Hi, I'm thinking of designing an SA-1110 based HW with alot of peripherials. In order 
to not be restricted by the number of pins on the SA-1110 I'm thinking of putting some 
sort of programmable logic between the CPU and some peripherials.
 
My question is: if i use the PLD to multiplex the interrupts from the peripherials 
into one interrupt connected to the CPU (the CPU has to read a status register in the 
PLD after an interrupt in order to find out which unit caused it), will that work at 
all? My thought was to write some Interrupt wrapper in Linux that caught all these 
interrupts and found out which unit it was, and then would issue an SWI that the 
actual device driver would catch. 
 
Is this conceptually sane or would it introduce a severe performance penalty on for 
instance a 10Mbit ethernet circuit? Is the concept of Software Interrupts enough 
supported in the SA-1110 and armlinux 2.4 to achieve this?
 
Any input is welcome 
 
Cheers
/J�rgen
 

_______________________________________________
http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm
http://www.arm.linux.org.uk/armlinux/mailinglists.php
Please visit the above addresses for information on this list.

Reply via email to