Implemented kDLCPUPinned (cudaMallocHost) #4985
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
tqchen
approved these changes
Mar 10, 2020
|
Thanks @jmorrill ! this is merged |
trevor-m
pushed a commit
to trevor-m/tvm
that referenced
this pull request
Apr 16, 2020
* implement kDLCPUPinned * Fix line endings * Fix whitespace for linter * cleanup up allocdataspace method
zhiics
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Apr 17, 2020
* implement kDLCPUPinned * Fix line endings * Fix whitespace for linter * cleanup up allocdataspace method
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Outdated suggestions cannot be applied.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.
Data allocated via cudaMallocHost is supposed to be faster at transferring data to/from a cuda device, and it was not implemented in the tvm runtime.
The DeviceAPIs treat DLDeviceTypes as their own device, so kDLCPUPinned felt a little bit out of place because it was sort of a kDLCPU (host memory) but really it was owned by a kDLGPU (cuda api).
I felt the least complicated path was to register an alias for "device_api.gpu" as "device_api.cpu_pinned" and implement the kDLCPUPinned logic in CUDADeviceAPI.
Some small checks also needed to be modified. Not sure if I missed any.
Open to suggestions if my implementation is way off.