From: Rusty Russell <rusty@rustcorp.com.au>
To: Anthony Liguori <anthony@codemonkey.ws>,
"Michael S. Tsirkin" <mst@redhat.com>,
Thomas Lendacky <tahm@linux.vnet.ibm.com>
Cc: Sasha Levin <levinsasha928@gmail.com>,
virtualization@lists.linux-foundation.org,
linux-kernel@vger.kernel.org, avi@redhat.com,
kvm@vger.kernel.org, netdev@vger.kernel.org
Subject: Re: [PATCH 0/3] virtio-net: inline header support
Date: Thu, 04 Oct 2012 13:04:56 +0930 [thread overview]
Message-ID: <87391u3o67.fsf@rustcorp.com.au> (raw)
In-Reply-To: <87mx033u74.fsf@codemonkey.ws>
Anthony Liguori <anthony@codemonkey.ws> writes:
> Rusty Russell <rusty@rustcorp.com.au> writes:
>
>> "Michael S. Tsirkin" <mst@redhat.com> writes:
>>
>>> Thinking about Sasha's patches, we can reduce ring usage
>>> for virtio net small packets dramatically if we put
>>> virtio net header inline with the data.
>>> This can be done for free in case guest net stack allocated
>>> extra head room for the packet, and I don't see
>>> why would this have any downsides.
>>
>> I've been wanting to do this for the longest time... but...
>>
>>> Even though with my recent patches qemu
>>> no longer requires header to be the first s/g element,
>>> we need a new feature bit to detect this.
>>> A trivial qemu patch will be sent separately.
>>
>> There's a reason I haven't done this. I really, really dislike "my
>> implemention isn't broken" feature bits. We could have an infinite
>> number of them, for each bug in each device.
>
> This is a bug in the specification.
>
> The QEMU implementation pre-dates the specification. All of the actual
> implementations of virtio relied on the semantics of s/g elements and
> still do.
lguest fix is pending in my queue. lkvm and qemu are broken; lkvm isn't
ever going to be merged, so I'm not sure what its status is? But I'm
determined to fix qemu, and hence my torture patch to make sure this
doesn't creep in again.
> What's in the specification really doesn't matter when it doesn't agree
> with all of the existing implementations.
>
> Users use implementations, not specifications. The specification really
> ought to be changed here.
I'm sorely tempted, except that we're losing a real optimization because
of this :(
The specification has long contained the footnote:
The current qemu device implementations mistakenly insist that
the first descriptor cover the header in these cases exactly, so
a cautious driver should arrange it so.
I'd like to tie this caveat to the PCI capability change, so this note
will move to the appendix with the old PCI layout.
Cheers,
Rusty.
next prev parent reply other threads:[~2012-10-04 3:34 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-28 9:26 [PATCH 0/3] virtio-net: inline header support Michael S. Tsirkin
2012-09-28 9:26 ` Michael S. Tsirkin
2012-09-28 9:26 ` [PATCH 1/3] virtio: add API to query ring capacity Michael S. Tsirkin
2012-09-28 9:26 ` Michael S. Tsirkin
2012-09-28 9:26 ` [PATCH 2/3] virtio-net: correct capacity math on ring full Michael S. Tsirkin
2012-09-28 9:26 ` Michael S. Tsirkin
2012-10-04 0:24 ` Rusty Russell
2012-09-28 9:26 ` [PATCH 3/3] virtio-net: put virtio net header inline with data Michael S. Tsirkin
2012-09-28 9:26 ` Michael S. Tsirkin
2012-10-03 6:44 ` [PATCH 0/3] virtio-net: inline header support Rusty Russell
2012-10-03 10:53 ` Paolo Bonzini
2012-10-03 10:53 ` Paolo Bonzini
2012-10-04 0:11 ` Rusty Russell
2012-10-04 7:09 ` Paolo Bonzini
2012-10-04 12:51 ` Rusty Russell
2012-10-04 13:23 ` Paolo Bonzini
2012-10-05 5:43 ` Rusty Russell
2012-10-05 5:43 ` Rusty Russell
2012-10-06 12:54 ` Paolo Bonzini
2012-10-06 12:54 ` Paolo Bonzini
2012-10-09 4:59 ` Rusty Russell
2012-10-09 4:59 ` Rusty Russell
2012-10-09 7:27 ` Paolo Bonzini
2012-10-09 7:27 ` Paolo Bonzini
2012-10-11 0:03 ` Rusty Russell
2012-10-11 0:03 ` Rusty Russell
2012-10-11 11:04 ` Michael S. Tsirkin
2012-10-11 11:04 ` Michael S. Tsirkin
2012-10-11 22:37 ` Rusty Russell
2012-10-11 22:37 ` Rusty Russell
2012-10-12 7:38 ` Paolo Bonzini
2012-10-12 7:38 ` Paolo Bonzini
2012-10-12 11:52 ` Cornelia Huck
2012-10-12 11:52 ` Cornelia Huck
2012-10-03 6:44 ` Rusty Russell
2012-10-03 7:10 ` Rusty Russell
2012-10-04 1:24 ` Anthony Liguori
2012-10-04 3:34 ` Rusty Russell [this message]
2012-10-04 4:29 ` Anthony Liguori
2012-10-04 7:44 ` Rusty Russell
2012-10-05 7:47 ` Paolo Bonzini
2012-10-05 7:47 ` Paolo Bonzini
2012-10-08 21:31 ` Michael S. Tsirkin
2012-10-08 21:31 ` Michael S. Tsirkin
2012-10-04 1:35 ` Anthony Liguori
2012-10-04 5:17 ` Rusty Russell
2012-10-08 20:41 ` Michael S. Tsirkin
2012-10-08 20:41 ` Michael S. Tsirkin
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=87391u3o67.fsf@rustcorp.com.au \
--to=rusty@rustcorp.com.au \
--cc=anthony@codemonkey.ws \
--cc=avi@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=levinsasha928@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=tahm@linux.vnet.ibm.com \
--cc=virtualization@lists.linux-foundation.org \
/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.