Re: [avr-libc-dev] eeproms and interrupts

[Joerg Wunsch]

As Branislav Katreniak wrote:

> But what about applications using preemtive kernel? Library
> functions which do not access periferials are all thread safe
> (except that using malloc() and free()). Why should the
> documentation say something different?

See Bernard's reply and my followup.

Seriously, I'd expect any tiny OS to rather supply their own EEPROM
functions based on interrupts.  That's IMHO the way better approach,
as it avoids blocking other threads for too long, and any thread that
waits for EEPROM IO (which is comparably slow) can easily give up its
remainder of the time slice anyway.  The avr-libc way is only there to
have at least a minimal approach of handling the EEPROM, as avr-libc
normally doesn't want to go too much into hardware (like supplying
their own interrupt-driven implementations for anything).

-- 
cheers, J"org               .-.-.   --... ...--   -.. .  DL8DTL

http://www.sax.de/~joerg/                        NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)