On Wed, Jun 07, 2017 at 03:04:11PM -0700, Omar Sandoval wrote:

On Sat, Jun 03, 2017 at 04:04:02PM +0200, Christoph Hellwig wrote:

quoted

Currently we only create hctx for online CPUs, which can lead to a lot
of churn due to frequent soft offline / online operations.  Instead
allocate one for each present CPU to avoid this and dramatically simplify
the code.

Signed-off-by: Christoph Hellwig <hch@lst.de>

Oh man, this cleanup is great. Did you run blktests on this? block/008
does a bunch of hotplugging while I/O is running.

I haven't run blktests yet, in fact when I did the work blktests didn't
exist yet.   But thanks for the reminder, I'll run it now.