Signed-off-by: onever23 <oleg.neverovitch@hpe.com>#18167
Conversation
Modified dc_array_io() to handle large iod vectors by splitting them into batches
|
Errors are component not formatted correctly,Ticket number prefix incorrect,Ticket number suffix is not a number. See https://daosio.atlassian.net/wiki/spaces/DC/pages/11133911069/Commit+Comments,Unable to load ticket data |
|
Test stage Build on EL 9 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/1/execution/node/272/log |
|
Test stage Build on EL 8 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/1/execution/node/280/log |
|
Test stage Build on Leap 15 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/1/execution/node/327/log |
|
Test stage NLT completed with status UNSTABLE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos//view/change-requests/job/PR-18167/2/testReport/ |
Signed-off-by: onever23 <oleg.neverovitch@hpe.com>
Signed-off-by: onever23 <oleg.neverovitch@hpe.com>
|
Test stage Build on EL 9 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/3/execution/node/261/log |
|
Test stage Build on EL 8 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/3/execution/node/269/log |
|
Test stage Build on EL 9 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/4/execution/node/262/log |
|
Test stage Build on EL 8 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/4/execution/node/270/log |
|
Test stage Build on Leap 15 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/4/execution/node/345/log |
Typos and syntax fix
|
Test stage NLT completed with status UNSTABLE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos//view/change-requests/job/PR-18167/5/testReport/ |
Added env variable to control splits
|
Test stage Functional on EL 9 completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-18167/6/execution/node/1108/log |
| daos_array_iod_t *iod = &splits.iod_v[sn]; | ||
| d_sg_list_t *sgl = &splits.sgl_v[sn]; | ||
| daos_size_t nr = splits.cnt > 1 ? splits.nrec[sn] : total_nr; | ||
| rc = dc_array_io_int(array, th, iod, sgl, op_type, task, nr); |
There was a problem hiding this comment.
i don't think this will work.
let's say your split into 5. if you call dc_array_io_int (... task) with the same task on all 5, the first 1 that completes will mark the whole operation as complete even if the others are in flight.
the other problem is you are launching all of them at the same time too, which defeats the purpose of the change.
Bug fixes in iod_split
3 participants
Modified dc_array_io() to handle large iod vectors by splitting them into batches
Steps for the author:
After all prior steps are complete: