Re: [PATCHSET] block, mempool, percpu: implement percpu mempool andfix blkcg percpu alloc deadlock

[Andrew Morton]

On Thu, 22 Dec 2011 20:40:43 -0500 Vivek Goyal <vgoyal@xxxxxxxxxx> wrote:

> That's why the need of per cpu data structures to make stat collection
> lockless.

btw, (and this is a common refrain): was there any reason for avoiding
using percpu_counters here?

afacit these stats don't have the transient-negative-number problem
because they're purely upcounters, but percpu_counters would be helpful
when num_online_cpus is much less than num_possible_cpus.  Plus it would
avoid implementing yet another hand-coded percpu_counter...



--
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/