On Fri, Dec 01, 2017 at 11:12:50AM +0000, Konstantin Ananyev wrote: > Simple functional test for rte_smp_mb() implementations. > Also when executed on a single lcore could be used as rough > estimation how many cycles particular implementation of rte_smp_mb() > might take. > > Signed-off-by: Konstantin Ananyev <konstantin.anan...@intel.com> > --- > test/test/Makefile | 1 + > test/test/test_mb.c | 278 > ++++++++++++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 279 insertions(+) > create mode 100644 test/test/test_mb.c > First pass comment on this patch is that it really could do with some comments explaining what is happening and why. Although it's meant to be a "simple functional test", it doesn't appear to be so on first reading, without comments as to how the test is meant to work and what the expected output values are.
/Bruce
