On Tue, Dec 06, 2016 at 11:12:45AM +0100, Christian Borntraeger wrote: > On 12/06/2016 10:20 AM, Stefan Hajnoczi wrote: > > On Mon, Dec 05, 2016 at 09:06:17PM +0100, Christian Borntraeger wrote: > >> On 12/01/2016 08:26 PM, Stefan Hajnoczi wrote: > >>> This patch is based on the algorithm for the kvm.ko halt_poll_ns > >>> parameter in Linux. The initial polling time is zero. > >>> > >>> If the event loop is woken up within the maximum polling time it means > >>> polling could be effective, so grow polling time. > >>> > >>> If the event loop is woken up beyond the maximum polling time it means > >>> polling is not effective, so shrink polling time. > >>> > >>> If the event loop makes progress within the current polling time then > >>> the sweet spot has been reached. > >>> > >>> This algorithm adjusts the polling time so it can adapt to variations in > >>> workloads. The goal is to reach the sweet spot while also recognizing > >>> when polling would hurt more than help. > >>> > >>> Two new trace events, poll_grow and poll_shrink, are added for observing > >>> polling time adjustment. > >>> > >>> Signed-off-by: Stefan Hajnoczi <stefa...@redhat.com> > >> > >> Not sure way, but I have 4 host ramdisks with the same iothread as guest > >> virtio-blk. running fio in the guest on one of these disks will poll, as > >> soon as I have 2 disks in fio I almost always see shrinks (so polling > >> stays at 0) and almost no grows. > > > > Shrinking occurs when polling + ppoll(2) time exceeds poll-max-ns. > > > > What is the value of poll-max-ns > > I used 50000ns as poll value. When using 500000ns it is polling again. > > > and how long is run_poll_handlers_end - run_poll_handlers_begin? > > Too long. I looked again and I realized that I used cache=none without > io=native. After adding io=native things are better. Even with 4 disks > polling still happens. So it seems that the mileage will vary depending > on the settings
Okay, it could be the things I mentioned. The self-tuning can be too conservative and poll less than we'd hope for. On the other hand it's hard to predict the actual ideal polling amount so I guess being conservative is alright. Maybe the algorithm can be refined later. Stefan
signature.asc
Description: PGP signature
