From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp3.clb.oleane.net (smtp3.clb.oleane.net [213.56.31.19]) by ozlabs.org (Postfix) with ESMTP id D52E3679FC for ; Wed, 15 Feb 2006 03:09:33 +1100 (EST) From: "Laurent Lagrange" To: "'Pantelis Antoniou'" Subject: RE: Gianfar is slower than fcc_enet on MPC8541 ??? Date: Tue, 14 Feb 2006 17:14:59 +0100 Message-ID: <000201c63181$cf2c3180$5201a8c0@GEG2400> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0003_01C6318A.30F09980" In-Reply-To: <449c76a00602140728g3a9eb46sd70f23b894c4b96@mail.gmail.com> Cc: linuxppc-embedded@ozlabs.org List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , This is a multi-part message in MIME format. ------=_NextPart_000_0003_01C6318A.30F09980 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit Hi Pantelis, Thanks for the express reply. I know that what I say seems incredible. But I don't understand what NAPI does. My measure is very simple. I display a message on the client when 1000 exchanges are done. I already check the ifconfig stats on the board after some seconds. The measures seems the same with or without NAPI. More details ? Thanks again Laurent -----Message d'origine----- De : Pantelis Antoniou [mailto:pantelis.antoniou@gmail.com] Envoyé : mar. 14 février 2006 16:29 À : Laurent Lagrange Cc : linuxppc-embedded@ozlabs.org Objet : Re: Gianfar is slower than fcc_enet on MPC8541 ??? Hi Laurent, I found that pretty hard to believe. What are you measuring exactly? Speed of replies? If so it's explainable since the TSECs use NAPI. Regards Pantelis On 2/14/06, Laurent Lagrange wrote: Hello, I work on a cutom MPC8541 board with Linux 2.6.9. The kernel activates the L1 cache (instructions and data) and the L2 cache (entirely used as cache and not as sram). I configure 1 FCC (FCC1), 2 TSECs with or without NAPI (no effect) but without stashing in L2 sram. All PHYs are automatically configured in 100MB full duplex. eth0: Gianfar Ethernet Controller Version 1.1, 00:10:cd:48:48:e0 eth0: Running with NAPI disabled eth0: 64/64 RX/TX BD ring size eth1: Gianfar Ethernet Controller Version 1.1, 00:10:cd:48:48:e1 eth1: Running with NAPI disabled eth1: 64/64 RX/TX BD ring size eth2: FCC ENET Version custom, 00:10:cd:48:48:e2 Then I launch 3 simple TCP servers, one on each ports. From remote machines I runs 3 TCP clients. The client sends messages of 1000 bytes, The server receives and echoes the message The client receives the echoed message, check the content and sends a new message again. The result is that the 2 TSECs are 2 times slower than the FCC. If I run a "top" application on the board, I use less than 10% of the CPU Each port consumes about 1/3 of the CPU. Any idea on how to configure the gianfar driver ? Thanks Laurent _______________________________________________ Linuxppc-embedded mailing list Linuxppc-embedded@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-embedded ------=_NextPart_000_0003_01C6318A.30F09980 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hi=20 Pantelis,

Thanks=20 for the express reply.

I know=20 that what I say seems incredible. But I don't=20 understand what NAPI does.

My=20 measure is very simple. I display a message on the client=20 when 1000 exchanges are done.

I=20 already check the ifconfig stats on the board after some=20 seconds.

The=20 measures seems the same with or without NAPI.

More=20 details ?

Thanks=20 again

Laurent

-----Message d'origine-----
De : Pantelis = Antoniou=20 [mailto:pantelis.antoniou@gmail.com]
Envoy=E9 : mar. 14 = f=E9vrier=20 2006 16:29
=C0 : Laurent Lagrange
Cc :=20 linuxppc-embedded@ozlabs.org
Objet : Re: Gianfar is = slower than=20 fcc_enet on MPC8541 ???

Hi Laurent,

I found = that=20 pretty hard to believe.

What are you measuring = exactly?

Speed of=20 replies? If so it's explainable since the TSECs=20 use
NAPI.

Regards

Pantelis

On 2/14/06, Laurent=20 Lagrange <lagrange@fr.oleane.com>=20 wrote:

Hello,

I=20 work on a cutom MPC8541 board with Linux 2.6.9.
The kernel = activates the=20 L1 cache (instructions and data)
and the L2 cache (entirely used = as cache=20 and not as sram).

I configure
1 FCC (FCC1),
2 TSECs = with or=20 without NAPI (no effect) but without stashing in L2 sram.
All = PHYs are=20 automatically configured in 100MB full=20 duplex.

        eth0: = Gianfar=20 Ethernet Controller Version 1.1,=20 = 00:10:cd:48:48:e0
        eth0= :=20 Running with NAPI=20 disabled
        eth0: = 64/64=20 RX/TX BD ring = size
        eth1:=20 Gianfar Ethernet Controller Version 1.1,=20 = 00:10:cd:48:48:e1
        eth1= :=20 Running with NAPI=20 disabled
        eth1: = 64/64=20 RX/TX BD ring size =
        eth2:=20 FCC ENET Version custom, 00:10:cd:48:48:e2

Then I launch 3 = simple TCP=20 servers, one on each ports.

From remote machines I runs 3 TCP = clients.
The client sends messages of 1000 bytes,
The server = receives=20 and echoes the message
The client receives the echoed message, = check the=20 content
and sends a new message again.

The result is that = the 2=20 TSECs are 2 times slower than the FCC.

If I run a "top" = application=20 on the board, I use less than 10% of the CPU
Each port consumes = about=20 1/3 of the CPU.

Any idea on how to configure the gianfar = driver=20 = ?

Thanks
Laurent

___________________________________= ____________
Linuxppc-embedded=20 mailing list
Linuxppc-embedded@ozlabs.org=
https://oz= labs.org/mailman/listinfo/linuxppc-embedded

------=_NextPart_000_0003_01C6318A.30F09980--