I can get rid of this error if I decrease my heap size by 1 in the assembly startup file. The linker then reports: Program Size: Code=2928 RO-data=96 RW-data=12 ZI-data=13356
But during a debug session uVision reports: *** Restricted Version with 16384 Byte Code Size Limit *** Currently used: 3036 Bytes (18%)
My concern is the linker is adding up the total program size including ZI-data which too restrictive. Is there an eval version of the linker or linker option that just looks at Code size? I bought this eval kit because I believed I could actually get 16K Code size (not total program size).
I bought this eval kit Bought???
the eval is free. If it bundled with some hardware you pay only for the hardware.
Erik
Stop your whining Erik:) I purchased the hardware with the understanding that the software tools were restricitive but not so restrictive. How much did you pay?
Basically I can work around the linker restrictions using a smaller heap and smaller problem sizes. Thank goodness whatever gets past the linker, the debugger will debug all the way up to the full 16K *Code* size limitation.
Stop your whining Erik:) accepting the :) I continue: how can stating facts be 'whining' How much did you pay? I (i.e. my company) paid for 3 sets of full keil
sir erac
i now also be evaleuate the arm and too is hard for code and small ...... yes.
but simpel projetcs are good and be working. i simulator the small codes very good you be knowing.
i no complain for you #)
you be please do not worry for the money ))-
kalib
You do know that you're not suppose to SELL binaries built by evaluation software, don't you?
You do know that you're not suppose to SELL binaries built by evaluation software, don't you? I have no doubt that the den of thieves that use the eval for commercial purposes are aware of that. They just choose to ignore it.
just so that there is no doubt I will rephrase the above: "using the eval for anything that produce an income is outright thieving"
On the issue, I did once compare the various toolsets before we bought 8 'seats' and had to get the full versions of all to do a real evaluation. The word 'evaluation' really should be replaced by 'demonstration'. All, except Tasking were willing to loan a full set for the evaluation since the company was a reputable (large $1B++ annually) business.
erac,
you is right you be knowing.
thank you for yo the kindly helping.
My concern is the linker is adding up the total program size including ZI-data which too restrictive.
On an ARM, nothing stops you from putting code in your "data" memory and running it there. If the linker didn't add this up, you could, without too much trouble, build an application with a code size larger than 16 kB.
On the other hand, a creative thief can find any number of ways to avoid paying. The most creative must still be to post on this forum to ask if soneone has any cracks...
On the other hand, a creative thief can find any number of ways to avoid paying.
Of course. That doesn't mean that you need to leave an easily-prevented method available for use.
I (...) paid for 3 sets of full keil
we knowing sir erac is very good yes
but he is 3 more better with me because i be having only one fulll keil
he is 300% better yes than everyone!!!!!!
I could also get another eval board that comes with software that allows larger code:
www.st.com/.../contentid-54-101-STR91X_SK__IAR.html
Then I could ship code with every ARM processor that I steal:)
Have you checked that it doesn't make the same interpretation (or "misinterpretation") of "code size"...?
;-)