On Friday 12 July 2002 21:24, Arnd Bergmann wrote:
> Daniel Phillips wrote:
>
> > Any idea how one might implement NEVER_SLEEPS()? Maybe as:
>
> Why would you want that? AFAICS there are two kinds of "never
> sleeping" functions: 1. those that don't sleep but don't care
> about it and 2. those that must not sleep because a lock is
> held.
>
> For 1. no point marking it because it might change without
> being a bug. You also don't want to mark every function
> in the kernel SLEEPS() or NEVER_SLEEPS().
>
> For 2. we already have MUST_HOLD(foo) or similar, which implicitly
> means it can never sleep. The same is true for functions
> with spinlocks or preempt_disable around their body.
Thanks for that.
So far, only the MUST_HOLD style of executable locking documentation has
really survived scrutiny. It needs some variants: MUST_HOLD_READ,
MUST_HOLD_WRITE, MUST_HOLD_SEM, MUST_HOLD_READ_SEM and MUST_HOLD_WRITE_SEM,
or names to that effect.
-- Daniel - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jul 15 2002 - 22:00:23 EST