Alex T

I believe this should rather be reworked using `_replace` approach, as this is specific to one board we support. @SyrianSpock, do you think you'd want to do that?

Thanks and no rush :smile: I just believe this is a better way to go and looking at the code and @arfoll's comments I think it will increase its chances...

So @SyrianSpock, would you be willing to modify this one as requested?

> In short, Yocto thinks that libmraa which uses CMake already creates the RPMs. If the lines of code from the PR are not included, the RPMs will not be...

Okay, thanks Tom. @ckalluri, please clarify why you need this, otherwise I'd personally be inclined to reject this for no good justification.

Thanks @b49020 for reporting this and finding the reason. Are you up to creating a PR to fix this, now that we have the root cause?

I haven't built docs much to be honest, but looking at the CMakeLists.txt and that mraa2.i file in the python2 dir, this looks rather like it's not finding the autogenerated...

Hmm, okay - so the log looks generally ok to me, nothing apparently wrong... No further ideas at this point, sorry. Generally the docs are being build-tested in our CI,...

@swfFrg, could you please do the sign-off mentioned above? We'd be happy to merge your patch, but we can't do this without a signoff (see CONTRIBUTING.md as to why).

The syslog way of debugging is exactly what you've pasted as a `logread` :) Anyway, looks like that new OS version disables `/dev/mem` - as many modern distros do -...