* [Qemu-devel] [vhost-net] handle_tx: extract data from vq when !zcopy_used
@ 2015-04-10 12:58 Catalin Vasile
2015-04-10 15:14 ` Greg Kurz
0 siblings, 1 reply; 2+ messages in thread
From: Catalin Vasile @ 2015-04-10 12:58 UTC (permalink / raw)
To: qemu-devel
[-- Attachment #1: Type: text/plain, Size: 194 bytes --]
I'm really confused.
Where does vhost-net (handle_tx) actually extract data from vq when zero
copy is not used?
I just see it popping a vq element, but not actually copying anything from
there.
[-- Attachment #2: Type: text/html, Size: 251 bytes --]
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [Qemu-devel] [vhost-net] handle_tx: extract data from vq when !zcopy_used
2015-04-10 12:58 [Qemu-devel] [vhost-net] handle_tx: extract data from vq when !zcopy_used Catalin Vasile
@ 2015-04-10 15:14 ` Greg Kurz
0 siblings, 0 replies; 2+ messages in thread
From: Greg Kurz @ 2015-04-10 15:14 UTC (permalink / raw)
To: Catalin Vasile; +Cc: qemu-devel
On Fri, 10 Apr 2015 15:58:33 +0300
Catalin Vasile <catalinvasile92@gmail.com> wrote:
> I'm really confused.
> Where does vhost-net (handle_tx) actually extract data from vq when zero
> copy is not used?
> I just see it popping a vq element, but not actually copying anything from
> there.
vhost-net just passes the pointers to the socket, whether zero copy is used
or not:
err = sock->ops->sendmsg(NULL, sock, &msg, len);
BTW, I think virtualization@lists.linux-foundation.org would have been more
appropriate since this isn't a about QEMU.
Cheers.
--
Greg
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2015-04-10 15:14 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-04-10 12:58 [Qemu-devel] [vhost-net] handle_tx: extract data from vq when !zcopy_used Catalin Vasile
2015-04-10 15:14 ` Greg Kurz
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).