From: Anthony Liguori <anthony@codemonkey.ws>
To: Blue Swirl <blauwirbel@gmail.com>
Cc: kvm-devel@lists.sourceforge.net, qemu-devel@nongnu.org,
Marcelo Tosatti <marcelo@kvack.org>,
Paul Brook <paul@codesourcery.com>,
Aurelien Jarno <aurelien@aurel32.net>
Subject: Re: [Qemu-devel] [PATCH 1/5] PCI DMA API (v3)
Date: Sat, 19 Apr 2008 15:02:34 -0500 [thread overview]
Message-ID: <480A4FDA.3070706@codemonkey.ws> (raw)
In-Reply-To: <f43fc5580804191240i4dc7f820s77a36678e30e7dae@mail.gmail.com>
Blue Swirl wrote:
> On 4/17/08, Anthony Liguori <aliguori@us.ibm.com> wrote:
>
>> Yes, the vector version of packet receive is tough. I'll take a look at
>> your patch. Basically, you need to associate a set of RX vectors with each
>> VLANClientState and then when it comes time to deliver a packet to the VLAN,
>> before calling fd_read, see if there is an RX vector available for the
>> client.
>>
>> In the case of tap, I want to optimize further and do the initial readv()
>> to one of the clients RX buffers and then copy that RX buffer to the rest of
>> the clients if necessary.
>>
>
> The vector versions should also help SLIRP to add IP and Ethernet
> headers to the incoming packets.
>
Yeah, I'm hoping that with my posted linux-aio interface, I can add
vector support since linux-aio has a proper asynchronous vector function.
Are we happy with the DMA API? If so, we should commit it now so we can
start adding proper vector interfaces for net/block.
Regards,
Anthony Liguori
> I made an initial version of the vectored AIO SCSI with ESP. It does
> not work, but I can see that just using the vectors won't give too
> much extra performance, because at least initially the vector length
> is 1. Collecting the statuses may be tricky.
>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
> Don't miss this year's exciting event. There's still time to save $100.
> Use priority code J8TL2D2.
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> ------------------------------------------------------------------------
>
> _______________________________________________
> kvm-devel mailing list
> kvm-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/kvm-devel
>
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
WARNING: multiple messages have this Message-ID (diff)
From: Anthony Liguori <anthony@codemonkey.ws>
To: Blue Swirl <blauwirbel@gmail.com>
Cc: Anthony Liguori <aliguori@us.ibm.com>,
kvm-devel@lists.sourceforge.net, qemu-devel@nongnu.org,
Marcelo Tosatti <marcelo@kvack.org>,
Paul Brook <paul@codesourcery.com>,
Aurelien Jarno <aurelien@aurel32.net>
Subject: Re: [kvm-devel] [Qemu-devel] [PATCH 1/5] PCI DMA API (v3)
Date: Sat, 19 Apr 2008 15:02:34 -0500 [thread overview]
Message-ID: <480A4FDA.3070706@codemonkey.ws> (raw)
In-Reply-To: <f43fc5580804191240i4dc7f820s77a36678e30e7dae@mail.gmail.com>
Blue Swirl wrote:
> On 4/17/08, Anthony Liguori <aliguori@us.ibm.com> wrote:
>
>> Yes, the vector version of packet receive is tough. I'll take a look at
>> your patch. Basically, you need to associate a set of RX vectors with each
>> VLANClientState and then when it comes time to deliver a packet to the VLAN,
>> before calling fd_read, see if there is an RX vector available for the
>> client.
>>
>> In the case of tap, I want to optimize further and do the initial readv()
>> to one of the clients RX buffers and then copy that RX buffer to the rest of
>> the clients if necessary.
>>
>
> The vector versions should also help SLIRP to add IP and Ethernet
> headers to the incoming packets.
>
Yeah, I'm hoping that with my posted linux-aio interface, I can add
vector support since linux-aio has a proper asynchronous vector function.
Are we happy with the DMA API? If so, we should commit it now so we can
start adding proper vector interfaces for net/block.
Regards,
Anthony Liguori
> I made an initial version of the vectored AIO SCSI with ESP. It does
> not work, but I can see that just using the vectors won't give too
> much extra performance, because at least initially the vector length
> is 1. Collecting the statuses may be tricky.
>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
> Don't miss this year's exciting event. There's still time to save $100.
> Use priority code J8TL2D2.
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> ------------------------------------------------------------------------
>
> _______________________________________________
> kvm-devel mailing list
> kvm-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/kvm-devel
>
next prev parent reply other threads:[~2008-04-19 20:02 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-04-15 22:11 [PATCH 1/5] PCI DMA API (v3) Anthony Liguori
2008-04-15 22:11 ` [Qemu-devel] " Anthony Liguori
2008-04-15 22:11 ` [PATCH 2/5] virtio for QEMU (v3) Anthony Liguori
2008-04-15 22:11 ` [Qemu-devel] " Anthony Liguori
2008-04-15 22:11 ` [PATCH 3/5] virtio network driver (v3) Anthony Liguori
2008-04-15 22:11 ` [Qemu-devel] " Anthony Liguori
2008-04-15 22:11 ` [PATCH 4/5] virtio block " Anthony Liguori
2008-04-15 22:11 ` [Qemu-devel] " Anthony Liguori
2008-04-15 22:11 ` [PATCH 5/5] virtio balloon " Anthony Liguori
2008-04-15 22:11 ` [Qemu-devel] " Anthony Liguori
2008-04-16 19:51 ` [Qemu-devel] [PATCH 1/5] PCI DMA API (v3) Blue Swirl
2008-04-16 19:51 ` Blue Swirl
2008-04-16 19:54 ` Anthony Liguori
2008-04-16 19:54 ` Anthony Liguori
2008-04-17 19:27 ` Blue Swirl
2008-04-17 19:27 ` Blue Swirl
2008-04-17 20:05 ` Anthony Liguori
2008-04-17 20:05 ` Anthony Liguori
2008-04-19 19:40 ` Blue Swirl
2008-04-19 19:40 ` Blue Swirl
2008-04-19 20:02 ` Anthony Liguori [this message]
2008-04-19 20:02 ` [kvm-devel] " Anthony Liguori
2008-04-20 6:42 ` Blue Swirl
2008-04-20 6:42 ` [kvm-devel] " Blue Swirl
2008-04-20 19:29 ` Anthony Liguori
2008-04-20 19:29 ` [kvm-devel] " Anthony Liguori
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=480A4FDA.3070706@codemonkey.ws \
--to=anthony@codemonkey.ws \
--cc=aurelien@aurel32.net \
--cc=blauwirbel@gmail.com \
--cc=kvm-devel@lists.sourceforge.net \
--cc=marcelo@kvack.org \
--cc=paul@codesourcery.com \
--cc=qemu-devel@nongnu.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.