----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/67827/#review206365 -----------------------------------------------------------
Ship it! src/master/allocator/mesos/hierarchical.hpp Lines 685 (patched) <https://reviews.apache.org/r/67827/#comment289305> const Resources& per comment below src/master/allocator/mesos/hierarchical.cpp Lines 1812-1813 (original), 1812-1813 (patched) <https://reviews.apache.org/r/67827/#comment289306> What is "normal offer behavior"? In this patch we can just adjust the comment to reflect that we're not doing an explicit copy anymore, and if we want to document *why* the current approach is used, we can do that in a separate patch after hearing from Yan? src/master/allocator/mesos/hierarchical.cpp Lines 2044-2047 (original), 2024-2027 (patched) <https://reviews.apache.org/r/67827/#comment289307> Ditto here. src/master/allocator/mesos/hierarchical.cpp Lines 2689 (patched) <https://reviews.apache.org/r/67827/#comment289304> This can be `const Resources&` to avoid a copy? - Benjamin Mahler On July 20, 2018, 11:30 p.m., Meng Zhu wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/67827/ > ----------------------------------------------------------- > > (Updated July 20, 2018, 11:30 p.m.) > > > Review request for mesos and Benjamin Mahler. > > > Bugs: MESOS-9104 > https://issues.apache.org/jira/browse/MESOS-9104 > > > Repository: mesos > > > Description > ------- > > This helper removes any resources that the framework is not > capable of receiving based on the given framework capability. > > > Diffs > ----- > > src/master/allocator/mesos/hierarchical.hpp > 702e7c0aa84b4b672d82c759c25a28a77c78ad50 > src/master/allocator/mesos/hierarchical.cpp > 707dd6bd0f255a64d759ce87cbf75be57d86b392 > > > Diff: https://reviews.apache.org/r/67827/diff/3/ > > > Testing > ------- > > make check > > > Thanks, > > Meng Zhu > >
