feat(s2n-quic-core): use scatter buffers for encryption #1899
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
camshaft
force-pushed
the
camshaft/vectored-encrypt
branch
from
a2e344f
to
7831906
Compare
camshaft
force-pushed
the
camshaft/vectored-encrypt
branch
from
7831906
to
d200e0a
Compare
camshaft
force-pushed
the
camshaft/vectored-encrypt
branch
from
d200e0a
to
d6e6ffc
Compare
|
|
||
| /// Initializes the buffer with extra bytes | ||
| /// | ||
| /// NOTE the EncoderBuffer position should not include the extra bytes. This ensures the bytes |
There was a problem hiding this comment.
any way to debug assert this?
There was a problem hiding this comment.
Not really, unfortunately. Once we've split up the buffer into its parts, we've lost the state required to keep track of where the cursor should be.
Comment on lines
+599
to
+601
| .unwrap_or_else(|err| { | ||
| panic!("encryption error; opener={opener_name}, sealer={sealer_name} - {err:?}") | ||
| }); |
There was a problem hiding this comment.
would this be equivalent to (I think the err gets appended):
Suggested change
| .unwrap_or_else(|err| { | |
| panic!("encryption error; opener={opener_name}, sealer={sealer_name} - {err:?}") | |
| }); | |
| .expect("encryption error; opener={opener_name}, sealer={sealer_name}"); |
There was a problem hiding this comment.
Formatting requires a macro, since the arguments are variadic. So expect only allows for a static string
There was a problem hiding this comment.
Formatting requires a macro, since the arguments are variadic. So expect only allows for a static string
WesleyRosenblum
approved these changes
Aug 23, 2023
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Outdated suggestions cannot be applied.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.
Description of changes:
In preparation of aws/aws-lc-rs#206 being available, this change refactors the crypto Key interface to take a
scatter::Buffer, which allows the encryption method to defer copying of the last chunk ofBytes. This has some pretty big wins for bulk transfer, since the majority of the payload is aBytesand comes as the last chunk. Since encryption is itself a copy, this sees about a 8% increase in efficiency overall.Call-outs:
The AWS-LC interface only takes a single extra bytes at the end, instead of an arbitrary list. Our new
scatter::Bufferalso does the same which makes it so we don't have to allocate a list ofBytes. This could change in the future, but it's unclear if having an arbitrary list of chunks would even help all that much.Testing:
Without this change
Throughput is ~20Gbps. You can see the
memcpytaking about 8% of CPU cycles in the flamegraph:With this change without AWS-LC-RS scatter method
You can see in the flamegraph that the
memcpyhas moved to inside theencryptmethod, since the copy was deferred until encrypt time:With this change + AWS-LC-RS scatter method
Throughput is ~21Gbps. You can see in the flamegraph that the
memcpyis gone:By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.