From: Jiri Pirko <jiri@resnulli.us>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtualization@lists.linux.dev, jasowang@redhat.com,
xuanzhuo@linux.alibaba.com, eperezma@redhat.com,
parav@nvidia.com, feliu@nvidia.com
Subject: Re: [PATCH virtio 0/8] virtio_pci_modern: allow parallel admin queue commands execution
Date: Mon, 24 Jun 2024 15:36:31 +0200 [thread overview]
Message-ID: <Znl2X2ScZMKwXmo9@nanopsycho.orion> (raw)
In-Reply-To: <20240624091517-mutt-send-email-mst@kernel.org>
Mon, Jun 24, 2024 at 03:16:06PM CEST, mst@redhat.com wrote:
>On Mon, Jun 24, 2024 at 03:09:20PM +0200, Jiri Pirko wrote:
>> Mon, Jun 24, 2024 at 01:07:23PM CEST, mst@redhat.com wrote:
>> >On Mon, Jun 24, 2024 at 11:04:43AM +0200, Jiri Pirko wrote:
>> >> From: Jiri Pirko <jiri@nvidia.com>
>> >>
>> >> Currently the admin queue command execution is serialized by a lock.
>> >> This patchsets lifts this limitation allowing to execute admin queue
>> >> commands in parallel. To do that, admin queue processing needs to be
>> >> converted from polling to interrupt based completion.
>> >>
>> >> Patches #1-#6 are preparations, making things a bit smoother as well.
>> >> Patch #7 implements interrupt based completion for admin queue.
>> >> Patch #8 finally removes the admin queue serialization lock.
>> >
>> >Okay ... I assume this is infrastructure for some other need?
>> >I'll review, but I'd like to see this actually used.
>>
>> No other need. The currently, admin queue cmds are serialized. Note that
>> admin queue is currently created for PF. If you have for example
>> multiple VFs in legacy mode reading and writing pci bar, that virtio
>> vfio driver translates it into admin queue commands. So for example
>> probe on multiple VFs gets serialized. This patchset makes it possible
>> to run in parallel.
>
>
>Well probe is mostly serialized but I think I get what you mean -
I mean probe of multiple VFs that belong under one PF.
>I am guessing you mean something else like legacy access commands.
Yes. For transitional VF, the legacy bar0 access is translated by
virtio vfio driver to admin commands. See virtiovf_pci_bar0_rw().
>>
>> >
>> >
>> >> Jiri Pirko (8):
>> >> virtio_pci: push out single vq find code to vp_find_one_vq_msix()
>> >> virtio_pci_modern: treat vp_dev->admin_vq.info.vq pointer as static
>> >> virtio: push out code to vp_avq_index()
>> >> virtio: create admin queues alongside other virtqueues
>> >> virtio_pci_modern: create admin queue of queried size
>> >> virtio_pci_modern: pass cmd as an identification token
>> >> virtio_pci_modern: use completion instead of busy loop to wait on
>> >> admin cmd result
>> >> virtio_pci_modern: remove admin queue serialization lock
>> >>
>> >> drivers/virtio/virtio.c | 28 +----
>> >> drivers/virtio/virtio_pci_common.c | 109 ++++++++++++++------
>> >> drivers/virtio/virtio_pci_common.h | 9 +-
>> >> drivers/virtio/virtio_pci_modern.c | 160 ++++++++++++-----------------
>> >> include/linux/virtio.h | 2 +
>> >> include/linux/virtio_config.h | 2 -
>> >> 6 files changed, 150 insertions(+), 160 deletions(-)
>> >>
>> >> --
>> >> 2.45.1
>> >
>
prev parent reply other threads:[~2024-06-24 13:36 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-24 9:04 [PATCH virtio 0/8] virtio_pci_modern: allow parallel admin queue commands execution Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 1/8] virtio_pci: push out single vq find code to vp_find_one_vq_msix() Jiri Pirko
2024-06-24 10:52 ` Heng Qi
2024-06-24 13:11 ` Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 2/8] virtio_pci_modern: treat vp_dev->admin_vq.info.vq pointer as static Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 3/8] virtio: push out code to vp_avq_index() Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 4/8] virtio: create admin queues alongside other virtqueues Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 5/8] virtio_pci_modern: create admin queue of queried size Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 6/8] virtio_pci_modern: pass cmd as an identification token Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 7/8] virtio_pci_modern: use completion instead of busy loop to wait on admin cmd result Jiri Pirko
2024-06-24 11:30 ` Michael S. Tsirkin
2024-06-24 13:10 ` Jiri Pirko
2024-06-25 11:07 ` Jiri Pirko
2024-06-25 12:53 ` Parav Pandit
2024-06-25 14:29 ` Jiri Pirko
2024-06-24 11:34 ` Michael S. Tsirkin
2024-06-24 13:10 ` Jiri Pirko
2024-06-24 9:04 ` [PATCH virtio 8/8] virtio_pci_modern: remove admin queue serialization lock Jiri Pirko
2024-06-24 9:53 ` [PATCH virtio 0/8] virtio_pci_modern: allow parallel admin queue commands execution Heng Qi
2024-06-24 11:23 ` Michael S. Tsirkin
2024-06-24 13:46 ` Jiri Pirko
2024-06-24 13:55 ` Michael S. Tsirkin
2024-06-24 14:51 ` Jiri Pirko
2024-06-24 15:16 ` Michael S. Tsirkin
2024-06-25 2:11 ` Heng Qi
2024-06-25 6:38 ` Jiri Pirko
2024-06-25 6:41 ` Parav Pandit
2024-06-25 7:29 ` Michael S. Tsirkin
2024-06-25 8:01 ` Jiri Pirko
2024-06-25 7:20 ` Michael S. Tsirkin
2024-06-25 7:25 ` Parav Pandit
2024-06-24 15:10 ` Jiri Pirko
2024-06-24 15:23 ` Michael S. Tsirkin
2024-06-24 15:45 ` Jiri Pirko
2024-06-24 11:07 ` Michael S. Tsirkin
2024-06-24 13:09 ` Jiri Pirko
2024-06-24 13:16 ` Michael S. Tsirkin
2024-06-24 13:36 ` Jiri Pirko [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=Znl2X2ScZMKwXmo9@nanopsycho.orion \
--to=jiri@resnulli.us \
--cc=eperezma@redhat.com \
--cc=feliu@nvidia.com \
--cc=jasowang@redhat.com \
--cc=mst@redhat.com \
--cc=parav@nvidia.com \
--cc=virtualization@lists.linux.dev \
--cc=xuanzhuo@linux.alibaba.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).