Re: [vfio-users] Any advice how to minimize vfio device latency *without* virtual interrupt controller?
On Sat, 4 Jun 2022, at 4:37 PM, Roman Mamedov wrote: > On Sat, 4 Jun 2022 16:25:32 +0100 > James Dutton wrote: > >> I ran into problems some time ago, with UDP packets being dropped during >> sustained disk accesses, so the sound device might also be affected by disk >> accesses. > > Did you check to ensure that you're using iothreads? > http://kvmonz.blogspot.com/p/knowledge-disk-performance-hints-tips.html > And native aio as the article also suggests. Thank you, my disks are configured for native IO. My problem is different, how to shorten the interrupt handling path from PCI USB card to VM, without virtual interrupt handling. I suspect that my isolcpus and having all vCPU on the same node where USB card is attached might not be helping. It would be probably appropriate if I had virtual interrupt controller, but nope, there's no AVIC feature in EPYC 7003. I guess this means that interrupts must be handled by the host kernel and then routed to VM. Which host core is doing this handling? B. -- Bronek Kozicki b...@incorrekt.com ___ vfio-users mailing list vfio-users@redhat.com https://listman.redhat.com/mailman/listinfo/vfio-users
Re: [vfio-users] Any advice how to minimize vfio device latency *without* virtual interrupt controller?
On Sat, 4 Jun 2022 16:25:32 +0100 James Dutton wrote: > I ran into problems some time ago, with UDP packets being dropped during > sustained disk accesses, so the sound device might also be affected by disk > accesses. Did you check to ensure that you're using iothreads? http://kvmonz.blogspot.com/p/knowledge-disk-performance-hints-tips.html And native aio as the article also suggests. -- With respect, Roman ___ vfio-users mailing list vfio-users@redhat.com https://listman.redhat.com/mailman/listinfo/vfio-users
Re: [vfio-users] Any advice how to minimize vfio device latency *without* virtual interrupt controller?
On Sat, 4 Jun 2022 at 14:00, Bronek Kozicki wrote: > Hi > > My CPU/motherboard combo does not seem to support virtual interrupt > controller and so far I am not getting much help to fix this (the > motherboard is Gigabyte MZ72-HB0 rev 3.0 and the CPU is EPYC 7413 - > apparently AVIC no longer works on EPYC 7003 CPUs; there is 256GB RAM > attached to each of two sockets, total 512GB). > > I am trying to figure out configuration to minimize the sound latency in > the virtual machines (i.e. minimize crackling) without AVIC. Sound is > provided by an USB audio interface attached to USB card which is passed > through with vfio. USB card model is > https://www.startech.com/en-gb/cards-adapters/pexusb3s42v (it has two > independent USB controllers; a different variant of this card has 4) > > > As a test, try to re-run your tests without any disk access. HDD or SSD. Please tell us if that improves the situation or not. If that improves the audio, you can then look into the disk access area to try to minimise the duration of sustained disk access. For the tests, you can use a ramdisk / tmpfs instead of a physical disk if you need to store/read data somewhere temporarily, and then copy it to a real disk after the test completes. I ran into problems some time ago, with UDP packets being dropped during sustained disk accesses, so the sound device might also be affected by disk accesses. Kind Regards James ___ vfio-users mailing list vfio-users@redhat.com https://listman.redhat.com/mailman/listinfo/vfio-users