5-stable kernel hang during boot

[Paul Keusemann]

On Tue, Nov 08, 2005 at 10:27:33PM -0600, Billy Newsom wrote:
> See my posts and others on these topics:
>  critical BOOT failure updating to latest 5-Stable (5.4)
>  5.3 -> 5.4 breaks ATA (Intel ICH2)
>  [PATCH] option to re-enable aggressive ATA probing

Thanks for the pointers.  I have not been able to figure out how to get
the mail list archive search to give me anything useful yet.

So far, neither the patch to /usr/src/sys/dev/ata/ata-lowlevel.c or 
reverting to the 5.3-Release acpi.ko has solved my problem.  It appears I
have the same problem as Julian Dunn.  The hang is occurring in the same
place anyway.

Has anyone who has had a problem with 5-Stable tried 6.0?

> 
> Those are from September 2005, when I and others started having problems 
> with the latest (poor) edits to the 5-Stable code.
> 
> There was obviously a regression for many of us with certain hard drives 
> or systems.  As an important side note, notice that I am using a similar 
> computer to yours, an IBM PC365 with dual Pentium Pros. Although I have 
> not tried the patch due to lack of time, I hope that will fix your 
> issue. Also, I wonder if we share a similar bad BIOS issue in which our 
> systems seem to choke on this new code for IDE detection during boot 
> time.  See my dmesg output in the archives.
> 
> BTW, this has effectively kept me from upgrading to the latest 5-Stable 
> since September, as it is not a quick and easy task on my old system.
> 
> Also, there might be another mailing list - ACPI or something where 
> there may be more answers.
> 
> I don't know if there is a PR for this, yet. I had thought of filing 
> one, but I sort of forgot. Check the archives to see if someone else did.
> 
> Billy
> 
> Paul Keusemann wrote:
> >Hi,
> >
> >I've got an IBM Intellistation M Pro with dual 600 Mhz PIIIs, an IBM
> >DNES-309170W (da0 on ahc0) as the root drive and HighPoint RocketRaid 1520
> >with two WD1600JD drives mirrored as data drives.  There is nothing on
> >ata0 and the CD-ROM drive is the master on ata1.  I have been running a
> >custom kernel (GENERIC with SMP) since installation.  The verbose output
> >from dmesg from the a 5.4 Release kernel (booted from CD-ROM) is attached.
> >
> >I installed 5.3 Release on this machine a while back and just decided
> >to try to upgrade to 5.4 Stable.  (This is my first attempt at 
> >doing a buildworld upgrade.)  I cvsup'ed /usr/src using the stable-supfile
> >in /usr/share/examples/cvsup.  I then followed the instructions in
> >/usr/src/Makefile to build world, build a new custom kernel and install
> >the new kernel.  Then I attempted to reboot into single user mode.  The
> >boot hangs while processing ata1, right after the following:
> >
> >	ata0: [MPSAFE]
> >	ata1: channel #1 on atapci0
> >	atapci0: Reserved 0x8 bytes for rid 0x18 type 4 at 0x170
> >	atapci0: Reserved 0x1 bytes for rid 0x1c type 4 at 0x376
> >
> >The really interesting part is that after installing the new kernel,
> >I can't boot from kernel.old or another backup of my custom kernel.
> >They both hang in the same way as the new kernel.  If I boot off the 5.4
> >Release CD-ROM, remove the new kernel and rename kernel.old to kernel,
> >the system comes up just fine.
> >
> >At this point I guess I'm looking for any help I can get.
> >
> >Could the fact that there is nothing on ata0 be causing a problem?
> >
> >I know I can boot the 5.4 Release kernel, would I be better off upgrading
> >from the install?
> >
> >I have had no problems with this machine so I can just scratch
> >the upgrade, but it would be nice to be able to get it up to 5-stable
> >and eventually 6-stable.
> >
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at freebsd.org"

-- 
Paul Keusemann			                               pkeusem at visi.com
4266 Joppa Court		                               (952) 894-7805
Savage, MN  55378