On Tue, May 29, 2018 at 1:34 PM, Andre McCurdy <[email protected]> wrote: > On Tue, May 29, 2018 at 12:13 PM, Khem Raj <[email protected]> wrote: >> On Tue, May 29, 2018 at 11:41 AM, Andre McCurdy <[email protected]> wrote: >>> This change looks wrong (or at least incomplete and lacking a decent >>> explanation): >> >> why do you think it is wrong ? I would be happy to make the >> explanation more decent >> if you have ideas please share. > > I would be happy to try... if I had any idea why you think you need it! > > What causes the conflict in flags and why hasn't it been seen before? > Why do you only need to fix Cortex A7 and not when building for other > armv7ve cores? >
gcc-runtime is a bit special where it gets itself configured when building cross compiler but then we tame it to re-target itself for target using some shims e.g. setting generic arch values so it can plug itself amicably when we build gcc for target and use it ondevice because it needs to rely on right combination of options, and gcc for target is also made to target more common ISAs via doing things like below. EXTRA_OECONF_append_armv7ve = " --with-arch=armv7-a" these options cause additional march values to be added. If you are interested. Plug this patch out and use gcc8 patches I have floating around and you can see whats going on. I am also interested in finding a solution where we do not need this patch. >>> >>> >>> http://git.openembedded.org/openembedded-core/commit/?h=master-next&id=fbec01f01fdad23d95271db063ee0a4d0ab44568 >>> >>> It was never posted to the list but seems to have made it as far as >>> master-next. >>> -- >>> _______________________________________________ >>> Openembedded-core mailing list >>> [email protected] >>> http://lists.openembedded.org/mailman/listinfo/openembedded-core -- _______________________________________________ Openembedded-core mailing list [email protected] http://lists.openembedded.org/mailman/listinfo/openembedded-core
