reduce vertex metadata size and remove vertex locks #8
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
|
Are you sure that it doesn't do that to prevent false sharing? |
|
The lock needs to be cacheline padded. I don't think yours does that? |
meyerzinn
force-pushed
the
meyer/fix-vertex-size
branch
from
12f1e97 to
67a249c
Compare
meyerzinn
changed the title
|
Realized we don't need to lock vertices if each batch has only one thread write to each vertex. |
meyerzinn
force-pushed
the
meyer/fix-vertex-size
branch
from
67a249c to
1579a30
Compare
meyerzinn
force-pushed
the
meyer/fix-vertex-size
branch
from
1579a30 to
d6cace4
Compare
AdityaAtulTewari
approved these changes
Mar 21, 2024
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Outdated suggestions cannot be applied.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.
Apparently,
galois::SpinLockis a massive class that takes >128 bytes to represent. This replaces the use ofSpinLockwith a simple atomic byte. In theory, we could combine the buffer field and the lock into one flags field, but we already have spare bits in the begin and end fields so that wouldn't change the size.Also, in each batch only one thread writes to each vertex, so we do not need a lock on vertices.