Am 18.12.23 um 15:49 schrieb Paolo Bonzini: > On Mon, Dec 18, 2023 at 1:41 PM Fiona Ebner <f.eb...@proxmox.com> wrote: >> I think it's because of nested drains, because when additionally >> checking that the drain count is zero and only executing the loop then, >> that issue doesn't seem to manifest > > But isn't virtio_scsi_drained_end only run if bus->drain_count == 0? > > if (bus->drain_count-- == 1) { > trace_scsi_bus_drained_end(bus, sdev); > if (bus->info->drained_end) { > bus->info->drained_end(bus); > } > } >
You're right. Sorry, I must've messed up my testing yesterday :( Sometimes the CPU spikes are very short-lived. Now I see the same issue with both variants. Unfortunately, I haven't been able to figure out why it happens yet. Best Regards, Fiona