linux-mips
[Top] [All Lists]

Re: AutoNegotiation could not complete in PB1500 platform

To: He Jin <thesistarball@yahoo.com.cn>
Subject: Re: AutoNegotiation could not complete in PB1500 platform
From: Ed Okerson <eokerson@texasconnect.net>
Date: Wed, 17 Dec 2003 22:55:37 -0600 (CST)
Cc: linux-mips <linux-mips@linux-mips.org>
In-reply-to: <20031218033717Z8225373-16706+1429@linux-mips.org>
Original-recipient: rfc822;linux-mips@linux-mips.org
Sender: linux-mips-bounce@linux-mips.org
Did you set the EM bit in the mac_control register to match the endian
mode you are running the CPU in?

Ed Okerson

On Thu, 18 Dec 2003, He Jin wrote:

> Hi, Dear all,
>
> I'm porting my own firmware code to PB1500, however some troubles ocurred 
> with the Au1x NIC and LSI PHY. The dump of PHY register (the NIC 
> initialization code is borrowed from the driver inside YAMON) is as follow:
>
> ////////////////////////////////////
>  MII status: Link is up
>  1th 0x3000
>  2th 0x7809                   => should be 0x7829
>               ^^^^^^^^^^
>  3th 0x16
>  4th 0xf840
>  5th 0x1e1
>  6th 0x 0                     => should be 0x45e1
>               ^^^^^^^^^^
> 17th 0x22
> 18th 0xffc0
> ////////////////////////////////////
>
> It shows the MII interface couldn't finish the AutoNegotiation process. The 
> values I think right is the output of some small probe programs using YAMON 
> to boot the board and running those small probe programs . Besides, the LED 
> in NIC interface could flash normally and we observed the LSI PHY chips 
> should have been reset sucessfully in our firmware code using oscillograph 
> device to probe the 'reset' pin in the PHY chips.
>
> If I use YAMON, everything OK. Could somebody tell me why ?
>
> Thanks a lot !
>
>
>
>
>
>
>
>


<Prev in Thread] Current Thread [Next in Thread>