On 2007-02-12, Bart Veer <[EMAIL PROTECTED]> wrote:
> Grant> Thanks. I've heard from a couple others who also use the
> Grant> i2c module in RedBoot.
>
> Grant> I glanced at the souce code and noticed calls to mutex
> Grant> routines and wasn't sure if that would cause problems or
> Grant> not.
>
> Those calls involve the driver API rather than the kernel API,
> allowing the generic I2C code to work correctly in both a
> multi-threaded kernel configuration and in a polled-only
> configuration like RedBoot. However note that it is not just
> the generic I2C support CYGPKG_IO_I2C you need to worry about.
> Typically there will also be a hardware-specific device driver
> and that needs to do the right thing as well. Usually that is
> not hard - in the tx and rx routines check whether or not
> interrupts are enabled and then either poll or wait for the
> interrupt - but some device drivers may not bother.
>
> If instead of a driver for some I2C bus master hardware you
> are using the bitbanging support in the generic package, that
> is usable in both interrupt-driven and polled environments.
Yes, I'm planning on using the bit-banging driver (I should
have mentioned that), so it sounds like there shouldn't be any
problems.
--
Grant Edwards grante Yow! Now my EMOTIONAL
at RESOURCES are heavily
visi.com committed to 23
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
