K60 upper flash block not working?

[BrentW]

K60N512 0m33Z versus K60N512 1N30D.

I have working code developed on the 0m33Z that no longer works on the 1N30D. The upper block of flash (0x40000 - 0x7FFFF) appears to be totally inaccessible, almost as if the device was a 256K instead of a 512K.

I have tried several things with IAR EW (6.3) – erasing the entire block, filling a sector, and loading a debug workspace located at 0x40000. The erase acted like it worked but after checking, it did not. Filling a sector caused errors, as did the debug app.

If I was dealing with rev 2 silicon I would have a few ideas to try.

Thanks in Advance for any ideas, help, etc.

Brent

[comsosysarch]

There is a errata on 1Nblah silicon (v1.2) preliminary devices that the PGMSEC (program sector) flash controller command does not work. You have to use program phrase or words or whatever and takes a long time to program a huge data table or something.

This worked on 0m33z (v1.0) but was busted on v1.2. Reportedly it now works on v1.4 (4n-blah) production devices.

This does not sound exactly like your problem but it is close enough to a part of it that it bears mentioning.

[BrentW]

Thanks comsosysarch. I'm only doing 32 bit writes. A good stab at it though.

iafg, true story - neither have FlexNVM.

The only other thing I've found is that the PFLSH bit in the SIM_FCFG2 register is set to 1 in the 0M33 device and set to 0 in the 1N30 device. I have found no way yet to modify that setting.

Thanks,

Brent

[comsosysarch · Accepted solution]

I vaguely recall we had some obscure alert from the distributor where we got some K60's that some had the flash size configured incorrectly, apparently it is the same hardware and just a configuration bit is flipped at the factory to allow access (saving them alot of money making two different sets of Silicon so not as odd as it sounds, though, yeah, funny huh).

Our devices were supposed to have been effected but were not.

Maybe yours were? Can you exchange them? Doubt you can change that bit yourself, but hey who knows...

Sorry, it has been so long I no longer seem to have the email archived to refer back.

And I do not recall seeing any actual errata for that, although I may not have looked since we did not have the issue on our chips in house and it was only one particular batch of them if I recall, presuming you already looked over the errata sheet anyhow.

[BrentW]

Thanks again! I've got a Service Request in now. We'll see what they have to say. Exchanging would be a bit difficult since they are installed. Fortunately still in R&D so low quantities but it could delay development a bit.

No worries on the reply time - it's all relative.

[egoodii]

You never returned the full chip marking.  I will just ask -- 1121?

[iafg]

You'll probably want to include the full part number as it is marked on the chip...