From: Po-Yu Chuang <ratbert.chuang@gmail.com>
To: David Miller <davem@davemloft.net>
Cc: mirqus@gmail.com, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org, bhutchings@solarflare.com,
eric.dumazet@gmail.com, joe@perches.com, dilinger@queued.net,
ratbert@faraday-tech.com
Subject: Re: [PATCH v4] net: add Faraday FTMAC100 10/100 Ethernet driver
Date: Thu, 24 Feb 2011 15:27:55 +0800 [thread overview]
Message-ID: <AANLkTimorcNAEbSpuE_dUXKuOBZDcDcGSJ2cGefniThc@mail.gmail.com> (raw)
In-Reply-To: <20110131.203556.193730771.davem@davemloft.net>
Hi David,
On Tue, Feb 1, 2011 at 12:35 PM, David Miller <davem@davemloft.net> wrote:
> From: Po-Yu Chuang <ratbert.chuang@gmail.com>
> Date: Tue, 1 Feb 2011 11:56:16 +0800
>
>> If I simply allocate a page for each rx ring entry, I still need to allocate
>> an skb and copy at least packet header in first page to skb->data. Then
>> add the page of rest of payload to skb by skb_fill_page_desc().
>
> You should attach the pages, the use __pskb_pull_tail() to bring in the
> headers to the linear skb->data area.
>
> See drivers/net/niu.c:niu_process_rx_pkt().
I tried two ways to implement zero-copy.
One is to preallocate skb big enough for any rx packet and use the skb
as rx buffer.
The other is use page as rx buffer, use skb_fill_page_desc() to add a
data page to
skb and then pull only header to skb by __pskb_pull_tail() as you suggested.
Two implementations are slower than the original memcpy version.
(benchmarked with iperf)
I guess the problem is because a HW restriction that the rx buffer must be
64 bits aligned. Since I cannot make rx buffer starts at offset 2 bytes, the
IP header, TCP header and data are not 4 bytes aligned. The performance
drops drastically.
Therefore, I will submit later a v6 which is still using memcpy().
best regards,
Po-Yu Chuang
next prev parent reply other threads:[~2011-02-24 7:27 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-13 11:49 [PATCH] net: add Faraday FTMAC100 10/100 Ethernet driver Po-Yu Chuang
2011-01-13 14:03 ` Ben Hutchings
2011-01-14 5:37 ` Po-Yu Chuang
2011-01-13 14:22 ` Eric Dumazet
2011-01-13 16:29 ` Andres Salomon
2011-01-14 6:44 ` Po-Yu Chuang
2011-01-14 6:56 ` Po-Yu Chuang
2011-01-13 15:39 ` Joe Perches
2011-01-14 6:35 ` Po-Yu Chuang
[not found] ` <1294959948.4114.189.camel@Joe-Laptop>
2011-01-14 6:49 ` Po-Yu Chuang
2011-01-17 9:21 ` [PATCH v2] " Po-Yu Chuang
2011-01-17 17:19 ` Joe Perches
2011-01-19 9:40 ` Po-Yu Chuang
2011-01-19 12:46 ` Ben Hutchings
2011-01-19 16:41 ` Joe Perches
2011-01-20 5:30 ` Po-Yu Chuang
2011-01-20 8:46 ` Joe Perches
2011-01-17 17:29 ` Eric Dumazet
2011-01-17 18:58 ` Ben Hutchings
2011-01-17 20:39 ` Eric Dumazet
2011-01-17 18:21 ` Eric Dumazet
2011-01-18 3:08 ` Po-Yu Chuang
2011-01-19 9:20 ` Po-Yu Chuang
2011-01-20 15:30 ` [PATCH v3] " Po-Yu Chuang
2011-01-20 15:35 ` Eric Dumazet
2011-01-20 15:43 ` Po-Yu Chuang
2011-01-20 15:41 ` Eric Dumazet
2011-01-20 15:54 ` Po-Yu Chuang
2011-01-20 17:56 ` Joe Perches
2011-01-21 3:35 ` Po-Yu Chuang
2011-01-20 19:00 ` Joe Perches
2011-01-21 5:03 ` Po-Yu Chuang
[not found] ` <1295592411.6795.10.camel@Joe-Laptop>
2011-01-21 7:06 ` Po-Yu Chuang
2011-01-20 19:01 ` Michał Mirosław
2011-01-21 3:37 ` Po-Yu Chuang
2011-01-21 7:55 ` [PATCH v4] " Po-Yu Chuang
2011-01-21 9:08 ` Eric Dumazet
2011-01-24 8:07 ` Po-Yu Chuang
2011-01-21 12:26 ` Michał Mirosław
2011-01-24 8:26 ` Po-Yu Chuang
2011-01-24 20:22 ` Michał Mirosław
2011-01-25 2:46 ` Po-Yu Chuang
2011-02-01 3:56 ` Po-Yu Chuang
2011-02-01 4:35 ` David Miller
2011-02-24 7:27 ` Po-Yu Chuang [this message]
2011-02-24 7:51 ` David Miller
2011-02-24 8:07 ` Po-Yu Chuang
2011-02-24 8:22 ` Eric Dumazet
2011-02-24 9:29 ` [PATCH ref0] " Po-Yu Chuang
2011-02-24 17:39 ` Eric Dumazet
2011-02-24 17:48 ` Eric Dumazet
2011-02-25 2:32 ` Po-Yu Chuang
2011-02-25 9:45 ` Po-Yu Chuang
2011-02-25 10:52 ` Eric Dumazet
2011-02-25 18:34 ` David Miller
2011-02-25 18:45 ` Eric Dumazet
2011-02-25 18:47 ` Eric Dumazet
2011-03-01 5:45 ` Po-Yu Chuang
2011-03-01 5:53 ` Eric Dumazet
2011-02-25 9:57 ` [PATCH v6] " Po-Yu Chuang
2011-02-25 11:40 ` Eric Dumazet
2011-03-01 5:20 ` Po-Yu Chuang
2011-03-01 5:26 ` Eric Dumazet
2011-03-01 5:45 ` Eric Dumazet
2011-03-01 5:51 ` Po-Yu Chuang
2011-03-01 5:54 ` Eric Dumazet
2011-03-01 5:59 ` Eric Dumazet
2011-03-01 6:48 ` [PATCH] " Po-Yu Chuang
2011-03-01 7:27 ` Eric Dumazet
2011-03-03 20:19 ` David Miller
2011-02-24 18:43 ` [PATCH v4] " David Miller
2011-01-24 12:39 ` [PATCH v5] " Po-Yu Chuang
2011-01-24 15:07 ` Eric Dumazet
2011-01-25 2:46 ` Po-Yu Chuang
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=AANLkTimorcNAEbSpuE_dUXKuOBZDcDcGSJ2cGefniThc@mail.gmail.com \
--to=ratbert.chuang@gmail.com \
--cc=bhutchings@solarflare.com \
--cc=davem@davemloft.net \
--cc=dilinger@queued.net \
--cc=eric.dumazet@gmail.com \
--cc=joe@perches.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mirqus@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=ratbert@faraday-tech.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).