Other Parts Discussed in Thread: ARM-CGT,
Hi,
we're facing some dramatic performance issues when using the newer v20 ARM-CGT compiler for AM3358.
As far as I remember, there was an option in previous versions of that compiler to mark some specific functions in code to be executed in cache of the SoC. This option seem to be gone in newer versions - causing some nameable speed issues with exactly these functions, which have been marked for cache-usage in our existing code.
So my question: was this option removed completely? Or is it just undocumented? Can anybody shed some light on what happened with this optimisation-parameter?
Thanks!