Incomplete MCF52235RM (Rev. 1 06/2006)

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Incomplete MCF52235RM (Rev. 1 06/2006)

6,883 Views
Jaux
Contributor I
We are in the process of porting Forth (SwiftX) to the MCF52235. The reference manual is incomplete with omissions, obvious errors and inconsistencies. I have been able to fill some of the gaps looking at the flood of code generated by CodeWarrior 6.3

Specific:

1. The latest published reference manual ( MCF52235RM (06/2006, Rev 1 )) is a mess, specifically Appendix A. There are entire address ranges missing. And in the main body of the manual, there are conflicts, errors, and omissions. Tables 15-13 and 15-14 (Interrupt Sources) are full of errors. There are duplicate EPORT sources (and there is no EPF0 !!) in 15-14.

2. The chip status is Active and the RM is still preliminary! Is there a less-preliminary version?

3. I have made a comparison between the header files created by CodeWarrior 6.3 ( which was shipped with the M52235EVB ) and the register addresses from the manual and it seems the CodeWarrior people worked from a different manual than the one published on your web site. I cannot help to wonder what inside source the CodeWarrior people used. It surely was not this manual!
Labels (1)
0 Kudos
18 Replies

714 Views
DrSeuss
Contributor I
Mr. Vagnoni,
At 1:33am last night I finally figured out your comment of chapter22. There are only two PITs in this device.
I will fix this and any identified errors that are brought to my attention. Please don't keep me up at night with seek and find puzzles. I spent days wondering what was wrong with the title: Programmable Interrupt Timer Modules.
 
All,
I'm really not sure this is the best place to post typos with the manual, as this thread will explode. The proper procedure is to open a service request on our help site. I know this is a pain, but I know of no better answer right now.
 
Understand that my time is multi-tasked with manual and other technical issues and creating exciting new MCUs.
With this family of devices, a decision was make to release the device with preliminary documentation, so customers that wished could play with this exciting new device. This decision is a difficult one as some customers could care less about the docs, others care more of the docs than the device.
 
This device was released with a ton of example software that has helped most customer get over the doc issue. I did not foresee that these software examples would not help all customers get over the doc.
 
I will be meeting with the doc people next week and I think we will make one more pass though the manual and then release it. (bookmark that SR page!) This should be in about a month or two.
 
Thanks to all for the support of this device. (and its manual)
0 Kudos

714 Views
Jaux
Contributor I
Amen to that.

We on the outside really appreciate all the effort going into the design and developing of a new chip, and we actually have sympathy for the docs lagging the actual device. The advantages of having the device with preliminary documentation far outweigh the other issues, and if we do grumble in frustration not getting the chip to work with the advance information - this is called synergetic growth.

I have the advantage of having an interactive link into the chip and a lot of the problems was solved using reverse engineered sample programs, snippets from CodeWarrior and one or two clever posts on the forum - my thanks for that input.

I think this issue has now been put to bed, and as the originator of this subject, I declare it closed with my thanks and appreciation to the Freescale Documentation Dept for all the hard work.
0 Kudos

714 Views
DrSeuss
Contributor I
I talk with swiftX today, I will work with them to provide all info necessary and to get them on our website.
 
FYI, Nice Guys. All they wanted was to help make freesacle customers happy. My kind of folk!
0 Kudos

714 Views
Jaux
Contributor I
People of the Planet Freescale,

This is why we remained loyal for more than 20 years despite little glitches here and there.

DrSeuss, I salute you
0 Kudos

714 Views
DrSeuss
Contributor I
While proofing the document we discovered more issues with Appendix A. I have been promised it will go live tomorrow.
 
Is there a specific question or issue that I can help with?
0 Kudos

714 Views
Jaux
Contributor I
Hello Dr Seuss,

I wish there was only one issue. As I explained to you, I am using embedded Forth (SwiftX) to do the software development, ( our friends at Metrowerks in the past has completely killed my enthusiasm for CodeWarrior ) and although we got the s/w running on the 52235 via BDM I have to get going with the development of the intended application. I have been twiddling my thumbs since June and just got frustrated trying to figure the maze in the chip without a map.

If you say the rev 2 will be published tomorrow, I am happy. I am sure a lot of other people is waiting with the same anticipation as me for this momentous event.
0 Kudos

714 Views
DrSeuss
Contributor I
THE NEW PHONE BOOK IS HERE!
THE NEW PHONE BOOK IS HERE!
 
 
I also need to credit the Pubs folks:
 
I have also contacted SwiftX (no reply) to discuss why they are not contacted freescale for support of their product.
 
While I have also had issues with CodeWarrior (and I am biased), this is the main release method for my device config and header files. I would give them another try. They are on my "good" list. ( I think that makes the list equal 1)   
 
 :-)
 
 
I will open a new thread if I hear from swiftX.
 
 
 
0 Kudos

714 Views
p_vagnoni
Contributor III
Hello,
 
I am new on this forum, i am looking for news in last 2 months because I found many errors on the CFM52235RM manual and now, in the rev. 2 some of them are present. (title of chapter 22 is a big exaple...).
 
I need this manual because I am working on this new micro. What revision of manual will be a real Reference Manual?
 
Bye,
 
Paolo Vagnoni
0 Kudos

714 Views
Jaux
Contributor I
I have reported several errors in rev 1 of the RM, and as you have noted there are some more errors. The latest is rev 2 ( released on Friday Sept 9 )

I use a different tool for software engineering than CodeWarrior - it is called SwiftX and is an embedded Forth. ( www.forth.com )

My advise to you is to use rev 2 of the manual and try to read over the mistakes. I have been able to get the chip running and I can talk to it via BDM module. I am in the process of verifying the chip/RM rev2.

Is there anything spesific you have problems with?

J
0 Kudos

714 Views
Jaux
Contributor I
Oh Dr, Dr, I feel so much better already!

Wonderful news - and please convey my thanks to the documentation people. I am sure this is a new Freescale all-time record.

Forth Inc's contact details: ( you might have the wrong tel no )

Leon Wagner (CEO)

FORTH, Inc.
5155 W. Rosecrans Avenue
Suite 1018
Hawthorne, CA 90250 USA

Telephone
310.491.3356

Toll-free (U.S. and Canada)
800.55.FORTH
888.55.FORTH (sales hotline)

Fax
310.978.9454


I think you are right - I cannot blame the CodeWarrior guys for something Metrowerks has done. I will give it another look-at (very precautiously) Maybe they can turn this lemon into an orange.
0 Kudos

714 Views
DrSeuss
Contributor I

Thank you for your recent service request (1-310643355) concerning problems with the MCF52235 reference manual.  We appreciate your comments and thank you for alerting us to the issues you mentioned.  Furthermore, we apologize for the inconvenience that these issues have caused you.  We are aware of some inconsistencies and errors in the 52235 reference manual, and devote considerable internal effort to fixing these errors. 

 Customer feedback, such as yours, is invaluable in helping us identify and solve the inconsistencies.  The presence of these inconsistencies is one reason why the manual is still labeled "preliminary".  The next reference manual revision will include numerous fixes that will address at least some of your concerns.  For example, we are in the process of making sure that Appendix A (memory map) and the body of the reference manul are consistent.  The interrupt source tables (15-13 and 15-14) have also been updated.  The current Edge Port information (Section 16.1) is in fact correct; there are two Edge Port modules, the first of which has only 7 inputs (i.e., missing EPF0).  The CodeWarrior documentation team uses the same reference manuals that are published on our external Web site; we will look into any inconsistencies between the manual and the header files.

 Please look for the new revision of the manual (Rev 2) on our external Web site at the end of August 2006.

0 Kudos

714 Views
Jaux
Contributor I
Dr Suess,

You mentioned that rev 2 of the reference manual would have been published Setember 1 - what happened to it?
0 Kudos

714 Views
Ricardo_RauppV
Contributor I
Dr Suess
We have 4 designs been designed right now with MCF52235.We really appreciate Freescale provide the final revision of such manual.We´ve searched for this since September 1 and we didin´t find it...We really  hope Freescale can solve it soon ...
0 Kudos

714 Views
Jaux
Contributor I
Ricardo,

It does not look like we are going to get any help from Freescale soon. I suggest we pool resources and maybe we can solve each other's problems. My direct e-mail is julyate@openmrt.co.za, BR John Ulyate
0 Kudos

714 Views
Jaux
Contributor I
-> DrSeuss

Good morning - thanks for taking the time to reply, I take note of your comments and will check for the revised RM.

This is typical why Freescale is such a great company to "work" for. The people are dedicated and share a deep commitment of producing the finest in silicon.

Keep up the great work!
0 Kudos

714 Views
Jaux
Contributor I
--> Tom

Thanks. You people at Freescale sometimes don't realize this but a lot of us are religiously loyal to the company ( in my case for the past 25 years )

We spent countless and thankless hours developing the little gadgets that people take for granted and which make you a nice heap of money.

As Ricardo said, it is hard enough competing, and the MCF52235 has enormous potential, please make sure that when you release a chip, the support documentation is there and up to date. I have to agree - you actually made the chip...


BR,

John Ulyate
CTO : OpenMRT (Pty) Ltd
0 Kudos

714 Views
Ricardo_RauppV
Contributor I

Dear Freescale Team

Although MCF52235 is a very promising device, it is in fact very hard and unproductive start a design with such Reference Manual. It is in fact very very poor. Instead of add resources to new designs, we spend time trying to guess some specs.  We believe the hardest part, you have made: the chip...

0 Kudos

714 Views
J2MEJediMaster
Specialist I
I'm working on passing this info on to the appropriate people. Development work is tough enough without the documentation hindering rather than helping you.

---Tom
0 Kudos