public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: "André Schwarz" <andre.schwarz@matrix-vision.de>
To: u-boot@lists.denx.de
Subject: [U-Boot] network bootp/tftp hang with tsec on MPC8343
Date: Mon, 21 Sep 2009 15:06:04 +0200	[thread overview]
Message-ID: <1253538364.4803.15.camel@swa-e6500> (raw)
In-Reply-To: <5a775e870909190751j3b07dd1fjcc4d5c499dcc94fb@mail.gmail.com>

Siva,


> Myself Siva, working on a custom board using MCF5485. We had a typical
> problem with tftp. We are using u-boot(1.3.3) from Freescale on our

first of all I'd try to use latest code.

>  board and trying to do tftp of filesize 12MB. Sometime file is
> getting downloaded into external DDR SDRAM correctly. Sometime u-boot
> get?s hanged. This problem is observed during cold bootup. When we
> switch off the board and powered it up, we are facing this problem. If
> we wait for 1 or 2 minutes after the board get?s powered up, try to do
> tftp it goes well. After going through your post on internet
> (http://www.mail-archive.com/u-boot at lists.denx.de/msg00719.html) , I
> am suspecting this could be problem with our DDR SDRAM controller
> initialization.
> 

Using memory with single read/write access is very relaxed compared to
burst mode access during DMA ops. I've had a plain simple timing
violation ... there are lots of dynamic parameters that have to be met.
There's no general advice - sorry.

Are you sure that it is a memory problem ? Any obscure crashes running
linux or other apps ? If not it's more likely to be a networking issue.
R/S-GMII also has strict timing requirements. What about signal
integrity issues - are you sure the interface is working fine from an
electrical point of view ?

> U-boot Ethernet driver uses DMA to transfer the frames. But the serail
> console log shows some timeouts (T) when the U-boot hangs during
> transfer.
> 
ok - could be anything.

>  
> 
> If we wait for 1 or 2 minutes before doing tftp to get the image into
> RAM , it is getting downloaded properly. Could you please suggest what
> exactly you did in your SRAM configuration to fix the problem?

looks like a thermal issue, i.e. electrical problem.
Again, this might be either local memory or MII. What kind of network
are you using ? I'm not familiar with MCF54xx ...

Get you memory device datasheet and match the dynamic parameters with
that of your CPU's memory controller.

Maybe there are others having similar problems with your particular CPU.


Regards,
Andr?

>  
>  
> 
> Please find the attached log:
>  
>  
> U-Boot 1.3.3 (Sep 18 2009 - 10:24:03)
> CPU:   Freescale MCF5485
>        CPU CLK 200 Mhz BUS CLK 100 Mhz
> Board: Freescale FireEngine 5485 EVB
> DRAM:  128 MB
> POST: Checking Ram...PASSED.
> FLASH: 16 MB
> In:    serial
> Out:   serial
> Err:   serial
> Net:   rxbd f0012000 txbd f0012100
> rxbd f0012870 txbd f0012970
> FEC0, FEC1
> Hit any key to stop autoboot:  5
>  
>  
> -> printenv
> bootdelay=5
> baudrate=115200
> netmask=255.255.255.0
> hostname=M548xEVB
> netdev=eth0
> loadaddr=1000
> u-boot=u-boot.bin
> load=tftp ${loadaddr) ${u-boot}
> kernel=u-boot.bin
> load=tftp ${loadaddr) ${kernel}
> upd=run load; run prog
> progboot=prot off 1:0-2;era e0000000 e005ffff;cp.b ${loadaddr}
> e0000000 ${filesize};
> progkern=prot off 1:4-51;era e0080000 e067ffff;cp.b ${loadaddr}
> e0080000 ${filesize};save
> bootargs=console=/dev/console,115200 root=/dev/ram0 rootfstype=romfs
> ro
> bootcmd=cp.b e0080000 ${loadaddr} ${filesize};go 2000
> ethact=FEC0
> ethaddr=00:25:F7:10:00:1F
> eth1addr=00:25:F7:10:00:20
> stdin=serial
> stdout=serial
> stderr=serial
> mem=130560k
> gatewayip=169.254.0.102
> ipaddr=169.254.0.101
> serverip=169.254.0.253
> filesize=C7FFFF
> Environment size: 720/131068 bytes
> -> tftp 10000 image.bin;prot off 1:4-103;era E0080000 E0CFFFFF;cp.b
> 1000 E0080000 BEEC0F;reset;
> Using FEC0 device
> TFTP from server 169.254.0.253; our IP address is 169.254.0.101
> Filename 'image.bin'.
> Load address: 0x10000
> Loading:
> *\b#################################################################
>   #################################################################
>   #################################################################
>   ########################################T #########################
>   #################################################################
>   #########################
>  
>  
> Thanks,
> Siva
>  
> 



MATRIX VISION GmbH, Talstrasse 16, DE-71570 Oppenweiler
Registergericht: Amtsgericht Stuttgart, HRB 271090
Geschaeftsfuehrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner, Hans-Joachim Reich

       reply	other threads:[~2009-09-21 13:06 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <5a775e870909190751j3b07dd1fjcc4d5c499dcc94fb@mail.gmail.com>
2009-09-21 13:06 ` André Schwarz [this message]
2008-08-21 14:21 [U-Boot] network bootp/tftp hang with tsec on MPC8343 Andre Schwarz
2008-08-27 15:07 ` Andre Schwarz
2008-08-27 15:30   ` JerryVanBaren
2008-08-27 15:57     ` Andre Schwarz

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=1253538364.4803.15.camel@swa-e6500 \
    --to=andre.schwarz@matrix-vision.de \
    --cc=u-boot@lists.denx.de \
    /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