Introduce QMC_DISABLE_HIP_HOST_REGISTER. ON by default. - #4674
Merged
Conversation
prckent
previously approved these changes
Jul 14, 2023
Contributor
|
It will be interesting to see if this helps or hinders the full range of system sizes run in the nightlies. |
Contributor
|
Test this please |
ye-luo
force-pushed
the
introduce-QMC_HIP_HOST_REGISTER
branch
from
July 14, 2023 19:39
00f2e9c to
6c5d36f
Compare
Contributor
Author
|
Test this please |
Merged
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Proposed changes
Since rocm 5.6. hipHostRegister becomes terribly slow https://github.com/ye-luo/miniqmc/wiki/hip-hsa-issues.
Introduce QMC_DISABLE_HIP_HOST_REGISTER. which can be used to make hipHostRegister no-op.
I set it ON by default for now. Will go through a few nights of performance test runs.
My guess is that it makes runs slower because we lose more from not using pinned memory than slower hipHostRegister in close to production benchmarks.
What type(s) of changes does this code introduce?
Does this introduce a breaking change?
What systems has this change been tested on?
epyc-server
Checklist