Re: [RFC 0/8] virtio: Improve boot time of virtio-scsi-pci and virtio-blk-pci

[Greg Kurz]

On Mon, 29 Mar 2021 18:35:16 +0100
Stefan Hajnoczi <stefanha@redhat.com> wrote:

> On Thu, Mar 25, 2021 at 04:07:27PM +0100, Greg Kurz wrote:
> > Now that virtio-scsi-pci and virtio-blk-pci map 1 virtqueue per vCPU,
> > a serious slow down may be observed on setups with a big enough number
> > of vCPUs.
> > 
> > Exemple with a pseries guest on a bi-POWER9 socket system (128 HW threads):
> > 
> > 1           0m20.922s       0m21.346s
> > 2           0m21.230s       0m20.350s
> > 4           0m21.761s       0m20.997s
> > 8           0m22.770s       0m20.051s
> > 16          0m22.038s       0m19.994s
> > 32          0m22.928s       0m20.803s
> > 64          0m26.583s       0m22.953s
> > 128         0m41.273s       0m32.333s
> > 256         2m4.727s        1m16.924s
> > 384         6m5.563s        3m26.186s
> > 
> > Both perf and gprof indicate that QEMU is hogging CPUs when setting up
> > the ioeventfds:
> > 
> >  67.88%  swapper         [kernel.kallsyms]  [k] power_pmu_enable
> >   9.47%  qemu-kvm        [kernel.kallsyms]  [k] smp_call_function_single
> >   8.64%  qemu-kvm        [kernel.kallsyms]  [k] power_pmu_enable
> > =>2.79%  qemu-kvm        qemu-kvm           [.] 
> > memory_region_ioeventfd_before
> > =>2.12%  qemu-kvm        qemu-kvm           [.] 
> > address_space_update_ioeventfds
> >   0.56%  kworker/8:0-mm  [kernel.kallsyms]  [k] smp_call_function_single
> > 
> > address_space_update_ioeventfds() is called when committing an MR
> > transaction, i.e. for each ioeventfd with the current code base,
> > and it internally loops on all ioventfds:
> > 
> > static void address_space_update_ioeventfds(AddressSpace *as)
> > {
> > [...]
> >     FOR_EACH_FLAT_RANGE(fr, view) {
> >         for (i = 0; i < fr->mr->ioeventfd_nb; ++i) {
> > 
> > This means that the setup of ioeventfds for these devices has
> > quadratic time complexity.
> > 
> > This series introduce generic APIs to allow batch creation and deletion
> > of ioeventfds, and converts virtio-blk and virtio-scsi to use them. This
> > greatly improves the numbers:
> > 
> > 1           0m21.271s       0m22.076s
> > 2           0m20.912s       0m19.716s
> > 4           0m20.508s       0m19.310s
> > 8           0m21.374s       0m20.273s
> > 16          0m21.559s       0m21.374s
> > 32          0m22.532s       0m21.271s
> > 64          0m26.550s       0m22.007s
> > 128         0m29.115s       0m27.446s
> > 256         0m44.752s       0m41.004s
> > 384         1m2.884s        0m58.023s
> 
> Excellent numbers!
> 
> I wonder if the code can be simplified since
> memory_region_transaction_begin/end() supports nesting. Why not call
> them directly from the device model instead of introducing callbacks in
> core virtio and virtio-pci code?
> 

It seems a bit awkward that the device model should assume a memory
transaction is needed to setup host notifiers, which are ioeventfds
under the hood but the device doesn't know that.

> Also, do you think there are other opportunities to have a long
> transaction to batch up machine init, device hotplug, etc? It's not
> clear to me when transactions must be ended. Clearly it's necessary to

The transaction *must* be ended before calling
virtio_bus_cleanup_host_notifier() because
address_space_add_del_ioeventfds(), called when
finishing the transaction, needs the "to-be-closed"
eventfds to be still open, otherwise the KVM_IOEVENTFD 
ioctl() might fail with EBADF.

See this change in patch 3:

@@ -315,6 +338,10 @@ static void 
virtio_bus_unset_and_cleanup_host_notifiers(VirtioBusState *bus,
 
     for (i = 0; i < nvqs; i++) {
         virtio_bus_set_host_notifier(bus, i + n_offset, false);
+    }
+    /* Let address_space_update_ioeventfds() run before closing ioeventfds */
+    virtio_bus_set_host_notifier_commit(bus);
+    for (i = 0; i < nvqs; i++) {
         virtio_bus_cleanup_host_notifier(bus, i + n_offset);
     }
 }

Maybe I should provide more details why we're doing that ?

> end the transaction if we need to do something that depends on the
> MemoryRegion, eventfd, etc being updated. But most of the time there is
> no immediate need to end the transaction and more code could share the
> same transaction before we go back to the event loop or vcpu thread.
> 

I can't tell for all scenarios that involve memory transactions but
it seems this is definitely not the case for ioeventfds : the rest
of the code expects the transaction to be complete.

> Stefan

Thanks for the review !

Cheers,

--
Greg

pgpfdfb3jqazL.pgp
Description: OpenPGP digital signature