1
1
Граф коммитов

26368 Коммитов

Автор SHA1 Сообщение Дата
Jeff Squyres
b02d8c48f5 usnic: make the releasing safer
Since the usnic BTL is single-threaded in this area, there really is
no danger, but don't use one of the pointers hanging off the frag
after we return it to the freelist.  Instead, save the endpoint
pointer before returning the frag.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-01-10 12:03:53 -08:00
Jeff Squyres
e25b860627 usnic: clarify types
The types are technically typedef equivalent, but it's less confusing
to use the types that agree with the name of the constructor.

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-01-10 12:03:53 -08:00
Jeff Squyres
40fe575132 usnic: trivial updates (no code/logic changes)
- Add more explanatory comments
- Trivial whitespace / style updates
- Rename opal_btl_usnic_force_retrans() -> opal_btl_usnic_fast_retrans()

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-01-10 10:40:02 -08:00
Nathan Hjelm
3593fad4d2 Merge pull request #2679 from hjelmn/cpuid_fix
master: amd64: save/restore all 64 bits of rbx around cpuid
2017-01-10 09:12:28 -07:00
Gilles Gouaillardet
6d59b476de Merge pull request #2686 from ggouaillardet/topic/pmix2x_ptl_base_sendrecv
pmix2x: ptl/base: send header and message data together via writev()
2017-01-10 16:26:10 +09:00
Gilles Gouaillardet
44c1ff60f1 Merge pull request #2672 from ggouaillardet/topic/misc_memory_leaks
Plug misc memory leaks
2017-01-10 13:16:04 +09:00
Gilles Gouaillardet
a01960bee5 pmix2x: ptl/base: send header and message data together via writev()
on Linux, sending the header and then the message data does severely
impact performances of ptl/tcp :
on the receiver, reading the data can often result in an PMIX_ERR_RESOURCE_BUSY
or PMIX_ERR_WOULD_BLOCK, which ends up degrading performances)
this commit send both header and message data at the same time via writev()
and makes ptl/tcp virtually as efficient as ptl/usock.

Short writev generally occur when the kernel buffer is full, so there is no
point for retrying in this case.

fwiw, no such degradation was observed on OSX.

Refs open-mpi/ompi#2657

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-10 13:07:39 +09:00
Nathan Hjelm
b320882932 Merge pull request #2688 from hjelmn/cvar_fix
mca/base: account for NULL string_value in verbose set
2017-01-09 14:27:31 -07:00
Nathan Hjelm
d6bd69dc93 mca/base: account for NULL string_value in verbose set
The MCA variable code calls the string from value function with a NULL
string to verify values. The verbosity enumerator was not correctly
checking for a non-NULL value before trying to set the string.

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-01-09 11:52:31 -07:00
Joshua Ladd
3e23380bba Merge pull request #2675 from artpol84/orte/state/exit_1_fix
orte/odls: Fix ORTE state machine for the non-zero exit case
2017-01-09 12:32:37 -05:00
Ralph Castain
67fce2861b Merge pull request #2685 from rhc54/topic/cov
Resolve Coverity issues
2017-01-07 13:11:40 -08:00
Ralph Castain
84ce7eed2a Merge pull request #2683 from rhc54/topic/nits
Cleanup some configure stuff for static builds
2017-01-07 11:33:20 -08:00
Ralph Castain
e25e69dc2f Resolve Coverity issues
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-01-07 10:45:52 -08:00
Ralph Castain
822e2680ba Cleanup some configure stuff for static builds - still can't get wrapper extra libs to be recognized
Signed-off-by: Ralph Castain <rhc@open-mpi.org>

pmix2x: minor configure updates

Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
2017-01-07 08:37:36 -08:00
Nathan Hjelm
5b70ae3ec0 amd64: save/restore all 64 bits of rbx around cpuid
This commit fixes a bug in the timer check. When -fPIC is used we need
to save/restore ebx. The code copied from patcher was meant for 32-bit
systems and did not work correctly on 64-bit systems. This commit
updates the save/restore to use rbx instead of ebx.

Fixes #2678

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
2017-01-06 18:54:20 -07:00
Joshua Ladd
dc7d2f5b6a Merge pull request #2571 from alex-mikheev/topic/sshmem_prio_fix
oshmem: sshmem: make mmap allocator a default instead of verbs
2017-01-06 17:39:37 -05:00
Joshua Ladd
7fc9f9bbac Merge pull request #2620 from karasevb/fix_rmaps_mindist
rmaps/mindist: fix pmix errors
2017-01-06 17:26:48 -05:00
Ralph Castain
ca16f3f9ed Merge pull request #2676 from rhc54/topic/alps
Minor cleanups to eliminate warnings
2017-01-06 12:43:42 -08:00
Ralph Castain
684e69695f Minor cleanups to eliminate warnings
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-01-06 08:44:10 -08:00
Ralph Castain
39d880f65d Merge pull request #2673 from rhc54/topic/usock
Raise the priority of the usock component so it gets preferentially picked
2017-01-06 08:24:35 -08:00
Artem Polyakov
3eb6c98542 orte/odls: Fix ORTE state machine for the non-zero exit case
This commit fixes rare race condition that occurs when the process
that is calling `exit(-1)` has delay between fd cleanup and actual
OS-level exit. This may happen if the process has some work to do
`on_exit()`.

**Problem description**:
Consider an application process that has called `exit(nonzero)`, it's
fd's was closed
but it's actual termination at OS level is delayed by some cleanups (eg.
in callbacks registered via `on_exit()`).
Observed sequence of events was the following:

* orted gets stdio disconnection and activating `IOF COMPLETE` state.
* parallel OOB disconnection causes `COMMUNICATION FAILURE` state to be
activated.
* during `COMMUNICATION FAILURE` processing `odls_base_default_wait_local_proc`
is called even though real waitpid wasn't yet called (code mentions that
waitpid might not be called for unspecified reason). Because of that real exit
code is unknown and set to 0. `odls_base_default_wait_local_proc` callback sees
`IOF COMPLETE` flag and in conjunction with 0-exit-code it activates
`WAITPID FIRED` state.
* processing of `WAITPID FIRED` leads to `NORMALLY TERMINATED` to be
activated.
* `NORMALLY TERMINATED` state in particular leads `ORTE_PROC_FLAG_ALIVE` flag
for this proc to be dropped.
* when application process finally exits and `wait_signal_callback` is
launched. It sets real exit code and calls `odls_base_default_wait_local_proc`
again but at this time since the process has `ORTE_PROC_FLAG_ALIVE` flag
dropped `WAITPID FIRED` state is activated (instead of `EXITED WITH NON-ZERO`)
leading to a hang that was observed.

Signed-off-by: Artem Polyakov <artpol84@gmail.com>
2017-01-06 11:12:55 +02:00
Gilles Gouaillardet
189d7b9480 opal/dss: revamp opal_value_unload() to keep valgrind happy
reorder tests to avoid valgrind complaining about uninitialized variables

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 17:10:39 +09:00
Ralph Castain
444f5fa35d Raise the priority of the usock component so it gets preferentially picked
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2017-01-05 22:53:04 -08:00
Gilles Gouaillardet
a1a0e324b3 util/hostfile: plug a memory leak
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:45 +09:00
Gilles Gouaillardet
6b9343a966 plm/rsh: plug a memory leak
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:45 +09:00
Gilles Gouaillardet
8ba92d7516 iof/base: plug a memory leak in orte_iof_base_close()
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:45 +09:00
Gilles Gouaillardet
e396b17a7f orte/orted: plug a memory leak
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:45 +09:00
Gilles Gouaillardet
6b90b03c28 orted/pmix: plug a memoy leak in pmix_server_fencenb_fn()
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:45 +09:00
Gilles Gouaillardet
7fe6840232 state/hnp: plug a memory leak
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:45 +09:00
Gilles Gouaillardet
4d58b8dcae ess/pmi: plug a memory leak
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:45 +09:00
Gilles Gouaillardet
c0c5dd8ccc orte: plug a memory leak in orte_rml.recv_cancel
do not invoke orte_rml.recv_cancel after the orte progress thread has gone

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:44 +09:00
Gilles Gouaillardet
17fac4bfd1 grpcomm/base: get rid of the seq_num field of the orte_grpcomm_signature_t struct
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:44 +09:00
Gilles Gouaillardet
fe25f50871 grpcomm/base: plug a memory leak on finalize
manually allocate sequence numbers to be stored into the
orte_grpcomm_base.sig_table hash table, and manually release
them on orte_grpcomm_base_close()

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:44 +09:00
Gilles Gouaillardet
2189c5bcc3 ompi/dpm: plug a memory leak in disconnect_waitall()
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:44 +09:00
Gilles Gouaillardet
a988ad24eb orte/runtime: plug a leak in orte_finalize()
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 15:38:44 +09:00
Gilles Gouaillardet
c2ddb1e2fc mca/base: plug a memory leak
register mca_base_var_enum_value_flag_t so they can be free'd
upon finalize

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:36 +09:00
Gilles Gouaillardet
cf534d0c95 ompi/proc: plug a memory leak in ompi_proc_finalize()
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:35 +09:00
Gilles Gouaillardet
6d5cb9fe0d event: plug a leak when closing the event framework
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:35 +09:00
Gilles Gouaillardet
b3a2bdda7b opal/threads: manually invoke thread-specific key destructors on the main thread.
there is no such thing as pthread_join(main_thread), so key destructors
are never invoked on the main thread, which causes valgrind report
some memory leaks. Manually store and then invoke the key destructors and
make valgrind happy.

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:35 +09:00
Gilles Gouaillardet
6ef281e163 pmix/base: fix misc memory leaks
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:35 +09:00
Gilles Gouaillardet
0ee5d56ab1 grpcomm/direct: plug a memory leak in barrier_release()
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:35 +09:00
Gilles Gouaillardet
a59dfd7b14 sec/munge: plug a memory leak
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:35 +09:00
Gilles Gouaillardet
f2d6584189 grpcomm/base: plug misc memory leaks
- add a destructor to orte_grpcomm_caddy_t in order to plug a memory leak

- plug a memory leak in barrier_release()

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 13:46:21 +09:00
Gilles Gouaillardet
c4a47ae9a9 orte/orted: plug misc memory leaks
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 11:35:59 +09:00
Gilles Gouaillardet
88535b6200 orte/util: revamp orte_attr_unload() to keep valgrind happy
reorder tests to avoid valgrind complaining about uninitialized variables

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 11:35:59 +09:00
Gilles Gouaillardet
c612499bc1 opal: mca/base: fix a memory leak in the mca_base_var_enum_flag_t destructor
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 11:35:59 +09:00
Gilles Gouaillardet
58f2a764f9 ess/hnp: plug memory leaks
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 11:35:59 +09:00
Gilles Gouaillardet
24c61b0625 oob/tcp: plug a memory leak in mca_oob_tcp_component_lost_connection()
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 11:35:59 +09:00
Gilles Gouaillardet
7e5da7382e btl/tcp: plug leaks when closing component
remove tcp_local from the tcp_procs table, and release it

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 11:35:59 +09:00
Gilles Gouaillardet
c7d9e62d47 rml/base: plug a memory leak
add a destructor to orte_rml_send_request_t in order
to plug a memory leak

Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
2017-01-06 11:35:59 +09:00