On Mon, Feb 03, 2020 at 06:46:41PM +0100, Christoph Hellwig wrote:
> On Sat, Jan 18, 2020 at 08:28:38PM +1100, Dave Chinner wrote:
> > I think it's pretty gross, actually. It makes the same mistake made
> > with locking in the old direct IO code - it encodes specific lock
> > operations via flags
On Sat, Jan 18, 2020 at 08:28:38PM +1100, Dave Chinner wrote:
> I think it's pretty gross, actually. It makes the same mistake made
> with locking in the old direct IO code - it encodes specific lock
> operations via flags into random locations in the DIO path. This is
> a very slippery slope,
On Thu, Jan 16, 2020 at 03:00:04PM +0100, Jan Kara wrote:
> I'd like to note that using i_dio_count has also one advantage you didn't
> mention. For AIO case, if you need to hold i_rwsem in exclusive mode,
> holding the i_rwsem just for submission part is a significant performance
> advantage