Re: shared interrupts

On Wed, 25 Jan 2006, Frank Mori Hess wrote:

> On Tuesday 24 January 2006 09:08 pm, Calin A. Culianu wrote:
>
>> So  you are basically describing it as: A pended (to Linux) interrupt
>> will be masked out until it is serviced by linux (otherwise you get
>> infinite calls to the rt handler)... thus priority inverting the RT ISR
>> you actually want to run...?
>
> Yes.

Ouch.  That really is bad if you want to keep latency low.  I am glad I am 
sticking to dedicated interrupts in my applications here.  Priority 
inversion for ISRs would really put a wrench in the works for us..

Hey thanks for clearing this up for me!

-Calin

Received on 2006-01-26Z16:31:42