[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <64d39c34bbd92_26add629414@willemb.c.googlers.com.notmuch>
Date: Wed, 09 Aug 2023 10:01:24 -0400
From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
To: "Erdogan, Tahsin" <trdgn@...zon.com>,
"herbert@...dor.apana.org.au" <herbert@...dor.apana.org.au>,
"jasowang@...hat.com" <jasowang@...hat.com>,
"willemdebruijn.kernel@...il.com" <willemdebruijn.kernel@...il.com>,
"davem@...emloft.net" <davem@...emloft.net>,
"kuba@...nel.org" <kuba@...nel.org>,
"edumazet@...gle.com" <edumazet@...gle.com>,
"pabeni@...hat.com" <pabeni@...hat.com>
Cc: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3] tun: avoid high-order page allocation for packet
header
Erdogan, Tahsin wrote:
> On Wed, 2023-08-09 at 09:18 -0400, Willem de Bruijn wrote:
> > Tun sendmsg is a special case, only used by vhost-net from inside the
> > kernel. Arguably consistency with packet_snd/packet_alloc_skb would
> > be
> > more important. That said, this makes sense to me. I assume your
> > configuring a device with very large MTU?
>
> That's right. I am setting MTU to 9100 in my test.
Makes sense. That's not even that large.
Please address the commit message points about virtio_net_hdr.hdr_len
and write() vs writev().
A writev() specific solution could even take the first iov length as
hint. Note that I'm not suggesting that. IFF_NAPI_FRAGS already does
exactly that, plus the geometry of subsequent frags.
Powered by blists - more mailing lists