Re: dm-crypt parallelization patches

[Vivek Goyal]

On Tue, Apr 09, 2013 at 11:57:21AM -0700, Tejun Heo wrote:

[..]
> And destroy all per-ioc and cgroup logics in block layer in the
> process.

Oh, I am in no-way suggesting don't use bio_associate_current(). I am 
just trying to analyze the performance issue right now and saying that
as far as performance is concenred, one will get it back even if one
does not use bio_associate_current().

Yes but one needs to use bio_associate_current() to make sure bio's
are attributed to right cgroup and associate bio to right task. This
should help solving the long standing issue of task losing its ioprio
if dm-crypt target is in the stack.

Thanks
Vivek
--
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/