 RE: But *SHOULD* I? Cpt. Vince Andy, I do appreciate your response. Especailly from you. Since the overall design is in constant flux for one of my latest projects, the question of building a generic platform for 'any' ARM/Cortex processor comes into play. Of course I don't want to box myself in, and after [you] told me about the new ST libraries, I have to determine if it is worth the effort to vet the new library that is CMSIS compliant and incorporate it. Using the CMSIS rendition of ST's library, I wrote a rather basic RTX model, and then I added a generic time-base (using TIM2), and found that the basic RTX freaks out... and dies. The methods used worked just fine using my 'old way' (precisely, the old 2.0x version of STM32F103 libraries--after I vetted them). I now 'get' to either vet the new libraries, and wonder if they are going to be fruitful or not. That is, IF I used the CMSIS methodology/topology. (e.g. use v2.0x or v3.0 of ST's library). So, I was kind of curious if anybody used it... and if they did, if they used it using the Keil RTX OS. I consider this issue forum rather closed at this point. I didn't see a bunch of people saying "YES!" etc. Thus, I'll determine if I use CMSIS on my own, and for my own rational... or not. (Yes, I was hoping for an easy answer... if I got a lot of "NO!" or "YES!" then I could make my decision a bit easier). Andy, thanks once again. --Cpt. Vince Foster 2nd Cannon Place Fort Marcy Park, VA |