-
Couldn't load subscription status.
- Fork 928
coll/libnbc: add bruck algorithm for MPI_Iallgather #6573
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Signed-off-by: Alex Anenkov <[email protected]>
|
Can one of the admins verify this patch? |
|
ok to test. |
|
bot:ompi:retest |
1 similar comment
|
bot:ompi:retest |
| return res; | ||
| } | ||
|
|
||
| /* allocate temporary buffers */ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is getting ugly. Why can't we have a nice function based init ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I agree, that this approach is ugly, but I adhered overall design of libnbc (allreduce, alltoall, etc). I am ready to fix this PR, but it is required an additional discussion of memory allocation in libnbc.
|
Ping - now almost a year old. Any update on this pr? |
|
Adding this to the 5.0 milestone for tracking. Since this seems more like a feature than a bug fix, if you want it for 5.0 please target getting this into master by the end of April. Thanks. |
|
@alex-anenkov - How was this tested? Is this safe for v5.0? |
|
Hello @gpaulsen. Two years have passed since the opening of this PR. I'm not involved with OpenMPI development, so I'm not sure if PR is safe for v5.0. I think it will be better if I close it. Thanks. |
|
@alex-anenkov That's a shame, but thank you. |
Implements Bruck algorithm for MPI_Iallgather.
Signed-off-by: Alex Anenkov [email protected]