On Wed, Jun 03, 2015 at 11:20:23AM +0100, Stuart Haslam wrote: > >Yup, I too am not sure that it's a good idea to install tests via >"make install", but that's what was being done before your change and it >(kind of) worked.
Oh, hmmm. >We do need some method of installing test binaries, to cope with >cross-building, where the tests can't be run on the machine they were >built on. As discussed yesterday I think the right thing to do is create >another target named install-test which installs just the test binaries >and any supporting scripts, but that can be done later and for this >series all that's required is not to introduce a regression. There's also maybe some use in installing the *source* for some of the tests as examples alongside documentation. But definitely the tests themselves shouldn't be installed using "make install"... Cheers, -- Steve McIntyre [email protected] <http://www.linaro.org/> Linaro.org | Open source software for ARM SoCs _______________________________________________ lng-odp mailing list [email protected] https://lists.linaro.org/mailman/listinfo/lng-odp
