OMPI: Notified RMA ops Public API#9
Conversation
|
Hello! The Git Commit Checker CI bot found a few problems with this PR: 8d7ea3b: Public APis to makefile
0b6b3c3: Public APIs for put_withnotify. get_with_notify
Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks! |
joe-explr
force-pushed
the
notified-rma-sm
branch
3 times, most recently
from
5a9c010 to
f79b94b
Compare
|
Hello! The Git Commit Checker CI bot found a few problems with this PR: 8ba738e: Edits
Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks! |
joe-explr
force-pushed
the
notified-rma-sm
branch
from
8ba738e to
80fef14
Compare
|
Hello! The Git Commit Checker CI bot found a few problems with this PR: dff9ea5: Public APIs for:
Please fix these problems and, if necessary, force-push new commits back up to the PR branch. Thanks! |
joe-explr
force-pushed
the
notified-rma-sm
branch
from
dff9ea5 to
6519bd0
Compare
This commit adds notification support to the OSC SM component by implementing the put_with_notify, get_with_notify, rput_with_notify, and rget_with_notify functions. These functions perform the same operations as their non-notify counterparts but also increment notification counters after the data transfer completes. The changes include: - Added function pointer types for notify variants in osc.h - Added function prototypes in osc_sm.h - Implemented the notify functions in osc_sm_comm.c - Updated the module template to register the new functions - Removed TODO comments that have been addressed Signed-off-by: Joseph Antony <jajoseph.antony18@gmail.com>
Signed-off-by: Joseph Antony <jajoseph.antony18@gmail.com>
put_with_notify get_with_notify Signed-off-by: Joseph Antony <jajoseph.antony18@gmail.com>
put_with_notify
get_with_notify
Signed-off-by: Joseph Antony <jajoseph.antony18@gmail.com>
…for a single and multi rank window. Signed-off-by: Joseph Antony <jajoseph.antony18@gmail.com>
joe-explr
force-pushed
the
notified-rma-sm
branch
2 times, most recently
from
4e90525 to
f0ee69f
Compare
Signed-off-by: Joseph Antony <jajoseph.antony18@gmail.com>
Signed-off-by: Joseph Antony <jajoseph.antony18@gmail.com>
| if (notify < 0 || (uint32_t) notify >= module->node_states[target].notify_counter_count) { | ||
| return OMPI_ERR_BAD_PARAM; | ||
| } | ||
|
|
There was a problem hiding this comment.
Very good! 👍 I think we have a different error defined for it though (MPI_ERR_RMA_NOTIFICATION). We might need to add that first, so take this as a reminder for later :)
|
|
||
| opal_atomic_rmb(); | ||
| opal_atomic_add(&module->notify_counters[notify], 1); | ||
| opal_atomic_add(&osc_sm_target_notify_base(module, target)[notify], 1); |
There was a problem hiding this comment.
| opal_atomic_add(&osc_sm_target_notify_base(module, target)[notify], 1); | |
| opal_atomic_add(osc_sm_target_notify_base(module, target)[notify], 1); |
|
|
||
| /* initialize my state shared */ | ||
| module->my_node_state = &module->node_states[ompi_comm_rank(module->comm)]; | ||
| memset (module->my_node_state, 0, sizeof(*module->my_node_state)); |
There was a problem hiding this comment.
Maybe we move this to before we fill the counter information? That way we can keep the memset and don't run the risk of missing an init if we later add new fields.
2 participants
Summary of Changes: