Trap D on fresh MCP2 R51 install

Fresh MCP2 install on a new R51 fails at the initial boot to OS/2 with a
Trap D as follows:

Exception in module:
TRAP 000d ERRCD=0000 ERACC=**** ERLIM=********
EAX=0000530e EBX=f9800000 ECX=00000102 EDX=00000741
ESI=fa1f04c0 EDI=00000001 EBP=00004e52 FLG=00013202
CS:EIP=0c08:00000001 CSACC=109b CSLIM=0000fffe
SS:ESP=0030:00004e40 SSACC=1097 SSLIM=00003fff
DS=0bf8 DSACC=0093 DSLIM=0000043f CR0=8001001b
ES=0030 ESACC=1097 ESLIM=00003fff CR2=fabd4027
FS=0ca8 FSACC=0093 FSLIM=000019ee
GS=0d00 GSACC=0093 GSLIM=0000018a

The system detected an internal processing error at
location ##0168:fff1eb1c - 000e:cb1c.
60000, 9084

078606d6
Internal revision 14.089_W4

I can boot to the command line on this R51 at that point with <ALT F1>.
From there, I can acutally get briefly to the Desktop and see the
initial install Desktop by simply entering the command PMSHELL at C:\.
Which then is quickly followed by the 'standard' OS/2 install screen and
we're right back at the CD-ROM install at the Desktop screen.


I chose not to install any Network Adapter, nor Audio, nor Printer for
the initial install. I can use TEDIT to edit CONFIG.SYS, and note that
the IBMLANLK and CLEANUP are still it it at this point. The default
SVGA video is still there, as would be expected. And since it actually
can sbow, for a moment, the correct desktop and revert to the PM install
screen, one would think this isn't a video problem.


Any suggestion on how to debug what driver is doing this via hand
manipulation of the CONFIG.SYS file, and/or import into the system from
a valid CD-ROM drive which it can see at command line, of a utility to
debug and fix this?


Thanks!



--


--> Sleep well; OS2's still awake! ;)

Mike Luther

Apparently 'solved' ..

Mike Luther wrote:
[color=blue]
> Fresh MCP2 install on a new R51 fails at the initial boot to OS/2 with
> a Trap D as follows:
>
> Exception in module:
> TRAP 000d ERRCD=0000 ERACC=**** ERLIM=********
> EAX=0000530e EBX=f9800000 ECX=00000102 EDX=00000741
> ESI=fa1f04c0 EDI=00000001 EBP=00004e52 FLG=00013202
> CS:EIP=0c08:00000001 CSACC=109b CSLIM=0000fffe
> SS:ESP=0030:00004e40 SSACC=1097 SSLIM=00003fff
> DS=0bf8 DSACC=0093 DSLIM=0000043f CR0=8001001b
> ES=0030 ESACC=1097 ESLIM=00003fff CR2=fabd4027
> FS=0ca8 FSACC=0093 FSLIM=000019ee
> GS=0d00 GSACC=0093 GSLIM=0000018a
>
> The system detected an internal processing error at
> location ##0168:fff1eb1c - 000e:cb1c.
> 60000, 9084
>
> 078606d6
> Internal revision 14.089_W4[/color]


It appears to be related to APM as released in at least the MCP2 install
version latest on the CD-ROM's now distributed with brand new OS/2
membership material, the ones produced in Ireland. I doubt if this is a
regression either. During the install you'll notice that the Advanced
Power Management icon is shown as active in the pane with serial port
and other options. I didn't turn it off.


When the Trap D occurs on that second reboot, jsut before OS/2 finishes
the install from MCP2, on this R51, I can choose the option to use <ALT
F1> at the OS/2 Boot Blob. Choose <F2> to boot to a command line. At
the command line use the IBM provided TEDIT text file editor to edit
CONFIG.SYS. Add a REM in front of these two lines as shown below:


REM DEVICE=C:\OS2\BOOT\APM.SYS
REM DEVICE=C:\OS2\MDOS\VAPM.SYS

The R51 here then immediated rebooted normally. Yes, there is a very
recent update via Passport Advantage which, IIRC, addresses power
management problems on Thinkpad units. I have not installed and tried
it here yet. In any event, even if it works, unless you are using the
modified CD install technique and the fix becomes a part of the original
install, it won't do any good for an original install.


I'll report back on the fix when I get data on it later as seen here
from the distributed fix which I already had.


--


--> Sleep well; OS2's still awake! ;)

Mike Luther

On Thu, 15 Sep 2005 04:50:06 UTC, Mike Luther <mike.luther@ziplog.com>
wrote:
[color=blue]
> Fresh MCP2 install on a new R51 fails at the initial boot to OS/2 with a
> Trap D as follows:[/color]

Did you REM out the APM.SYS lines? That was the only problem I had
installing eCS 1.2 on an R51.
Jayant
--
Jayant Murthy
Indian Institute of Astrophysics
Koramangala, Bangalore 34
India

I finally figured this out ..

Jayant Murthy wrote:
[color=blue]
>On Thu, 15 Sep 2005 04:50:06 UTC, Mike Luther <mike.luther@ziplog.com>
>wrote:
>
>
>[color=green]
>>Fresh MCP2 install on a new R51 fails at the initial boot to OS/2 with a
>>Trap D as follows:
>>
>>[/color]
>
>Did you REM out the APM.SYS lines? That was the only problem I had
>installing eCS 1.2 on an R51.
>Jayant
>
>[/color]

And got it to finish the install run Jayant. After which I've
discovered that a number of others have the same solution.


While you are here and report that you have experience with the R51,
would you share another bit of experience with us? Which of the models
do you have; the Intel dual chip video model or the ATI video model?
Can you open a DOS-VDM window session on any advanced screen resolution
to the complete windowed session on the normal desktop without total
initial screen corruption that requires an <ALT HOME> that then pops a
working full screen DOS session, then a subsequent <ALT HOME> that will
then produced the desired DIS_VDM window session?


The model that produces this flaw here is the Intel 825/855 video
chipset model. I would dearly like to know if the ATI 7500 video
version of the R51 has this same issue or not? At first I thought that
some versions of the "T" series Tpads also used that Intel 825/855 video
chipset, but on further research I don't think that is true. If you (or
anyone else!) have the R51 with the ATI 7500 chipset and it does not
have this bug it would be very helpful to know this.


Then .. armed with this information, and which model you have of the
R51, Veit Kannegieser has posted a suggestion to try his APMBIOS.SYS as
a possible fix for the APM.SYS error on the R51. That in another thread
on this. His suggestion is on my list of things to try. But two other
things which might also be involved in the near-done but still not
working R51-OS/2 project here, as I see it, including the video
corruption issue need to be drilled down into before the APM borehole,
as I see it.


I could be totally wrong there also! Don't claim to be a guru at
anything but breaking things that ought to work... chuckle.


--


--> Sleep well; OS2's still awake! ;)

Mike Luther