----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/35631/#review88472 -----------------------------------------------------------
Ship it! Great, thanks Jie! src/tests/hierarchical_allocator_tests.cpp (line 976) <https://reviews.apache.org/r/35631/#comment141009> I think we were avoiding {} here before we had gcc 4.8 as an assumption, so should be safe to do now, up to you. src/tests/hierarchical_allocator_tests.cpp (line 980) <https://reviews.apache.org/r/35631/#comment141008> Can you do {} instead of explicit hashmap construction? src/tests/hierarchical_allocator_tests.cpp (line 986) <https://reviews.apache.org/r/35631/#comment141010> Maybe just s/slaveCount/GetParam()/ and remove the 'slaveCount' variable? Then below you could replace 'slaveCount' with 'slaves.size()' or a foreach loop. Whichever you think is cleaner! src/tests/hierarchical_allocator_tests.cpp (line 994) <https://reviews.apache.org/r/35631/#comment141007> Might be nice to have a bit more port fragmentation, to match the pathological scenario. :) - Ben Mahler On June 19, 2015, 12:15 a.m., Jie Yu wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/35631/ > ----------------------------------------------------------- > > (Updated June 19, 2015, 12:15 a.m.) > > > Review request for mesos, Ben Mahler and Vinod Kone. > > > Bugs: MESOS-2892 > https://issues.apache.org/jira/browse/MESOS-2892 > > > Repository: mesos > > > Description > ------- > > Added a performance benchmark for hierarchical allocator addSlave. > > > Diffs > ----- > > src/tests/hierarchical_allocator_tests.cpp > 85bb29e7cfc00579ff8f6d62d6c75e1b99ffcdba > > Diff: https://reviews.apache.org/r/35631/diff/ > > > Testing > ------- > > bin/mesos-test.sh --benchmark > > 0.23.0 current master > > > 0.23.0 before MESOS-2373 > > > Thanks, > > Jie Yu > >
