Orivej Desh
78b7e342bd
Fix oob_tcp tcp_component_close segfault with active listeners
...
oob_tcp in non-HNP mode shares libevent event_base with oob_base [1].
orte_oob_base_close calls:
(1) oob_tcp component_shutdown, then
(2) opal_progress_thread_finalize, then
(3) oob_tcp tcp_component_close [2].
opal_progress_thread_finalize calls tracker_destructor [3] that frees the
event_base [4]. If any oob_tcp event listeners are active at this time, oob_tcp
will crash trying to delete them at [5] [6].
This change moves oob_tcp event listener cleanup from component_close to
component_shutdown so that it happens before the event_base is freed.
[1] https://github.com/open-mpi/ompi/blob/v4.0.1/orte/mca/oob/tcp/oob_tcp_listener.c#L160
[2] https://github.com/open-mpi/ompi/blob/v4.0.1/orte/mca/oob/base/oob_base_frame.c#L95
[3] https://github.com/open-mpi/ompi/blob/v4.0.1/opal/runtime/opal_progress_threads.c#L232
[4] https://github.com/open-mpi/ompi/blob/v4.0.1/opal/runtime/opal_progress_threads.c#L65
[5] https://github.com/open-mpi/ompi/blob/v4.0.1/orte/mca/oob/tcp/oob_tcp_component.c#L192
[6] https://github.com/open-mpi/ompi/blob/v4.0.1/orte/mca/oob/tcp/oob_tcp_listener.c#L955
Signed-off-by: Orivej Desh <orivej@gmx.fr>
2019-07-04 20:45:47 +00:00
Nysal Jan K A
5d51b2310d
Merge pull request #6791 from nysal/ucx_maxtag_fix
...
pml/ucx: Fix the max tag and context id values
2019-07-03 19:04:39 +05:30
Nysal Jan K.A
fe4ef147f8
pml/ucx: Fix the max tag and context id values
...
Signed-off-by: Nysal Jan K.A <jnysal@in.ibm.com>
2019-07-03 14:33:01 +05:30
Gilles Gouaillardet
07830d05a7
fortran/mpif-h: correctly handle array_of_errcodes in ompi_comm_spawn[_multiple]_f
...
Since array_of_errcodes is only allocated when MPI_ERRCODES_IGNORE is not used,
it should not be cleaned when MPI_ERRCODES_IGNORE is used.
Correctly allocate array_of_errcodes with the right size (e.g. maxprocs).
Thanks Gyevi-Nagy Laszlo for reporting this issue.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2019-07-03 09:53:46 +09:00
Geoff Paulsen
f1b2a09675
Merge pull request #6649 from devreal/rdma-fetchop-local
...
OSC rdma: make sure accumulating in shared memory is safe
2019-06-28 14:46:21 -05:00
bosilca
7005e46eed
Merge pull request #6782 from ggouaillardet/topic/neighbor_alltoallv
...
mpi/c: fix param checks in [I]Neighbor_alltoall{v,w}
2019-06-28 11:02:18 -04:00
Gilles Gouaillardet
5655d64bd3
mpi/c: fix param checks in [I]Neighbor_alltoall{v,w}
...
do not check some input parameters when an {in,out}degree is zero
Thanks Junchao Zhang for analyzing and reporting this issue.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2019-06-28 13:39:28 +09:00
Orivej Desh
de522545c0
Fix ORTE_FORCED_TERMINATE message
...
The format string expects to see the file and line before the error text and code.
Signed-off-by: Orivej Desh <orivej@gmx.fr>
2019-06-28 00:46:55 +00:00
Gilles Gouaillardet
5cb393292e
Merge pull request #6780 from ggouaillardet/topic/pmix_refresh
...
pmix/pmix4x: refresh to the latest PMIx master
2019-06-27 03:43:44 +02:00
Gilles Gouaillardet
63aa156bb0
pmix/pmix4x: refresh to the latest PMIx master
...
refresh to pmix/pmix@99971222ce
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2019-06-27 09:35:49 +09:00
Gilles Gouaillardet
f1ae036466
Merge pull request #6771 from ggouaillardet/topic/pmix_refresh
...
pmix/pmix4x: refresh to the latest PMIx master
2019-06-24 04:12:24 +02:00
Gilles Gouaillardet
5679a88867
pmix/pmix4x: refresh to the latest PMIx master
...
refresh to pmix/pmix@f67efc835c
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2019-06-24 10:17:23 +09:00
Artem Polyakov
6433da7ba8
Merge pull request #6754 from artpol84/osc/fixes_all
...
OSC/UCX: Fix early cleanup + race conditions
2019-06-21 18:45:37 -07:00
Artem Polyakov
6678ac0f55
osc/ucx: Fix possible win creation/destruction race condition
...
To avoid fully initializing the osc/ucx component for MPI application
that are not using One-Sided functionality, the initialization happens
at the first MPI window creation.
This commit ensures atomicity of global state modifications.
Signed-off-by: Artem Polyakov <artpol84@gmail.com>
2019-06-20 09:05:03 -07:00
Artem Polyakov
0857742624
osc/ucx: Fix worker pool finalization
...
Signed-off-by: Artem Polyakov <artpol84@gmail.com>
2019-06-20 09:05:03 -07:00
Ralph Castain
fd27d76130
Merge pull request #6764 from rhc54/topic/flux
...
Fix finalize of flux component
2019-06-19 05:58:48 -07:00
Ralph Castain
d4070d5f58
Fix finalize of flux component
...
Per patches from @SteVwonder and @garlick
Signed-off-by: Ralph Castain <rhc@pmix.org>
2019-06-18 21:14:04 -07:00
Nathan Hjelm
560886f095
Merge pull request #6746 from devreal/osc_winalloc_err
...
OSC rdma win allocate: propagate errors to avoid deadlocks
2019-06-18 17:57:53 -07:00
Edgar Gabriel
d621e3f0bb
Merge pull request #6759 from haraldkl/ompi_aggregators-cost_calc_intfix
...
Suggestion to fix division by zero in file view.
2019-06-18 09:55:11 -05:00
Geoffrey Paulsen
54a286ee9d
Revert "ompi_info: report MPI1 compat is disabled"
...
This reverts commit 61ccc65302
2019-06-14 13:22:21 -05:00
Geoffrey Paulsen
76dfaceedc
Revert "configury: Remove --enable-mpi1-compatibility"
...
This reverts commit 6f904aff2c
2019-06-14 13:22:21 -05:00
Geoffrey Paulsen
ca4b70913e
Revert "man: remove man pages of removed MPI1 subroutines"
...
This reverts commit 26c1b833c7
2019-06-14 13:22:21 -05:00
Geoffrey Paulsen
ed9a670074
Revert "mpi.h.in: delete removed MPI1 functions/datatypes (API change!)"
...
This reverts commit a6d6be2853
2019-06-14 13:22:21 -05:00
Geoffrey Paulsen
5cc0141675
Revert "MPI_Type_get_envelope: remove MPI-1 deleted names"
...
This reverts commit 65eb118e08
2019-06-14 13:22:21 -05:00
Geoffrey Paulsen
e036941ab5
Revert "mpi.h: remove MPI_UB/MPI_LB when not enabling MPI-1 compat"
...
This reverts commit 7223334d4d
2019-06-14 13:22:21 -05:00
Geoffrey Paulsen
6de263fc29
Revert "mpi: make C++ bindings compile when MPI-1 compat is disabled"
...
This reverts commit b323655809
2019-06-14 13:22:21 -05:00
Harald Klimach
e222a04ae5
Suggestion to fix division by zero in file view.
...
In common_ompi_aggregators calc_cost routine:
do not cast the real division to an int intermediately.
This patch removes the obsolete int variable c and assigns
the result of the P_a/P_x division directly to n_as.
With the intermediate int c variable, n_as gets 0 if P_a < P_x,
resulting in a division by 0 when computing n_s.
Signed-off-by: Harald Klimach <harald.klimach@uni-siegen.de>
2019-06-13 18:47:32 +02:00
Jeff Squyres
7c3aeb3061
Merge pull request #6686 from alex-anenkov/coll-iallreduce-recursivedoubling
...
coll/libnbc: add recursive doubling algorithm for MPI_Iallreduce
2019-06-10 10:09:51 -04:00
Gilles Gouaillardet
80e0ac7379
Merge pull request #6751 from ggouaillardet/topic/pmix_refresh
...
pmix/pmix4x: refresh to the latest PMIx master
2019-06-10 16:47:21 +09:00
Gilles Gouaillardet
d9326ff2ca
pmix/pmix4x: refresh to the latest PMIx master
...
refresh to pmix/pmix@186dca196c
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2019-06-10 15:17:43 +09:00
Yossi Itigin
a46e5da3ca
Merge pull request #6744 from brminich/topic/all2all_linear_sync_fix
...
COLL/BASE: Fix linear sync all2all
2019-06-09 21:23:38 +03:00
Joseph Schuchart
8f27cc26d9
OSC rdma win allocate: synchronize error codes across shared memory group
...
Signed-off-by: Joseph Schuchart <schuchart@hlrs.de>
2019-06-07 11:03:21 +02:00
KAWASHIMA Takahiro
85c3311b7d
Merge pull request #6726 from yanagibashi/pr/add-f08-procedure-names
...
mpiext/pcollreq: Add `_f08` to procedure names
2019-06-07 09:10:58 +09:00
Mikhail Brinskii
79006f4e5a
COLL/BASE: Fix linear sync all2all
...
Signed-off-by: Mikhail Brinskii <mikhailb@mellanox.com>
2019-06-06 19:22:42 +03:00
Yossi Itigin
8535dd570b
Merge pull request #6732 from dmitrygladkov/topic/pml/ucx_init
...
PML/UCX: Don't destroy UCP worker if it wasn't created
2019-06-06 10:41:33 +03:00
Geoff Paulsen
4f5e96d1f0
Merge pull request #6724 from sam6258/shmem_wait_types
...
shmem/c: Fix shmem type for calls to shmem_test and shmem_wait_until …
2019-06-05 12:37:34 -05:00
KAWASHIMA Takahiro
2b856573b2
Merge pull request #6699 from t-kurita/pr/java-alltoallw-arrays
...
java: Fix compilation error in allToAllw using Java arrays
2019-06-04 11:33:17 +09:00
Dmitry Gladkov
c864ca51d2
PML/UCX: Don't destroy UCP worker if it wasn't created
...
Signed-off-by: Dmitry Gladkov <dmitrygla@mellanox.com>
2019-06-03 10:49:36 +03:00
Tsubasa Yanagibashi
3148b0cfaa
mpiext/pcollreq: Add _f08 to procedure names
...
The procedure names don't contain "_f08" of Fortran 2008 bindings of
Persistent Collective Operations(mpiext/pcollreq/use-mpi-f08).
This fix adds "_f08" to the procedure names of pcollreq/use-mpi-f08,
same as other Fortran 2008 routines in `ompi/mpi/fortran/use-mpi-f08/mod`.
Signed-off-by: Tsubasa Yanagibashi <fj2505dt@aa.jp.fujitsu.com>
2019-05-31 15:22:42 +09:00
bosilca
e3eb6b5897
Merge pull request #6683 from bosilca/topic/spc_init
...
Fixed SPC/MPI_T initialization error.
2019-05-30 17:50:58 -04:00
Scott Miller
ca59cabc67
shmem/c: Fix shmem type for calls to shmem_test and shmem_wait_until with [u]int32_t and [u]int64_t
...
Signed-off-by: Scott Miller <scott.miller1@ibm.com>
2019-05-30 17:20:30 -04:00
markalle
008ab98946
Merge pull request #6531 from markalle/patcher_additions
...
shmat/shmdt additions for patcher
2019-05-30 12:16:05 -05:00
Nathan Hjelm
8961daae4a
opal/atomic: work around memory barrier bug in older gcc
...
This commit fixes an issue seem with some older versions of gcc
(verified to occur in gcc 6.x) where on x86_64 systems the
acquire memory barrier in C11 atomics acts as a no-op. On these
systems the three memory barriers should all be equivalent.
This is related to the error fixed in open-mpi/ompi@30119ee .
References #6655 .
Signed-off-by: Nathan Hjelm <hjelmn@google.com>
2019-05-30 06:58:28 -07:00
Yossi Itigin
5f37fe654a
Merge pull request #6694 from hoopoepg/topic/pci-flush-on-quiet
...
SPML/UCX: added synchronized flush on quiet
2019-05-29 11:19:59 +03:00
George Bosilca
6c75334162
Use the correct counter name in the example.
...
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
2019-05-29 00:54:56 -04:00
George Bosilca
a0fce4eac2
Fix the man pages for some of the MPI_T_* functions.
...
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
2019-05-29 00:23:35 -04:00
George Bosilca
eed770ce5c
Fix the SPC initialization.
...
Use the PVAR ctx to save the SPC index, so that no lookup nor
restriction on the SPC vars position is imposed.
Make sure the PVAR are always registered.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
2019-05-29 00:23:18 -04:00
George Bosilca
7dab8c002b
Fixed SPC/MPI_T initialization error.
...
Signed-off-by: Yong Qin <yongq@mellanox.com>
2019-05-28 15:10:32 -04:00
valentin petrov
2469f6ca13
Merge pull request #6708 from vspetrov/master
...
Coll/hcoll: adding scatterv interface
2019-05-27 20:22:00 +03:00
Sergey Oblomov
0b108411f8
SPML/UCX: added synchronized flush on quiet
...
- added synchronized flush operation on quiet call.
- flush is implemented using get operation
Signed-off-by: Sergey Oblomov <sergeyo@mellanox.com>
2019-05-27 16:07:04 +03:00
