Re: kref refcnt and false positives

[Eric W. Biederman]

Andrew Morton <akpm@xxxxxxxx> writes:

> Guys, we have about 1000000000000000000000000000000 reports of weirdo
> crashes, smashes, bashes and splats in the kref code.  The last thing we
> need is some obscure, tricksy little optimisation which leads legitimate
> uses of the API to mysteriously fail.  
>
> If we are allocating and freeing kref-counted objects at a sufficiently
> high frequency for this thing to make a difference then we should fix that
> instead of trying to suck faster.

Agreed. Correct code maintenance certainly trumps performance.

For the same reason someone reusing the data structure shouldn't
assume the kref code left it in any particular state.

So both sides should be liberal in what they accept.

Eric
-
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/