From: Greg Kurz <groug@kaod.org>
To: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Fam Zheng <fam@euphon.net>, Kevin Wolf <kwolf@redhat.com>,
qemu-block@nongnu.org, "Michael S. Tsirkin" <mst@redhat.com>,
qemu-devel@nongnu.org, Max Reitz <mreitz@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>,
David Gibson <david@gibson.dropbear.id.au>
Subject: Re: [RFC 0/8] virtio: Improve boot time of virtio-scsi-pci and virtio-blk-pci
Date: Tue, 30 Mar 2021 15:15:27 +0200 [thread overview]
Message-ID: <20210330151527.0b5d2870@bahia.lan> (raw)
In-Reply-To: <YGIP1HKa5NsHqnj7@stefanha-x1.localdomain>
[-- Attachment #1: Type: text/plain, Size: 4296 bytes --]
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
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
prev parent reply other threads:[~2021-03-30 13:17 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-25 15:07 [RFC 0/8] virtio: Improve boot time of virtio-scsi-pci and virtio-blk-pci Greg Kurz
2021-03-25 15:07 ` [RFC 1/8] memory: Allow eventfd add/del without starting a transaction Greg Kurz
2021-03-29 17:03 ` Stefan Hajnoczi
2021-03-30 7:47 ` Greg Kurz
2021-03-30 10:16 ` Stefan Hajnoczi
2021-03-25 15:07 ` [RFC 2/8] virtio: Introduce virtio_bus_set_host_notifiers() Greg Kurz
2021-03-29 17:06 ` Stefan Hajnoczi
2021-03-25 15:07 ` [RFC 3/8] virtio: Add API to batch set host notifiers Greg Kurz
2021-03-29 17:10 ` Stefan Hajnoczi
2021-03-30 10:17 ` Greg Kurz
2021-03-30 13:55 ` Stefan Hajnoczi
2021-03-30 14:17 ` Greg Kurz
2021-03-31 14:47 ` Stefan Hajnoczi
2021-03-31 16:21 ` Greg Kurz
2021-03-25 15:07 ` [RFC 4/8] virtio-pci: Batch add/del ioeventfds in a single MR transaction Greg Kurz
2021-03-29 17:24 ` Stefan Hajnoczi
2021-03-30 10:29 ` Greg Kurz
2021-03-25 15:07 ` [RFC 5/8] virtio-blk: Fix rollback path in virtio_blk_data_plane_start() Greg Kurz
2021-03-29 17:25 ` Stefan Hajnoczi
2021-03-25 15:07 ` [RFC 6/8] virtio-blk: Use virtio_bus_set_host_notifiers() Greg Kurz
2021-03-29 17:26 ` Stefan Hajnoczi
2021-03-25 15:07 ` [RFC 7/8] virtio-scsi: Set host notifiers and callbacks separately Greg Kurz
2021-03-25 15:07 ` [RFC 8/8] virtio-scsi: Use virtio_bus_set_host_notifiers() Greg Kurz
2021-03-29 17:28 ` Stefan Hajnoczi
2021-03-25 17:05 ` [RFC 0/8] virtio: Improve boot time of virtio-scsi-pci and virtio-blk-pci Michael S. Tsirkin
2021-03-25 17:43 ` Stefan Hajnoczi
2021-03-25 20:51 ` Greg Kurz
2021-03-25 18:05 ` Greg Kurz
2021-03-29 17:35 ` Stefan Hajnoczi
2021-03-30 13:15 ` Greg Kurz [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210330151527.0b5d2870@bahia.lan \
--to=groug@kaod.org \
--cc=david@gibson.dropbear.id.au \
--cc=fam@euphon.net \
--cc=kwolf@redhat.com \
--cc=mreitz@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).