Bug 32631

Summary: installing 4.4 "mounting root from ufs:/dev/md0c" and won't continue
Product: Base System Reporter: Chris Sears <cbsears_sf>
Component: miscAssignee: Warner Losh <imp>
Status: Closed FIXED    
Severity: Affects Only Me    
Priority: Normal    
Version: Unspecified   
Hardware: Any   
OS: Any   

Description Chris Sears 2001-12-08 22:00:01 UTC 
Trying to install 4.4, the mount hangs after

  mounting root from ufs:/dev/md0c

The same hardware installs 4.1, NetBSD 1.5
Similar to problem report i386/30902
except that was for an Award BIOS.
The solution for that was "assign IRQ for USB option".
That doesn't seem to be available for this BIOS.
Comment 1 Sheldon Hearn 2001-12-10 12:49:52 UTC 
On Sat, 08 Dec 2001 13:59:10 PST, Chris Sears wrote:

> >Number:         32631
> >Category:       misc
> >Synopsis:       installing 4.4 "mounting root from ufs:/dev/md0c" and won't continue

Hi David,

Any ideas on this one (since you were involved in the related i386/30902)?

Pheonix BIOS... *shudder*

Ciao,
Sheldon.
Comment 2 Craig Boston 2002-03-08 00:40:38 UTC 
I have identical symptoms on a Vaio PCG-FX170K (hangs right after trying to
mount root partition).  Problem is present in everything after
4.4-PRERELEASE, including 4.4-RC, all the way through 4.5-STABLE (still
happening).  This happens both when the kernel is used to boot a live
system, as well as the install CDs.  Affects both GENERIC and custom
kernels.

I did a little playing around with cvsup and was able to track the problem
to this commit on 2001/08/15:

http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/pccard/pcic_pci.c#rev1.54.2.6
(about 12 hours before newvers.sh was changed to 4.4-RC)

With pcic_pci.c and .h files prior to this commit, system boots fine.  With
the new version, it hangs exactly as the original post described.  I was
going to try to figure out what the change was, but this scared me off:

    Changes since 1.54.2.5: +505 -193 lines

Yikes...  This is Warner's MFC of some of the new cardbus code
from -current, it looks like.  Should we get his opinion on it?

I can provide bootverbose or whatever debugging output that's needed, but
it's a hard lock -- no DDB :(

--
Craig
Comment 3 Sheldon Hearn freebsd_committer freebsd_triage 2002-03-08 11:55:06 UTC 
Responsible Changed
From-To: freebsd-bugs->imp

Craig's detective work with rolling back cvsup dates seems to have pointed 
to a cardbus MFC.
Comment 4 Warner Losh freebsd_committer freebsd_triage 2002-09-29 23:16:02 UTC 
State Changed
From-To: open->closed

This was almost certainly the power bug that we had for a while, 
that's sense been corrected.