Messages in this thread | | | From | Linus Torvalds <> | Date | Thu, 23 Feb 2012 14:45:53 -0800 | Subject | Re: [PATCH 00/10] jump label: introduce very_[un]likely + cleanups + docs |
| |
On Thu, Feb 23, 2012 at 2:33 PM, Ingo Molnar <mingo@elte.hu> wrote: > > At the risk of being flamed some more, where does the confusion > stem from?
From the fact that "very" is an English word that means "very", and "unlikely()" and "likely()" are already used in their obvious meaning.
So quite frankly, "very_unlikely()" quite naturally means something else than you are trying to make it mean.
The fact that EVERY SINGLE OTHER OPERATION that worked on that data structure used to be named "jump_label_xyz()" and is now named "static_key_xyz()" is also a big clue, I think. Naming it anything else was always a mistake.
Seriously, I don't understand why you don't just use the obvious name. The data structure is named "static_key". The things that change it are named "static_key_inc()" or something. So a name like "static_key_true()" is simply *better*, isn't it?
It's not just about less confusion, it's actually about just having consistent naming.
Linus
| |