This repository has been archived by the owner on Mar 11, 2020. It is now read-only.
Cherry-pick: Add option to force TCP_QUICKACK. #1
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
|
Upstream PR: haskell-distributed#67 |
It is well known that Nagle's algorithm (the absence of TCP_NODELAY) and delayed acknowledgements (the absence of TCP_QUICKACK) cause performance issues on TCP; see for example: * https://eklitzke.org/the-caveats-of-tcp-nodelay * http://jvns.ca/blog/2015/11/21/why-you-should-understand-a-little-about-tcp/ TCP_NODELAY was already implemented; this commit adds TCP_QUICKACK. This change breaks the external API, because `socketToEndPoint` (for reasons not obvious to me) takes all socket options as individual arguments, instead of as a settings object.
nh2
changed the base branch from
cardano-sl-0.4
to
cardano-sl-0.4-tcp-quickacks-setting
nh2
changed the base branch from
cardano-sl-0.4-tcp-quickacks-setting
to
cardano-sl-0.4
nh2
force-pushed
the
csl-0.4-tcp-quickacks-setting
branch
from
5e0dd61
to
83aa874
Compare
Feel free to correct this! I'm sure it would be warmly welcomed upstream. |
OK, I've just asked on haskell-distributed#67 (comment) |
avieth
suggested changes
May 17, 2017
| @@ -1129,6 +1155,10 @@ handleIncomingMessages params (ourEndPoint, theirEndPoint) = do | |||
| Left err -> prematureExit err | |||
| Right sock -> tryIO (go sock) >>= either prematureExit return | |||
| where | |||
| settings = defaultTCPInternalSettings | |||
| { tcpInternalForceQuickAck = tcpForceQuickAck params | |||
| } | |||
There was a problem hiding this comment.
I suggest we put a field transportInternalSettings into TCPTransport and create it once at createTransport time.
Or, how about moving TCPParameters into Network.Transport.TCP.Internal, and doing away with TCPInternalSettings completely?
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
It is well known that Nagle's algorithm (the absence of TCP_NODELAY)
and delayed acknowledgements (the absence of TCP_QUICKACK) cause
performance issues on TCP; see for example:
TCP_NODELAY was already implemented; this commit adds TCP_QUICKACK.
This change breaks the external API, because
socketToEndPoint(for reasons not obvious to me) takes all socket options as individual
arguments, instead of as a settings object.