mpirun default for coll_hcoll_enable is 1
fixed by Igor, reviewed by miked
cmr=v1.7.4:reviewer=ompi-rm1.7
This commit was SVN r30156.
configury/Makefile.am changes; this commit renames the internal
installdirs.h framework struct field names to match the configry macro
names:
* pkgdatdir -> ompidatadir
* pkglibdir -> ompilibdir
* pkgincludedir -> ompiincludedir
This commit was SVN r30145.
The following SVN revision numbers were found above:
r30140 --> open-mpi/ompi@8b778903d8
pkg{data,lib,includedir}, use our own ompi{data,lib,includedir}, which is
always set to {datadir,libdir,includedir}/openmpi. This will keep us from
having help files in prefix/share/open-rte when building without Open MPI,
but in prefix/share/openmpi when building with Open MPI.
This commit was SVN r30140.
Complements r30073: tighten up the string parsing of the vendor parts
ID MCA param a bit. Also fix a small memory leak: ensure to free the
array uint32_t's parsed out of the MCA param.
This commit was SVN r30128.
The following SVN revision numbers were found above:
r30073 --> open-mpi/ompi@6003702a51
The following Trac tickets were found above:
Ticket 4301 --> https://svn.open-mpi.org/trac/ompi/ticket/4301
This commit adds support for placing the send memory segment in a
traditional shared memory segment when XPMEM is not available. The
current default is to reserve 4MB for shared memory on each process.
The latest benchmarks show vader performing better than sm on both
Intel and AMD CPUs.
For large messages vader will now use CMA if it is available (and
XPMEM is not).
cmr=v1.7.5:reviewer=jsquyres
This commit was SVN r30123.
Per RFC which expired two weeks ago:
We are planning to make a change to Open MPI to always set up the btls. This
means the btl init will be called even if add_procs is never called for that
btl. In the openib btl free lists fragments are currently allocated in btl_init.
To avoid wasting that memory this commit moves that final device setup to
the add_procs function. This included allocating free lists, and starting the
async event thread.
At this time this change is safe since we have a barrier after add_procs in
MPI_Init. If this changes we will need to re-think some of the initialization
since we might have the possibility of a connection request before add_procs
is called.
Tested with Mellanox ConnectX2 and QLogic HCAs.
Commit also cleans up tabs in btl_openib_async.c.
cmr=v1.7.5:reviewer=miked
This commit was SVN r30122.
1. Fix ompi_info memory leak in usnic BTL: do not allocate memory in
the component register function, because ompi_info only calls the
component register function and then dlclose's the component -- it
does not call component finalize. Instead, defer parsing the MCA
param (and alloc'ing memory) until the component init function so
that any allocated memory can be freed in the component close
function.
1. Also add a new check to ensure that we actually have some part
numbers to check. Add a show_help message if we don't find any
vendor part IDs to check.
1. Add a verbose output if usnic disqualifies itself from selection
because THREAD_MULTIPLE was specified.
cmr=v1.7.5:reviewer=dgoodell
This commit was SVN r30073.
- Modifications to coll/hcoll component related to the changes in the libhcoll API.
Now, hcoll_destroy_context accepts one more parameter that indicates if the context was
really destroyed as a result of the call.
This new "non-blocking" context destruction fixes hang discovered in IMB with mcast enabled.
- Clean up all the left contexts (if any) on the comm_world destruction.
fixed by Val, reviewed by miked
cmr=v1.7.4:reviewer=ompi-rm1.7
This commit was SVN r30055.
