On Fri, Aug 8, 2014 at 6:01 AM, Heikki Linnakangas
<hlinnakangas@vmware.com> wrote:
> It's possible that two backends arrive at phase 3 at the same time, with
> different values. For example, backend A wants to update the minimum to
> contain 10, and and backend B wants to update it to 5. Now, if backend B
> gets to update the tuple first, to 5, backend A will update the tuple to 10
> when it gets the lock, which is wrong.
>
> The simplest solution would be to get the buffer lock in exclusive mode to
> begin with, so that you don't need to release it between steps 2 and 5. That
> might be a significant hit on concurrency, though, when most of the
> insertions don't in fact have to update the value. Another idea is to
> re-check the updated values after acquiring the lock in exclusive mode, to
> see if they match the previous values.
No, the simplest solution is to re-check the bounds after acquiring
the exclusive lock. So instead of doing the addValue with share lock,
do a consistency check first, and if it's not consistent, do the
addValue with exclusive lock.