Merged
Conversation
We may have reserved keys that are pushed by the application. Signed-off-by: Ralph Castain <rhc@pmix.org> (cherry picked from commit daf5369)
Need to propagate recent fix to gds/hash to the shmem2 component. Add an example and CI test to ensure we catch any future breakage. Signed-off-by: Ralph Castain <rhc@pmix.org> (cherry picked from commit 1b2644a)
Cleanup some of the checks on return values Signed-off-by: Ralph Castain <rhc@pmix.org> (cherry picked from commit 9f502cd)
Bunch of minor fixes Signed-off-by: Ralph Castain <rhc@pmix.org> (cherry picked from commit a965d18)
Indicates hit end of unpack buffer Signed-off-by: Ralph Castain <rhc@pmix.org> (cherry picked from commit 7343eda)
Minor cleanups to examples Signed-off-by: Ralph Castain <rhc@pmix.org> (cherry picked from commit 2f4492f)
We know v5.0 doesn't support them Signed-off-by: Ralph Castain <rhc@pmix.org>
If we have an unpack mismatch, then let's output the two types we found to better enable debugging without forcing oodles of verbose output Signed-off-by: Ralph Castain <rhc@pmix.org>
The resolve functions were being constrained to look at only information already known by the client process - they would do a local "get" for the "local peers" key (or equivalent), but that only returns what the client has been told about. If the client is asking about other nodes or other nspaces, then it could be that the client hasn't been informed about those nodes/nspaces. So we have to go up to the server (if we are connected) to see if the server has broader knowledge. However, not all runtimes will inform every server about every job being executed - e.g., if there are no local procs for that job. So if the server sees that the host supports the "query" upcall, then ask the host to respond to the request. If the host doesn't support the upcall, or returns a non-success status (e.g., it might not support that query), then have the server respond to the best of its knowledge. If the client isn't connected, or the server has no knowledge, then the client will respond to the best of its knowledge. Signed-off-by: Ralph Castain <rhc@pmix.org> (cherry picked from commit a2e6cfc)
Signed-off-by: Ralph Castain <rhc@pmix.org>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Correct the fetch procedure
We may have reserved keys that are pushed by the application.
Signed-off-by: Ralph Castain rhc@pmix.org
(cherry picked from commit daf5369)
Propagate shmem fix and add new example plus test
Need to propagate recent fix to gds/hash to the shmem2
component. Add an example and CI test to ensure we catch
any future breakage.
Signed-off-by: Ralph Castain rhc@pmix.org
(cherry picked from commit 1b2644a)
Silence Coverity warning in fd support
Cleanup some of the checks on return values
Signed-off-by: Ralph Castain rhc@pmix.org
(cherry picked from commit 9f502cd)
Silence Coverity warnings
Bunch of minor fixes
Signed-off-by: Ralph Castain rhc@pmix.org
(cherry picked from commit a965d18)
Silence unnecessary error log
Indicates hit end of unpack buffer
Signed-off-by: Ralph Castain rhc@pmix.org
(cherry picked from commit 7343eda)
Silence some more Coverity nits
Minor cleanups to examples
Signed-off-by: Ralph Castain rhc@pmix.org
(cherry picked from commit 2f4492f)
Cleanup the "resolve" functions
The resolve functions were being constrained to look at only
information already known by the client process - they would
do a local "get" for the "local peers" key (or equivalent),
but that only returns what the client has been told about.
If the client is asking about other nodes or other nspaces,
then it could be that the client hasn't been informed about
those nodes/nspaces. So we have to go up to the server (if
we are connected) to see if the server has broader knowledge.
However, not all runtimes will inform every server about
every job being executed - e.g., if there are no local procs
for that job. So if the server sees that the host supports
the "query" upcall, then ask the host to respond to the
request. If the host doesn't support the upcall, or returns
a non-success status (e.g., it might not support that query),
then have the server respond to the best of its knowledge.
If the client isn't connected, or the server has no knowledge,
then the client will respond to the best of its knowledge.
Signed-off-by: Ralph Castain rhc@pmix.org
(cherry picked from commit a2e6cfc)
Add resolve doc and minor cleanup
Signed-off-by: Ralph Castain rhc@pmix.org
bot:notacherrypick