5213 bootloader

[Patrick Flannery-2]

All,

 

I need to create a bootloader for a mcf5213 application. I hope to have it reside on a sector of the onboard flash. Can anyone point me in the right direction to sample code and documentation?

 

Thanks in advance,

Pat

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[wrw]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi Pat:

 

Here are a couple of links that may help you:

 

http://forums.freescale.com/freescale/board/message?board.id=CFCOMM&message.id=3113&query.id=239157#M3113

http://www.linuxdevices.com/articles/AT5085702347.html

 

If you haven’t already done so, the first step is to get your code to copy from Flash to RAM.  Then you can insert a small piece of code in the Flash that can accept applications over the USB, Ethernet etc.  The trick is either running it (the bootloader) entirely out of Flash or using a small amount of RAM that is then overwritten once the application is installed.  Depending on the RTOS (if you are using one) this can be somewhat problematic.  But check out the above links.

 

Regards,

 

Wendi

 

---

From: coldfire-bounce@... [mailto:coldfire-bounce@...] On Behalf Of Patrick Flannery
Sent: Thursday, January 24, 2008 8:39 AM
To: wrw
Subject: [ColdFire] 5213 bootloader

 

All,

 

I need to create a bootloader for a mcf5213 application. I hope to have it reside on a sector of the onboard flash. Can anyone point me in the right direction to sample code and documentation?

 

Thanks in advance,

Pat

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Mark Butcher]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi Pat

 

The uTasker project includes a bootloader for the M5223X (almost identical to 5213, but with Ethernet). It occupies 1 x 2k FLASH blocks (or 2 x 2k if supporting SPI FLASH and encryption). It is described in the following links:

http://www.utasker.com/docs/uTasker/uTaskerBoot_003.PDF
http://www.utasker.com/docs/uTasker/BM-Booloader_for_M5223X.PDF

 

You may find it useful as a starting point.

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Patrick Flannery
Gesendet: Donnerstag, 24. Januar 2008 17:39
An: mark@...
Betreff: [ColdFire] 5213 bootloader

 

All,

 

I need to create a bootloader for a mcf5213 application. I hope to have it reside on a sector of the onboard flash. Can anyone point me in the right direction to sample code and documentation?

 

Thanks in advance,

Pat

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Sanjay Morab]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Dear Mark

Hi,

 

We are using MCF52230 for our one of the project. We had developed all source code using the EVB. The code works fine. The EVB is for 52235. But when we transfer the coder to target board which has MCF 52230 there are few problems we are facing.

 

1. The Does not boot up all the time even with the sample code. It does ot detect the BDM also.

2. When it starts , for some time it works fine and then it starts giving the PC ERROR = 0x000294 or similar.

 

 

I am attaching the pdf of the schematic if it has any problems.

 

Please let me know currently the Board is Two layer, is it strongly recomended to have 4 layers.

 

Sanjay Morab

 

Accord Electronics
50, Rashmi Industrial Estate,
Vasai Industrial Area,
Navghar, Vasai Road(e),
Dist. Thane
Pin - 401210
Ph -[phone]
      [phone]
Fax - [phone]
H.P. - [phone]

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Mark Butcher]

Some parts of this message have been removed. Learn more about Nabble's security policy.

X-SpamDetect-Info: ------------- Start ASpam results ---------------
X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information
X-SpamDetect: ****: 4.300000 From3consonants=0.3, Some html comments=1.0, Aspam=3.0
X-SpamDetect-Info: ------------- End ASpam results -----------------

Hi Sanjay

 

I don’t see the standard pull-up resistors on the BDM interace lines TMS, TRST, TDI and TDO.

I wonder whether these are floating and so the chip is sometimes booting to DBM mode or during operation switched to BDM mode.

 

Regards

 

Mark

 

I have used 2 layers with NE64 (almost identical footprint and LAN interface) so I expect that it is possible also with the M52230. 4 layers will of course be better but I wouldn’t expect extreme changes.

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Sanjay Morab
Gesendet: Freitag, 25. Januar 2008 03:13
An: mark@...
Betreff: [ColdFire] 52235/52230 problem

 

Dear Mark

Hi,

 

We are using MCF52230 for our one of the project. We had developed all source code using the EVB. The code works fine. The EVB is for 52235. But when we transfer the coder to target board which has MCF 52230 there are few problems we are facing.

 

1. The Does not boot up all the time even with the sample code. It does ot detect the BDM also.

2. When it starts , for some time it works fine and then it starts giving the PC ERROR = 0x000294 or similar.

 

 

I am attaching the pdf of the schematic if it has any problems.

 

Please let me know currently the Board is Two layer, is it strongly recomended to have 4 layers.

 

Sanjay Morab

 

Accord Electronics
50, Rashmi Industrial Estate,
Vasai Industrial Area,
Navghar, Vasai Road(e),
Dist. Thane
Pin - 401210
Ph -[phone]
      [phone]
Fax - [phone]
H.P. - [phone]

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[John Bodnar]

Hi folks,

Mark Butcher wrote:
> I have used 2 layers with NE64 (almost identical footprint and LAN
> interface) so I expect that it is possible also with the M52230. 4
> layers will of course be better but I wouldn’t expect extreme changes.

Mark makes an important point here. We intentionally matched the
MCF5223x pin out to that of the 9S12NE64 as closely as possible. The
VDDs and VSSs are in the same place, as are EXTAL and XTAL, RST_IN, the
Ethernet PHY, and basic I/O functionality like the UARTs, QSPI, and
timers. The 5223x and the NE64 have completely different BDM interfaces,
so those could not be matched exactly, but, otherwise, the plan was to
allow as much re-use of existing NE64 PCB designs as possible.

You'll see on forthcoming ColdFire designs that we're trying to make our
pin outs more conducive to simpler board designs. I can't promise any
major breakthroughs but hopefully enough care and forethought to make
everyone's life a little easier.

Best regards,

John Bodnar
Systems Engineer
Microcontroller Solutions Group
Freescale Semiconductor, Inc.

--
POPI Classification
[ ]General Business Information
[x]Freescale Internal Use
[ ]Freescale Confidential Proprietary

---
coldfire@...              Send a post to the list.
coldfire-join@...        Join the list.
coldfire-digest@...    Join the list in digest mode.
coldfire-leave@...     Leave the list.

[Brian McCombs]

John do you have an updated roadmap for the coldfire family? How soon
should we expect a V4 core with a display driver onboard?

-----Original Message-----
From: coldfire-bounce@... [mailto:coldfire-bounce@...]
On Behalf Of John Bodnar
Sent: Friday, January 25, 2008 10:26 AM
To: Brian McCombs
Subject: Re: AW: [ColdFire] 52235/52230 problem


Hi folks,

Mark Butcher wrote:
> I have used 2 layers with NE64 (almost identical footprint and LAN
> interface) so I expect that it is possible also with the M52230. 4
> layers will of course be better but I wouldn't expect extreme changes.

Mark makes an important point here. We intentionally matched the
MCF5223x pin out to that of the 9S12NE64 as closely as possible. The
VDDs and VSSs are in the same place, as are EXTAL and XTAL, RST_IN, the
Ethernet PHY, and basic I/O functionality like the UARTs, QSPI, and
timers. The 5223x and the NE64 have completely different BDM interfaces,

so those could not be matched exactly, but, otherwise, the plan was to
allow as much re-use of existing NE64 PCB designs as possible.

You'll see on forthcoming ColdFire designs that we're trying to make our

pin outs more conducive to simpler board designs. I can't promise any
major breakthroughs but hopefully enough care and forethought to make
everyone's life a little easier.

Best regards,

John Bodnar
Systems Engineer
Microcontroller Solutions Group
Freescale Semiconductor, Inc.

--
POPI Classification
[ ]General Business Information
[x]Freescale Internal Use
[ ]Freescale Confidential Proprietary

---
coldfire@...              Send a post to the list.
coldfire-join@...        Join the list.
coldfire-digest@...    Join the list in digest mode.
coldfire-leave@...     Leave the list.

---
coldfire@...              Send a post to the list.
coldfire-join@...        Join the list.
coldfire-digest@...    Join the list in digest mode.
coldfire-leave@...     Leave the list.

[John Bodnar]

Hi Sanjay,

It just occurred to me that Mark replied to this, and I promptly deleted
his message without seeing which questions he had answered, but I did
find one easy answer.

Sanjay Morab wrote:
> We are using MCF52230 for our one of the project. We had developed all
> source code using the EVB. The code works fine. The EVB is for 52235.
> But when we transfer the coder to target board which has MCF 52230
> there are few problems we are facing.
>  
> 1. The Does not boot up all the time even with the sample code. It
> does ot detect the BDM also.

The BDM doesn't come up enabled because you have JTAG_EN pulled to VDD.  
JTAG_EN must be tied low in order to use the BDM.  When it's tied high,
the muxes on the BDM interface switch to the JTAG TAP controller, and
the BDM is inaccessible.

John Bodnar
Systems Engineer
Microcontroller Solutions Group
Freescale Semiconductor, Inc.

--
POPI Classification
[ ]General Business Information
[x]Freescale Internal Use
[ ]Freescale Confidential Proprietary

---
coldfire@...              Send a post to the list.
coldfire-join@...        Join the list.
coldfire-digest@...    Join the list in digest mode.
coldfire-leave@...     Leave the list.

[Steve Strobel-2]

At 09:39 AM 1/24/2008, you wrote:
>All,
>
>I need to create a bootloader for a mcf5213 application. I hope to
>have it reside on a sector of the onboard flash. Can anyone point me
>in the right direction to sample code and documentation?

We use Ed Sutter's MicroMonitor to start an application that uses the
RTEMS RTOS.  It can also be used to start uClinux, although I think
U-Boot is more common for embedded Linux systems.

Steve



---
Steve Strobel
Link Communications, Inc.
1035 Cerise Rd
Billings, MT 59101-7378
[phone] ext 102
[phone] (fax)
WWW: http://www.link-comm.com
MailTo:steve.strobel@...

---
coldfire@...              Send a post to the list.
coldfire-join@...        Join the list.
coldfire-digest@...    Join the list in digest mode.
coldfire-leave@...     Leave the list.

[Sanjay Morab]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Dear Mark

 

One information also required.

 

When we are using the two layer board for the ETHRNET what should be trace width and separation required from out put of the Magnatics to RJ45 connector as those two pairs are connected as differntial pair.

 

 

regards

Sanjay Morab

----- Original Message -----

From: Mark@...

To: sanjay@...

Sent: Friday, January 25, 2008 2:59 PM

Subject: AW: [ColdFire] 52235/52230 problem

X-SpamDetect-Info: ------------- Start ASpam results ---------------
X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information 
X-SpamDetect: ****: 4.300000 From3consonants=3, Some html comments=0, Aspam=0
X-SpamDetect-Info: ------------- End ASpam results -----------------

Hi Sanjay

 

I don’t see the standard pull-up resistors on the BDM interace lines TMS, TRST, TDI and TDO.

I wonder whether these are floating and so the chip is sometimes booting to DBM mode or during operation switched to BDM mode.

 

Regards

 

Mark

 

I have used 2 layers with NE64 (almost identical footprint and LAN interface) so I expect that it is possible also with the M52230. 4 layers will of course be better but I wouldn’t expect extreme changes.

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Sanjay Morab
Gesendet: Freitag, 25. Januar 2008 03:13
An: mark@...
Betreff: [ColdFire] 52235/52230 problem

 

Dear Mark

Hi,

 

We are using MCF52230 for our one of the project. We had developed all source code using the EVB. The code works fine. The EVB is for 52235. But when we transfer the coder to target board which has MCF 52230 there are few problems we are facing.

 

1. The Does not boot up all the time even with the sample code. It does ot detect the BDM also.

2. When it starts , for some time it works fine and then it starts giving the PC ERROR = 0x000294 or similar.

 

 

I am attaching the pdf of the schematic if it has any problems.

 

Please let me know currently the Board is Two layer, is it strongly recomended to have 4 layers.

 

Sanjay Morab

 

Accord Electronics
50, Rashmi Industrial Estate,
Vasai Industrial Area,
Navghar, Vasai Road(e),
Dist. Thane
Pin - 401210
Ph -[phone]
      [phone]
Fax - [phone]
H.P. - [phone]

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Mark Butcher]

Some parts of this message have been removed. Learn more about Nabble's security policy.

X-SpamDetect-Info: ------------- Start ASpam results ---------------
X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information
X-SpamDetect: ****: 4.300000 From3consonants=0.3, Some html comments=1.0, Aspam=3.0
X-SpamDetect-Info: ------------- End ASpam results -----------------

Sanjay

 

I recommend using the Layout guide in the following Freescale Application Note: AN2759 [it is for the NE64 but suitable also for the Coldfire]
 

 http://www.freescale.com/files/microcontrollers/doc/app_note/AN2759.pdf?fpsp=1&WT_TYPE=Application%20Notes&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=pdf&WT_ASSET=Documentation

 

The MC9S12NE64 has a heat sink area on the bottom of the chip which is used to help dissipate heat (from the PHY) to the board. This isn’t available on the M5223X and so it does tend to run a little hot.

Check the core supply voltage on VDD1 and VDD2 – it should be decoupled with capacitors and measure about 2.5V. I did have a board (M52233) which suddenly stopped working and this voltage was too low – it became very hot. After replacing the chip the board worked correctly again.

 

Good luck

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Sanjay Morab
Gesendet: Montag, 28. Januar 2008 03:04
An: mark@...
Betreff: Re: [ColdFire] 52235/52230 problem

 

Dear Mark

 

One information also required.

 

When we are using the two layer board for the ETHRNET what should be trace width and separation required from out put of the Magnatics to RJ45 connector as those two pairs are connected as differntial pair.

 

 

regards

Sanjay Morab

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Patrick Flannery-2]

Wendi, Mark, Steve,

 

Thank you all for your timely responses.

 

I didn't think that the RTOS that the application is using made a difference to a bootloader program so I didn't mention it in my original email. We are using a MQX RTOS for the MCF5213 known as MQX-lite. Is the RTOS that the application is using important? I thought once it was compiled it became part of the S19 file produced by the Code Warrior compiler. The bootloader I had envisioned would receive the S19 file a sector at a time thru the serial port, store it in SRAM, and then flash it into the 5213 flash. Am I oversimplifying what's required?

 

1. the bootloader reads a minimum flash sector size (2k) of the new application into SRAM through the serial port;

2. it flashes that into the next sector of the 5213 flash.

3. steps 1 and 2 are repeated until the new application has been fully received and flashed.

 

The 5213 flash is 256k and my application is above 220k so there isn't room to keep a copy of the old application stored anywhere in case the flash of the new application is unsuccessful. Since the hardware design is fixed and can't be changed, adding an external memory isn't possible. I realize this isn't the optimum method but I inherited the design and it's fixed due to safety approvals required by the application.

 

Pat

 

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Mark Butcher]

Some parts of this message have been removed. Learn more about Nabble's security policy.

X-SpamDetect-Info: ------------- Start ASpam results ---------------
X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information
X-SpamDetect: ****: 4.727000 From3consonants=0.3, Some html comments=1.0, IP in URL=0.4, Aspam=3.0
X-SpamDetect-Info: ------------- End ASpam results -----------------

Hi All

Please note that the uTasker project for the M5223X has just released a service pack (SP6) which includes additional chip support in form of the M5222X and M5221X. Details of the SP6 release are here:
http://www.utasker.com/forum/index.php?topic=167.0

Since the original uTasker V1.3 release, the feature set has been increased by SMTP login, TFTP and VLAN, FTP passive mode, GNU and Eclipse project support, NetBios, dynamic web content generation support and file system support in internal FLASH and/or external (multiple) SPI  FLASH devices.

The original boot software support (which enables SW uploads via Ethernet/Internet) has also be extended to allow it to use external FLASH and encryption.

An on-line demo is running at http://84.75.9.2/ which also shows dynamic web generation (Menu item called "Open a statistics window" is being used for a different function...when running on the M52235 EVB).

For any one interested in the project, please visit its home page below. It is open-source and free for educational and other non-commercial use, including free support on the uTasker forum: http://www.uTasker.com/forum/

New users are always welcome!

Regards

Mark 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[John Ulyate]

Some parts of this message have been removed. Learn more about Nabble's security policy.

X-SpamDetect-Info: ------------- Start ASpam results ---------------
X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information
X-SpamDetect: *****: 5.427000 ImageNoInfo=2.0, IP in URL=0.4, Aspam=3.0
X-SpamDetect-Info: ------------- End ASpam results -----------------

Mark,

 

I had a look at http://84.75.9.2/ - very nice, well done. We might have a project, I will contact you privately.

 

 

 

John Ulyate

Mobile : [phone]

e-Mail : julyate@...

Skype : john.ulyate

 

 

 

From: coldfire-bounce@... [mailto:coldfire-bounce@...] On Behalf Of Mark Butcher
Sent: 29 January 2008 01:47
To: John Ulyate
Subject: [ColdFire] uTasker Service Pack 6 released with support for M5223X, M5222X, M5221X

 

X-SpamDetect-Info: ------------- Start ASpam results ---------------

X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information 

X-SpamDetect: ****: 4.727000 From3consonants=3, Some html comments=0, IP in URL=4, Aspam=0

X-SpamDetect-Info: ------------- End ASpam results -----------------

Hi All

Please note that the uTasker project for the M5223X has just released a service pack (SP6) which includes additional chip support in form of the M5222X and M5221X. Details of the SP6 release are here:
http://www.utasker.com/forum/index.php?topic=167.0

Since the original uTasker V1.3 release, the feature set has been increased by SMTP login, TFTP and VLAN, FTP passive mode, GNU and Eclipse project support, NetBios, dynamic web content generation support and file system support in internal FLASH and/or external (multiple) SPI  FLASH devices.

The original boot software support (which enables SW uploads via Ethernet/Internet) has also be extended to allow it to use external FLASH and encryption.

An on-line demo is running at http://84.75.9.2/ which also shows dynamic web generation (Menu item called "Open a statistics window" is being used for a different function...when running on the M52235 EVB).

For any one interested in the project, please visit its home page below. It is open-source and free for educational and other non-commercial use, including free support on the uTasker forum: http://www.uTasker.com/forum/

New users are always welcome!

Regards

Mark 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Mark Butcher]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi Pat

 

If you are using a serial interface for loading new code there is no reason for keeping the old code also in FLASH (this is only useful if SW uploads are remote – via Internet for example – where a failure could otherwise leave the device in a state which inhibits further communication).

A boot loader via serial interface will require about 4..6k of space so you will have no problems with your application size.

 

The boot loader is independent of the OS used. It must simply ensure that the new code is loaded to the correct location (this is defined in the SREC) and then passes control to it (it needs to be able to jump to the start address of the new code – which can be a fixed address for simplicity – eg. start of first unused FLASH sector after the boot code). The application will need to have its interrupt vectors in SRAM (which is probably the case with MQX-Lite anyway) because the boot code must have its vector table fixed in FLASH starting at 0x00000000 so that it can handle the reset. VBR is set by the application to where its own interrupt vectors are stored.

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Patrick Flannery
Gesendet: Dienstag, 29. Januar 2008 00:31
An: mark@...
Betreff: [ColdFire] RE: 5213 bootloader

 

Wendi, Mark, Steve,

 

Thank you all for your timely responses.

 

I didn't think that the RTOS that the application is using made a difference to a bootloader program so I didn't mention it in my original email. We are using a MQX RTOS for the MCF5213 known as MQX-lite. Is the RTOS that the application is using important? I thought once it was compiled it became part of the S19 file produced by the Code Warrior compiler. The bootloader I had envisioned would receive the S19 file a sector at a time thru the serial port, store it in SRAM, and then flash it into the 5213 flash. Am I oversimplifying what's required?

 

1. the bootloader reads a minimum flash sector size (2k) of the new application into SRAM through the serial port;

2. it flashes that into the next sector of the 5213 flash.

3. steps 1 and 2 are repeated until the new application has been fully received and flashed.

 

The 5213 flash is 256k and my application is above 220k so there isn't room to keep a copy of the old application stored anywhere in case the flash of the new application is unsuccessful. Since the hardware design is fixed and can't be changed, adding an external memory isn't possible. I realize this isn't the optimum method but I inherited the design and it's fixed due to safety approvals required by the application.

 

Pat

 

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[wrw]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi Pat:

 

I think I read in your previous email that you must comply with some sort of regulatory agency as well.  If this is true you will most likely need to insure the code is good in some way (i.e checksum).  This should be run by the “bootloader”.  Additionally I would like to clarify on how you want to run this “bootloader”.  Do you want to always load the app or sometimes want to Flash the new code with the bootloader?

 

Wendi

 

---

From: coldfire-bounce@... [mailto:coldfire-bounce@...] On Behalf Of Mark Butcher
Sent: Tuesday, January 29, 2008 3:19 AM
To: wrw
Subject: AW: [ColdFire] RE: 5213 bootloader

 

Hi Pat

 

If you are using a serial interface for loading new code there is no reason for keeping the old code also in FLASH (this is only useful if SW uploads are remote – via Internet for example – where a failure could otherwise leave the device in a state which inhibits further communication).

A boot loader via serial interface will require about 4..6k of space so you will have no problems with your application size.

 

The boot loader is independent of the OS used. It must simply ensure that the new code is loaded to the correct location (this is defined in the SREC) and then passes control to it (it needs to be able to jump to the start address of the new code – which can be a fixed address for simplicity – eg. start of first unused FLASH sector after the boot code). The application will need to have its interrupt vectors in SRAM (which is probably the case with MQX-Lite anyway) because the boot code must have its vector table fixed in FLASH starting at 0x00000000 so that it can handle the reset. VBR is set by the application to where its own interrupt vectors are stored.

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Patrick Flannery
Gesendet: Dienstag, 29. Januar 2008 00:31
An: mark@...
Betreff: [ColdFire] RE: 5213 bootloader

 

Wendi, Mark, Steve,

 

Thank you all for your timely responses.

 

I didn't think that the RTOS that the application is using made a difference to a bootloader program so I didn't mention it in my original email. We are using a MQX RTOS for the MCF5213 known as MQX-lite. Is the RTOS that the application is using important? I thought once it was compiled it became part of the S19 file produced by the Code Warrior compiler. The bootloader I had envisioned would receive the S19 file a sector at a time thru the serial port, store it in SRAM, and then flash it into the 5213 flash. Am I oversimplifying what's required?

 

1. the bootloader reads a minimum flash sector size (2k) of the new application into SRAM through the serial port;

2. it flashes that into the next sector of the 5213 flash.

3. steps 1 and 2 are repeated until the new application has been fully received and flashed.

 

The 5213 flash is 256k and my application is above 220k so there isn't room to keep a copy of the old application stored anywhere in case the flash of the new application is unsuccessful. Since the hardware design is fixed and can't be changed, adding an external memory isn't possible. I realize this isn't the optimum method but I inherited the design and it's fixed due to safety approvals required by the application.

 

Pat

 

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Sanjay Morab]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Dear Mark

 

I had cheked the the VDD1 and VDD2 supplies. Vdd2 it is 2.5 and on VDD1 I am getting frequency instead of clean DC. I had used 0.22 uF decoupling as used in EVB. This consistant problem I am facing is The File gets downloaded to the MCF52230 but does not execute. All three board I had assembled does the same.

 

 

Sanjay Morab

 

Accord Electronics
50, Rashmi Industrial Estate,
Vasai Industrial Area,
Navghar, Vasai Road(e),
Dist. Thane
Pin - 401210
Ph -[phone]
      [phone]
Fax - [phone]
H.P. - [phone]

----- Original Message -----

From: Mark@...

To: sanjay@...

Sent: Monday, January 28, 2008 4:02 PM

Subject: AW: [ColdFire] 52235/52230 problem

X-SpamDetect-Info: ------------- Start ASpam results ---------------
X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information 
X-SpamDetect: ****: 4.300000 From3consonants=3, Some html comments=0, Aspam=0
X-SpamDetect-Info: ------------- End ASpam results -----------------

Sanjay

 

I recommend using the Layout guide in the following Freescale Application Note: AN2759 [it is for the NE64 but suitable also for the Coldfire]
 

 http://www.freescale.com/files/microcontrollers/doc/app_note/AN2759.pdf?fpsp=1&WT_TYPE=Application%20Notes&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=pdf&WT_ASSET=Documentation

 

The MC9S12NE64 has a heat sink area on the bottom of the chip which is used to help dissipate heat (from the PHY) to the board. This isn’t available on the M5223X and so it does tend to run a little hot.

Check the core supply voltage on VDD1 and VDD2 – it should be decoupled with capacitors and measure about 2.5V. I did have a board (M52233) which suddenly stopped working and this voltage was too low – it became very hot. After replacing the chip the board worked correctly again.

 

Good luck

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Sanjay Morab
Gesendet: Montag, 28. Januar 2008 03:04
An: mark@...
Betreff: Re: [ColdFire] 52235/52230 problem

 

Dear Mark

 

One information also required.

 

When we are using the two layer board for the ETHRNET what should be trace width and separation required from out put of the Magnatics to RJ45 connector as those two pairs are connected as differntial pair.

 

 

regards

Sanjay Morab

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Seymour David-ra2693]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi Sanjay,

Are you certain you are programming the flash?  After programming best to peek at memory to verify that the core really was programmed into flash.

Also good to test using small piece of code to load directly to SRAM and try running/debugging to test the tools.

Regards,

David

 

David E Seymour

---

From: coldfire-bounce@... [mailto:coldfire-bounce@...] On Behalf Of Sanjay Morab
Sent: Wednesday, January 30, 2008 2:37 AM
To: Seymour David
Subject: Re: [ColdFire] 52235/52230 problem

 

Dear Mark

 

I had cheked the the VDD1 and VDD2 supplies. Vdd2 it is 2.5 and on VDD1 I am getting frequency instead of clean DC. I had used 0.22 uF decoupling as used in EVB. This consistant problem I am facing is The File gets downloaded to the MCF52230 but does not execute. All three board I had assembled does the same.

 

 

Sanjay Morab

 

Accord Electronics
50, Rashmi Industrial Estate,
Vasai Industrial Area,
Navghar, Vasai Road(e),
Dist. Thane
Pin - 401210
Ph -[phone]
      [phone]
Fax - [phone]
H.P. - [phone]

----- Original Message -----

From: Mark@...

To: sanjay@...

Sent: Monday, January 28, 2008 4:02 PM

Subject: AW: [ColdFire] 52235/52230 problem

 

X-SpamDetect-Info: ------------- Start ASpam results ---------------

X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information 

X-SpamDetect: ****: 4.300000 From3consonants=3, Some html comments=0, Aspam=0

X-SpamDetect-Info: ------------- End ASpam results -----------------

Sanjay

 

I recommend using the Layout guide in the following Freescale Application Note: AN2759 [it is for the NE64 but suitable also for the Coldfire]
 

 http://www.freescale.com/files/microcontrollers/doc/app_note/AN2759.pdf?fpsp=1&WT_TYPE=Application%20Notes&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=pdf&WT_ASSET=Documentation

 

The MC9S12NE64 has a heat sink area on the bottom of the chip which is used to help dissipate heat (from the PHY) to the board. This isn’t available on the M5223X and so it does tend to run a little hot.

Check the core supply voltage on VDD1 and VDD2 – it should be decoupled with capacitors and measure about 2.5V. I did have a board (M52233) which suddenly stopped working and this voltage was too low – it became very hot. After replacing the chip the board worked correctly again.

 

Good luck

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Sanjay Morab
Gesendet: Montag, 28. Januar 2008 03:04
An: mark@...
Betreff: Re: [ColdFire] 52235/52230 problem

 

Dear Mark

 

One information also required.

 

When we are using the two layer board for the ETHRNET what should be trace width and separation required from out put of the Magnatics to RJ45 connector as those two pairs are connected as differntial pair.

 

 

regards

Sanjay Morab

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Sanjay Morab]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Dear David

 

Yes the flash is programmed and varified. Also I had written small code and tested. I had checked with SRAM also but the result is same. The Code is downloaded to SRAM but no execution. On EVB same codes are working in SRAM as well as FLASH

Sanjay Morab

 

Accord Electronics
50, Rashmi Industrial Estate,
Vasai Industrial Area,
Navghar, Vasai Road(e),
Dist. Thane
Pin - 401210
Ph -[phone]
      [phone]
Fax - [phone]
H.P. - [phone]

----- Original Message -----

From: David.Seymour@...

To: sanjay@...

Sent: Wednesday, January 30, 2008 8:25 PM

Subject: RE: [ColdFire] 52235/52230 problem

Hi Sanjay,

Are you certain you are programming the flash?  After programming best to peek at memory to verify that the core really was programmed into flash.

Also good to test using small piece of code to load directly to SRAM and try running/debugging to test the tools.

Regards,

David

 

David E Seymour

---

From: coldfire-bounce@... [mailto:coldfire-bounce@...] On Behalf Of Sanjay Morab
Sent: Wednesday, January 30, 2008 2:37 AM
To: Seymour David
Subject: Re: [ColdFire] 52235/52230 problem

 

Dear Mark

 

I had cheked the the VDD1 and VDD2 supplies. Vdd2 it is 2.5 and on VDD1 I am getting frequency instead of clean DC. I had used 0.22 uF decoupling as used in EVB. This consistant problem I am facing is The File gets downloaded to the MCF52230 but does not execute. All three board I had assembled does the same.

 

 

Sanjay Morab

 

Accord Electronics
50, Rashmi Industrial Estate,
Vasai Industrial Area,
Navghar, Vasai Road(e),
Dist. Thane
Pin - 401210
Ph -[phone]
      [phone]
Fax - [phone]
H.P. - [phone]

----- Original Message -----

From: Mark@...

To: sanjay@...

Sent: Monday, January 28, 2008 4:02 PM

Subject: AW: [ColdFire] 52235/52230 problem

 

X-SpamDetect-Info: ------------- Start ASpam results ---------------

X-SpamDetect-Info: This message may be spam. This message BODY has been altered to show you the spam information 

X-SpamDetect: ****: 4.300000 From3consonants=3, Some html comments=0, Aspam=0

X-SpamDetect-Info: ------------- End ASpam results -----------------

Sanjay

 

I recommend using the Layout guide in the following Freescale Application Note: AN2759 [it is for the NE64 but suitable also for the Coldfire]
 

 http://www.freescale.com/files/microcontrollers/doc/app_note/AN2759.pdf?fpsp=1&WT_TYPE=Application%20Notes&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=pdf&WT_ASSET=Documentation

 

The MC9S12NE64 has a heat sink area on the bottom of the chip which is used to help dissipate heat (from the PHY) to the board. This isn’t available on the M5223X and so it does tend to run a little hot.

Check the core supply voltage on VDD1 and VDD2 – it should be decoupled with capacitors and measure about 2.5V. I did have a board (M52233) which suddenly stopped working and this voltage was too low – it became very hot. After replacing the chip the board worked correctly again.

 

Good luck

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Sanjay Morab
Gesendet: Montag, 28. Januar 2008 03:04
An: mark@...
Betreff: Re: [ColdFire] 52235/52230 problem

 

Dear Mark

 

One information also required.

 

When we are using the two layer board for the ETHRNET what should be trace width and separation required from out put of the Magnatics to RJ45 connector as those two pairs are connected as differntial pair.

 

 

regards

Sanjay Morab

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.

[Patrick Flannery-2]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi Wendi,

 

I would like to use the bootloader to flash new applications only when a firmware upgrade has been created. This will allow the customer to perform firmware upgrades in the field.

Also I had intended to perform a checksum operation during the flashing of the new application.

 

Pat

---

From: coldfire-bounce@... [mailto:coldfire-bounce@...] On Behalf Of Wendi Whitcomb
Sent: Tuesday, January 29, 2008 12:11 PM
To: Patrick Flannery
Subject: RE: [ColdFire] RE: 5213 bootloader

Hi Pat:

 

I think I read in your previous email that you must comply with some sort of regulatory agency as well.  If this is true you will most likely need to insure the code is good in some way (i.e checksum).  This should be run by the “bootloader”.  Additionally I would like to clarify on how you want to run this “bootloader”.  Do you want to always load the app or sometimes want to Flash the new code with the bootloader?

 

Wendi

 

---

From: coldfire-bounce@... [mailto:coldfire-bounce@...] On Behalf Of Mark Butcher
Sent: Tuesday, January 29, 2008 3:19 AM
To: wrw
Subject: AW: [ColdFire] RE: 5213 bootloader

 

Hi Pat

 

If you are using a serial interface for loading new code there is no reason for keeping the old code also in FLASH (this is only useful if SW uploads are remote – via Internet for example – where a failure could otherwise leave the device in a state which inhibits further communication).

A boot loader via serial interface will require about 4..6k of space so you will have no problems with your application size.

 

The boot loader is independent of the OS used. It must simply ensure that the new code is loaded to the correct location (this is defined in the SREC) and then passes control to it (it needs to be able to jump to the start address of the new code – which can be a fixed address for simplicity – eg. start of first unused FLASH sector after the boot code). The application will need to have its interrupt vectors in SRAM (which is probably the case with MQX-Lite anyway) because the boot code must have its vector table fixed in FLASH starting at 0x00000000 so that it can handle the reset. VBR is set by the application to where its own interrupt vectors are stored.

 

Regards

 

Mark

 

 

 

Mark Butcher Bsc (Hons) MPhil MIEE

 

M.J.Butcher Consulting               

Obere Bahnhofstr. 13,  CH-5507 Mellingen,

Switzerland         www.mjbc.ch / www.uTasker.com

056 491 48 10 / 079 402 26 37 / Skype: M_J_Butcher

 

 

 

-----Ursprüngliche Nachricht-----
Von: coldfire-bounce@... [mailto:coldfire-bounce@...] Im Auftrag von Patrick Flannery
Gesendet: Dienstag, 29. Januar 2008 00:31
An: mark@...
Betreff: [ColdFire] RE: 5213 bootloader

 

Wendi, Mark, Steve,

 

Thank you all for your timely responses.

 

I didn't think that the RTOS that the application is using made a difference to a bootloader program so I didn't mention it in my original email. We are using a MQX RTOS for the MCF5213 known as MQX-lite. Is the RTOS that the application is using important? I thought once it was compiled it became part of the S19 file produced by the Code Warrior compiler. The bootloader I had envisioned would receive the S19 file a sector at a time thru the serial port, store it in SRAM, and then flash it into the 5213 flash. Am I oversimplifying what's required?

 

1. the bootloader reads a minimum flash sector size (2k) of the new application into SRAM through the serial port;

2. it flashes that into the next sector of the 5213 flash.

3. steps 1 and 2 are repeated until the new application has been fully received and flashed.

 

The 5213 flash is 256k and my application is above 220k so there isn't room to keep a copy of the old application stored anywhere in case the flash of the new application is unsuccessful. Since the hardware design is fixed and can't be changed, adding an external memory isn't possible. I realize this isn't the optimum method but I inherited the design and it's fixed due to safety approvals required by the application.

 

Pat

 

 

coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list. coldfire@... Send a post to the list. coldfire-join@... Join the list. coldfire-digest@... Join the list in digest mode. coldfire-leave@... Leave the list.