Avoid synchronization by cloning a prototype Mac and reallocating buffers
#42
Conversation
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
jchambers
force-pushed
the
clone_mac
branch
3 times, most recently
from
4cb1e82 to
2a09306
Compare
jchambers
changed the title
Mac and reallocating buffers
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Outdated suggestions cannot be applied.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.
I've been thinking back on the conversation in #34, and I think I've found a best-of-both-worlds approach. To recap, the suggestion in #34 was to allocate a new buffer and instantiate a new
Macon each call togenerateOneTimePasswordto avoid synchronization, leading to higher throughput. The problem was that instantiating a newMacwas pretty slow.It turns out that cloning a
Macis actually pretty fast, though! We can't guarantee that cloning aMacwill be possible in all setups—some provider might have a non-cloneableHmacSHA512implementation, for example—but it'll probably be true in most setups. This lets us makegenerateOneTimePasswordnon-synchronized while still retaining thread safety, which seems like a win.This may come at the cost of a modest throughput decrease in the single-threaded case, but if it does, it's on the order of 1%, and I'd say that's a worthwhile price to pay.