Re: [GIT PULL] slab fixes for 3.2-rc4
From: Tejun Heo
Date: Tue Dec 20 2011 - 15:29:07 EST
Hello, Linus.
On Tue, Dec 20, 2011 at 11:28:25AM -0800, Linus Torvalds wrote:
> > Description for 7340a0b152 "this_cpu: Introduce this_cpu_ptr() and
> > generic this_cpu_* operations" should explain the above three.
>
> I don't think that's relevant.
>
> Sure, they have semantics, but the semantics are stupid and wrong.
> Whether they are documented or not isn't even the issue.
I was trying to point Pekka to documentation so that at least the
existing semantics are clear.
> Being "generic" is not actually a good thing. Not when we're talking
> about random details like this.
Yeah, I generally agree that reducing the API would be great. Given
the usage, I think (or at least hope) dropping preemption protected
ones wouldn't hurt much but it might be worthwhile to keep
__this_cpu_*() - the ones which expect the caller to take care of
synchronization - w/ assertion on irq disabled.
Christoph, what do you think? What would be the minimal set that you
can work with?
Thanks.
--
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/