Todd Lipcon has posted comments on this change. Change subject: rpc: periodic timers ......................................................................
Patch Set 2: (2 comments) http://gerrit.cloudera.org:8080/#/c/7733/2/src/kudu/rpc/periodic-test.cc File src/kudu/rpc/periodic-test.cc: Line 88: TEST_P(PeriodicTimerTest, TestReset) { > When I looped 1000 times with TSAN all I got were data races on access to c yea, of course it will be timing-dependent. But perhaps just running the whole thing slower (eg with period_ms_ being 500ms instead of 50) would mean that the same absolute time slownesses would be much less likely to cause a failure. http://gerrit.cloudera.org:8080/#/c/7733/2/src/kudu/rpc/periodic.h File src/kudu/rpc/periodic.h: PS2, Line 59: periodic_period_jitter_pct > I think the interface would be cleaner as ok, I'm alright with a gflag on the argument that they're easier to change than constants, so long as it's experimental -- To view, visit http://gerrit.cloudera.org:8080/7733 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: I23f80631a5591b1ac023974f3d7d9eb1576cdb86 Gerrit-PatchSet: 2 Gerrit-Project: kudu Gerrit-Branch: master Gerrit-Owner: Adar Dembo <[email protected]> Gerrit-Reviewer: Adar Dembo <[email protected]> Gerrit-Reviewer: Dan Burkert <[email protected]> Gerrit-Reviewer: David Ribeiro Alves <[email protected]> Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Mike Percy <[email protected]> Gerrit-Reviewer: Tidy Bot Gerrit-Reviewer: Todd Lipcon <[email protected]> Gerrit-HasComments: Yes
