> Dan Ports wrote:
> Yes, that comment was supposed to be attached to
> possibleUnsafeConflicts.
> Actually, I think that "other" hash no longer exists
> The comment above SERIALIZABLEXACT also needs to be updated since
> it refers to said hash table. And if I'm not mistaken (Kevin?), we
> can eliminate SERIALIZABLEXACTTAG altogether and not bother putting
> the vxid in the sxact.
I think we still need the vxid. It shows in the pg_locks view, and
we might possibly need it to find the right process to cancel once we
have some way to do that. But there's no point with having the tag
level anymore.
Otherwise, I agree with Dan on all counts.
Dan, if you're not already working on these, I'll take them, so you
can focus on 2PC.
-Kevin
