[
https://issues.apache.org/jira/browse/CASSANDRA-9472?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15116758#comment-15116758
]
Stefania commented on CASSANDRA-9472:
-------------------------------------
I've pushed one more commit
[here|https://github.com/stef1927/cassandra/commit/f035a5a0e69c745d2257186258a1ffa113427120]
where I fixed a compilation problem (a leftover of my changes yesterday) and
removed the {{peer == 0}} guards.
bq. I assumed that (as it used to be) the peer could be set to zero on free,
and that this was how you were encountering the problem. I see now it is only
the fake NativeCell that causes problems; I do wonder if we shouldn't simply
allocate it for real, to avoid the check for every single call to a cell. It's
also highly unlikely to ever appear in anyone's debugger if it's only that
specific cell, so we could simply ignore the problem entirely.
Yes it was only the fake native cell and clustering causing problems but I was
unable to reproduce these segmentation faults today so either some code
underneath has changed or the root cause must have been something else.
I've launched the tests on my branch, if they are clear I think we are good to
commit:
|[testall|http://cassci.datastax.com/view/Dev/view/stef1927/job/stef1927-9472-updated-testall/]|
|[dtest|http://cassci.datastax.com/view/Dev/view/stef1927/job/stef1927-9472-updated-dtest/]|
> Reintroduce off heap memtables
> ------------------------------
>
> Key: CASSANDRA-9472
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9472
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Benedict
> Assignee: Benedict
> Fix For: 3.x
>
>
> CASSANDRA-8099 removes off heap memtables. We should reintroduce them ASAP.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
