ALLOC_WITH_HINT: added inplace realloc - v4.0 #6707
Conversation
- in some cases realloc operation may be completed without allocation of new buffer (and without additional data copy) - added logic to reallocate buffer inplace if possible Signed-off-by: Sergey Oblomov <[email protected]> (cherry picked from commit 277c2a9)
Signed-off-by: Sergey Oblomov <[email protected]> (cherry picked from commit a51badd)
Signed-off-by: Sergey Oblomov <[email protected]> (cherry picked from commit d6a0912)
|
Can one of the admins verify this patch? |
|
@hppritcha ok to merge? |
|
@AboorvaDevarajan Not quite sure what you were asking...? The phrase "Can one of the admins verify this patch?" is usually posted by our Jenkins system when a github user who is not part of the open-mpi community submits a pull request. When a member of the github open-mpi community replies with "ok to test", Jenkins then actually submits the CI job. It's an anti-DOS system (i.e., vs. overwhelming our CI system with a flood of junk PRs). |
|
@jsquyres I failed to disable the default notifications in one of the Jenkins CI job bot that I was running, the job polled and added comments in a few PRs automatically, I disabled it right away. Will be extra cautious when enabling such things. Thanks. |
|
Oh, gotcha -- that wasn't you posting those comments; it was a bot running with your credentials. That makes much more sense! 😄 |
|
bot:retest |
|
The UH CI appears to be offline. Let's see if a new round of CI will fix the issue... bot:retest |
added logic to reallocate buffer inplace if possible
backport from #6667