cvs commit: src/sys/alpha/alpha interrupt.c src/sys/alpha/isa isa.c src/sys/amd64/amd64 intr_machdep.c src/sys/amd64/include intr_machdep.h src/sys/amd64/isa atpic.c src/sys/arm/arm intr.c src/sys/dev/sio sio.c src/sys/dev/uart uart_kbd_sun.c uart_tty.c ...

[John Baldwin]

On Wednesday 26 October 2005 07:48 pm, Peter Wemm wrote:
> On Wednesday 26 October 2005 10:20 am, John Baldwin wrote:
> > On Tuesday 25 October 2005 06:45 pm, Mark Linimon wrote:
> > > On Tue, Oct 25, 2005 at 07:48:48PM +0000, John Baldwin wrote:
> > > >   Reorganize the interrupt handling code a bit to make a few
> > > > things cleaner and increase flexibility to allow various
> > > > different approaches to be tried in the future.
> > >
> > > Wow.  Having (in a previous existance) worked on real-time OS
> > > interrupt handlers I can likely guess how much work went into this.
> > >  Nice.
> >
> > Actually, this diff wasn't a huge deal, it was mostly just a
> > refactoring of what was already there.
>
> Of course the real challenge is to make things like the puc device do
> the right thing automatically instead of needing 'options
> PUC_FASTINTR'.

You mean like sio(4) tried to?  The problem is that with the previosu code if 
sio(4) went first, it would register INTR_FAST and some later PCI device 
wouldn't be able to register its interrupt.  There's not an easy solution to 
that problem if you want to keep the semantics that INTR_FAST implies 
INTR_EXCL.

-- 
John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org