On Sat, 15 Nov 2008, T. Makinen wrote:
On Fri, Nov 14, 2008 at 9:29 PM, David Brownlee <abs%netbsd.org@localhost> wrote:On Fri, 14 Nov 2008, T. Makinen wrote: I think the best option would be to change REAL_DMA from a compiletime to a runtime test and default it on for TT and off for Falcon. Do you want to take a pass at that? :)Sure, it seems that adding return value of 0 for Falcon from scsi_dmaok() fixes this as you suggested :)
Looks good, committed :)
Sounds like a plan. Do you want to cleanup the code and params and maybe build a kernel for DavidR to test as well?Here's link to diff that includes pmap related changes and Falcon DMA change: http://koti.welho.com/tmakinen/atari/changes.diff
Committed! thanks.
On Sat, 15 Nov 2008, David Ross wrote:
Nope, still get at least the "lock error" panic with the netbsd-24 kernel: http://i37.tinypic.com/2psn3pl.jpg http://i37.tinypic.com/14xvn4.jpg http://i37.tinypic.com/2d7f2gn.jpg Hope the spam filters don't kick in due to the tinypic.com URLs. I figure it's better than trying to attach 400kb jpegs. :-)
Hmm, was this the same panic that you saw when testing a current
kernel on your 1.6.x install?
Could you try a kernel with DMA disabled for both TT030 and Falcon
http://mono.org/abs/netbsd-nodma.gz
Tuomo - do you have a spare disk onto which you could test an
install on one of your Falcons?
--
David/absolute -- www.NetBSD.org: No hype required --