> Il giorno 09 gen 2018, alle ore 20:53, Jens Axboe <axboe@xxxxxxxxx> ha scritto: > > On 1/9/18 12:52 PM, Tejun Heo wrote: >> Hello, Paolo. >> >> On Thu, Jan 04, 2018 at 08:00:02PM +0100, Paolo Valente wrote: >>> The solution for the second type of parameters may prove useful to >>> unify also the computation of statistics for the throttling policy. >>> >>> Does this proposal sound reasonable? >> >> So, the above should work too but I wonder whether we could do this >> simpler. Frankly, I wouldn't mind if cfq and bfq can't be mixed on a >> system - e.g. they can be built together but you can't enable bfq on >> some devides and cfq on others. If we do that, all we need to do is >> just removing / adding cftypes when either gets activated which cgroup >> already does. > > Not sure that would fly, since cfq is legacy and bfq is mq. You don't > always have a free choice of which one to use... > Yep. So, do you guys think that our proposal may be ok? We are waiting just for the green light to start implementing it. Thanks, Paolo > -- > Jens Axboe >
