Re: Gayle support reworked

[Frank Wille <frank%phoenix.owl.de@localhost>]

On Sun, 4 Jan 2015 22:47:37 +0100
Radoslaw Kujawa <radoslaw.kujawa%c0ff33.net@localhost> wrote:

> > On 22 Dec 2014, at 19:14, Michael L. Hitch
> > <mhitch%lightning.msu.montana.edu@localhost> wrote:
> > [...]
> >  My netbsd-7 and netbsd-current kernels both crash in gayle_intr_status
> > ().  I've just started looking at these changes.
> 
> Sorry for the late reply, I way away for christmas / new year?s. 
> 
> I think I didn?t test on an A4000. I probably only tested on A600, A1200
> with FastATA and A500 with ACA500. It is quite possible that there is
> some A4000-related bug.

I was testing some recent NetBSD kernels on my A4000 these days and can
confirm Michael's observation. It turned out to be an uninitialised bus space
tag and handle for Gayle's interrupt status register. I committed gayle.c 1.8, which fixes that.

My A4000 works now with netbsd7 and current. Will request pullup-7.

-- 
Frank Wille