To SoftICE's developers at Numega

I found (I think programmer’s) error in NMTRANS.DLL.
Version info: 4.05, file size 413766, date time 25.01.2000 4:05
Adress of error: 1000EDC4
there are two jumps to this location after returning zero from
the routine below (also called in the never ending loop)
Description:
mov esi, 1
1000EDC9:
lea eax, [esp+xx]
mov ecx, ebx ;this pointer
push eax
push esi
call 1000EC90 ;return value ignored !!!
inc esi
jmp 1000EDC9 -> Never ending loop

This never ending loop occurs sometimes when translating symbols
from W2K (which have both .dbg and .pdb files). Typical example is
for CRYPT32 symbols from W2K SP1 Free Build.

As a simplest workaround I have used replacing the final jump
by two nop instructions. Then everything seems OK so I think this
never ending loop must be either:

  1. relic from some previous version with additional code (eg. for loop
    exiting)
  2. intentional bug or bug originating in disregard

Please contact me with information how this bug will be resolved.
Paul

There is a such thing known as support. To get immediate attention that’s
where to go.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com http:</http:>

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Tuesday, December 05, 2000 1:26 PM
To: NT Developers Interest List
Subject: [ntdev] To SoftICE’s developers at Numega
Importance: High

I found (I think programmer’s) error in NMTRANS.DLL.
Version info: 4.05, file size 413766, date time 25.01.2000 4:05
Adress of error: 1000EDC4
there are two jumps to this location after returning zero from
the routine below (also called in the never ending loop)
Description:
mov esi, 1
1000EDC9:
lea eax, [esp+xx]
mov ecx, ebx;this pointer
push eax
push esi
call 1000EC90 ;return value ignored !!!
inc esi
jmp 1000EDC9-> Never ending loop

This never ending loop occurs sometimes when translating symbols
from W2K (which have both .dbg and .pdb files). Typical example is
for CRYPT32 symbols from W2K SP1 Free Build.

As a simplest workaround I have used replacing the final jump
by two nop instructions. Then everything seems OK so I think this
never ending loop must be either:
1. relic from some previous version with additional code (eg. for loop
exiting)
2. intentional bug or bug originating in disregard

Please contact me with information how this bug will be resolved.
Paul


You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</mailto:xxxxx>

To SoftICE’s developers at Numega
Oh, Is this happen every time?
I mean for CRYPT32 symbol.

Fei
----- Original Message -----
From: Hrdina Pavel
To: NT Developers Interest List
Sent: Tuesday, December 05, 2000 10:25 AM
Subject: [ntdev] To SoftICE’s developers at Numega

I found (I think programmer’s) error in NMTRANS.DLL.
Version info: 4.05, file size 413766, date time 25.01.2000 4:05
Adress of error: 1000EDC4
there are two jumps to this location after returning zero from
the routine below (also called in the never ending loop)
Description:
mov esi, 1
1000EDC9:
lea eax, [esp+xx]
mov ecx, ebx;this pointer
push eax
push esi
call 1000EC90 ;return value ignored !!!
inc esi
jmp 1000EDC9-> Never ending loop

This never ending loop occurs sometimes when translating symbols
from W2K (which have both .dbg and .pdb files). Typical example is
for CRYPT32 symbols from W2K SP1 Free Build.

As a simplest workaround I have used replacing the final jump
by two nop instructions. Then everything seems OK so I think this
never ending loop must be either:

  1. relic from some previous version with additional code (eg. for loop exiting)
  2. intentional bug or bug originating in disregard

Please contact me with information how this bug will be resolved.
Paul


You are currently subscribed to ntdev as: xxxxx@hotmail.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)

It is always good idea to search for updates on NuMega web. If nothing is
found it is a good thing to report such problems to their tech support. You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody will
ask you for more information and if problem is reproducible enough, they fix
will it and inform you about it. Finally, if you’re really lucky, the update
released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]


From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
Reply To: NT Developers Interest List
Sent: Tuesday, December 05, 2000 7:25 PM
To: NT Developers Interest List
Subject: [ntdev] To SoftICE’s developers at Numega
Importance: High

I found (I think programmer’s) error in NMTRANS.DLL.
Version info: 4.05, file size 413766, date time 25.01.2000 4:05
Adress of error: 1000EDC4
there are two jumps to this location after returning zero from
the routine below (also called in the never ending loop)
Description:
mov esi, 1
1000EDC9:
lea eax, [esp+xx]
mov ecx, ebx;this pointer
push eax
push esi
call 1000EC90 ;return value ignored !!!
inc esi
jmp 1000EDC9-> Never ending loop

This never ending loop occurs sometimes when translating symbols
from W2K (which have both .dbg and .pdb files). Typical example is
for CRYPT32 symbols from W2K SP1 Free Build.

As a simplest workaround I have used replacing the final jump
by two nop instructions. Then everything seems OK so I think this
never ending loop must be either:

  1. relic from some previous version with additional code (eg. for loop
    exiting)
  2. intentional bug or bug originating in disregard

Please contact me with information how this bug will be resolved.
Paul


You are currently subscribed to ntdev as: xxxxx@rkk.cz
To unsubscribe send a blank email to $subst(‘Email.Unsub’)

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f**k**g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing
is
found it is a good thing to report such problems to their tech support.
You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody
will
ask you for more information and if problem is reproducible enough, they
fix
will it and inform you about it. Finally, if you’re really lucky, the
update
released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]


From: Hrdina Pavel[SMTP:xxxxx@compelson.com]

Reply To: NT Developers Interest List
Sent: Tuesday, December 05, 2000 7:25 PM
To: NT Developers Interest List
Subject: [ntdev] To SoftICE’s developers at Numega
Importance: High

I found (I think programmer’s) error in NMTRANS.DLL.
Version info: 4.05, file size 413766, date time 25.01.2000 4:05
Adress of error: 1000EDC4
there are two jumps to this location after returning zero from
the routine below (also called in the never ending loop)
Description:
mov esi, 1
1000EDC9:
lea eax, [esp+xx]
mov ecx, ebx;this pointer
push eax
push esi
call 1000EC90 ;return value ignored !!!
inc esi
jmp 1000EDC9-> Never ending loop

This never ending loop occurs sometimes when translating symbols
from W2K (which have both .dbg and .pdb files). Typical example is
for CRYPT32 symbols from W2K SP1 Free Build.

As a simplest workaround I have used replacing the final jump
by two nop instructions. Then everything seems OK so I think this
never ending loop must be either:

  1. relic from some previous version with additional code (eg. for loop
    exiting)
  2. intentional bug or bug originating in disregard

Please contact me with information how this bug will be resolved.
Paul


You are currently subscribed to ntdev as: xxxxx@rkk.cz
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)

As it goes, the answer to many bugs in DS 1.5 is to install DS 2.0, AFAIK
this particular issue does not happen with DriverStudio 2.0.

Note that before a bug can be addressed, it must be reproducible! Meaning,
in more than one machine. What often happens is, some problem only happens
in your machine, or it only happens when your driver is on it, or only after
you install the latest beta of that 100-grand library, and we can’t have
your source code or your library, consequently we can’t duplicate the
problem. When we can duplicate the problem, it will typically get fixed. But
if the bug only happens in that one machine at 5pm during an 8 hour stress
test and only when it rains, it isn’t likely that the bug’s going to be
addressed unless a whole lot of work on location is done.

Reminds me, years ago I worked for a mainframe company and one of our
customers was an European country’s equivalent of the American CIA. In those
days bugs were diagnosed from core dumps. Well, the dumps we got were all
cut with scissor, because the censors wouldn’t let any information out
unless they considered it harmless. Needless to say, many bugs were left
unresolved.

The process is, or should be, simple. If you have a problem with a Numega
piece of software, please contact tech support at once. They’re going to do
their best to address your issue. If it escalates to bug status, developers
will expect it to be reproducible, that is, outside the customer’s location:
this means that we must have duplicates of your equipment, your operating
environment, all your installed software, and so on. Developers will then
assess the bug, prioritize it, and possibly fix it. Once the bug is fixed, a
note goes up on the web site and a system that fixes it is made available
for downloading.

But remember, reproducibility is the golden rule! If we can’t reproduce your
problem inhouse, it’s going to be tough to get it fixed. Right in front of
me, in my office, I have a white board with the current status of all the
high priority bugs in the system. You guys would be surprised to see how
many of those have a “can’t reproduce” tag attached!

Alberto Moreira
Development Manager, SoftIce and BoundsChecker

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 3:47 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f**k**g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[ mailto:xxxxx@lists.osr.com
mailto:xxxxx ]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing is
found it is a good thing to report such problems to their tech support. You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody will

ask you for more information and if problem is reproducible enough, they fix

will it and inform you about it. Finally, if you’re really lucky, the update

released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com http: ,
http://www.skytale.com http:]

> — You are currently subscribed to ntdev as:
xxxxx@compuware.com To unsubscribe send a blank email to
$subst(‘Email.Unsub’) ----------
> From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> Reply To: NT Developers Interest List
> Sent: Tuesday, December 05, 2000 7:25 PM
> To: NT Developers Interest List
> Subject: [ntdev] To SoftICE’s developers at Numega
> Importance: High
>
> I found (I think programmer’s) error in NMTRANS.DLL.
> Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> Adress of error: 1000EDC4
> there are two jumps to this location after returning zero from
> the routine below (also called in the never ending loop)
> Description:
> mov esi, 1
> 1000EDC9:
> lea eax, [esp+xx]
> mov ecx, ebx;this pointer
> push eax
> push esi
> call 1000EC90 ;return value ignored !!!
> inc esi
> jmp 1000EDC9-> Never ending loop
>
> This never ending loop occurs sometimes when translating symbols
> from W2K (which have both .dbg and .pdb files). Typical example is
> for CRYPT32 symbols from W2K SP1 Free Build.
>
> As a simplest workaround I have used replacing the final jump
> by two nop instructions. Then everything seems OK so I think this
> never ending loop must be either:
> 1. relic from some previous version with additional code (eg. for loop
> exiting)
> 2. intentional bug or bug originating in disregard
>
> Please contact me with information how this bug will be resolved.
> Paul
>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx>

You both are very unfair and untruthful. Our technical support is not of
serial numbers but of engineers who know their stuff. We’re different. I’ve
seen patches released sometimes in a week. So please stop spreading FUD.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com http:</http:>

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 3:47 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f k g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com
mailto:xxxxx]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing is
found it is a good thing to report such problems to their tech support. You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody will

ask you for more information and if problem is reproducible enough, they fix

will it and inform you about it. Finally, if you’re really lucky, the update

released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com http: ,
http://www.skytale.com http:]

> — You are currently subscribed to ntdev as: xxxxx@compuware.com To
unsubscribe send a blank email to $subst(‘Email.Unsub’) ----------

> From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> Reply To: NT Developers Interest List
> Sent: Tuesday, December 05, 2000 7:25 PM
> To: NT Developers Interest List
> Subject: [ntdev] To SoftICE’s developers at Numega
> Importance: High
>
> I found (I think programmer’s) error in NMTRANS.DLL.
> Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> Adress of error: 1000EDC4
> there are two jumps to this location after returning zero from
> the routine below (also called in the never ending loop)
> Description:
> mov esi, 1
> 1000EDC9:
> lea eax, [esp+xx]
> mov ecx, ebx;this pointer
> push eax
> push esi
> call 1000EC90 ;return value ignored !!!
> inc esi
> jmp 1000EDC9-> Never ending loop
>
> This never ending loop occurs sometimes when translating symbols
> from W2K (which have both .dbg and .pdb files). Typical example is
> for CRYPT32 symbols from W2K SP1 Free Build.
>
> As a simplest workaround I have used replacing the final jump
> by two nop instructions. Then everything seems OK so I think this
> never ending loop must be either:
> 1. relic from some previous version with additional code (eg. for loop
> exiting)
> 2. intentional bug or bug originating in disregard
>
> Please contact me with information how this bug will be resolved.
> Paul
>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx></mailto:xxxxx>

Support of yours is really unpleasant.
You hear right ! I’m talking about Numega.
?
If you can demonstrate me (and to others of course) Numega’s support
is really great, so tell me how to simply download repaired nmtrans.dll.
If you don’t satify me, my meaning (and of some others) will remain the
same.
?
Paul

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Gentile, Sam
Sent: Wednesday, December 06, 2000 3:47 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

You both are very unfair and untruthful. Our technical support is not of
serial numbers but of engineers who know their stuff. We’re different.
I’ve seen patches released sometimes in a week. So please stop spreading
FUD.
?
Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com
http://www.numega.com
?

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 3:47 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f**k**g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
??? considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing
is
found it is a good thing to report such problems to their tech support.
You
will receive a kind e-mail informing you that a problem was submitted to

engineers. If you’re lucky, after several urgencies and months somebody
will
ask you for more information and if problem is reproducible enough, they
fix
will it and inform you about it. Finally, if you’re really lucky, the
update
released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]

— You are currently subscribed to ntdev as:
xxxxx@compuware.com To unsubscribe send a blank email to
$subst(‘Email.Unsub’) ----------
From: ??? Hrdina Pavel[SMTP:xxxxx@compelson.com]
Reply To: ??? NT Developers Interest List
Sent: ??? Tuesday, December 05, 2000 7:25 PM
To: ? NT Developers Interest List
Subject: ??? [ntdev] To SoftICE’s developers at Numega
Importance: ? High

I found (I think programmer’s) error in NMTRANS.DLL.
Version info: 4.05, file size 413766, date time 25.01.2000 4:05
Adress of error: 1000EDC4
?? there are two jumps to this location after returning zero from
?? the routine below (also called in the never ending loop)
Description:
??? mov esi, 1
? 1000EDC9:
??? lea eax, [esp+xx]
??? mov ecx, ebx;this pointer
??? push eax
??? push esi
??? call 1000EC90??? ;return value ignored !!!
??? inc esi
??? jmp 1000EDC9-> Never ending loop

This never ending loop occurs sometimes when translating symbols
from W2K (which have both .dbg and .pdb files). Typical example is
for CRYPT32 symbols from W2K SP1 Free Build.

As a simplest workaround I have used replacing the final jump
by two nop instructions. Then everything seems OK so I think this
never ending loop must be either:

  1. relic from some previous version with additional code (eg. for loop

exiting)
2. intentional bug or bug originating in disregard

Please contact me with information how this bug will be resolved.
Paul


You are currently subscribed to ntdev as: xxxxx@rkk.cz
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)

Third time makes all, as I said already twice, this issue is fixed in
DriverStudio 2.0. But if you want a DS 1.5 nmtrans.dll, please talk to our
support people, and we’ll take it from there, and see what we can do. Did
you ever pass this issue to our support people ? If I have an issue number,
I can track it down and see who was “really unpleasant” to you, this is
obvious something we’d like to correct.

Alberto.

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 10:50 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

Support of yours is really unpleasant.
You hear right ! I’m talking about Numega.

If you can demonstrate me (and to others of course) Numega’s support
is really great, so tell me how to simply download repaired nmtrans.dll.
If you don’t satify me, my meaning (and of some others) will remain the
same.

Paul

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Gentile, Sam
Sent: Wednesday, December 06, 2000 3:47 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

You both are very unfair and untruthful. Our technical support is not of
serial numbers but of engineers who know their stuff. We’re different. I’ve
seen patches released sometimes in a week. So please stop spreading FUD.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com http:</http:>

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 3:47 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f k g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com
mailto:xxxxx]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing is
found it is a good thing to report such problems to their tech support. You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody will

ask you for more information and if problem is reproducible enough, they fix

will it and inform you about it. Finally, if you’re really lucky, the update

released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com http: ,
http://www.skytale.com http:]

> — You are currently subscribed to ntdev as: xxxxx@compuware.com To
unsubscribe send a blank email to $subst(‘Email.Unsub’) — You
are currently subscribed to ntdev as: xxxxx@compuware.com To
unsubscribe send a blank email to $subst(‘Email.Unsub’) ----------

> From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> Reply To: NT Developers Interest List
> Sent: Tuesday, December 05, 2000 7:25 PM
> To: NT Developers Interest List
> Subject: [ntdev] To SoftICE’s developers at Numega
> Importance: High
>
> I found (I think programmer’s) error in NMTRANS.DLL.
> Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> Adress of error: 1000EDC4
> there are two jumps to this location after returning zero from
> the routine below (also called in the never ending loop)
> Description:
> mov esi, 1
> 1000EDC9:
> lea eax, [esp+xx]
> mov ecx, ebx;this pointer
> push eax
> push esi
> call 1000EC90 ;return value ignored !!!
> inc esi
> jmp 1000EDC9-> Never ending loop
>
> This never ending loop occurs sometimes when translating symbols
> from W2K (which have both .dbg and .pdb files). Typical example is
> for CRYPT32 symbols from W2K SP1 Free Build.
>
> As a simplest workaround I have used replacing the final jump
> by two nop instructions. Then everything seems OK so I think this
> never ending loop must be either:
> 1. relic from some previous version with additional code (eg. for loop
> exiting)
> 2. intentional bug or bug originating in disregard
>
> Please contact me with information how this bug will be resolved.
> Paul
>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx></mailto:xxxxx>

http://www.numega.com/drivercentral/ http:</http:>

It’s been posted up there for months, available to the general public, free
of charge & without having to contact numega tech support directly, or send
any serial numbers.

http://www.numega.com/drivercentral/icecentral.asp#nmtrans
http:

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 9:50 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

Support of yours is really unpleasant.
You hear right ! I’m talking about Numega.

If you can demonstrate me (and to others of course) Numega’s support
is really great, so tell me how to simply download repaired nmtrans.dll.
If you don’t satify me, my meaning (and of some others) will remain the
same.

Paul

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Gentile, Sam
Sent: Wednesday, December 06, 2000 3:47 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

You both are very unfair and untruthful. Our technical support is not of
serial numbers but of engineers who know their stuff. We’re different. I’ve
seen patches released sometimes in a week. So please stop spreading FUD.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com http:</http:>

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 3:47 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f k g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com
mailto:xxxxx]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing is
found it is a good thing to report such problems to their tech support. You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody will

ask you for more information and if problem is reproducible enough, they fix

will it and inform you about it. Finally, if you’re really lucky, the update

released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com http: ,
http://www.skytale.com http:]

> — You are currently subscribed to ntdev as: xxxxx@compuware.com To
unsubscribe send a blank email to $subst(‘Email.Unsub’) — You
are currently subscribed to ntdev as: xxxxx@timesn.com To unsubscribe send a
blank email to $subst(‘Email.Unsub’) ----------
> From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> Reply To: NT Developers Interest List
> Sent: Tuesday, December 05, 2000 7:25 PM
> To: NT Developers Interest List
> Subject: [ntdev] To SoftICE’s developers at Numega
> Importance: High
>
> I found (I think programmer’s) error in NMTRANS.DLL.
> Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> Adress of error: 1000EDC4
> there are two jumps to this location after returning zero from
> the routine below (also called in the never ending loop)
> Description:
> mov esi, 1
> 1000EDC9:
> lea eax, [esp+xx]
> mov ecx, ebx;this pointer
> push eax
> push esi
> call 1000EC90 ;return value ignored !!!
> inc esi
> jmp 1000EDC9-> Never ending loop
>
> This never ending loop occurs sometimes when translating symbols
> from W2K (which have both .dbg and .pdb files). Typical example is
> for CRYPT32 symbols from W2K SP1 Free Build.
>
> As a simplest workaround I have used replacing the final jump
> by two nop instructions. Then everything seems OK so I think this
> never ending loop must be either:
> 1. relic from some previous version with additional code (eg. for loop
> exiting)
> 2. intentional bug or bug originating in disregard
>
> Please contact me with information how this bug will be resolved.
> Paul
>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx></mailto:xxxxx></http:>

Paul,

I gotta step in here … Ever filled a support position? I have, and it is a
pain in the ass. The stories you hear about “No sir, you can’t fold an 8”
floppy to fit in a 5" floppy drive." are true. In the last year, I have
spent hours listening to a customer breathing and mumbling to himself while
he figures out how to write C code. Oh joy, I LOVE it when I hear his voice
on the line.

Methinks I would love to hear your customers comment about the support you
give them.

I’m not a fan of Compuware/NuMega … in fact the last time I used SoftIce
was 8 years ago, and that was for DOS. Is Numega’s support “really great”, I
have no idea, but seeing the caliber of responses they have posted to the
news groups I’d estimate it is adequate, competent, and in some situations
probably even brilliant.

Gary

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 7:50 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

Support of yours is really unpleasant.
You hear right ! I’m talking about Numega.

If you can demonstrate me (and to others of course) Numega’s support
is really great, so tell me how to simply download repaired nmtrans.dll.
If you don’t satify me, my meaning (and of some others) will remain the
same.

Paul

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Gentile, Sam
Sent: Wednesday, December 06, 2000 3:47 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

You both are very unfair and untruthful. Our technical support is not of
serial numbers but of engineers who know their stuff. We’re different. I’ve
seen patches released sometimes in a week. So please stop spreading FUD.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com http:</http:>

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 3:47 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f k g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com
mailto:xxxxx]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing is
found it is a good thing to report such problems to their tech support. You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody will

ask you for more information and if problem is reproducible enough, they fix

will it and inform you about it. Finally, if you’re really lucky, the update

released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com http: ,
http://www.skytale.com http:]

> — You are currently subscribed to ntdev as: xxxxx@compuware.com To
unsubscribe send a blank email to $subst(‘Email.Unsub’) — You
are currently subscribed to ntdev as: xxxxx@delphieng.com To unsubscribe
send a blank email to $subst(‘Email.Unsub’) ----------
> From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> Reply To: NT Developers Interest List
> Sent: Tuesday, December 05, 2000 7:25 PM
> To: NT Developers Interest List
> Subject: [ntdev] To SoftICE’s developers at Numega
> Importance: High
>
> I found (I think programmer’s) error in NMTRANS.DLL.
> Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> Adress of error: 1000EDC4
> there are two jumps to this location after returning zero from
> the routine below (also called in the never ending loop)
> Description:
> mov esi, 1
> 1000EDC9:
> lea eax, [esp+xx]
> mov ecx, ebx;this pointer
> push eax
> push esi
> call 1000EC90 ;return value ignored !!!
> inc esi
> jmp 1000EDC9-> Never ending loop
>
> This never ending loop occurs sometimes when translating symbols
> from W2K (which have both .dbg and .pdb files). Typical example is
> for CRYPT32 symbols from W2K SP1 Free Build.
>
> As a simplest workaround I have used replacing the final jump
> by two nop instructions. Then everything seems OK so I think this
> never ending loop must be either:
> 1. relic from some previous version with additional code (eg. for loop
> exiting)
> 2. intentional bug or bug originating in disregard
>
> Please contact me with information how this bug will be resolved.
> Paul
>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx></mailto:xxxxx>

I guess he’s not willing to do a little looking. Rather complain.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com http:</http:>

-----Original Message-----
From: Jose Flores [mailto:xxxxx@timesn.com]
Sent: Wednesday, December 06, 2000 2:06 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

http://www.numega.com/drivercentral/ http:</http:>

It’s been posted up there for months, available to the general public, free
of charge & without having to contact numega tech support directly, or send
any serial numbers.

http://www.numega.com/drivercentral/icecentral.asp#nmtrans
http:

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 9:50 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

Support of yours is really unpleasant.
You hear right ! I’m talking about Numega.

If you can demonstrate me (and to others of course) Numega’s support
is really great, so tell me how to simply download repaired nmtrans.dll.
If you don’t satify me, my meaning (and of some others) will remain the
same.

Paul

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Gentile, Sam
Sent: Wednesday, December 06, 2000 3:47 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

You both are very unfair and untruthful. Our technical support is not of
serial numbers but of engineers who know their stuff. We’re different. I’ve
seen patches released sometimes in a week. So please stop spreading FUD.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com http:</http:>

-----Original Message-----
From: Hrdina Pavel [mailto:xxxxx@compelson.com]
Sent: Wednesday, December 06, 2000 3:47 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I think you’ve put really great explanation of
my reason not to contact Numega’s tech. support,
but immediately interested developers.

I know such a Byzantine methods of official ways.

Additionally the technical support is full of things
like f k g serial numbers, names and other
information not really needed to converse about
such errors. Everything needed is correct version
specification and reliable symptoms of the bug.

Paul

PS: In my opinion Numega’s web is something making
considerable headaches to everyone.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com
mailto:xxxxx]On Behalf Of Vodicka, Michal
Sent: Wednesday, December 06, 2000 3:19 AM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

It is always good idea to search for updates on NuMega web. If nothing is
found it is a good thing to report such problems to their tech support. You
will receive a kind e-mail informing you that a problem was submitted to
engineers. If you’re lucky, after several urgencies and months somebody will

ask you for more information and if problem is reproducible enough, they fix

will it and inform you about it. Finally, if you’re really lucky, the update

released after next several months will contain the fix.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com http: ,
http://www.skytale.com http:]

> — You are currently subscribed to ntdev as: xxxxx@compuware.com To
unsubscribe send a blank email to $subst(‘Email.Unsub’) — You
are currently subscribed to ntdev as: xxxxx@timesn.com To unsubscribe send a
blank email to $subst(‘Email.Unsub’) ----------
> From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> Reply To: NT Developers Interest List
> Sent: Tuesday, December 05, 2000 7:25 PM
> To: NT Developers Interest List
> Subject: [ntdev] To SoftICE’s developers at Numega
> Importance: High
>
> I found (I think programmer’s) error in NMTRANS.DLL.
> Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> Adress of error: 1000EDC4
> there are two jumps to this location after returning zero from
> the routine below (also called in the never ending loop)
> Description:
> mov esi, 1
> 1000EDC9:
> lea eax, [esp+xx]
> mov ecx, ebx;this pointer
> push eax
> push esi
> call 1000EC90 ;return value ignored !!!
> inc esi
> jmp 1000EDC9-> Never ending loop
>
> This never ending loop occurs sometimes when translating symbols
> from W2K (which have both .dbg and .pdb files). Typical example is
> for CRYPT32 symbols from W2K SP1 Free Build.
>
> As a simplest workaround I have used replacing the final jump
> by two nop instructions. Then everything seems OK so I think this
> never ending loop must be either:
> 1. relic from some previous version with additional code (eg. for loop
> exiting)
> 2. intentional bug or bug originating in disregard
>
> Please contact me with information how this bug will be resolved.
> Paul
>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compelson.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx></mailto:xxxxx></http:></mailto:xxxxx>

Sam,

I don’t agree with all what Paul wrote but persist on everything what I
wrote myself. It is based on my personal experience with your tech support.
I don’t want to blame tech support persons, they always replied to my
reports and always were polite and willing. Unfortunately, most of my
problems ended with a status “added to bug database”. During this year I
filled about 20 - 30 reports and only few problems are fixed in current
version i.e. Driver Studio 2.0 (search for both xxxxx@rkk.cz and
xxxxx@veridicom.cz if you’re interested). I expressed my disappointment
directly to NuMega several times and received many excuses and promises.
Actually, I received more excuses than fixes. After this complaining some
developers contacted me and were able to fix some problems. Unfortunately,
these fixes weren’t released, yet (about 3 months). Also, I have no
information about status of other problems reported months before and
perfectly reproducible. One example:

SoftICE isn’t able to work with process and thread IDs above 0x400 and also
object handles. To reproduce this problem simply start SoftICE on w2k
system, list all threads using THREAD command, find some with TID above
0x400 and execute thread command with this TID as a parameter. SI will
respond with “Thread: Invalid argument XXX message”. The same is for
processes and process handle table. Try “proc -o system” which will display
system process handle table. If you examine output thoroughly, you’ll see
that information for handles above 0x400 are invalid (it mirrors handles
from 0). You can compare it with System Internals NTHandle utility output.
These aren’t fatal problems but become really annoying when one solve
something where these information are necessary. I reported this sometimes
in August and still have no fix and no information about current status.

So what is unfair and untruthful?

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]


From: Gentile, Sam[SMTP:xxxxx@compuware.com]
Reply To: NT Developers Interest List
Sent: Wednesday, December 06, 2000 15:46
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

You both are very unfair and untruthful. Our technical support is not of
serial numbers but of engineers who know their stuff. We’re different.
I’ve seen patches released sometimes in a week. So please stop spreading
FUD.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
> http:</http:>
>
>
> -----Original Message-----
> From: Hrdina Pavel [mailto:xxxxx@compelson.com]
> Sent: Wednesday, December 06, 2000 3:47 AM
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
>
>
> I think you’ve put really great explanation of
> my reason not to contact Numega’s tech. support,
> but immediately interested developers.
>
> I know such a Byzantine methods of official ways.
>
> Additionally the technical support is full of things
> like f k g serial numbers, names and other
> information not really needed to converse about
> such errors. Everything needed is correct version
> specification and reliable symptoms of the bug.
>
> Paul
>
> PS: In my opinion Numega’s web is something making
> considerable headaches to everyone.
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx]On Behalf Of Vodicka,
> Michal
> Sent: Wednesday, December 06, 2000 3:19 AM
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
>
> It is always good idea to search for updates on NuMega web. If
> nothing is
> found it is a good thing to report such problems to their tech
> support. You
> will receive a kind e-mail informing you that a problem was
> submitted to
> engineers. If you’re lucky, after several urgencies and months
> somebody will
> ask you for more information and if problem is reproducible enough,
> they fix
> will it and inform you about it. Finally, if you’re really lucky,
> the update
> released after next several months will contain the fix.
>
> Best regards,
>
> Michal Vodicka
> Veridicom
> (RKK - Skytale)
> [WWW: http: , http:]
>
>
>
> > — You are currently subscribed to ntdev as:
> xxxxx@compuware.com To unsubscribe send a blank email to
> $subst(‘Email.Unsub’) ----------
> > From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> > Reply To: NT Developers Interest List
> > Sent: Tuesday, December 05, 2000 7:25 PM
> > To: NT Developers Interest List
> > Subject: [ntdev] To SoftICE’s developers at Numega
> > Importance: High
> >
> > I found (I think programmer’s) error in NMTRANS.DLL.
> > Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> > Adress of error: 1000EDC4
> > there are two jumps to this location after returning zero from
> > the routine below (also called in the never ending loop)
> > Description:
> > mov esi, 1
> > 1000EDC9:
> > lea eax, [esp+xx]
> > mov ecx, ebx;this pointer
> > push eax
> > push esi
> > call 1000EC90 ;return value ignored !!!
> > inc esi
> > jmp 1000EDC9-> Never ending loop
> >
> > This never ending loop occurs sometimes when translating symbols
> > from W2K (which have both .dbg and .pdb files). Typical example is
>
> > for CRYPT32 symbols from W2K SP1 Free Build.
> >
> > As a simplest workaround I have used replacing the final jump
> > by two nop instructions. Then everything seems OK so I think this
> > never ending loop must be either:
> > 1. relic from some previous version with additional code (eg. for
> loop
> > exiting)
> > 2. intentional bug or bug originating in disregard
> >
> > Please contact me with information how this bug will be resolved.
> > Paul
> >
> > —
> > You are currently subscribed to ntdev as: xxxxx@rkk.cz
> > To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>
> >
>
> —
> You are currently subscribed to ntdev as: xxxxx@compelson.com
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
></http:></http:></mailto:xxxxx></mailto:xxxxx>

I’m sure Alberto will see this and can do something about it. I’m not in
that group. Sorry.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com

-----Original Message-----
From: Vodicka, Michal [mailto:xxxxx@rkk.cz]
Sent: Wednesday, December 06, 2000 6:07 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

Sam,

I don’t agree with all what Paul wrote but persist on everything what I
wrote myself. It is based on my personal experience with your tech support.
I don’t want to blame tech support persons, they always replied to my
reports and always were polite and willing. Unfortunately, most of my
problems ended with a status “added to bug database”. During this year I
filled about 20 - 30 reports and only few problems are fixed in current
version i.e. Driver Studio 2.0 (search for both xxxxx@rkk.cz and
xxxxx@veridicom.cz if you’re interested). I expressed my disappointment
directly to NuMega several times and received many excuses and promises.
Actually, I received more excuses than fixes. After this complaining some
developers contacted me and were able to fix some problems. Unfortunately,
these fixes weren’t released, yet (about 3 months). Also, I have no
information about status of other problems reported months before and
perfectly reproducible. One example:

SoftICE isn’t able to work with process and thread IDs above 0x400 and also
object handles. To reproduce this problem simply start SoftICE on w2k
system, list all threads using THREAD command, find some with TID above
0x400 and execute thread command with this TID as a parameter. SI will
respond with “Thread: Invalid argument XXX message”. The same is for
processes and process handle table. Try “proc -o system” which will display
system process handle table. If you examine output thoroughly, you’ll see
that information for handles above 0x400 are invalid (it mirrors handles
from 0). You can compare it with System Internals NTHandle utility output.
These aren’t fatal problems but become really annoying when one solve
something where these information are necessary. I reported this sometimes
in August and still have no fix and no information about current status.

So what is unfair and untruthful?

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]

> ----------
> From: Gentile, Sam[SMTP:xxxxx@compuware.com]
> Reply To: NT Developers Interest List
> Sent: Wednesday, December 06, 2000 15:46
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
> You both are very unfair and untruthful. Our technical support is not of
> serial numbers but of engineers who know their stuff. We’re different.
> I’ve seen patches released sometimes in a week. So please stop spreading
> FUD.
>
> Sam Gentile
> Principal Software Engineer
> Compuware NuMega
> Core Technologies - Shared Components
> “Love Yourself. Love your neighbor. Love your COM.”
> xxxxx@numega.com mailto:xxxxx
> http:</http:>
>
>
> -----Original Message-----
> From: Hrdina Pavel [mailto:xxxxx@compelson.com]
> Sent: Wednesday, December 06, 2000 3:47 AM
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
>
>
> I think you’ve put really great explanation of
> my reason not to contact Numega’s tech. support,
> but immediately interested developers.
>
> I know such a Byzantine methods of official ways.
>
> Additionally the technical support is full of things
> like f k g serial numbers, names and other
> information not really needed to converse about
> such errors. Everything needed is correct version
> specification and reliable symptoms of the bug.
>
> Paul
>
> PS: In my opinion Numega’s web is something making
> considerable headaches to everyone.
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx]On Behalf Of Vodicka,
> Michal
> Sent: Wednesday, December 06, 2000 3:19 AM
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
>
> It is always good idea to search for updates on NuMega web. If
> nothing is
> found it is a good thing to report such problems to their tech
> support. You
> will receive a kind e-mail informing you that a problem was
> submitted to
> engineers. If you’re lucky, after several urgencies and months
> somebody will
> ask you for more information and if problem is reproducible enough,
> they fix
> will it and inform you about it. Finally, if you’re really lucky,
> the update
> released after next several months will contain the fix.
>
> Best regards,
>
> Michal Vodicka
> Veridicom
> (RKK - Skytale)
> [WWW: http: , http:]
>
>
>
> > — You are currently subscribed to ntdev as:
> xxxxx@compuware.com To unsubscribe send a blank email to
> $subst(‘Email.Unsub’) ----------
> > From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> > Reply To: NT Developers Interest List
> > Sent: Tuesday, December 05, 2000 7:25 PM
> > To: NT Developers Interest List
> > Subject: [ntdev] To SoftICE’s developers at Numega
> > Importance: High
> >
> > I found (I think programmer’s) error in NMTRANS.DLL.
> > Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> > Adress of error: 1000EDC4
> > there are two jumps to this location after returning zero from
> > the routine below (also called in the never ending loop)
> > Description:
> > mov esi, 1
> > 1000EDC9:
> > lea eax, [esp+xx]
> > mov ecx, ebx;this pointer
> > push eax
> > push esi
> > call 1000EC90 ;return value ignored !!!
> > inc esi
> > jmp 1000EDC9-> Never ending loop
> >
> > This never ending loop occurs sometimes when translating symbols
> > from W2K (which have both .dbg and .pdb files). Typical example is
>
> > for CRYPT32 symbols from W2K SP1 Free Build.
> >
> > As a simplest workaround I have used replacing the final jump
> > by two nop instructions. Then everything seems OK so I think this
> > never ending loop must be either:
> > 1. relic from some previous version with additional code (eg. for
> loop
> > exiting)
> > 2. intentional bug or bug originating in disregard
> >
> > Please contact me with information how this bug will be resolved.
> > Paul
> >
> > —
> > You are currently subscribed to ntdev as: xxxxx@rkk.cz
> > To unsubscribe send a blank email to
$subst(‘Email.Unsub’)
>
> >
>
> —
> You are currently subscribed to ntdev as: xxxxx@compelson.com
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)

>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx></mailto:xxxxx></mailto:xxxxx>

> ----------

From: Moreira, Alberto[SMTP:xxxxx@compuware.com]
Reply To: NT Developers Interest List
Sent: Wednesday, December 06, 2000 16:28
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

As it goes, the answer to many bugs in DS 1.5 is to install DS 2.0, AFAIK
this particular issue does not happen with DriverStudio 2.0.

Note that before a bug can be addressed, it must be reproducible! Meaning,
in more than one machine.

As a driver developer I cannot agree. A bug reproducible on developer’s
machine is the easiest and the least common case. Some non-reproducible bugs
can be solved using crashdumps, some using symptom examination, thinking and
simulation, some using cooperation with affected user. But the developer
must be interested about it.

What often happens is, some problem only happens in your machine, or it
only happens when your driver is on it, or only after you install the
latest beta of that 100-grand library, and we can’t have your source code
or your library, consequently we can’t duplicate the problem. When we can
duplicate the problem, it will typically get fixed.

I’m sorry but my experience is different. Most of problems I submitted are
reproducible and aren’t fixed, yet.

But if the bug only happens in that one machine at 5pm during an 8 hour
stress test and only when it rains, it isn’t likely that the bug’s going
to be addressed unless a whole lot of work on location is done.

Well, I’m always willing to cooperate. When some of your developer contacted
me, we were always able to find what causes a problem. Sometimes, I had to
write a test driver which created error conditions. However, mostly nobody
was interested.

Reminds me, years ago I worked for a mainframe company and one of our
customers was an European country’s equivalent of the American CIA. In
those days bugs were diagnosed from core dumps. Well, the dumps we got
were all cut with scissor, because the censors wouldn’t let any
information out unless they considered it harmless. Needless to say, many
bugs were left unresolved.

The process is, or should be, simple. If you have a problem with a Numega
piece of software, please contact tech support at once. They’re going to
do their best to address your issue. If it escalates to bug status,
developers will expect it to be reproducible, that is, outside the
customer’s location: this means that we must have duplicates of your
equipment, your operating environment, all your installed software, and so
on. Developers will then assess the bug, prioritize it, and possibly fix
it. Once the bug is fixed, a note goes up on the web site and a system
that fixes it is made available for downloading.

You described an ideal case how it should work. Unfortunately, the reality
is quite different. Most of my reports were added to bug database and
nothing else happened. At least I never heard about them again. Only fatal
problems when SoftICE crashed were solved in timely manner.

There is a big problem on your side with informing users about status of
reported problem. Only initial phase is good but after agreenment with tech
support that there is a problem and should be fixed I’m missing next
information. I’d like to know assigned priority, current status of problem,
expected timeframe and responsible person.

Also, when a bug is fixed I’d expect to receive the patch soon. Currently,
I’m still waiting for fixes which were made before DS 2.0 release (August?)
but weren’t included there.

But remember, reproducibility is the golden rule! If we can’t reproduce
your problem inhouse, it’s going to be tough to get it fixed. Right in
front of me, in my office, I have a white board with the current status of
all the high priority bugs in the system. You guys would be surprised to
see how many of those have a “can’t reproduce” tag attached!

Really? And did you inform affected users about it? Did you ask them for
more information?

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]

I certainly will. Note for example that the TID>400 problem is now solved,
we will soon have a patch available on the web site. Some of the issues
Michal Vodicka has reported have also been fixed in that patch. However, in
his particular case, reproducibility has been a nontrivial problem; if I
have it right, due to the kind of work he does, we seldom have access to
configurations that match those he uses.

And remember, if we can’t reproduce a problem, that problem has a high
chance of not getting fixed. Reproducibility is the first and foremost
necessary ingredient.

But you see, you cannot pin the issue of how fast the development
organization answers your requests to how good our tech support people are.
Two different issues. If you come to me and say, “your development group
stinks”, I may take it as a valid criticism and see what’s going wrong. But
it is not fair, as I see it, to come and say, “your support people stink”,
just because some of your requests aren’t been answered at the speed you’d
like to see them answered.

But keep the bugs flowing, we really appreciate hearing from you!

Alberto.

-----Original Message-----
From: Gentile, Sam [mailto:xxxxx@compuware.com]
Sent: Wednesday, December 06, 2000 6:30 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I’m sure Alberto will see this and can do something about it. I’m not in
that group. Sorry.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
http://www.numega.com

-----Original Message-----
From: Vodicka, Michal [mailto:xxxxx@rkk.cz]
Sent: Wednesday, December 06, 2000 6:07 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

Sam,

I don’t agree with all what Paul wrote but persist on everything what I
wrote myself. It is based on my personal experience with your tech support.
I don’t want to blame tech support persons, they always replied to my
reports and always were polite and willing. Unfortunately, most of my
problems ended with a status “added to bug database”. During this year I
filled about 20 - 30 reports and only few problems are fixed in current
version i.e. Driver Studio 2.0 (search for both xxxxx@rkk.cz and
xxxxx@veridicom.cz if you’re interested). I expressed my disappointment
directly to NuMega several times and received many excuses and promises.
Actually, I received more excuses than fixes. After this complaining some
developers contacted me and were able to fix some problems. Unfortunately,
these fixes weren’t released, yet (about 3 months). Also, I have no
information about status of other problems reported months before and
perfectly reproducible. One example:

SoftICE isn’t able to work with process and thread IDs above 0x400 and also
object handles. To reproduce this problem simply start SoftICE on w2k
system, list all threads using THREAD command, find some with TID above
0x400 and execute thread command with this TID as a parameter. SI will
respond with “Thread: Invalid argument XXX message”. The same is for
processes and process handle table. Try “proc -o system” which will display
system process handle table. If you examine output thoroughly, you’ll see
that information for handles above 0x400 are invalid (it mirrors handles
from 0). You can compare it with System Internals NTHandle utility output.
These aren’t fatal problems but become really annoying when one solve
something where these information are necessary. I reported this sometimes
in August and still have no fix and no information about current status.

So what is unfair and untruthful?

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]

> ----------
> From: Gentile, Sam[SMTP:xxxxx@compuware.com]
> Reply To: NT Developers Interest List
> Sent: Wednesday, December 06, 2000 15:46
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
> You both are very unfair and untruthful. Our technical support is not of
> serial numbers but of engineers who know their stuff. We’re different.
> I’ve seen patches released sometimes in a week. So please stop spreading
> FUD.
>
> Sam Gentile
> Principal Software Engineer
> Compuware NuMega
> Core Technologies - Shared Components
> “Love Yourself. Love your neighbor. Love your COM.”
> xxxxx@numega.com mailto:xxxxx
> http:</http:>
>
>
> -----Original Message-----
> From: Hrdina Pavel [mailto:xxxxx@compelson.com]
> Sent: Wednesday, December 06, 2000 3:47 AM
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
>
>
> I think you’ve put really great explanation of
> my reason not to contact Numega’s tech. support,
> but immediately interested developers.
>
> I know such a Byzantine methods of official ways.
>
> Additionally the technical support is full of things
> like f k g serial numbers, names and other
> information not really needed to converse about
> such errors. Everything needed is correct version
> specification and reliable symptoms of the bug.
>
> Paul
>
> PS: In my opinion Numega’s web is something making
> considerable headaches to everyone.
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx]On Behalf Of Vodicka,
> Michal
> Sent: Wednesday, December 06, 2000 3:19 AM
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
>

> It is always good idea to search for updates on NuMega web. If
> nothing is
> found it is a good thing to report such problems to their tech
> support. You
> will receive a kind e-mail informing you that a problem was
> submitted to
> engineers. If you’re lucky, after several urgencies and months
> somebody will
> ask you for more information and if problem is reproducible enough,
> they fix
> will it and inform you about it. Finally, if you’re really lucky,
> the update
> released after next several months will contain the fix.
>
> Best regards,
>
> Michal Vodicka
> Veridicom
> (RKK - Skytale)
> [WWW: http: , http:]
>
>
>
> > — You are currently subscribed to ntdev as:
> xxxxx@compuware.com To unsubscribe send a blank email to
> $subst(‘Email.Unsub’) ----------
> > From: Hrdina Pavel[SMTP:xxxxx@compelson.com]
> > Reply To: NT Developers Interest List
> > Sent: Tuesday, December 05, 2000 7:25 PM
> > To: NT Developers Interest List
> > Subject: [ntdev] To SoftICE’s developers at Numega
> > Importance: High
> >
> > I found (I think programmer’s) error in NMTRANS.DLL.
> > Version info: 4.05, file size 413766, date time 25.01.2000 4:05
> > Adress of error: 1000EDC4
> > there are two jumps to this location after returning zero from
> > the routine below (also called in the never ending loop)
> > Description:
> > mov esi, 1
> > 1000EDC9:
> > lea eax, [esp+xx]
> > mov ecx, ebx;this pointer
> > push eax
> > push esi
> > call 1000EC90 ;return value ignored !!!
> > inc esi
> > jmp 1000EDC9-> Never ending loop
> >
> > This never ending loop occurs sometimes when translating symbols
> > from W2K (which have both .dbg and .pdb files). Typical example is
>
> > for CRYPT32 symbols from W2K SP1 Free Build.
> >
> > As a simplest workaround I have used replacing the final jump
> > by two nop instructions. Then everything seems OK so I think this
> > never ending loop must be either:
> > 1. relic from some previous version with additional code (eg. for
> loop
> > exiting)
> > 2. intentional bug or bug originating in disregard
> >
> > Please contact me with information how this bug will be resolved.
> > Paul
> >
> > —
> > You are currently subscribed to ntdev as: xxxxx@rkk.cz
> > To unsubscribe send a blank email to
$subst(‘Email.Unsub’)
>
> >
>
> —
> You are currently subscribed to ntdev as: xxxxx@compelson.com
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)

>
> —
> You are currently subscribed to ntdev as: xxxxx@rkk.cz
> To unsubscribe send a blank email to $subst(‘Email.Unsub’)
>


You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)


You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to $subst(‘Email.Unsub’)</http:></http:></mailto:xxxxx></mailto:xxxxx></mailto:xxxxx>

I tend to think that “this is off-topic” messages are self-contradictory and
a waste of time, but this “numega sucks” thread has become annoying and is
off-topic. If you don’t like numega’s products or their support take it up
with numega.

Mark Roddy
xxxxx@hollistech.com
www.hollistech.com
WindowsNT Windows 2000 Consulting Services

It is great to hear it from you who were screaming here about windbag bugs
and bad support many times. If you haven’t noticed it, SoftICE is one of two
main NT debuggers and many people use it. That’s why I think it is relevant
to discuss quality of their tech support here. Of course, I tried to solve
these problems with NuMega before.

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]


From: Roddy, Mark[SMTP:xxxxx@stratus.com]
Reply To: NT Developers Interest List
Sent: Thursday, December 07, 2000 6:59 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I tend to think that “this is off-topic” messages are self-contradictory
and
a waste of time, but this “numega sucks” thread has become annoying and is
off-topic. If you don’t like numega’s products or their support take it up
with numega.

Mark Roddy
xxxxx@hollistech.com
www.hollistech.com
WindowsNT Windows 2000 Consulting Services


You are currently subscribed to ntdev as: xxxxx@rkk.cz
To unsubscribe send a blank email to $subst(‘Email.Unsub’)

> ----------

From: Moreira, Alberto[SMTP:xxxxx@compuware.com]
Reply To: NT Developers Interest List
Sent: Thursday, December 07, 2000 6:31 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I certainly will. Note for example that the TID>400 problem is now solved,
we will soon have a patch available on the web site. Some of the issues
Michal Vodicka has reported have also been fixed in that patch.

Great to know. But why wasn’t I informed about it before? It should be
obvious to inform all people who reported a problem about the fix.

However, in
his particular case, reproducibility has been a nontrivial problem; if I
have it right, due to the kind of work he does, we seldom have access to
configurations that match those he uses.

The only unusual thing which caused some of my problems is monochrome
monitor used for SoftICE display. I believe it is one of main SoftICE
advantages and should be supported. Also, it is recommended in SI
documentation :slight_smile:

And remember, if we can’t reproduce a problem, that problem has a high
chance of not getting fixed. Reproducibility is the first and foremost
necessary ingredient.

Again, if something isn’t reproducible in your environment, you should try
to work with user who is able to reproduce it.

But you see, you cannot pin the issue of how fast the development
organization answers your requests to how good our tech support people
are.
Two different issues.

I agree and tried to write it. I never complained about tech support people
only about the developers’ speed and their willingness to make fixes.

If you come to me and say, “your development group
stinks”, I may take it as a valid criticism and see what’s going wrong.
But
it is not fair, as I see it, to come and say, “your support people stink”,
just because some of your requests aren’t been answered at the speed you’d
like to see them answered.

Never said it.

But keep the bugs flowing, we really appreciate hearing from you!

Well, you’ll see next bugreport from me soon. Let’s stop this thread now to
save Mark Roddy’s sanity :wink:

Best regards,

Michal Vodicka
Veridicom
(RKK - Skytale)
[WWW: http://www.veridicom.com , http://www.skytale.com]

Alberto.

-----Original Message-----
From: Gentile, Sam [mailto:xxxxx@compuware.com]
Sent: Wednesday, December 06, 2000 6:30 PM
To: NT Developers Interest List
Subject: [ntdev] RE: To SoftICE’s developers at Numega

I’m sure Alberto will see this and can do something about it. I’m not in
that group. Sorry.

Sam Gentile
Principal Software Engineer
Compuware NuMega
Core Technologies - Shared Components
“Love Yourself. Love your neighbor. Love your COM.”
xxxxx@numega.com mailto:xxxxx
> http://www.numega.com
>
>
>
> -----Original Message-----
> From: Vodicka, Michal [mailto:xxxxx@rkk.cz]
> Sent: Wednesday, December 06, 2000 6:07 PM
> To: NT Developers Interest List
> Subject: [ntdev] RE: To SoftICE’s developers at Numega
>
>
> Sam,
>
> I don’t agree with all what Paul wrote but persist on everything what I
> wrote myself. It is based on my personal experience with your tech
> support.
> I don’t want to blame tech support persons, they always replied to my
> reports and always were polite and willing. Unfortunately, most of my
> problems ended with a status “added to bug database”. During this year I
> filled about 20 - 30 reports and only few problems are fixed in current
> version i.e. Driver Studio 2.0 (search for both xxxxx@rkk.cz and
> xxxxx@veridicom.cz if you’re interested). I expressed my disappointment
> directly to NuMega several times and received many excuses and promises.
> Actually, I received more excuses than fixes. After this complaining some
> developers contacted me and were able to fix some problems. Unfortunately,
> these fixes weren’t released, yet (about 3 months). Also, I have no
> information about status of other problems reported months before and
> perfectly reproducible. One example:
>
> SoftICE isn’t able to work with process and thread IDs above 0x400 and
> also
> object handles. To reproduce this problem simply start SoftICE on w2k
> system, list all threads using THREAD command, find some with TID above
> 0x400 and execute thread command with this TID as a parameter. SI will
> respond with “Thread: Invalid argument XXX message”. The same is for
> processes and process handle table. Try “proc -o system” which will
> display
> system process handle table. If you examine output thoroughly, you’ll see
> that information for handles above 0x400 are invalid (it mirrors handles
> from 0). You can compare it with System Internals NTHandle utility output.
> These aren’t fatal problems but become really annoying when one solve
> something where these information are necessary. I reported this sometimes
> in August and still have no fix and no information about current status.
>
> So what is unfair and untruthful?
>
> Best regards,
>
> Michal Vodicka
> Veridicom
> (RKK - Skytale)
> [WWW: http://www.veridicom.com , http://www.skytale.com]
>
></mailto:xxxxx>

>Great to know. But why wasn’t I informed about it before? It should be

obvious to inform all people who reported a problem about the fix.

You weren’t notified because we didn’t release it yet. I am just now giving
a tidbit of advance information. Watch this channel!

The only unusual thing which caused some of my problems is monochrome
monitor used for SoftICE display. I believe it is one of main SoftICE
advantages and should be supported. Also, it is recommended in SI
documentation :slight_smile:

Some modern video boards preclude the use of a monochrome monitor, so it’s
sort of on its way into oblivion. But dual monitor functionality
compensates, unless of course you’re a video developer.

Well, you’ll see next bugreport from me soon. Let’s stop this thread now to
save Mark Roddy’s sanity :wink:

Well, I don’t complain when they flame Windbg. :wink:

Alberto.