From: Stephen Hemminger <shemminger@vyatta.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Masayuki Ohtake <masa-korg@dsn.okisemi.com>,
Sam Ravnborg <sam@ravnborg.org>, Joe Perches <joe@perches.com>,
LKML <linux-kernel@vger.kernel.org>,
ML netdev <netdev@vger.kernel.org>,
Greg Rose <gregory.v.rose@intel.com>,
Maxime Bizon <mbizon@freebox.fr>,
Kristoffer Glembo <kristoffer@gaisler.com>,
Ralf Baechle <ralf@linux-mips.org>,
John Linn <john.linn@xilinx.com>,
Randy Dunlap <randy.dunlap@oracle.com>,
"David S. Miller" <davem@davemloft.net>,
MeeGo <meego-dev@meego.com>, "Wang, Qi" <qi.wang@intel.com>,
"Wang, Yong Y" <yong.y.wang@intel.com>,
Andrew <andrew.chih.howe.khor@intel.com>,
Intel OTC <joel.clark@intel.com>,
"Foster, Margie" <margie.foster@intel.com>,
Toshiharu Okada <okada533@dsn.okisemi.com>,
Tomoya Morinaga <morinaga526@dsn.okisemi.com>,
Takahiro Shimizu <shimizu394@dsn.okisemi.com>
Subject: Re: [PATCH] Gigabit Ethernet driver of Topcliff PCH
Date: Thu, 2 Sep 2010 08:10:07 -0700 [thread overview]
Message-ID: <20100902081007.2f910bc0@nehalam> (raw)
In-Reply-To: <1283434847.2454.726.camel@edumazet-laptop>
On Thu, 02 Sep 2010 15:40:47 +0200
Eric Dumazet <eric.dumazet@gmail.com> wrote:
> Le jeudi 02 septembre 2010 à 21:39 +0900, Masayuki Ohtake a écrit :
> > Hi Eric
> >
> > Thank you for your comments.
> >
> > > I find hard to believe this driver needs to copy all outgoing frames on
> > > pre-allocated skbs.
> > >
> > > + /* [Header:14][payload] ---> [Header:14][paddong:2][payload] */
> > > + memcpy(tmp_skb->data, skb->data, ETH_HLEN);
> > > + tmp_skb->data[ETH_HLEN] = 0x00;
> > > + tmp_skb->data[ETH_HLEN + 1] = 0x00;
> > > + tmp_skb->len = skb->len;
> > > + memcpy(&tmp_skb->data[ETH_HLEN + 2], &skb->data[ETH_HLEN],
> > > + (skb->len - ETH_HLEN));
> > > + buffer_info->kernel_skb = skb;
> > > + skb = tmp_skb;
> > >
> > > Whats the deal here please ?
> >
> > This processing depends on hardware specification.
> >
> > At the time of transmission.
> > Hardware accepts a packet in the following format.
> > [Header: 14octet] + [padding: 2octet] + [payload]
> > Also, it is necessary to align the head of a [Header: 14octet] at 64byte.
> >
> > In my knowledge, SKB received by kernel are the following format.
> > [padding: 2octet] + [Header:14octet] + [payload]
> > Also, The head of [payload] has aligned at 16 byte.
> >
> > So, it has adjusted with the format of hardware by a copy.
>
> The two bytes padding can be handled by network stack, if you
> force in your setup phase :
> dev->hard_header_len = ETH_HLEN + 2;
>
> then, you can do a single memcpy:
>
> memcpy(tmp_skb->data, skb->data, skb->len);
>
> About the 64 byte alignement, it might be a bit more complex :)
With Eric's suggestion, you might find most packets are going
to be aligned but the code should not depend on it always being true.
Packets that get forwarded have header determined by the receiving
interface.
Something like:
skb = skb_realloc_headroom(skb, 2);
if (!skb)
goto drop;
__skb_push(skb, 2);
memmove(skb->data, skb->data + 2, ETH_HLEN);
skb->data[ETH_HLEN] = 0;
skb->data[ETH_HLEN+1] = 0;
if (!IS_ALIGNED(skb->data, 16)) {
nskb = netdev_alloc_skb(dev, skb->len + 16);
if (!nskb)
goto drop2;
skb_reserve(nskb, PTR_ALIGN(skb->data, 16) - skb->data);
skb_put(skb, skb->len);
memcpy(nskb->data, skb->data, skb->len);
dev_kfree_skb(skb);
skb = nskb;
}
--
next prev parent reply other threads:[~2010-09-02 15:10 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-26 9:56 [PATCH] Gigabit Ethernet driver of Topcliff PCH Masayuki Ohtake
2010-08-26 10:28 ` Sam Ravnborg
2010-08-26 12:47 ` Masayuki Ohtake
2010-08-26 14:44 ` Joe Perches
2010-08-26 15:34 ` Stephen Hemminger
2010-08-26 15:40 ` Stephen Hemminger
2010-08-26 15:41 ` Stephen Hemminger
2010-08-26 15:42 ` Stephen Hemminger
2010-08-26 15:43 ` Stephen Hemminger
2010-08-26 15:45 ` Stephen Hemminger
2010-08-26 15:47 ` Stephen Hemminger
2010-08-26 15:57 ` Stephen Hemminger
2010-08-26 16:05 ` Stephen Hemminger
2010-08-26 16:16 ` Joe Perches
2010-08-26 16:29 ` Stephen Hemminger
2010-08-26 17:02 ` Joe Perches
2010-08-31 14:15 ` Masayuki Ohtake
2010-08-31 14:51 ` Eric Dumazet
2010-09-02 12:39 ` Masayuki Ohtake
2010-09-02 13:40 ` Eric Dumazet
2010-09-02 15:10 ` Stephen Hemminger [this message]
2010-09-03 13:32 ` Masayuki Ohtake
2010-09-03 13:43 ` Eric Dumazet
2010-09-03 14:11 ` Masayuki Ohtake
2010-08-31 15:08 ` Randy Dunlap
2010-08-31 16:10 ` Joe Perches
2010-08-31 17:25 ` [PATCH] drivers/net/pch_gbe: Use bool not unsigned char Joe Perches
2010-08-31 18:38 ` [PATCH] drivers/net/pch_gbe: Cleanup stats use Joe Perches
2010-09-01 1:33 ` Stephen Hemminger
2010-09-01 1:38 ` Joe Perches
2010-09-03 2:23 ` [PATCH] Gigabit Ethernet driver of Topcliff PCH FUJITA Tomonori
2010-09-07 1:13 ` Masayuki Ohtake
2010-09-07 3:21 ` FUJITA Tomonori
2010-09-07 4:06 ` Masayuki Ohtake
-- strict thread matches above, loose matches on Subject: below --
2010-09-03 14:09 Masayuki Ohtake
2010-09-03 16:35 ` Jiri Slaby
2010-09-07 1:13 ` Masayuki Ohtake
2010-09-08 13:52 ` Masayuki Ohtake
2010-09-08 14:16 ` Jiri Slaby
2010-09-08 14:54 ` Stephen Hemminger
2010-09-08 14:55 ` Stephen Hemminger
2010-09-09 13:37 ` Masayuki Ohtake
2010-09-09 13:38 ` Masayuki Ohtake
2010-09-03 20:00 ` Joe Perches
2010-09-07 2:42 ` Masayuki Ohtake
2010-09-08 20:36 ` David Miller
2010-09-15 12:19 ` Masayuki Ohtake
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=20100902081007.2f910bc0@nehalam \
--to=shemminger@vyatta.com \
--cc=andrew.chih.howe.khor@intel.com \
--cc=davem@davemloft.net \
--cc=eric.dumazet@gmail.com \
--cc=gregory.v.rose@intel.com \
--cc=joe@perches.com \
--cc=joel.clark@intel.com \
--cc=john.linn@xilinx.com \
--cc=kristoffer@gaisler.com \
--cc=linux-kernel@vger.kernel.org \
--cc=margie.foster@intel.com \
--cc=masa-korg@dsn.okisemi.com \
--cc=mbizon@freebox.fr \
--cc=meego-dev@meego.com \
--cc=morinaga526@dsn.okisemi.com \
--cc=netdev@vger.kernel.org \
--cc=okada533@dsn.okisemi.com \
--cc=qi.wang@intel.com \
--cc=ralf@linux-mips.org \
--cc=randy.dunlap@oracle.com \
--cc=sam@ravnborg.org \
--cc=shimizu394@dsn.okisemi.com \
--cc=yong.y.wang@intel.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 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).