On Sun, Jun 28, 2015 at 3:27 AM, Joseph Mingrone <jrm at ftfl.ca> wrote:
> Kevin Oberman <rkoberman at gmail.com> writes:
> > I have narrowed it to between 283990 and 284074. Have to crash, so
maybe
> > someone else will track it down by tomorrow. I only see a handfull
(<10)
> > commits in there that look like possible causes. 280434 is compiling
> right
> > now as the prime candidate.
>
> Cool. I think you mean 284034 and not 280434. It certainly looks
> suspicious. I'm trying 284028, but it'll be awhile. It
wouldn't boot
> with just the kernel, so I'll try building world as well.
>
Yes, I did.
I can now confirm that 284034 is the culprit. Unfortunately, that single
commit MFCed 10 updates of which 6 may be the problem. I tend to suspect
the PCI changes rather than the CardBus patch, but I will spend a little
time looking at the individual commits in HEAD to see which, if any, touch
both PCI and CBB.
Adrian, I don't have the ability to test on head ATM, so, if you have a
chance, go for it. Most if this is way beyond my ken, so I am extremely
unlikely to spot it.
I am adding jhb@ to the CC as 284034 was his commit.
--
Kevin Oberman, Network Engineer, Retired
E-mail: rkoberman at gmail.com
PGP Fingerprint: D03FB98AFA78E3B78C1694B318AB39EF1B055683