All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Fam Zheng <famz@redhat.com>
Cc: kwolf@redhat.com, stefanha@redhat.com,
	Alberto Garcia <berto@igalia.com>,
	qemu-devel@nongnu.org, qemu-block@nongnu.org
Subject: Re: [Qemu-devel] [Qemu-block] [PATCH v2 11/11] block: Only poll block layer fds in bdrv_aio_poll
Date: Fri, 11 Sep 2015 12:46:06 +0200	[thread overview]
Message-ID: <55F2B0EE.7030407@redhat.com> (raw)
In-Reply-To: <20150911104055.GB9241@ad.nay.redhat.com>



On 11/09/2015 12:40, Fam Zheng wrote:
> On Fri, 09/11 11:54, Paolo Bonzini wrote:
>>
>>
>> On 11/09/2015 11:44, Fam Zheng wrote:
>>>>>> That would be a step back.  Using GSource is useful because it lets
>>>>>> you integrate libraries such as GTK+.
>>>>>
>>>>> Can we move GTK to a separate GSource thread?
>>>>
>>>> I think that GTK should always run in the main thread, or at least the
>>>> one running the default main loop / GMainContext.
>>>
>>> Yeah it's basically GMainContext staying in the main thread and
>>> block/net/chardev I/O put in a new AioContext thread.
>>
>> Why?  The point of an event loop is that you can multiplex everything on
>> the same thread.  Unless we have specific needs (e.g. scalability) one
>> thread is the way to go and keep things simple.
> 
> The reason is scalability. :)

Scalability of what?  If virtio-net or virtio-serial needs to be more
scalable, putting all of them into a non-main-loop thread will not make
things more scalable, because you have a single thread anyway.  You'd
need to go BQL-free and allow an arbitrary number.

> Moving things to AIO isn't deviation, it's more about enabling of dataplane and
> epoll. That's why block was moved to AioContext, and I think we can do similar
> for net and serial, the difference is that as a start, they don't need to be
> fully BQL-free like virtio-blk and scsi. But by running in an aio_poll() loop,
> they can better performance because of epoll.

Isn't that what your "iohandler.c with AioHandler" already does?  True,
it would be epoll-within-poll, not pure poll.  But if you need epoll,
you might as well go BQL-free.

Paolo

  reply	other threads:[~2015-09-11 10:46 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-29  4:42 [Qemu-devel] [PATCH v2 00/11] aio: Introduce handler type to fix nested aio_poll for dataplane Fam Zheng
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 01/11] aio: Introduce "type" in aio_set_fd_handler and aio_set_event_notifier Fam Zheng
2015-08-27 13:50   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 02/11] aio: Save type to AioHandler Fam Zheng
2015-08-27 13:50   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 03/11] block: Mark fd handlers as "protocol" Fam Zheng
2015-08-27 13:53   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-09-07  4:43     ` Fam Zheng
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 04/11] nbd: Mark fd handlers client type as "nbd server" Fam Zheng
2015-08-27 14:02   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 05/11] aio: Mark ctx->notifier's client type as "context" Fam Zheng
2015-08-27 17:12   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 06/11] dataplane: Mark host notifiers' client type as "dataplane" Fam Zheng
2015-08-27 17:14   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 07/11] aio-posix: introduce aio_{disable, enable}_clients Fam Zheng
2015-08-27 17:23   ` Stefan Hajnoczi
2015-09-07  5:26     ` Fam Zheng
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 08/11] aio-win32: Implement " Fam Zheng
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 09/11] block: Introduce bdrv_aio_poll Fam Zheng
2015-08-27 17:25   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-08-28 11:50   ` Stefan Hajnoczi
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 10/11] block: Replace nested aio_poll with bdrv_aio_poll Fam Zheng
2015-08-28 11:50   ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-07-29  4:42 ` [Qemu-devel] [PATCH v2 11/11] block: Only poll block layer fds in bdrv_aio_poll Fam Zheng
2015-07-29  7:36   ` Paolo Bonzini
2015-07-29  7:37   ` Paolo Bonzini
2015-07-29 10:57     ` Fam Zheng
2015-07-29 11:02       ` Paolo Bonzini
2015-07-29 11:53         ` Fam Zheng
2015-07-29 12:03           ` Paolo Bonzini
2015-07-30  1:35             ` Fam Zheng
2015-07-30 13:22               ` Paolo Bonzini
2015-09-09  3:22             ` Fam Zheng
2015-09-11  8:15               ` Paolo Bonzini
2015-09-11  9:14                 ` Fam Zheng
2015-09-11  9:36                   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2015-09-11  9:43                     ` Daniel P. Berrange
2015-09-11  9:44                     ` Fam Zheng
2015-09-11  9:54                       ` Paolo Bonzini
2015-09-11 10:40                         ` Fam Zheng
2015-09-11 10:46                           ` Paolo Bonzini [this message]
2015-09-11 11:01                             ` Fam Zheng
2015-09-11 11:02                               ` Paolo Bonzini
2015-09-11 11:12                                 ` Fam Zheng
2015-09-11  9:45                     ` Paolo Bonzini
2015-07-29  7:38 ` [Qemu-devel] [PATCH v2 00/11] aio: Introduce handler type to fix nested aio_poll for dataplane Paolo Bonzini
2015-08-28 11:53 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-09-07  6:28   ` Fam Zheng
2015-09-11 10:39 ` [Qemu-devel] " Kevin Wolf
2015-09-11 11:46   ` Fam Zheng
2015-09-11 12:22     ` Kevin Wolf
2015-09-14  7:27       ` Fam Zheng
2015-09-14  8:40         ` Kevin Wolf
2015-09-28  9:31       ` Stefan Hajnoczi

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=55F2B0EE.7030407@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=berto@igalia.com \
    --cc=famz@redhat.com \
    --cc=kwolf@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.