Hi, Could you explain were you are going with this because this is a lot of churn that takes a lot of my reviewing time and this doesn't seem to end soon. I want that driver to stop growing, maybe by creating a library. Some of the supported RTCs doesn't share much more than the time and date layout and should be in a separate driver. On 25/08/2017 at 21:30:09 +0200, Heiner Kallweit wrote: > Final goal of the refactoring is to abstract everything that the chips > have in common and handle it in generic code. Then we can get rid of > all the "switch (chip)" and "if (chip == xxx)" code. > > To give one example: > A lot of chips have a bit for setting 12hr / 24hr mode. However some > chips have this config bit in the timekeeping registers, others in > a config register, and on some chips it's inverted. > But the functionality of the bit is always the same. > > Ultimately adding support for a chip just requires to add one config > structure member. > > The way to reach this goal is a long one and to faciliate reviewing > the patches I'll split them into series of 5 to 10 patches. > > Heiner Kallweit (5): > rtc: ds1307: factor out determining the chip type > rtc: ds1307: factor out trickle charger initialization > rtc: ds1307: factor out fixing the weekday > rtc: ds1307: introduce constants for the timekeeping register masks > rtc: ds1307: improve ds1307_set_time to respect config flag bits > > drivers/rtc/rtc-ds1307.c | 256 ++++++++++++++++++++++++++--------------------- > 1 file changed, 140 insertions(+), 116 deletions(-) > > -- > 2.14.1 > -- Alexandre Belloni, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com
