issues
search
ulfm-devel
/
ompi
Open MPI main development repository
https://www.open-mpi.org
Other
0
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Round of review
#56
abouteiller
opened
3 years ago
0
MPI_Finalize hangs when using -mca mpi_ft_detector_thread true
#55
abouteiller
closed
3 years ago
3
Thread-safe Agree
#54
abouteiller
closed
4 years ago
1
Optimize number of atomics in error cases during SYNC_WAIT rearming
#53
abouteiller
opened
4 years ago
0
OpenIB Finalize assert in reggache
#52
abouteiller
opened
4 years ago
0
Recursive error notification in ERA Agree
#51
abouteiller
closed
4 years ago
2
deadlock in recv/allreduce if participating process fails
#50
abouteiller
opened
5 years ago
4
Bugs in running SC18 tutorial code in the provided docker container
#49
abouteiller
opened
5 years ago
1
parallel I/O fails after a process failure
#48
abouteiller
closed
5 years ago
2
MPI_Abort kills only MPI processes after a fault
#47
abouteiller
opened
5 years ago
3
UCX support
#46
abouteiller
opened
5 years ago
1
Dev documentation on how to adapt components to FT
#45
abouteiller
opened
5 years ago
0
v4.0.x+ulfm: use mpi_ext not working
#44
abouteiller
closed
5 years ago
0
OSX: Send to failed process on BTL TCP may deadlock
#43
abouteiller
closed
4 years ago
1
v4.0.x+ulfm: pmix3x bug with RANGE_CUSTOM
#42
abouteiller
closed
5 years ago
2
MPI_Barrier with MPI_THREAD_MULTIPLE causes assertion in ompi_request_wait_completion
#41
abouteiller
closed
5 years ago
6
irecv rc asserts in get_rprocs
#40
abouteiller
closed
5 years ago
1
Man pages
#39
abouteiller
opened
5 years ago
0
Version based off stable release
#38
abouteiller
closed
4 years ago
3
COMM_SPAWN: spawnees INIT may fail creating proc_t too early
#37
abouteiller
opened
5 years ago
0
MPI_COMM_WORLD error handler invoked during finalize by internal ops
#36
abouteiller
closed
5 years ago
1
Intercomms collective not really fault tolerant
#35
abouteiller
opened
6 years ago
0
Dealing with node-level failures
#34
abouteiller
closed
5 years ago
3
Difficulties with spawning new processes on the victim's node
#33
abouteiller
closed
5 years ago
2
FT RMA
#32
abouteiller
opened
6 years ago
0
Open IB post-fault credit release is slow
#31
abouteiller
closed
3 years ago
1
Coll Base operations return "ERROR_IN_STATUS" again
#30
abouteiller
closed
6 years ago
1
Cori: Fallback to IBoGNI after a fault is reported and crash
#29
abouteiller
closed
5 years ago
1
Titan: ugni init error
#28
abouteiller
closed
6 years ago
1
Only 1 (one) local failure reported by PMIx/orted
#27
abouteiller
closed
6 years ago
1
MPI_Comm_spawn deadlock w/faults
#26
abouteiller
opened
6 years ago
3
nextcid_nb not interrupted by failures/revokes
#25
abouteiller
closed
6 years ago
1
coll_comm request cancellation takes a recursive mutex
#24
abouteiller
closed
6 years ago
2
Allreduce_ft_nb gets revoked
#23
abouteiller
closed
6 years ago
1
OpenIB UD timeout when messaging a failed process
#22
abouteiller
closed
6 years ago
1
TCP BTL triggers false detection
#21
abouteiller
closed
6 years ago
3
rename coll_ftbasic
#20
abouteiller
closed
5 years ago
1
Occasionally assert on req_complete when a failure is reported during wait
#19
abouteiller
closed
6 years ago
1
Shrink on intercoms
#18
abouteiller
opened
6 years ago
0
Agree ERA topology detection
#17
abouteiller
opened
6 years ago
0
Open IB BTL error spam (retry_exceeded_error)
#16
abouteiller
closed
6 years ago
5
FT Topo
#15
abouteiller
opened
6 years ago
0
Verify that topo components work after failure
#14
abouteiller
closed
6 years ago
1
Various error messages in otherwise "normal" error scenarios
#13
abouteiller
closed
6 years ago
1
review comments in #90d010a
#12
abouteiller
closed
5 years ago
1
instantaneous detection of failure from shared-memory siblings
#11
abouteiller
closed
6 years ago
1
auto mca-no-build when configuring --with-ft=mpi
#10
abouteiller
closed
7 years ago
1
FT Files
#9
abouteiller
opened
8 years ago
0
FT NBC
#8
abouteiller
opened
8 years ago
1
Failure Propagation and detection works only on comm_world
#7
abouteiller
opened
8 years ago
1
Next