[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference hydra::axp-developer

Title:Alpha Developer Support
Notice:[email protected], 800-332-4786
Moderator:HYDRA::SYSTEM
Created:Mon Jun 06 1994
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:3722
Total number of notes:11359

3113.0. "femsys" by RDGENG::HUDSON (That's what I think) Thu Jan 30 1997 11:09

    Company Name :  femsys
    Contact Name :  geoff tolton
    Phone        :  +44 116 254 1475
    Fax          :  +44 116 255 8982
    Email        :  [email protected]
    Date/Time in :  30-JAN-1997 16:08:58
    Entered by   :  Nick Hudson
    SPE center   :  REO

    Category     :  UNIX
    OS Version   :  3.2
    System H/W   :  alpha


    Brief Description of Problem:
    -----------------------------

rom:	WATNOW::"[email protected]" 30-JAN-1997 16:07:06.09
To:	Nick Hudson <[email protected]>
CC:	
Subj:	Help!!

Dear Nick

Sorry to bother you but I am at a bit of a loss as to how to solve this
problem. 
We have a customer with a DEC Alpha running OSF 3.2. We sent him a standard
distribution tape (the one I made during my last visit) but when he tries
to execute
the program he gets the following message

cygni> femgv
inst fault=4, status word=               8, pc=       12008f9d4
Illegal instruction (core dumped)

Do you have any idea as to what inst fault 4 is, or can you suggest any
possible causes for
this dismal crash?

I would be very grateful for any help or suggestions

Thank in advance

Geoff Tolton

Femsys Support
[email protected]

Femsys Sales
[email protected]

general
[email protected]

Phone +44 116 254 1475
fax   +44 116 255 8982
***********************

T.RTitleUserPersonal
Name
DateLines
3113.1KZIN::HUDSONThat&#039;s what I thinkThu Jan 30 1997 11:50125
From:	DEC:.REO.REOVTX::HUDSON       "[email protected] - UK Software
Partner Engineering 830-4121" 30-JAN-1997 16:49:43.37
To:	nm%vbormc::"[email protected]"
CC:	HUDSON
Subj:	RE: Help!!

Hi Geoff,

The error message could mean that the executable has been corrupted, so it may
be worth checking the same executable on another machine, but I think the patch
below looks quite promising, especially since I guess your app will be using
floating point.  Maybe you could get the customer to check the little example
program quoted in the following note.  If that shows the problem, then most
likely they need the patch.

The customer should be able to obtain the patch through the Customer Support
Centre.  I looked on the Digital public WWW patch page and can't find the patch
on there, and I believe the patch number below may differ depending on which
version of OSF the customer has.  But given the info below the CSC should be
able to get him the right one.

Regards

nick



PATCH ID: OSF365-019       SUBSET(s): OSFHWBIN365 OSFHWBIN350

*******************************************************************************

PROBLEM:	(QAR 40124)		(Patch ID: OSF365-019)
********
When executing with OSF PALcode revision 1.45, or greater, with Digital UNIX
V3.2, V3.2B, V3.2C, V3.2D-1, V3.2E-1, V3.2D-2, or V3.2E-2 some floating
point instructions fail.

Some floating point instructions fail and cause the message on the console:

	"unknown OPDEC from ieee emulation"

There may also be user output instruction fault messages such as:

       "inst fault=4, status word=               8, pc=     3ff80128e50"
       "Illegal instruction (core dumped)"

The following table identifies the machines that may have this problem.

      Platform                Firmware Rev
      --------                ------------
      DECpc AXP 150           SRM Console V2.2
      DEC 2000 AXP            SRM Console V2.2
      DEC 3000 AXP            Console V6.5
      DEC 4000 AXP            Console V3.7
      AlphaStation 2x0/400    SRM Console V6.0
      AlphaServer 400         SRM Console X4.4
      AlphaStation 255        SRM Console V6.0
      AlphaServer 1000        SRM Console V3.1-21
      AlphaServer 2000/2100   SRM Console V4.4
      DEC 7000/10000 AXP      LFU V11
      Digital AXPvme          Console V15.0
      DECAXPPC133             SRM Console X4.3-3114


To determine if a machine has OSF PALcode version 1.45, do one of these:

  1)  Look for the following line in the output from Digital UNIX as it boots.

             PALcode: OSF version 1.45

  2)  At the console prompt prior to booting Digital UNIX, type "show config".

>>>show config

   Console V3.7-3                    VMS PALcode V5.53A, OSF PALcode V1.45A

      Note: V1.45A is the same as V1.45.

Common floating point operations will reproduce the problem.  For example:

floor.c
-------

#include <math.h>

main()
{
   double dd = (double)0.019999999552965164;
   double int_dd;

   int_dd = floor(dd);

   printf("dd = %f    int_dd = %f\n", dd, int_dd);
}

Compiling with:
---------------

  cc   floor.c  -o floor

Executing:
---------

  floor

inst fault=4, status word=               8, pc=     3ff80128e50
Illegal instruction (core dumped)


FILE(s):

/usr/sys/BINARY/alpha_init.o 		subset OSFHWBIN365
CHECKSUM: 48845     65	  RCSfile: alpha_init.c  RCS: 1.2.94.2
/usr/sys/BINARY/softfp.o 		subset OSFHWBIN350
CHECKSUM: 58178      8	  RCSfile: softfp.c      RCS: 1.1.16.2
/usr/sys/BINARY/trap.o 			subset OSFHWBIN365
CHECKSUM: 53197     68	  RCSfile: trap.c	RCS: 1.2.70.2
/usr/sys/BINARY/fp_ieee_handler.o 	subset OSFHWBIN350
CHECKSUM: 56251      7	  RCSfile: fp_ieee_handler.c  RCS: 1.1.22.2
-----------------------------------------------------------------

INSTALLATION INSTRUCTIONS:

A kernel rebuild is required.

3113.2KZIN::HUDSONThat&#039;s what I thinkMon Feb 03 1997 11:0429
From:	WATNOW::"[email protected]"  3-FEB-1997 15:51:38.85
To:	Nick Hudson <[email protected]>
CC:	
Subj:	RE: Help!!

Nick,

Thanks for your speedy response to my plea. I have passed the information
on to our
customer and I am awaiting his reply.

Many thanks


Geoff Tolton

Femsys Support
[email protected]

Femsys Sales
[email protected]

general
[email protected]

Phone +44 116 254 1475
fax   +44 116 255 8982
***********************

3113.3KZIN::HUDSONThat&#039;s what I thinkMon Feb 03 1997 11:0511
From:	DEC:.REO.REOVTX::HUDSON       "[email protected] - UK Software
Partner Engineering 830-4121"  3-FEB-1997 16:05:17.81
To:	nm%vbormc::"[email protected]"
CC:	HUDSON
Subj:	RE: Help!!

Hi Geoff,

OK let me know what happens

nick
3113.4KZIN::HUDSONThat&#039;s what I thinkFri Feb 07 1997 09:3150
From:	WATNOW::"[email protected]"  7-FEB-1997 14:29:56.73
To:	Nick Hudson <[email protected]>
CC:	
Subj:	More Help!!

Dear Nick

Just to let you know how things are going with OSF 3.2 problem customer. He
ran the small test program
suggested in your e-mail but it worked correctly. Around the same time we
had reports from 3 other customers
saying that they could not run our software! - so it looks like we have got
a corrupted executable.

I did ask the customer to try and re-link the program but he said that it
failed with unsatisfied references. Unfortunately
he was not very forth coming as to what they were, other than they "looked
like Fortran library routines"

I managed to create a new executable at a friendly customers site and we
are now distributing this...

On a slightly different point, whilst typing this I have just had a call
from a DEC Alpha OpenVMS user who can't run
the executable I made last time at the porting centre!! It is unable to
find the X Toolkit library, DECW$XTLIBSHRR5.EXE

He is running OpenVMS 6.1, the executable was made under 6.2  Should he
have DECW$XTLIBSHRR5.EXE somewhere?
Will it be called something else under 6.1? Any ideas?

Thanks for your help

Geoff

Geoff Tolton

Femsys Support
[email protected]

Femsys Sales
[email protected]

general
[email protected]

Phone +44 116 254 1475
fax   +44 116 255 8982
***********************

3113.5KZIN::HUDSONThat&#039;s what I thinkFri Feb 07 1997 09:5253
From:	DEC:.REO.REOVTX::HUDSON       "[email protected] - UK Software
Partner Engineering 830-4121"  7-FEB-1997 14:52:23.48
To:	nm%vbormc::"[email protected]"
CC:	HUDSON
Subj:	RE: More Help!!

Hi Geoff,

>I managed to create a new executable at a friendly customers site and we
>are now distributing this...

It does sound like the problem is not the one referred to in the patch notes I
sent.  If the new executable works on the original customer's machine, then I'd
guess that most likely you somehow had a corrupted image.


>On a slightly different point, whilst typing this I have just had a call
>from a DEC Alpha OpenVMS user who can't run
>the executable I made last time at the porting centre!! It is unable to
>find the X Toolkit library, DECW$XTLIBSHRR5.EXE
>
>He is running OpenVMS 6.1, the executable was made under 6.2  Should he
>have DECW$XTLIBSHRR5.EXE somewhere?

While you should always be able to run an image linked on version "n" of VMS on
any version of VMS >="n", there is no guarantee that you'll be able to run it
on any version less than "n".  Sometimes it might work but usually it won't.

So if your image was built on 6.2, the chances are that you won't be able to
run it on 6.1.

The VMS image activator won't let you run a program if it thinks you're on a
lower version of VMS than the one where the LINK took place.  It will give you
a "SHRIDMISMATCH" errors when you try and RUN a program in this case.

I guess that the only reason you're not seeing SHRIDMISMATCH errors is that it
can't find DECW$XTLIBSHRR5.EXE on the customer system.  In other words, if you
did install that file, you'd only see SHRIDMISMATCH errors anyway.

I'm not sure where DECW$XTLIBSHRR5 comes from (maybe the MOTIF installation, or
maybe it comes with 6.2 of VMS).  You could try getting a copy of the image
from a 6.2 system and giving that to the customer, but I'd be surprised if you
get away with that.

What you really need to do is to re-link the application on a 6.1 machine
(maybe the customer's machine).  That way the resulting image would only be
relying on RTLs that the customer has.  Or get him to upgrade to 6.2.

Sorry if that's not very good news...

Regards
Nick

3113.6KZIN::HUDSONThat&#039;s what I thinkMon Feb 10 1997 04:2270
From:	WATNOW::"[email protected]"  7-FEB-1997 17:47:23.67
To:	Nick Hudson <[email protected]>
CC:	
Subj:	RE: More Help!!

Dear Nick

> While you should always be able to run an image linked on version "n" of
VMS 
> on
> any version of VMS >="n", there is no guarantee that you'll be able to
run it
> on any version less than "n".  Sometimes it might work but usually it
won't.
>  
> So if your image was built on 6.2, the chances are that you won't be able
to
> run it on 6.1.
>  
> The VMS image activator won't let you run a program if it thinks you're
on a
> lower version of VMS than the one where the LINK took place.  It will
give you
> a "SHRIDMISMATCH" errors when you try and RUN a program in this case.
>  
> I guess that the only reason you're not seeing SHRIDMISMATCH errors is
that it
> can't find DECW$XTLIBSHRR5.EXE on the customer system.  In other words,
if you
> did install that file, you'd only see SHRIDMISMATCH errors anyway.
>  
> I'm not sure where DECW$XTLIBSHRR5 comes from (maybe the MOTIF
installation, 
> or
> maybe it comes with 6.2 of VMS).  You could try getting a copy of the
image
> from a 6.2 system and giving that to the customer, but I'd be surprised
if you
> get away with that.
>  
> What you really need to do is to re-link the application on a 6.1 machine
> (maybe the customer's machine).  That way the resulting image would only
be
> relying on RTLs that the customer has.  Or get him to upgrade to 6.2.
>  
>  

Thanks for the quick reply. You have confirmed my suspicions. I will get
the user to try a relink, although I believe he is willing to upgrade to
OpenVMS 6.2

Many Thanks 

Geoff

Geoff Tolton

Femsys Support
[email protected]

Femsys Sales
[email protected]

general
[email protected]

Phone +44 116 254 1475
fax   +44 116 255 8982
***********************

3113.7KZIN::HUDSONThat&#039;s what I thinkTue Feb 11 1997 08:0351
From:	VBORMC::"[email protected]" "Femsys support" 11-FEB-1997
12:11:58.39
To:	"[email protected] - UK Software Partner Engineering 830-4121 
07-Feb-1997 1452 +0000" <[email protected]>
CC:	
Subj:	RE: More Help!!

Dear Nick

Thanks for the quick reply. You have confirmed my suspicions! I will get
the user
to try a relink, although I believe he is happy to upgrade to OpenVMS 6.2

Many thanks

Geoff

Geoff Tolton

Femsys Support
[email protected]

Femsys Sales
[email protected]

general
[email protected]

Phone +44 116 254 1475
fax   +44 116 255 8982
***********************


% ====== Internet headers and postmarks (see DECWRL::GATEWAY.DOC) ======
% Received: from mail.vbo.dec.com (mail.vbo.dec.com [16.36.208.34]) by
vbormc.vbo.dec.com (8.7.3/8.7) with ESMTP id NAA17368 for
<[email protected]>; Tue, 11 Feb 1997 13:08:56 +0100
% Received: from server21.digital.fr (server21.digital.fr [193.56.15.21]) by
mail.vbo.dec.com (8.7.3/8.7) with ESMTP id NAA18657 for
<[email protected]>; Tue, 11 Feb 1997 13:13:42 +0100 (MET)
% Received: from hil-img-1.compuserve.com (hil-img-1.compuserve.com
[149.174.177.131]) by server21.digital.fr (8.7.5/8.7) with SMTP id NAA30544 for
<[email protected]>; Tue, 11 Feb 1997 13:15:50 +0100 (MET)
% Received: by hil-img-1.compuserve.com (8.6.10/5.950515) id HAA18460; Tue, 11
Feb 1997 07:08:52 -050
% Date: Tue, 11 Feb 1997 07:08:37 -0500
% From: Femsys support <[email protected]>
% Subject: RE: More Help!!
% To: "[email protected] - UK Software Partner Engineering 830-4121 
07-Feb-1997 1452 +0000" <[email protected]>
% Message-ID: <[email protected]>