Update RMM tests based on deprecated CNMeM #359
Merged
Conversation
This file contains 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
As `get_default_resource_type` was dropped in RMM recently, use the newly introduce `rmm.mr.get_per_device_resource` instead to access the resource on device `0` (configured to a unique device for each worker). Since the memory resource itself is not realistically serializable, instead grab the type of each resource to send back. This is all done within a `lambda` to allow for a function that can be run on each worker.
As CNMeM has been dropped from Python and replaced with RMM's own pool resource, just check for that pool resource instead.
harrism
suggested changes
Aug 12, 2020
Simplify the resource type checks a bit. Thanks Mark! :) Co-authored-by: Mark Harris <[email protected]>
|
Much cleaner. Thanks Mark! :) |
Codecov Report
@@ Coverage Diff @@
## branch-0.15 #359 +/- ##
===============================================
+ Coverage 59.65% 60.04% +0.39%
===============================================
Files 17 17
Lines 1321 1334 +13
===============================================
+ Hits 788 801 +13
Misses 533 533
Continue to review full report at Codecov.
|
This was referenced Aug 12, 2020
Merged
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Outdated suggestions cannot be applied.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.
The recent RMM PR ( rapidsai/rmm#466 ) deprecated CNMeM and made some changes for handling memory resources for multiple devices (needed by XGBoost). As a result, we are seeing a few test failures in Dask-CUDA. This makes the necessary changes to update these tests.