This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

IAR upgrade problems

Other Parts Discussed in Thread: EK-LM3S811

My project was developed using IAR version 5.5 and is deployed and working well.  Now my client wants to add WiFi and I was out of flash space.  The LM3S9B92 has 256K and it was full.  Furthermore my IAR compiler was memory limited to 256K.  So I upgraded to the newest version (6.7) with unlimited memory.  I will be using the LM3S9D92 which is the same as the 9B but with double the flash (512K).  

The problem is I cannot get my old solid firmware to run using the new IAR software.  Some setup parameters have changed and I followed the migration directions but I can't get my old software to run with the new compiler.  Sometimes pointers are pointing to non-existing places and sometimes it just fault vector out depending on setup parameters.  sprintf() functions won't properly convert floating point numbers.  Tweaking setup parameters changes things and one configuration almost seems to work but its flaky and fragile.

Going back to the old WARM 5.5 compiler generates solid working code. 

The only thing that has changed is the compiler.  Source code and target platform are the same.

I've been pulling my hair out for 2 weeks trying to get this to work.

Anyone seen anything like this? Or have any suggestions?

Calls into IAR have not proven fruitful as yet.

  • We too have IAR - 256KB paid, multi-seat version (6.70.1.5794) - and it seems content under StellarisWare w/past (NRND) Cortex M3 and newer Cortex M4.  I do recall that at/around Ver 5.0 or 5.5 - the changes were large & may have caused some issues.  (and - sad to say - we knew we weren't "going back" - thus our "corrective actions/fixes" have dimmed w/passage of time...)

    We have reasonably current J-Links (Segger) and these update far more regularly than does IAR.  With luck - perhaps an update of your J-Link may provide comfort.  (that's certainly the 1st step I'd take - followed closely by "relaxing" your "optimizations" {None would be a good beginning...})

    We receive excellent product support from IAR.  As always - your ability to provide a tightly defined, "test case" -enables their best analysis. 

    Suggest that you get a simple project - employing limited & basic pointers - up/running.  And - use the older device - just to eliminate "issues" w/the newer MCU.  Be sure that you've linked in the older cm3 library - as I recall that caused some concern in our past.

    If you're still stuck - may make sense to, "beg, borrow, steal - God forbid purchase" an EK-LM3S811 eval board - so that your board and IDE may work in lockstep w/ours.    If you do that - we can exchange critical "screen caps" - and try to see what has, "run amuck." 

    Appears there's at least one, "lighthouse" (incandescently (40W)/illegally lit) which aims to shine some path thru your darkness...