----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/67190/#review203652 -----------------------------------------------------------
Ship it! The assumption here is that stringification of `Error` values is usually not happening on the hot path, so creating a new `string` does not have a significant performance impact. Not sure we should capture that in a comment here, though. Alternatively we could implement two overloads const string& stringify(const Error&); string stringify(Error&&); Maybe that is something we should capture in a `TODO`. - Benjamin Bannier On May 22, 2018, 10:28 p.m., Chun-Hung Hsiao wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/67190/ > ----------------------------------------------------------- > > (Updated May 22, 2018, 10:28 p.m.) > > > Review request for mesos, Benjamin Bannier, Benjamin Mahler, Michael Park, > and Zhitao Li. > > > Bugs: MESOS-8925 > https://issues.apache.org/jira/browse/MESOS-8925 > > > Repository: mesos > > > Description > ------- > > Overloaded `stringify` for `Error`s to reduce overheads. > > > Diffs > ----- > > 3rdparty/stout/include/stout/stringify.hpp > 49bf684a2aa18bdd9ab866eb76de5b938a3f3a4b > > > Diff: https://reviews.apache.org/r/67190/diff/2/ > > > Testing > ------- > > make check > > > Thanks, > > Chun-Hung Hsiao > >