Okay. This way with each commit the version can compile and I will be able to define the subsystem appropriately. Thank you for the clarification Lars. On Tue, Sep 12, 2017 at 08:14:18PM +0200, Lars-Peter Clausen wrote: > On 09/12/2017 08:06 PM, Julia Lawall wrote: > > > > > > On Tue, 12 Sep 2017, himanshi wrote: > > > >> Thanks for the review Daniel! I will change the imperative mood for the commit > >> message once the other changes are finalised too and as suggested by Julia, > >> would try to make the description specific than general. > >> > >> I tried to think of adding subsystem to the commit subject but could not > >> conclude any because of the files involved. > >> I like the idea of splitting the patch into 2 as you suggested but I > >> have a doubt that adding the new MACROS to different sysfs files can be put into 1 > >> patch with the subsystem you mentioned but changing the existing > >> IIO_DEVICE_ATTR_NAMED to use IIO_ATTR_NAMED (sysfs file again) would be included > >> in the second patch if I am not wrong. So would it be fine to keep the > >> subsystem as iio for the second patch? > > > > Indeed, the kernel has to compile after every commit. Unless you change > > the name of the macro, to allow the old and new versions to co-exist, it > > seems hard to break up such a patch. > > We can still split things into two parts. One patch introducing __ATTR_NAMED > in the device driver core and then another patch making use of that macro in > the IIO subsystem. _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
