Re-vendor dependencies #89
Conversation
This pulls in a new version of github.com/containerd/ttrpc from a fork to fix the deadlock issue in containerd/ttrpc#72. Will revert back to the upstream ttrpc vendor once the fix is merged (containerd/ttrpc#73). This fix also included some vendoring cleanup from running "vndr". Signed-off-by: Kevin Parsons <[email protected]>
kevpar
force-pushed
the
ttrpc-deadlock-fix
branch
from
9da51a7 to
16caa11
Compare
Yes, that's just some vendor cleanup that was performed automatically. |
|
Are we supposed to focus our reviewing efforts of the deadlock ttrpc changes here or the upstream PR? I take it this is a "this fixes the issue and barring any critical flaws we saw, just want a set of eyes on the vendored files", am I off? |
Mostly need focus on the upstream ttrpc PR. Comments here would be good if there are any concerns unique to this PR, such as from the vendoring or other ttrpc changes being brought in. |
|
Re-purposing this to be a general revendor PR and pulling in one more change. Will need another look in a moment. |
kevpar
changed the title
Signed-off-by: Kevin Parsons <[email protected]>
to fix the deadlock issue in Deadlock with multiple simultaneous requests from the same client containerd/ttrpc#72. Will revert back to
the upstream ttrpc vendor once the fix is merged (client/server: Don't block the main connection loop for transport IO containerd/ttrpc#73).
Signed-off-by: Kevin Parsons [email protected]