aweisberg commented on code in PR #26:
URL: https://github.com/apache/cassandra-accord/pull/26#discussion_r1094942728
##########
accord-core/src/main/java/accord/primitives/Txn.java:
##########
@@ -32,13 +32,26 @@
import javax.annotation.Nonnull;
import javax.annotation.Nullable;
-import static accord.primitives.Routables.Slice.Overlapping;
-
public interface Txn
{
enum Kind
{
- Read, Write;
+ Read,
+ Write,
+
+ /**
+ * A pseudo-transaction whose deps represent the complete set of
transactions that may execute before it,
+ * without interfering with their execution.
+ *
+ * A SyncPoint is unique in that it agrees not only an executeAt but
also a precise collection of dependencies,
Review Comment:
I am figuring through how regular transactions don't agree on their deps, so
with 1RT yeah that makes sense.
There are no changes to how `Accept` handles deps so I was trying to
understand what it means for these transactions to agree on deps, but regular
transactions don't. So one different is we always run accept, and I guess in
the future you mean that accept won't persist dependencies for regular
transactions (recovery will go find them again?), but we will continue to
persist them for barriers so when they recover the precise set of dependencies
is returned.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
