Re: [PATCH 15/20] block: merge struct block_device and struct hd_struct

[Jan Kara <jack@xxxxxxx>]

On Fri 20-11-20 10:15:46, Christoph Hellwig wrote:
> On Thu, Nov 19, 2020 at 03:39:21PM +0100, Jan Kara wrote:
> > This patch is kind of difficult to review due to the size of mostly
> > mechanical changes mixed with not completely mechanical changes. Can we
> > perhaps split out the mechanical bits? E.g. the rq->part => rq->bdev
> > renaming is mechanical and notable part of the patch. Similarly the
> > part->foo => part->bd_foo bits...
> 
> We'd end with really weird patches that way.  Never mind that I'm not
> even sure how we could mechnically do the renaming.

Well, I believe coccinelle should be able to do the renaming automatically.

> > Also I'm kind of wondering: AFAIU the new lifetime rules, gendisk holds
> > bdev reference and bdev is created on gendisk allocation so bdev lifetime is
> > strictly larger than gendisk lifetime. But what now keeps bdev->bd_disk
> > reference safe in presence device hot unplug? In most cases we are still
> > protected by gendisk reference taken in __blkdev_get() but how about
> > disk->lookup_sem and disk->flags dereferences before we actually grab the
> > reference?
> 
> Good question.  I'll need to think about this a bit more.

My thinking was that you could use

kobject_get_unless_zero(bdev->bd_device->kobj)

and after you hold this reference, you can do everything else safely. In
this case it is really useful that device is embedded in block_dev and
not in gendisk itself...

                                                                Honza
-- 
Jan Kara <jack@xxxxxxxx>
SUSE Labs, CR