UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Observer rhodeman
Observer
11,596 Views
Registered: ‎09-11-2013

Marvell 88E1111 Failing Auto-Negotiation on Spartan-6 (lights not coming on)

Hi all,

 

Alright, so I've been trying to debug this for a week now with a small amount of success, but I haven't been able to get a permanent solution just yet. Similar to others who have had this issue, the board (which has two 88E1111 PHYs) did successfully pass data during a short time frame. However, I powered down the board, came in the next day and nada... no RX/TX lights or anything.

 

Here is a quick rundown of the setup after some CONFIG/Reset hardware modifying:

  1. Custom Spartan-6 board with 2 88E1111 PHYs
  2. Desiring to run the PHYs in 1000BASE-T full-duplex mode
  3. ANEG[3:0] = "1111"
  4. HWCFG_MODE[3:0] = "1111"
  5. DIS_FC and DIS_SLEEP both are '0' (both changed from '1' when auto-neg was failing although it did work that one magical time)
  6. Resetn is now driven manually from a switch to make sure the reset is low for a while during power up, then toggled high

Since making these hardware adjustments, a small glimmer of hope has arisen where, if the PHYs are connected, the LED_DUPLEX LED on each PHY will periodically flash (flashing stops if the cable is unplugged). I'm not sure how to make use of this information, but the TX and RX lights still remain off. Obviously I'll be happy to provide any information I may have missed. It is my understanding that the RX/TX lights should come on after auto-negotiation is complete and that auto-negotiation does not rely on the FPGA to complete (aside from maybe the Reset pin).

 

Thanks for your help!

0 Kudos
6 Replies
Observer rhodeman
Observer
11,570 Views
Registered: ‎09-11-2013

Re: Marvell 88E1111 Failing Auto-Negotiation on Spartan-6 (lights not coming on)

Also for what it's worth... XTAL1 is 125MHz while output XTAL2 is left unconnected...
0 Kudos
Scholar kotir
Scholar
11,556 Views
Registered: ‎02-03-2010

Re: Marvell 88E1111 Failing Auto-Negotiation on Spartan-6 (lights not coming on)

Hi,

 

Have you tried to compare the schematics of SP601 and SP605 boards with your card ?

 

http://www.xilinx.com/support/documentation/boards_and_kits/xtp067_sp605_schematics.pdf

 

http://www.xilinx.com/support/documentation/boards_and_kits/xtp051_sp601_schematics.pdf

 

These boards also use the same PHY.

 

Regards,

KR

--------------------------------------------------​--------------------------------------------
Kindly note- Please mark the Answer as "Accept as solution" if information provided is helpful.

Give Kudos to a post which you think is helpful.
--------------------------------------------------​-------------------------------------------
0 Kudos
Observer rhodeman
Observer
11,543 Views
Registered: ‎09-11-2013

Re: Marvell 88E1111 Failing Auto-Negotiation on Spartan-6 (lights not coming on)

Hi, Thanks for the reply KR. We are currently playing with some of the hardware on our boards, but I know one of the problems was we were not keeping the PHY in reset for long enough. We're still seeing some strangeness, but at very least, we have some boards we can work on.

Out of curiosity, does anyone know if a blinking LED_DUPLEX can mean anything other than collision? We're still seeing that on several of our boards. Thanks again!
0 Kudos
Xilinx Employee
Xilinx Employee
11,536 Views
Registered: ‎02-06-2013

Re: Marvell 88E1111 Failing Auto-Negotiation on Spartan-6 (lights not coming on)

Hi

 

The Led behaviour depends on the mode configuration.

 

You can find the detailed description in the below document.

 

http://www.marvell.com/transceivers/assets/Marvell-Alaska-Ultra-88E1111-GbE.pdf

 

Also make sure that you are applying for phy reset for the minimum period required as per the doc.

Regards,

Satish

--------------------------------------------------​--------------------------------------------
Kindly note- Please mark the Answer as "Accept as solution" if information provided is helpful.

Give Kudos to a post which you think is helpful.
--------------------------------------------------​-------------------------------------------
0 Kudos
Scholar kotir
Scholar
11,531 Views
Registered: ‎02-03-2010

Re: Marvell 88E1111 Failing Auto-Negotiation on Spartan-6 (lights not coming on)

Hi,

 

The exact numbers of minimum reset pulse required will be in datasheet which you can receive from Marvell.

The valid initial reset pulse should be followed for proper operation.

If you get the data shee, check for T PU_REST 88E1111 as per datasheet.

 

Regards,

Koti Reddy

--------------------------------------------------​--------------------------------------------
Kindly note- Please mark the Answer as "Accept as solution" if information provided is helpful.

Give Kudos to a post which you think is helpful.
--------------------------------------------------​-------------------------------------------
0 Kudos
Observer rhodeman
Observer
11,183 Views
Registered: ‎09-11-2013

Re: Marvell 88E1111 Failing Auto-Negotiation on Spartan-6 (lights not coming on)

I thought I would give a quick update on what our issue was (but it's still being resolved). We are using power-over-ethernet in our design and it appeared that when I would source power from one board to another, the sourcing PHY (of sorts) was getting damaged. We are still debugging this issue, but this appears to be our culprit for the failing auto-negotiation on some boards while others work fine. Thanks!
0 Kudos