-
-
Notifications
You must be signed in to change notification settings - Fork 396
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Match GSO segment size to the first datagram, not the MTU #1837
Conversation
e7c47c4
to
a1add92
Compare
On second thought, maybe judging whether a datagram fits into a certain segment size in advance of beginning packet assembly isn't so bad. We already have the logic to make a conservative estimate in |
cbe6295
to
ce7f2f3
Compare
This is now feature-complete, but still needs testing. |
60c8054
to
28360a3
Compare
Went ahead and banged out the tests, finding and fixing a number of bugs in the process. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Initial round, got interrupted by some bored kids.
fae273a
to
79246f5
Compare
98b9664
to
733d798
Compare
Good stuff! |
37a5fd9
to
aee7545
Compare
Reduces the amount of padding required when sending homogeneously sized non-fragmentable data like application datagrams.
Fixes #1832.
Still working onhow to handle the case where a datagram is queued but won't fit with the current segment size. We need to either detect this in advance and finish the GSO batch before beginning another packet, or gracefully abandon packet assembly after failing to fit any frames.Currently leaning towards the latter.This will need careful testing.Also still TODO: End the GSO batch early if the amount of padding in the current packet would be excessive.