On 02/02/06, Jan Kiszka <[EMAIL PROTECTED]> wrote:
Ack.
I forgot to apply your earlier fixes so it will be done too.
That's good. Thanks. Keep me informed on progress.
Dmitry Adamushko wrote:
> Hi there,
>
> as I promised, here go the following patches (ordered as they have to be
> applied one by one) :
>
>
> 1) shirq-base.patch
>
> Adds the "name" field to the interrupt object of the nucleus layer.
> Reworks the related bits of the native skin (+ a few minor changes for
> posix and rtdm) to support the named interrupt objects.
>
>
> 2) shirq-v7.patch
>
> Generic support for shared interrupts.
>
>
> 3) shirq-proc.patch
>
> /proc support.
>
> Now /proc/xenomai/irq shows the names of handlers.
>
> The related code have been removed from the hal layer to nucleus.
>
>
> -----
>
> 4) shirq-isa.patch
>
> Trying to handle the edge-triggered stuff.
>
> This is a very preliminary version so the only thing I promise so far
> is that it compiles successfully.
Why not name it XN_ISR_EDGE or so (also the related functions etc.)?
Although it mostly targets ISA hardware, it's about the edge-triggered
characteristic of that devices after all.
Ack.
I forgot to apply your earlier fixes so it will be done too.
>
>
> The functionality added by first 3 patches seem to be working.
>
> --
> Best regards,
> Dmitry Adamushko
>
I'm now trying to organise a full test scenario with shared UART
ISA-IRQs at our institute. May take till beginning of next week, some
regressions in our own software need to be fixed first.
That's good. Thanks. Keep me informed on progress.
Jan
--
Best regards,
Dmitry Adamushko
