> On April 8, 2015, 6:44 p.m., Ian Downes wrote: > > src/tests/memory_test_helper.hpp, lines 69-70 > > <https://reviews.apache.org/r/32756/diff/1/?file=913065#file913065line69> > > > > How does this tie into the timing loop above in the test? Do we know > > within what time the kernel will mark pages inactive?
see my comment above > On April 8, 2015, 6:44 p.m., Ian Downes wrote: > > src/tests/cgroups_tests.cpp, line 1357 > > <https://reviews.apache.org/r/32756/diff/1/?file=913064#file913064line1357> > > > > Is it a problem of keeping up, or just that it's an asychronous process > > that periodically/as required marks pages? Did more reading and experiments. The kernel demotes the pages to inactive after some timeout time. But I am writing these pages in a loop, which should be able to beat the timeout. I get a constant stream of 15724544 bytes (14.996MB) printed out from the loop. The fact that the value doesn't change suggests it is not influenced by the timeout too. I don't quite get where the fixed %0.02 error is from - Chi ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/32756/#review79390 ----------------------------------------------------------- On June 3, 2015, 11:01 p.m., Chi Zhang wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/32756/ > ----------------------------------------------------------- > > (Updated June 3, 2015, 11:01 p.m.) > > > Review request for mesos, Ian Downes and Jie Yu. > > > Bugs: mesos-2572 > https://issues.apache.org/jira/browse/mesos-2572 > > > Repository: mesos > > > Description > ------- > > Added a memory statistics test for active anonymous memory. > > > Diffs > ----- > > src/tests/cgroups_tests.cpp 475f48a474eea708f98d8c0300862351a2d4379a > src/tests/memory_test_helper.hpp 11712d7f378d9426f160d53b0387c698a28a4207 > src/tests/memory_test_helper.cpp 8093e66c45cacc4bf0295a3448c128a1aa7c751a > > Diff: https://reviews.apache.org/r/32756/diff/ > > > Testing > ------- > > > Thanks, > > Chi Zhang > >
