> On June 16, 2015, 4:55 p.m., Niklas Nielsen wrote: > > src/hook/manager.cpp, lines 96-98 > > <https://reviews.apache.org/r/30339/diff/5/?file=984855#file984855line96> > > > > Won't some compilers potentially break with reaching end of non-void > > function? > > > > How about: > > > > ``` > > bool result = false; > > > > synchronized(mutex) { > > return !availableHooks.empty(); > > } > > > > return result; > > ``` > > > > @joris - any insights on what is safe to do here?
After mpark's changes to synchronized, it is now safe to do this :-) The key is informing the compiler that this path is always executed (which is what mpark's refactoring does to the synchronized macro). - Joris ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/30339/#review88085 ----------------------------------------------------------- On June 15, 2015, 10:26 p.m., Kapil Arya wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/30339/ > ----------------------------------------------------------- > > (Updated June 15, 2015, 10:26 p.m.) > > > Review request for mesos, Niklas Nielsen and Till Toenshoff. > > > Repository: mesos > > > Description > ------- > > Call hook manager only if hooks were specified on the commandline. > > > Diffs > ----- > > src/hook/manager.hpp 638e19f5062f070cd0ec94d830615a2c73fd357d > src/hook/manager.cpp b43b91878cc9f0f5f5d081b9b3af0768c07c282b > src/master/master.cpp 0137c37a42d4f7a9057568a8cd7b662ed92be353 > src/slave/containerizer/containerizer.cpp > e995ce602261c18373ac09c823638c4a252cca86 > src/slave/slave.cpp 67732a40ef683cb0188425f0bba8fe8ab83e461c > > Diff: https://reviews.apache.org/r/30339/diff/ > > > Testing > ------- > > make check > > > Thanks, > > Kapil Arya > >
