Messages in this thread | | | Date | Sat, 2 May 2015 23:09:12 +0800 | Subject | Re: [PATCH v6] block: loop: avoiding too many pending per work I/O | From | Ming Lei <> |
| |
On Fri, May 1, 2015 at 11:47 PM, Tejun Heo <tj@kernel.org> wrote: > On Fri, May 01, 2015 at 08:05:45AM -0700, Christoph Hellwig wrote: >> > Maybe just cap max_active to NR_OF_LOOP_DEVS * 16 or sth? But idk, >> > how many concurrent workers are we talking about and why are we >> > capping per-queue concurrency from worker pool side instead of command >> > tag side? >> >> Also we probably should have per device workqueues to start with.. > > Yeah, that's an option. The only thing is that each workqueue would
I guess we have to do that because of nested loop devices.
> have to be tagged WQ_RESCUER and end up with separate rescuer task, > which usually isn't big a deal but there are setups where a lot of > loop devices are used and it may sting a bit.
The work queue can be allocated just before the loop is to be used and destroyed when it needn't.
I will figure out one patch to do that.
> > Thanks. > > -- > tejun > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/
| |