v3.x: osc/rdma: use extent of the appropriate datatype in ompi_osc_rdma_rge… #3598
Conversation
…t_accumulate_internal() origin_datatype and target_datatype might be different and hence have different extent, so use either origin_extent or target_extent when appropriate. Refs open-mpi#3569 Signed-off-by: Gilles Gouaillardet <[email protected]> (cherry picked from commit 0f79259)
ggouaillardet
changed the title
|
@bwbarrett @edgargabriel CI failed at UH |
|
bot:ompi:retest |
|
@ggouaillardet, i don't understand your statement. This PR is against 3.x (well, now 3.0.x), where the libtool fixes haven't been merged. So whatever the failure was, it shouldn't have been because of the libtool versions. |
|
bot:ompi:retest |
1 similar comment
|
bot:ompi:retest |
|
@bwbarret when i read the logs, the error was caused by |
|
bot:ompi:retest |
1 similar comment
|
bot:ompi:retest |
|
@ggouaillardet, I think you're right; the Jenkins builders were not building exactly what we expected. I've reconfigured the sub-builders from the open-mpi.pull_request job to be more explicit in which sha1 to test. I've also added an output line in the builder script to print the sha1 of the commit being built to try and avoid confusion in the future. For example: Anyway, builds chugging along on this PR, hopefully this mess is behind us now... |
|
bot:ompi:retest Accidentally killed a very slow cray builder Note that the open-mpi.* tests were a configuration error and will never mark as complete. As soon as the Pull Request Build Checker goes green, we should be good to merge. |
a buffer defined by (buf, count, dt) will have data starting at buf+offset and ending len bytes later with len = opal_datatype_span(&dt.super, count, &offset); Signed-off-by: Gilles Gouaillardet <[email protected]> (cherry picked from commit open-mpi/ompi@e622ca8)
|
@hjelmn please review |
…t_accumulate_internal()
origin_datatype and target_datatype might be different and hence have different extent,
so use either origin_extent or target_extent when appropriate.
Refs #3569
Signed-off-by: Gilles Gouaillardet [email protected]
(cherry picked from commit 0f79259)