Em Wed, 21 Jun 2017 10:08:05 +0200 Johannes Thumshirn <jthumsh...@suse.de> escreveu:
> Currently the media subsystem has a very creative abuse of the > KERNEL_VERSION macro to encode an arbitrary version triplet for media > drivers and device hardware revisions. > > This series introduces a new macro called MEDIA_REVISION which encodes > a version triplet like KERNEL_VERSION does, but clearly has media > centric semantics and doesn't fool someone into thinking specific > parts are defined for a specific kernel version only like in out of > tree drivers. Sorry, but I can't see any advantage on it. On the downside, it includes the media controller header file (media.h) where it is not needed. > > Johannes Thumshirn (7): > [media] media: introduce MEDIA_REVISION macro > video: fbdev: don't use KERNEL_VERSION macro for MEDIA_REVISION > [media] media: document the use of MEDIA_REVISION instead of > KERNEL_VERSION > [media] cx25821: use MEDIA_REVISION instead of KERNEL_VERSION > [media] media: s3c-camif: Use MEDIA_REVISON instead of KERNEL_VERSION > [media] media: bcm2048: use MEDIA_REVISION isntead of KERNEL_VERSION > staging/atomisp: use MEDIA_VERSION instead of KERNEL_VERSION That's said, some of the above shouldn't be using KERNEL_VERSION at all. The V4L2 core sets the version already. So, drivers like cx25821, s3c-camif, bcm2048 and atomisp are likely doing the wrong thing. Thanks, Mauro
