On Fri, Apr 22, 2011 at 9:35 AM, Oliver Seitz <[email protected]> wrote:
> Sorry if I'm warming up old arguments, but I have not been here during that > discussion. > I wasn't either. Usually I'm writing one email with my ideas, then I discard the message...In this way I'm not dependent by other's answers. This time I've pushed on a wrong button, sorry. For me it's the same if the compiler or the user is handling the problem. thx, Vasile > > 1. the user must know there is an OSCCAL value stored somewhere in the > memory and the compiler has nothing to do with, else everything is written > in the micro definition file may be useless > > The datasheet states that on startup, the default value of the w register > is the calibration value. So you could argue that, by modifying the w > register, the definition file violates the rule not to change defaults. > > > 2. first time the user is programming the micro, he must save the OSCCAL > value by his own wish and store somewhere for further use, none of the > programmers are doing this automatically (as far I know) except maybe > generating a warning message about the existence of that value > > 3. within the compiler or without it, the stored value can be put again > in it's place and this is so simple that no matters how is done > > So the user should replace the MOVLW instruction by a RETLW instruction? Or > move the MOVLW to some other place, where it can be easily executed? This, > IMO, is urging the user to change default settings. > > The compiler supports bootloaders. That is no job for a compiler, that is > user program. But, for technical reasons, it is easy for the compiler to > handle the bootloader code, while it is hard or impossible to do that in a > user program. The same, IMO, applies to setting the OSCCAL value on PIC10xx > devices. I wouldn't bet on the compiler never to produce code before the > first user program statement. The compiler decides in which order machine > statements are executed, and setting OSCCAL must be the first thing done. > This is why I favour that the compiler should handle this. > > Greets, > Kiste > > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/jallib?hl=en. > > -- You received this message because you are subscribed to the Google Groups "jallib" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/jallib?hl=en.
