Topic/v2.x/ompi win peer invalid #3341
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
pmix/pmix112: Update to version 1.2.0
Revert the logic of io_ompio_sharedfp_lazy_open. The user now has to explicitely disable shared fp in order for the structures not to be allocated. Otherwise, resetting the shared fp e.g. in case the file was opened in append mode will not work correctly, the code could deadlock. Signed-off-by: Edgar Gabriel <[email protected]>
ess/hnp: add support for forwarding additional signals (open-mpi#2712)
Thanks Nicolas Joly for the report Fixes open-mpi#2782 Signed-off-by: Gilles Gouaillardet <[email protected]> (cherry picked from commit open-mpi/ompi@6f2ca58)
Fixes a bug reported on the mailing list. ompio did only reposition the individual file pointer when the file was opened in append mode. Set the shared file pointer also to point to the end of the file, similarly to the individual file pointer. This is the equivalent to commit d3a8d38 on master, cannot be cherry-picked because of differences in the organization of the ompio files ( usage of mca/common/ompio on master). Signed-off-by: Edgar Gabriel <[email protected]>
* The AC_LANG_PROGRAM macro adds the `main()` so it is erroneous to add it to the test program. * This was detected with the XL compilers which will fail to build the program in this situation. The GNU compiler does not error out or warn, but successfully compiles the program. Signed-off-by: Joshua Hursey <[email protected]> (cherry picked from commit 029964a) Signed-off-by: Joshua Hursey <[email protected]>
* Similar to commit open-mpi@029964a This removes an extra `int main` during configure. Signed-off-by: Joshua Hursey <[email protected]> (cherry picked from commit 72ac812) Signed-off-by: Joshua Hursey <[email protected]>
Fix libevent configure.ac
…-fix-v2.x Pr/sharedfp append fix v2.x
Remove ptmalloc2 hooks from v2.1.x
Signed-off-by: Jeff Squyres <[email protected]>
…-btl-comment openib: remove stale comment
Signed-off-by: Joshua Hursey <[email protected]>
MPI_T_pvar_get_index was returning an incorrect index. The index was never set correctly while registering the performance variables. Additionally fix a missing case in the mca_base_var_type_t to MPI datatype conversion. This type is currently used for control variables registered by mxm, fca and hcoll components. Signed-off-by: Nysal Jan K.A <[email protected]> (cherry picked from commit 25ba507) Signed-off-by: Joshua Hursey <[email protected]>
* MCA option to control how long we wait for stack traces:
- orte_timeout_for_stack_trace INTEGER
Default: 30
Setting to <= 0 will cause it to wait forever
* Useful when gathering stack traces from large jobs which might take
a long time.
Signed-off-by: Joshua Hursey <[email protected]>
Since the oob and connections systems do not work the same way they did in older versions of Open MPI these operations are no longer necessary. At best they do nothing and at worst they hurt performance by making us enter the event library more often in opal_progress(). Fixes open-mpi#2839 Signed-off-by: Nathan Hjelm <[email protected]> (cherry picked from commit 9f28c0a) Signed-off-by: Nathan Hjelm <[email protected]>
- This prevents us for accidentally masking a signal that was meant to terminate the application. Signed-off-by: Joshua Hursey <[email protected]> (cherry picked from commit f8918e3) Signed-off-by: Joshua Hursey <[email protected]>
- New MCA option: opal_stacktrace_output
- Specifies where the stack trace output stream goes.
- Accepts: none, stdout, stderr, file[:filename]
- Default filename 'stacktrace'
- Filename will be `stacktrace.PID`, or if VPID is available,
then the filename will be `stacktrace.VPID.PID`
- Update util/stacktrace to allow for different output avenues
including files. Previously this was hardcoded to 'stderr'.
- Since opal_backtrace_print needs to be signal safe, passing it a
FILE object that actually represents a file stream is difficult. This
is because we cannot open the file in the signal handler using
`fopen` (not safe), but have to use `open` (safe). Additionally, we
cannot use `fdopen` to convert the `int fd` to a `FILE *fh` since it
is also not signal safe.
- I did not want to break the backtrace.h API so I introduced a new
rule (documented in `backtrace.c`) that if the `FILE *file`
argument is `NULL` then look for the `opal_stacktrace_output_fileno`
variable to tell you which file descriptor to use for output.
Signed-off-by: Joshua Hursey <[email protected]>
(cherry picked from commit 6d98559)
Signed-off-by: Joshua Hursey <[email protected]>
verbs: remove extra event user increment/decrement operation
when a file is opened a second time for shared file pointer operations, avoid setting the create and exclusive flag. Signed-off-by: Edgar Gabriel <[email protected]>
it looks like disabling the lazy_open flag for sharedfp components revealead a bug that lead to a crash in file_close in some tests. Make sure the SHAREDFP_IS_SET flag is correctly set (and not overwritten again), and we use that to avoid a double-free of the communicator. This is equivalent to commit commit f5289a1 on master. The commit cannot be cherry-picked due to the differences in the organization of the ompio files (ompio uses now common/ompio for many files). Signed-off-by: Edgar Gabriel <[email protected]>
Signed-off-by: Ralph Castain <[email protected]> (cherry picked from commit c803af5)
The hook is called from memheap when memory range is going to be allocated by smalloc(), realloc() and others. ucx spml uses this hook to call ucp_mem_advise in order to speedup non blocking memory mapping. Signed-off-by: Alex Mikheev <[email protected]> (cherry picked from commit 986ca00)
Signed-off-by: Alex Mikheev <[email protected]> (cherry picked from commit 9da9e62)
Signed-off-by: Alex Mikheev <[email protected]>
Signed-off-by: Yossi Itigin <[email protected]>
this fixes a regression introduced in open-mpi/ompi@045d0c5 Fixes open-mpi#2879 Signed-off-by: Gilles Gouaillardet <[email protected]> (back-ported from commit open-mpi/ompi@9bcadbd)
…hed_chain coll/libnbc: fix the red_schain algo of ireduce with MPI_IN_PLACE
Signed-off-by: Gilles Gouaillardet <[email protected]> (cherry picked from commit open-mpi/ompi@ef10d3f)
PMIx_server_register_nspace() is an asynchronous operation, so the pmix glue wait for it completes before returning. Signed-off-by: Gilles Gouaillardet <[email protected]> (back-ported from commit open-mpi/ompi@dccb189)
….1.0rc4 VERSION: Bump to 2.1.0rc4
Signed-off-by: Jeff Squyres <[email protected]>
VERSION: Bump to v2.1.1a1
A 1 was missing when setting $1_LDFLAGS leading to erroneous items in the wrapper cflags. Signed-off-by: Nathan Hjelm <[email protected]> (cherry picked from commit open-mpi/ompi@8c20869)
You cannot include both pmi.h and pmi2.h as they have conflicting defines in them.
…ning-fixes v2.1.1: compiler warning fixes
v2.x: Fix plm/rsh runtime check
…usage_bugfix Fixes the coll_allgather usage bug
Thanks to Kevin Buckley for noticing the issue and supplying the patch. [skip ci] bot:notest Signed-off-by: Jeff Squyres <[email protected]> (cherry picked from commit 9e23c5e)
v2.1.x: openmpi.spec: also put the modulefile in /opt if install_in_opt==1
v2.x: config: fix typo in mxm configury
Add missing MPI_T_PVAR_SESSION_NULL to mpi.h v2.x
…n-page-fix MPI_Wtick: may return a higher resolution than 10e-6 these days
…_align osc/sm: fix MPI_Win_allocate_shared() alignment
…init_fix_v2.x Topic/pml ucx req init fix v2.x
orte-clean: fix bad username/uid usage, add orte-dvm
…configury-double-sized-complex v2.1.1: configury: consider double _Complex as a candidate for the C equivale…
osc/rdma: make locking code more robust
…ternal v2.1.x: --with-hwloc=external fix
…ILTIN_GCC v2.x: configury: fix asm atomic detection
Look for amd64 in addition to x86_64 as the platform type for x86_64 assembly. The FreeBSD-packaged Autoconf package has a patch to return amd64-unknown-freebsd11.0 instead of the x86_64-unknown-freebsd11.0 that a stock Autoconf package would return. Since we want to run Jenkins builds on FreeBSD, working around the FreeBSD patch is probably the easiest thing. Signed-off-by: Brian Barrett <[email protected]>
Fixes open-mpi#3267 Signed-off-by: Nathan Hjelm <[email protected]> (cherry picked from commit fad0803) Signed-off-by: Nathan Hjelm <[email protected]>
osc/rdma: fix typo in atomic code
build: Fix platform detection on FreeBSD
resolves open-mpi#3326 (open-mpi#3326) Signed-off-by: [email protected] (cherry picked from commit open-mpi/ompi@b3a2010)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
26 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.