[dpdk-dev] [PATCH v3] app: test: measure libipsec performance

[Thomas Monjalon]

10/03/2020 14:24, Ananyev, Konstantin:
> 
> > 05/03/2020 23:51, Ananyev, Konstantin:
> > >
> > > > 05/03/2020 12:45, Ananyev, Konstantin:
> > > > > I think the header need to be "app/test", or "test/ipsec".
> > > >
> > > > It should be "test/ipsec" as it is an IPsec test command
> > > > in the test application.
> > > >
> > > > > Apart from that:
> > > > > Acked-by: Konstantin Ananyev <konstantin.ananyev at intel.com>
> > > >
> > > > I wonder why we have a different maintainer for each IPsec test.
> > > > It looks a bit confusing.
> > > >
> > > > IPsec - EXPERIMENTAL
> > > > M: Konstantin Ananyev <konstantin.ananyev at intel.com>
> > > > T: git://dpdk.org/next/dpdk-next-crypto
> > > > F: lib/librte_ipsec/
> > > > M: Bernard Iremonger <bernard.iremonger at intel.com>
> > > > F: app/test/test_ipsec.c
> > > > F: doc/guides/prog_guide/ipsec_lib.rst
> > > > M: Savinay Dharmappa <savinay.dharmappa at intel.com>
> > > > F: app/test/test_ipsec_perf.c
> > > > M: Vladimir Medvedkin <vladimir.medvedkin at intel.com>
> > > > F: app/test/test_ipsec_sad.c
> > > > F: app/test-sad/
> > >
> > > I suppose mainly by historical reasons -
> > > each of these tests have different authors.
> > > Obviously each author claimed maintanership of his own code.
> > > Do you consider that as a problem?
> > 
> > Yes I consider a maintainer should know a whole area
> > and be the point of contact for this area.
> > For instance, when there is a bug with a test,
> > we don't know whether the problem is in the test or in the library.
> 
> But same problem could appear with any test/sample app.
> Let say you have a problem with testpmd.
> Is that testpmd itself, rte_ethdev or particular PMD?
> Obviously some preliminary investigation is needed before finger pointing.
> I don't see why having multiple persons as maintainers for the same area
> is a problem. In fact, it probably gives more chances for faster turnover
> (as long as they do their maintainer job properly).

Yes, having several maintainers is good.
But having one different maintainer per test file is quite ridiculous.
In general the maintainers of a lib are also maintaining the related app files.


> > Note that the original author can be found with the git history.
> > 
> > > > In general, only one person takes the responsibility to be
> > > > the main contact of any related question in the area.
> > > >
> > > > So the file pattern could be:
> > > > F: app/test/test_ipsec*
> > 
> > 
> 
>