Adar Dembo has posted comments on this change. ( http://gerrit.cloudera.org:8080/8345 )
Change subject: KUDU-2193 (part 1): switch to a waiting mutex in TSTabletManager ...................................................................... Patch Set 1: > Patch Set 1: > > I dunno, I'm sort of leaning more towards the idea that true spin-locks > should only be used for stuff as simple as "read this POD variable" and not > for anything that involves acquiring other locks or allocating memory. The issue being that even if you address the acquisition sites I listed, we still allocate memory while holding the TSTabletManager lock? I don't think that's a huge deal, or, if it is, it's certainly a pattern we use widely (i.e. take a spinlock, make a local copy of this map, release the spinlock). -- To view, visit http://gerrit.cloudera.org:8080/8345 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I763abddd74d8b1dabb618318dc84256b533077e3 Gerrit-Change-Number: 8345 Gerrit-PatchSet: 1 Gerrit-Owner: Todd Lipcon <t...@apache.org> Gerrit-Reviewer: Adar Dembo <a...@cloudera.com> Gerrit-Reviewer: David Ribeiro Alves <davidral...@gmail.com> Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon <t...@apache.org> Gerrit-Comment-Date: Fri, 20 Oct 2017 21:53:09 +0000 Gerrit-HasComments: No