Boot hang with JMicron Generic (197b:2363) device

[Michaelhz]

After first boot sequence eCS hangs during checkdisk session (see attached screenshot). System boots without problem when using DANIS506 (JMicron Adapter supports also traditional IDE). One HD is running on this controller (WD10EADS), this drive is AHCi-capable. Three other HD's are connected to the Mainboard's SB 700 controller, which is running in IDE mode.
A serial debug log is also attached to this ticket. It ends exactly after the third checkdisk should occur.
After pressing the reset-button on the machine, the second attached trap000d in OS2BOOT occurs. It does not occure during boot w/o serial debug log.

[David Azarewicz]

Please attach the output of the PCI.EXE command to this ticket.

Please do not change the system configuration, or report any different system configurations in this ticket.

Please do not attach any screen shots other than trap screens unless specifically requested to do so.

[Michaelhz]

Here's the PCI.EXE output.

[Michaelhz]

As requested, please find attached the pci.exe-output.

[David Azarewicz]

Please add "/A:2 /I" to the DANIS506 driver command line in your config.sys. So you will have something like:

BASEDRV=DANIS506.ADD /!BIOS /A:2 /I

Also, you can remove the "/c:1 /b:115200" from the AHCI driver command line. This is so that you can capture another debug log when the system is booted. The driver will output data to either the serial port or to the internal buffer, but not both at the same time.

Reboot the same disk as before. I expect that the system will boot OK. If it does not, please let me know. If the system does boot, please capture a Startup Debug Log (copy os2ahci$ ahci.log) and attach it to this ticket.

[Michaelhz]

This is the ahci.log-file. System boots only after DANIS506 is preceding.

[Michaelhz]

The system still does not boot, if OS2AHCI is in advance.

This is the boot order in config.sys. Of course OS2AHCI is not loaded, but the system works (at least in IDE-mode):

BASEDEV=DANIS506.ADD /!BIOS /A:2 /I
BASEDEV=OS2AHCI.ADD /d:2 /v:2

[David Azarewicz]

Ok, while booted this way, open a command window and type:

copy ibms506$ dani.txt

and attach the dani.txt file to this ticket.

also CD to \os2\boot on your boot disk and type

bldlevel danis506.add >daniver.txt

and attach the daniver.txt file to this ticket.

[Michaelhz]

OK, this is probably not what you want.

[Michaelhz]

This is the version of DANIS506.

[Michaelhz]

Now I'll reboot with

BASEDEV=DANIS506.ADD /!BIOS /A:2 /I /DEBUG:2
BASEDEV=OS2AHCI.ADD /v:2

and send a new dani.txt

[Michaelhz]

Ups, OS2AHCI is still putting information to ibms506$.

[Michaelhz]

After REMing out OS2AHCI I get the following dani3.txt

[David Azarewicz]

The problem is that the Dani driver and the Ahci driver are both claiming the same adapter. In this case it is the Dani driver causing the problem. You are using an outdated version of the Dani driver (1.8.7). The current version is 1.8.15. I suggest you update your Dani driver. Please read the "Interoperability With IDE Drivers" section in the AHCI readme file.

I attempted to resolve the conflict with the "/A:2 /I" switches for the Dani driver but it looks like I guessed incorrectly the adapter number the Dani driver was using.

[Michaelhz]

So please provide a current version of DANIS506 via a common channel (ecs software subscription or eC BetaZone).

[Michaelhz]

Booting with DANIS506 1.8.14 (from eCS 2.2 beta) works.

[Michaelhz]

Nevertheless please provide the most actual drive via eCS-software subscription.

BTW: New problem occurs, because disk content is not readable. Checkdisk is not "walking" on the drive, connected with JMicron-Chip. LVM is claiming wrong partition table. But this is a different story. I will open a new ticket and close this one. Meanwhile thanks for your support (even i can still not use OS2AHCI).