rtems-CVS-050303 i.MX/MX1/csb336 BUG and BUGFIX

[Jay Monkman]

On Fri, Mar 04, 2005 at 11:30:01PM +0100, Pavel Pisa wrote:

> I do not see so fundamental flaw in the fixed code. I think, that it would
> be better, if VECTOR_TABLE  macro is eliminated and real pointer variable
>   volatile rtems_irq_hdl** vectorTableAddress;
> to the start of the second level table is used in all ARM variants.

ARM interrupt handling is certainly a mess. Fixing it has been on my
todo list for too long, now.


> By the way, should I put bug and solution into RTEMS GNAT
> or someone other can do that? What is preferred way of login
> into RTEMS GNAT?

I just commited the change to CVS.

For future reference, you can log into GNATS with a username and
password of 'guest'.