Glitch in PLL1 source code

lpcware — Wed, 15 Jun 2016 19:19:38 GMT

Content originally posted in LPCWare by mch0 on Sat May 31 03:57:51 MST 2014
OK, back again after I realized my first solution did not work.

The problem is this:
The code supplied with LPCOpen (V2.12) to set up the core frequency has a fair chance to crash when executed from local SRAM.
For my project I need to execute from SRAM, since I have to modify the SPIFI flash content.
To do this I decided to relocate all code as early as possible from SPIFI to SRAM and then solely execute from there. I specically wanted to move everything, because then I don't have to care about later additions or whatever to the project and keep track of what resides where.

So I moved also the SysInit-Code and executed it from SRAM. Result: crash.
It turned out the original code (as supplied) would change the CPU frequency to 204MHz when executing from SPIFI, but not when executing from SRAM.

I feel that someone from NXP should look into this and check whether this is really a bug. In that case I'd suggest a correction of the code in V2.13.

I'll be glad to help if questions arise.
The attached report describes the battle and also contains a preliminary fix.

Somehow I feel I could do with a full license instead of the free license I use now :) (hint, hint).

Regards,

Mike

LPC MicrocontrollersのトピックGlitch in PLL1 source code

Glitch in PLL1 source code