Re: CLASSPNP.SYS load hang in kernel debugger

From: Slobodan Brcin \(eMVP\) (sbrcin_at_ptt.yu)
Date: 05/14/04 

Date: Fri, 14 May 2004 08:39:35 +0200

Konstantin,

You problem can be related to fact that some BIOS-es in rare occasions do
not get along with ntdetect.
So disk parameters passed from ntldr to kernel during transition are
missing. (whole disks are missing).
So ARC paths in kernel mode never get assigned since there is no match
between disks seen by BIOS and disks seen by drivers.

Regards,
Slobodan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Have an opinion on the effectiveness of Microsoft Embedded newsgroups? Tell
Microsoft!
https://www.windowsembeddedeval.com/community/newsgroups
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

"KM" <konstmor@nospam_yahoo.com> wrote in message
news:#wGOHtXOEHA.3300@TK2MSFTNGP09.phx.gbl...
> SAC,
>
> I think I misled you. Sorry about that.
>
> In the accident with the different platform macro I mentioned ealier I had
> the proper disk.sys but I also got a wrong (unnecessary) "FastTrak 376
> Controller" which falls into a "SCSI Miniport" driver group category if I
> understand correctly.
>
> Although I don't know how that RAID Array controller relates to
ClassPnP.sys
> and Disk.sys, I know that it has to be supported by BIOS. Since it was not
> the case on my test machine, the image hung. The interesting part was that
> it was hanging on ClassPnP.sys. I checked that with KD. Since I figured
out
> the problem pretty quickly I did not spend enough time in KD to understand
> why it hung on the ClassPnP.sys.
>
> Now about custom flash driver... I am not a hardware guys so please bear
> with my story.
> I played with an embedded device that had only CF adapter on the board
> (similar setup to yours). Instead of disk.sys a custom Flash driver
> (StartType=0, of course) was used there. With the disk.sys the image was
> giving me 7B right away. So the point was that depending on your CF
adapter
> and board (BIOS support) it may be needed for you to use (write) a custom
> flash disk driver.
>
> KM
>
> > KM,
> >
> > Your description of my problem does sound consistent with what I am
> > seeing.
> >
> > The only disk driver I have loaded is the "Disk Drive" which has the
> > GenDisk. I see a total of 8 Disk Drives components available, you are
> > suggesting I add them all?
> >
> > I have no way to run TAP, unfortunately.
> >
> > The CF adapter on my CPU board is somehow mapped in BIOS to look like
> > IDE Drive 0. I also have a floppy, and can FDISK and FORMAT the CF
> > as the C: drive from DOS.
> >
> > I do not have a custom flash driver with start at boot (StartType=0).
> > Do you think I might need something like this?
> >
> > Thanks in advance,
> > SAC
> >
> >
> >
> > "KM" <konstmor@nospam_yahoo.com> wrote in message
> news:<esv3vz9NEHA.1340@TK2MSFTNGP12.phx.gbl>...
> > > SAC,
> > >
> > > As you may already know, the classpnp.sys is the PnP (Kernel-mode).
> > >
> > > I have run into similar issues with the hang on ClassPnP.sys.
> > > I cannot tell for all the possibilities but in my case it was missing
a
> > > proper disk.sys driver. Basically I accidentially used a platform
macro
> (and
> > > XPe database) from another dev.machine where I had a different disk
> support
> > > (a diffirent scsi driver). As far as I could understand the issue was
> that
> > > ClassPnP.sys (PnP) was trying to pick up the wrong driver and, of
> course,
> > > that driver hung. This is close to 7B, I guess.
> > >
> > > Anyway... TAP input would be a cure for you as it supposes you run XP
on
> the
> > > target. If you can't get it, you will need to have to include all the
> > > drivers manually.
> > > Start with include all the disk driver and see if it helps.
> > >
> > > Also, you mentined you only got the CF adapter. Do you then have a
> (custom)
> > > flash driver in your image with the start at boot (StartType=0)?
> > >
> > > KM
> > >
> > > > Hello,
> > > >
> > > > I have a headless system attempting to boot from an on-board IDE
> > > > CompactFlash adapter. Using an RS-232 kernel debugger, I see it
hangs
> > > > during loading of CLASSPNP.SYS. I think this is pre-FBA.
> > > >
> > > > I assume this is probably something like the 7B error, but I've
> > > > definately got the IDE controller, Primary IDE, Secondary IDE, etc.
> > > > This same CF image will boot fine on a headed test PC.
> > > >
> > > > Ran TA.EXE on target device, seems to find relevant stuff. Can't
run
> > > > TAP.EXE b/c there is no real IDE connector on board, only the
> > > > CompactFlash adapter.
> > > >
> > > > Here is the output from the debugger:
> > > >
> > > >
> > > >
> > > > ModLoad: 804d4000 806c6980 ntoskrnl.exe
> > > > Windows XP Kernel Version 2600 UP Free x86 compatible
> > > > Built by: 2600.xpsp1.020828-1920
> > > > Kernel base = 0x804d4000 PsLoadedModuleList = 0x8054be30
> > > > System Uptime: not available
> > > > Loaded dbghelp extension DLL
> > > > Loaded ext extension DLL
> > > > Loaded exts extension DLL
> > > > Loaded kext extension DLL
> > > > Loaded kdexts extension DLL
> > > > Force unload of ntoskrnl.exe
> > > > ModLoad: 804d4000 806c6980 ntoskrnl.exe
> > > > ModLoad: 806c7000 806dfc00 hal.dll
> > > > ModLoad: fc9bb000 fc9bcb80 kdcom.dll
> > > > ModLoad: fc8cb000 fc8ce000 BOOTVID.dll
> > > > ModLoad: fc4bb000 fc4ca600 pci.sys
> > > > ModLoad: fc4cb000 fc4d3c00 isapnp.sys
> > > > ModLoad: fc4db000 fc4e4280 MountMgr.sys
> > > > ModLoad: fc47b000 fc499880 ftdisk.sys
> > > > ModLoad: fc9bd000 fc9be100 WMILIB.SYS
> > > > ModLoad: fc73b000 fc73f900 PartMgr.sys
> > > > ModLoad: fc9bf000 fc9c0280 intelide.sys
> > > > ModLoad: fc743000 fc748c80 PCIIDEX.SYS
> > > > ModLoad: fca83000 fca83d00 pciide.sys
> > > > ModLoad: fc4eb000 fc4f7000 volsnap.sys
> > > > ModLoad: fc465000 fc47a380 atapi.sys
> > > > ModLoad: fc441000 fc464700 Fastfat.sys
> > > > ModLoad: fc42d000 fc440780 KSecDD.sys
> > > > ModLoad: fc404000 fc42ce80 NDIS.sys
> > > > ModLoad: fc3ea000 fc403680 Mup.sys
> > > > ModLoad: fc4fb000 fc503400 disk.sys
> > > > ModLoad: fc50b000 fc516500 CLASSPNP.SYS
> > > >
> > > >
> > > > Nothing happens after this last line, it just hangs.
> > > >
> > > > Let me know if anyone has any ideas or questions.
> > > >
> > > > Thanks!
>
>

Relevant Pages

  • Re: [RFC] zero downtime upgrades to the kernel.
    ... > We could maybe start with just being able to upgrade kernel modules ... > Could we replace the hard disc driver while it was still being used, ... You can only upgrade a module that isn't in use. ... There are some trick you can use with disks: ...
    (Linux-Kernel)
  • Re: kernel panic, Areca, 6.2-REL with GENERIC [Was: Re: 6.2-st panic w/ ggate [Was: Re: kern
    ... On Feb 15, 2007, at 11:06 AM, Scott Long wrote: ... Aug 29 02:11:40 greengoose syslogd: kernel boot file is /boot/ kernel/kernel ... But if not, it seems whatever bug in this driver existed, did not go away... ... Gotta give a big thanks to the whoever is working on these drivers, once you get working with big disks, it takes A LONG TIME to figure out what's wrong... ...
    (freebsd-stable)
  • Re: Help with hardware please
    ... Anyhow, put the new board in, booted and got a kernel panic/not ... Are your disks plain IDE or are they SATA? ... IDE driver on the old motherboard? ...
    (comp.os.linux.misc)
  • Re: How to get a patched kernel into anaconda?
    ... The next thing I want to do is to boot from the disks and create my system on ... That means as far as I can see that anaconda must use the patched kernel. ... http://driverupdateprogram.com and using driver update disks. ...
    (Fedora)
  • hard disk not accessable via second ide controller
    ... I have a recently installed Debian with Linux 2.4.18-bf2.4. ... I have not yet moved any data to the 40 Gb disks. ... 00:01.0 PCI bridge: VIA Technologies, ... Serial driver version 5.05c with MANY_PORTS SHARE_IRQ ...
    (Debian-User)