Simon Kallweit <[EMAIL PROTECTED]> writes: > Hi > > I was looking for a macro to convert fix millisecond intervals into > tick counts to be used for cyg_thread_delay(). I didn't found it, and > was wondering if there is anything against introducing such a macro?
The problem with this is that while you want to convert to/from millisecond, others might want microseconds, or nanoseconds, or seconds+microseconds, or seconds+nanoseconds, or fixed or floating point fractions. Which of these do we supply? If we only supply a subset, then users still have to convert. Instead we supply the basic tools and let users do their own conversions. There are several examples of how to do this in the sources (in tests like tm_basic, and in the POSIX library) and there is support for creating time convertors to encapsulate the calculations. On a more general note, we do have to be careful in the eCos sources with what we include. If we just added every function and macro that anybody might find useful we would end up with an enormous API, like Windows. Nobody wants that. Instead we need to keep the API as small as possible. The result of that is that occasionally users have to write code to adapt the API to what they need. -- Nick Garnett eCos Kernel Architect eCosCentric Limited http://www.eCosCentric.com The eCos experts Barnwell House, Barnwell Drive, Cambridge, UK. Tel: +44 1223 245571 Registered in England and Wales: Reg No: 4422071 -- Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
