ch4/ipc: Simplify generic ipc caches #7320
Draft
+192
−197
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
|
test:mpich/ch4/ofi/gpu |
raffenet
force-pushed
the
new-mapped-ipc-cache
branch
from
fe953d9 to
d39ff45
Compare
|
test:mpich/ch4/gpu |
raffenet
force-pushed
the
new-mapped-ipc-cache
branch
from
d39ff45 to
1baf24d
Compare
raffenet
changed the title
|
test:mpich/ch4/gpu |
raffenet
force-pushed
the
new-mapped-ipc-cache
branch
from
1baf24d to
b13ac5f
Compare
Remove mapped buffer cache using AVL trees and replace with a hash table. The key for the hash is the combination of the sender base address and their local rank on the node. Each entry can store a mapped address for each device visible to the process doing the mapping.
Replace AVL tree structures with hash keyed on the allocation base pointer.
IPC send is synchronous by definition. The sender needs to be notified when the receiver is done reading from the send buffer to satisfy completion semantics. The syncflag parameter is redundant.
At send completion time, give the GPU layer a chance to cleanup resources used to perform the IPC transfer. This does not affect the default configuration with caching enabled.
raffenet
force-pushed
the
new-mapped-ipc-cache
branch
from
b13ac5f to
2003155
Compare
Avoid masking any errors here and also squash an unused label warning.
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Outdated suggestions cannot be applied.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.
Pull Request Description
Replace both AVL tree hashes used in GPU IPC. The sender-side GPU IPC handle cache and the receiver side mapped buffer cache can be implemented using hashable structures, which makes the resulting code much easier to understand and maintain.
Author Checklist
Particularly focus on why, not what. Reference background, issues, test failures, xfail entries, etc.
Commits are self-contained and do not do two things at once.
Commit message is of the form:
module: short descriptionCommit message explains what's in the commit.
Whitespace checker. Warnings test. Additional tests via comments.
For non-Argonne authors, check contribution agreement.
If necessary, request an explicit comment from your companies PR approval manager.