New MCF5225x manuals disappointing. Information flow?

[Obetz]

Hello all,

anybody else missing his error reports in the new MCF5225x manuals?

I submitted three documentation error reports (if someone from Freescale reads this: 1-690691025, 1-694827321, 1-694854494) last November but all three errors are in the new manual. I wonder why I put a lot of work into error reports if they are not published.

And I'm afraid there is even a new error (I currently can't check it): The CAN clock source changed from fsys/2 to fsys. Is it really fsys?

Also extremly annoying is the fact that putting 3,3V into EXTAL "may lead to a decrease in lifespan of the device". What about the large number of devices already shipped, using an external oscillator powered from Vcc?

Hard to stay calm.

Oliver

[TomE]

Please report your errors to this forum. That way others can find them by searching.

"MCF5329 with random SDRAM 1k "bit rot" corruptions" will show you the sort of information that we need that probably won't ever make it into the manuals. Likewise "MCF5329 (MCF5xxx) + SDR DRAM = lockup, anyone else?".

Search here for "Re: PDF "Version 3 ColdFire Core User’s Manual" document looks to be corrupted.' for reports of an essential manual with pages missing, where the response was "the pdf you mentioned is quite old and we do not have a new version of it". For the manual that describes how the CPU Core in all the MCF53 chips works, and they don't have a good copy?

Google for "freescale bear pit"for an example of cut-and-paste paragraphs that could have been written better, but have been cut-pasted from the MMC2107 manuals into everything else. AN3400 for the MCF5213 lists one of the traps, but you'd never find that App Note if working on any of the other parts. I don't really expect manuals for 12 year old parts to get updated. The team that worked on it is working on something newer...

I'm still waiting for some of my bug reports from 11 years ago to see results :smileyhappy:

Tom

[Obetz]

---

TomE wrote:

Google for "freescale bear pit"for an example of cut-and-paste paragraphs that could have been written better

---

I'm not sure which thread you mean.

Oliver

[TomE]

There are only two threads that show up with this search. Both are relevant.

Read "PIT hw boo-boo. Read if you need accurate PIT (CF)".

The two problems are:

1 - You have to ACK the PIT interrupt with a BYTE write, but the standard register definitions (in the manual and copied from there into header files) only defines the register as a 16-bit word. So you have to add your own definitions to be able to write this register "properly" and not reset the timer.

2 - In order to load the PIT with an initial value you have to write the OVR bit first. If you don't it downcounts from all-ones for the first cycle.

Both of these "Bear PITs" have been there for over 10 years.

[macl]

We at Freescale are investigating what happened with the documentation updates.  Your feedback and contributions to the community are very valuable.  Thanks for posting on this issue.

[FridgeFreezer]

macl - investigating is good, reporting back what you find & are going to do about it, and posting updated data sheets is even better...