Subject: Re: more magnum tests...
To: Warner Losh <firstname.lastname@example.org>
From: Mark Abene <email@example.com>
Date: 02/10/2001 23:46:22
On Sat, Feb 10, 2001 at 09:17:09AM -0700, Warner Losh wrote:
> In message <20010210073443.A20612@mail.radicalmedia.com> Mark Abene writes:
> : Processes just don't seem to be able to fork from init.
> One other thing to consider. Way back in OpenBSD/arc, there used to
> be a bug in the dma stuff that would produce similar symptoms on my
> rPC44. I'd read in the same blocks over and over from the floppy, or
> get strange errors from the disk. Turned out to be a cache coherency
> problem that was easy to fix (well, relatively easy, since I only
> fixed bt driver with an ugly kludge since I didn't know any better at
> the time).
> I know it is a longshot, but maybe it is back?
Could you elaborate on your kludge? What needed to be fixed exactly?
Was the cache coherency problem with L2 cache?