On 06/06/2016 09:54 AM, Jan Beulich wrote: >>>> On 06.06.16 at 15:09, <[email protected]> wrote: >> On 06/06/2016 04:47 AM, Jan Beulich wrote: >>> It's identical to bar_init() now. >>> >>> Signed-off-by: Jan Beulich <[email protected]> >>> --- >>> I'm sorry for this 3/2 - I only now noticed that this additional >>> simplification is now possible. >> I wonder whether we should also move content of read_dev_bar() into >> bar_init(). Especially given that it's not really reading a BAR but >> rather initializing the stashed value. > I had considered that too, but then thought the splitting out of > that logic could as well stay. If we were to do that, I'd in fact > prefer merging patches 2 and 3 (plus this additional adjustment).
If you could do that it would be great. (Again, mostly because I think the name is misleading and renaming it to something like dev_bar_init() would also not be good since we already have bar_init()). Thanks. -boris
