Re: [ECOS] spurious_IRQ on ARM7

[Jonathan Larmour <jlarmour at redhat dot com>]

Ilko Iliev wrote:
> 
> I make a simple test example - the spurious interrupt is generated every
> time when the timer interrupt appear.
> 
> Which options from "ROM monitor support" (in eCos configuration tool) must
> be activated to work the target with JEENI (or BDI2000) without ROM monitor?
> When all option in "ROM monitor support" are deactivated the program
> crashed in hal_if_diag_init() (in hal_if.c) by
> CYGACC_CALL_IF_SET_CONSOLE_COMM macro - the virtual vector table is not
> initialized maybe.
> My test example work only if I activate "Initialize whole of virtual table".

That's the correct thing to do.
 
> Do I need virtual vector table to debug with JEENI without ROM monitor?

The virtual vector table is a separate issue from using JEENI. You do need
it.

> Why is the spurious interrupt generated?

I'm not sure from the limited info here, but if you look at
hal/arm/arch/current/src/vectors.S and especially
hal/arm/at91/current/include/hal_platform_setup.h, there's a whole bunch of
stuff directly or indirectly dependent on the startup type. If your startup
type is RAM it may not work very well since this port was not designed to
support use with something like JEENI (since no-one's paid for it).

You may well need to either change the code, or perhaps you could even
kludge something up by changing to ROM startup type, but defining the
memory layout so that the program is actually in RAM, not ROM!

Jifl
-- 
Red Hat, Rustat House, Clifton Road, Cambridge, UK. Tel: +44 (1223) 271062
Maybe this world is another planet's Hell -Aldous Huxley || Opinions==mine

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss