Converting NT Driver to XP

NTFSD
21

My intent was to focus on the issue of interop and not to single out
those who cannot or would not pay for the IFS Kit. I’m speaking from
experience in dealing with interop issues with many filters and tracking
down who’s at fault - our driver or someone else’s. It’s very easy these
days to write software and distribute it all over the world and install
various components from services to drivers on any number of computers.

I’d dare to say that many who use NT log in via an administrative
account and therefore installing drivers or services is allowed. Given
that, you could silently install drivers and/or services and not know
for sure what you’re putting on a system. Of course, there’s driver
signing but that only applies to AV FS filters.

Which brings me back to my original point, what happens if I install an
application that silently inserts an FS filter in some IO stack without
my knowledge? What also happens if that filter is poorly written or
malicious?

I have spent countless hours crawling through crash dumps and debugging
deadlocked systems chasing after interop issues. It’s hard enough for
the commercial products to get all the issues addressed when you do this
full-time. Invariably, someone installs a filter with all good
intentions and an interop issue pops up. You have to be very careful
writing FS filters or some very bad things can happen.

Once again, these are my opinions. They may seem “dumb” to some because
they don’t agree with them but believe me, I’m not saying it because I
get a kickback on every kit sold or I just like spending money. I spoke
up because I believe some things are necessary to solidify these types
of drivers and reduce interop issues.

Besides, a little debate every now and then is healthy. :wink:

-----Original Message-----
From: xxxxx@earthling.net
Sent: Tue 11/27/2001 4:31 AM
To: File Systems Developers
Cc:
Subject: [ntfsd] RE: Converting NT Driver to XP

From: PeterB [mailto:xxxxx@inkvine.fluff.org]
On Mon, 26 Nov 2001, Kelley, Jerry wrote:
> Developing filters and file systems is not trivial or casual work for
> the hobbyist.
Why not? I’m genuinely curious. If a hobbyist wishes to write a
driver,
why shouldn’t they? It’s not like they’re going to force you to use
it,
is it?

Apart from the economic argument (why not charge, if all us people are
going to pay for it) there is a genuine reason for discouraging too many
filters and file systems out there. Filters in particular can cause
some really nasty system failures which can be really hard to pin down,
and sometimes only in particular configurations or when mixed with
particular other ones. I haven’t so much experience with NT file
systems per se, but can’t I imagine the situation is different. So by
making sure that only those of us who really need a filter produce one,
and that we are serious enough to put real money behind our development
(and presumably QA) they help keep NT’s reliability figures up. After
all, it’s hard enough to organise the plugfests that do get run, and
most of us know how odd problems get shown up there. Imagine if there
were 10 times as many attendees - say 200 companies, all wanting to test
with each of the other 199 filters/filesystems…

Rgds
Andy.

Sign-up for your own FREE Personalized E-mail at Mail.com

http://www.mail.com/?sr=signup

1 cent a minute calls anywhere in the U.S.!

http://www.getpennytalk.com/cgi-bin/adforward.cgi?p_key=RG9853KJ
http:tp://www.getpennytalk.com> &url=http://www.getpennytalk.com

—
You are currently subscribed to ntfsd as: xxxxx@nsisoftware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

???y˫???+.n?+???~?Z?˛??^r*D???kN???r??zǧu??jy???^j??? 0?j?b??(??(</http:>

22

> … Should I be taken out and beaten?

Of course not, but if you were ever assigned a semester project to do
something significant with NT IFS, you might find that you would *prefer* to
be taken out and beaten ;-).

Check out this thesis describing “FIFS”, a user-mode framework for
“dabbling” with NT IFS:
http://www.usenix.org/events/usenix-nt99/full_papers/almeida/almeida_html/.
I haven’t read the whole thing, so I can’t vouch for it, would be interested
to know if you find it at all useful.

Also, this paper describes a research project that (apparently) used FIFS:
http://citeseer.nj.nec.com/rimer99secure.html.

HTH …

– CCP

----- Original Message -----
From: “Argus”
To: “File Systems Developers”
Sent: Monday, November 26, 2001 6:28 PM
Subject: [ntfsd] RE: Converting NT Driver to XP

> What if we are still in school and can’t develop them professionally?
> Is that wrong? Should I be taken out and beaten? Sometimes interesting
> things come out of random dabbling. I think windows is a better operating
> system than linux, and it drives me crazy that MS has made it so hard to
> dabble in filesystems. I’ve no intentions of selling anything that I
> make–indeed, I doubt anyone would buy them anyhow. But I can’t for the
> life of me figure out what MS has to gain by keeping developers out of
file
> systems. And as to the source that’s included, the regular DDK has some
> samples from the OS too, what’s the difference?
>
> -brian
> ----- Original Message -----
> From: “Kelley, Jerry”
> To: “File Systems Developers”
> Sent: Monday, November 26, 2001 12:55 PM
> Subject: [ntfsd] RE: Converting NT Driver to XP
>
>
>
> It’s worth it for the documentation alone. Additionally, there are
> examples of how to use the new XP interfaces in SFilter so that’s got to
> be worth $109.
>
> It’s pretty interesting that a number of questions have come up about
> the price of the kit itself. The bottom line is, if you’re developing
> filters or file systems professionally, there should be no question
> about purchasing the IFS Kit. If you can’t afford the kit then you
> probably shouldn’t be in this line of work. Try developing web pages
> instead.
>
> Developing filters and file systems is not trivial or casual work for
> the hobbyist. I’m glad the kit is not free if, for nothing else, just to
> keep the development of this type of driver in the hands of the truly
> serious.
>
> My opinions alone.
>
> --jerry
>
> -----Original Message-----
> From: Ralph Shnelvar [mailto:xxxxx@dos32.com]
> Sent: Monday, November 26, 2001 7:53 AM
> To: File Systems Developers
> Subject: [ntfsd] RE: Converting NT Driver to XP
>
>
> Dear Jamey:
>
> On Sun, 25 Nov 2001 10:21:38 -0800, you wrote:
>
> >You do not need to buy the new IFS kit unless you want to have the new
> >samples. SFILTER sample in the IFS kit is not so good. There are many
> >problems that we have had to address.
>
> It is not clear from the English whether the XP version of SFILTER is
> better
> than the one I already have.
>
> In other words, is it worth the $109 to get the new SFILTER?
>
> >
> [snip]
> >
> >Jamey
>
> Ralph Shnelvar
>
> —
> You are currently subscribed to ntfsd as: xxxxx@nsisoftware.com
> To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com
>
> —
> You are currently subscribed to ntfsd as: argus@vt.edu
> To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com
>
>
> —
> You are currently subscribed to ntfsd as: xxxxx@acm.org
> To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com
>

—
You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

23

[ntfsd] RE: Converting NT Driver to XP Hmm. You make a very good point.
I still think MS could release a research only license, perhaps, which
expressly forbids distribution without a big red box reading “WARNING:
UNSAFE!” Or something like that. Oh well. I suppose I’ll just deal with
romfs until I get some money or the right job.

-brian
----- Original Message -----
From: Kelley, Jerry
To: File Systems Developers
Sent: Tuesday, November 27, 2001 9:34 AM
Subject: [ntfsd] RE: Converting NT Driver to XP

My intent was to focus on the issue of interop and not to single out those
who cannot or would not pay for the IFS Kit. I’m speaking from experience in
dealing with interop issues with many filters and tracking down who’s at
fault - our driver or someone else’s. It’s very easy these days to write
software and distribute it all over the world and install various components
from services to drivers on any number of computers.

I’d dare to say that many who use NT log in via an administrative account
and therefore installing drivers or services is allowed. Given that, you
could silently install drivers and/or services and not know for sure what
you’re putting on a system. Of course, there’s driver signing but that only
applies to AV FS filters.

Which brings me back to my original point, what happens if I install an
application that silently inserts an FS filter in some IO stack without my
knowledge? What also happens if that filter is poorly written or malicious?

I have spent countless hours crawling through crash dumps and debugging
deadlocked systems chasing after interop issues. It’s hard enough for the
commercial products to get all the issues addressed when you do this
full-time. Invariably, someone installs a filter with all good intentions
and an interop issue pops up. You have to be very careful writing FS filters
or some very bad things can happen.

Once again, these are my opinions. They may seem “dumb” to some because they
don’t agree with them but believe me, I’m not saying it because I get a
kickback on every kit sold or I just like spending money. I spoke up because
I believe some things are necessary to solidify these types of drivers and
reduce interop issues.

Besides, a little debate every now and then is healthy. :wink:

-----Original Message-----
From: xxxxx@earthling.net
Sent: Tue 11/27/2001 4:31 AM
To: File Systems Developers
Cc:
Subject: [ntfsd] RE: Converting NT Driver to XP

From: PeterB [mailto:xxxxx@inkvine.fluff.org]
On Mon, 26 Nov 2001, Kelley, Jerry wrote:
> Developing filters and file systems is not trivial or casual work for
> the hobbyist.
Why not? I’m genuinely curious. If a hobbyist wishes to write a driver,
why shouldn’t they? It’s not like they’re going to force you to use it,
is it?

Apart from the economic argument (why not charge, if all us people are going
to pay for it) there is a genuine reason for discouraging too many filters
and file systems out there. Filters in particular can cause some really
nasty system failures which can be really hard to pin down, and sometimes
only in particular configurations or when mixed with particular other ones.
I haven’t so much experience with NT file systems per se, but can’t I
imagine the situation is different. So by making sure that only those of us
who really need a filter produce one, and that we are serious enough to put
real money behind our development (and presumably QA) they help keep NT’s
reliability figures up. After all, it’s hard enough to organise the
plugfests that do get run, and most of us know how odd problems get shown up
there. Imagine if there were 10 times as many attendees - say 200
companies, all wanting to test with each of the other 199
filters/filesystems…

Rgds
Andy.

Sign-up for your own FREE Personalized E-mail at Mail.com

http://www.mail.com/?sr=signup

1 cent a minute calls anywhere in the U.S.!

http://www.getpennytalk.com/cgi-bin/adforward.cgi?p_key=RG9853KJ&url=http://
www.getpennytalk.com

You are currently subscribed to ntfsd as: xxxxx@nsisoftware.com
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

b‹š­箮·§¶\¬¹»®&޶ڧµ󝪆«‚믵箎‹§²欲¸›zǧu¦媹™¨¥¶‰^jí½Ÿíµ»ë½BX¬¶ʬ­ʦ

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

24

On Tue, 27 Nov 2001 xxxxx@earthling.net wrote:

>From: PeterB [mailto:xxxxx@inkvine.fluff.org]
>On Mon, 26 Nov 2001, Kelley, Jerry wrote:
>> Developing filters and file systems is not trivial or casual work for
>> the hobbyist.
>Why not? I’m genuinely curious. If a hobbyist wishes to write a driver,
>why shouldn’t they? It’s not like they’re going to force you to use it,
>is it?

Apart from the economic argument (why not charge, if all us people are
going to pay for it) there is a genuine reason for discouraging too
many filters and file systems out there. Filters in particular can
cause some really nasty system failures which can be really hard to
pin down, and sometimes only in particular configurations or when
mixed with particular other ones. I haven’t so much experience with
NT file systems per se, but can’t I imagine the situation is
different. So by making sure that only those of us who really need a
filter produce one, and that we are serious enough to put real money
behind our development (and presumably QA) they help keep NT’s
reliability figures up.
I think this is invalid. I don’t believe that those hobbyists will
distribute their drivers for use in production. Further, I believe that
they may well represent the difference between “potentially unstable code
that does what you want” and “no code at all to do what you want”. The
former is always preferable; a stable machine that can’t do what I want is
infinitely less useful than a potentially unstable machine (yet to be
proven, mind you; I’ve not seen anything particularly compelling to
suggest that hobbyist-developed code is any more or less unstable,
on average, than professionally-developed code) that can.

After all, it’s hard enough to organise the
plugfests that do get run, and most of us know how odd problems get
shown up there. Imagine if there were 10 times as many attendees -
say 200 companies, all wanting to test with each of the other 199
filters/filesystems…
If the hobbyist-developed driver is never used in conjunction with
those filters IRL anyway, how much does it matter that it might not work
properly with them? And given that commercial drivers are by no means
immune from such problems, by that same argument, shouldn’t one make the
development of third-party FSDs completely impossible?

–
Peter xxxxx@inkvine.fluff.org
http://www.inkvine.fluff.org/~peter/

logic kicks ass:
(1) Horses have an even number of legs.
(2) They have two legs in back and fore legs in front.
(3) This makes a total of six legs, which certainly is an odd number of
legs for a horse.
(4) But the only number that is both odd and even is infinity.
(5) Therefore, horses must have an infinite number of legs.

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

25

It’s a good point that new development should be encouraged. By no means
should anyone be discouraged from developing. My experience has been
that our customers would rather not have a feature than runs any risk of
destabilizing their systems. When you’re dealing with companies that
rely heavily on their systems staying up, they’ll tell (in no uncertain
terms) that if there’s any chance that a software product may bring down
their systems, you’ll be told what to do with that software.

I know that there are supposed to be strict rules for what goes on
systems. But I’ve seen cases where an administrator gets excited about
some neat new tool they found on TUCOWS and decide to install and test
it on their company’s server(s). The tool may be experimental or may be
rock solid. But it only takes one error in a filter to smoke the system.
(I know, I know, an administrator shouldn’t do that but it happens and
then we have to look at the crash dump.)

Regarding filter interop, many systems now come with an AV product
pre-installed. Additionally, on the XP flavors, you have the SR filter.
To say that you won’t have any other filters to worry about is not as
easy as you may think.

-----Original Message-----
From: PeterB [mailto:xxxxx@inkvine.fluff.org]
Sent: Wednesday, November 28, 2001 8:32 AM
To: File Systems Developers
Subject: [ntfsd] RE: Converting NT Driver to XP

On Tue, 27 Nov 2001 xxxxx@earthling.net wrote:

>From: PeterB [mailto:xxxxx@inkvine.fluff.org]
>On Mon, 26 Nov 2001, Kelley, Jerry wrote:
>> Developing filters and file systems is not trivial or casual work
for
>> the hobbyist.
>Why not? I’m genuinely curious. If a hobbyist wishes to write a
driver,
>why shouldn’t they? It’s not like they’re going to force you to use
it,
>is it?

Apart from the economic argument (why not charge, if all us people are
going to pay for it) there is a genuine reason for discouraging too
many filters and file systems out there. Filters in particular can
cause some really nasty system failures which can be really hard to
pin down, and sometimes only in particular configurations or when
mixed with particular other ones. I haven’t so much experience with
NT file systems per se, but can’t I imagine the situation is
different. So by making sure that only those of us who really need a
filter produce one, and that we are serious enough to put real money
behind our development (and presumably QA) they help keep NT’s
reliability figures up.
I think this is invalid. I don’t believe that those hobbyists will
distribute their drivers for use in production. Further, I believe that
they may well represent the difference between “potentially unstable
code
that does what you want” and “no code at all to do what you want”. The
former is always preferable; a stable machine that can’t do what I want
is
infinitely less useful than a potentially unstable machine (yet to be
proven, mind you; I’ve not seen anything particularly compelling to
suggest that hobbyist-developed code is any more or less unstable,
on average, than professionally-developed code) that can.

After all, it’s hard enough to organise the
plugfests that do get run, and most of us know how odd problems get
shown up there. Imagine if there were 10 times as many attendees -
say 200 companies, all wanting to test with each of the other 199
filters/filesystems…
If the hobbyist-developed driver is never used in conjunction with
those filters IRL anyway, how much does it matter that it might not work
properly with them? And given that commercial drivers are by no means
immune from such problems, by that same argument, shouldn’t one make the
development of third-party FSDs completely impossible?

–
Peter
xxxxx@inkvine.fluff.org
http://www.inkvine.fluff.org/~peter/

logic kicks ass:
(1) Horses have an even number of legs.
(2) They have two legs in back and fore legs in front.
(3) This makes a total of six legs, which certainly is an odd number of
legs for a horse.
(4) But the only number that is both odd and even is infinity.
(5) Therefore, horses must have an infinite number of legs.

You are currently subscribed to ntfsd as: xxxxx@nsisoftware.com
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

26

On Tue, 27 Nov 2001, Kelley, Jerry wrote:

I’d dare to say that many who use NT log in via an administrative
account and therefore installing drivers or services is allowed. Given
that, you could silently install drivers and/or services and not know
for sure what you’re putting on a system. Of course, there’s driver
signing but that only applies to AV FS filters.
Silently loading and unloading drivers doesn’t cause signing warnings
anyway; signatures are checked on installation, not on load; to avoid the
warnings, don’t bother installing the thing with a .inf and all is
well. Driver signing has never provided protection against this kind of
activity.

Which brings me back to my original point, what happens if I install an
application that silently inserts an FS filter in some IO stack without
my knowledge? What also happens if that filter is poorly written or
malicious?
what happens if I install a regular filter/driver? What happens if that
driver is poorly written or malicious?

To be consistent, you should advocate that all driver development
require NDAs and expensive toolkits. These problems are by no means
unique to that particular kind of driver.

My reply would be that if you’re going to run as an Administrator, you
must bear some responsibility, and *be careful*. Yes, Administrators can
install drivers and services. Reducing the ease with which one might
write an FSD on the off chance that someone might write a malicious FSD is
missing the point, IMO. The point should be – Administrators, be
careful; you can compromise or crash the machine.

I have spent countless hours crawling through crash dumps and debugging
deadlocked systems chasing after interop issues. It’s hard enough for
the commercial products to get all the issues addressed when you do this
full-time. Invariably, someone installs a filter with all good
intentions and an interop issue pops up. You have to be very careful
writing FS filters or some very bad things can happen.
FS filters are not the only kind of filter. FSDs are not the only kind of
driver. I’m not convinced that they’re significantly more dangerous than
any other kind of driver; other drivers/filters can suffer interop issues
too, other drivers/filters can crash your machine.

–
Peter xxxxx@inkvine.fluff.org
http://www.inkvine.fluff.org/~peter/

logic kicks ass:
(1) Horses have an even number of legs.
(2) They have two legs in back and fore legs in front.
(3) This makes a total of six legs, which certainly is an odd number of
legs for a horse.
(4) But the only number that is both odd and even is infinity.
(5) Therefore, horses must have an infinite number of legs.

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

27

On 11/26/01, “PeterB ” wrote:

> Personally, I have a hard time understanding why the IFS Kit is expensive,
> and I’m somewhat pissed off that the DDK is no longer downloadable; a
> “free” CD is not a good alternative, IMO. That they should wish to
> restrict people from writing windows software boggles the mind, to be
> honest.
>

Ok, I have to post something. Most folks have argued that the price of the
IFS kit serves as a barrier to “amateurs” who are just fooling around and
shouldn’t be allowed to develop filesystems for NT.

I have another argument in favor of the fees: It costs Microsoft real money
to develop the kit, and especially to write the documentation, to put on
plugfests that benefit us all, and sometimes to actually answer questions.
Why shouldn’t they charge us for it, to recover some of that expense?

Carl Appellof
VERITAS Software Corporation

—
You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

28

On Thu, 29 Nov 2001 xxxxx@veritas.com wrote:

On 11/26/01, “PeterB ” wrote:
> I have another argument in favor of the fees: It costs Microsoft real money
> to develop the kit,
Ditto the Platform SDK, but they give that away for free.
Ditto the DDK, but they give that away for “free”.

> and especially to write the documentation,
Given the abysmal level of documentation, I think this is a bit of a
stretch. They fairly obviously /don’t/ put a lot of effort into writing
it.

> to put on
> plugfests that benefit us all, and sometimes to actually answer questions.
> Why shouldn’t they charge us for it, to recover some of that expense?
'cos they make even more money off people making Windows a useful
platform that does everything they want it to do.

–
Peter xxxxx@inkvine.fluff.org
http://www.inkvine.fluff.org/~peter/

logic kicks ass:
(1) Horses have an even number of legs.
(2) They have two legs in back and fore legs in front.
(3) This makes a total of six legs, which certainly is an odd number of
legs for a horse.
(4) But the only number that is both odd and even is infinity.
(5) Therefore, horses must have an infinite number of legs.

—
You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

29

So, why do they charge $1000 for kit for file system only, and virtually
charge nothing for the DDK, or WinDBG, or MeDbg etc.?
I had to say something, too:-)
If they are worried about the code for FAT/CDFS, well, keep the code, just
give us the documentation and the header files, for free, that is - there are
many online samples.
What’s worse is that until recently, I couldn’t even order the IFS Kit! (Due
to export blah-blah:-)

Regards, Dejan.

xxxxx@veritas.com wrote:

On 11/26/01, “PeterB ” wrote:
>
> > Personally, I have a hard time understanding why the IFS Kit is expensive,
> > and I’m somewhat pissed off that the DDK is no longer downloadable; a
> > “free” CD is not a good alternative, IMO. That they should wish to
> > restrict people from writing windows software boggles the mind, to be
> > honest.
> >
>
> Ok, I have to post something. Most folks have argued that the price of the
> IFS kit serves as a barrier to “amateurs” who are just fooling around and
> shouldn’t be allowed to develop filesystems for NT.
>
> I have another argument in favor of the fees: It costs Microsoft real money
> to develop the kit, and especially to write the documentation, to put on
> plugfests that benefit us all, and sometimes to actually answer questions.
> Why shouldn’t they charge us for it, to recover some of that expense?
>
> Carl Appellof
> VERITAS Software Corporation
>
> —
> You are currently subscribed to ntfsd as: xxxxx@alfasp.com
> To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

–
Kind regards, Dejan M. CEO Alfa Co. www.alfasp.com
E-mail: xxxxx@alfasp.com
ICQ#: 56570367
Alfa File Monitor - File monitoring system for Win32 developers.
Alfa File Protector - File protection and hiding system for Win32 developers.

—
You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

30

> I have another argument in favor of the fees: It costs Microsoft real money

to develop the kit

Really?
Pick the FASTFAT/CDFS/RDBSS from the real world NT source and distribute it?
Costed money?

Max

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

31

> ----------

From: xxxxx@storagecraft.com[SMTP:xxxxx@storagecraft.com]
Reply To: xxxxx@lists.osr.com
Sent: Saturday, December 01, 2001 3:23 AM
To: xxxxx@lists.osr.com
Subject: [ntfsd] RE: Converting NT Driver to XP

> I have another argument in favor of the fees: It costs Microsoft real
money
> to develop the kit

Really?
Pick the FASTFAT/CDFS/RDBSS from the real world NT source and distribute
it?
Costed money?

Don’t forget they had to remove all secret stuff from ntifs.h :wink:

Best regards,

Michal Vodicka
STMicroelectronics Design and Application s.r.o.
[michal.vodicka@st.com, http:://www.st.com]

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

32

Well it does cost some money. A person has to package and make available
sources, the appropriate include files, and a build environment. It
might not seem like much, but it probably does cost Microsoft more to
make the IFS available than they get back in revenue from the 75 or so
people who actually need the thing.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Maxim
S. Shatskih
Sent: Friday, November 30, 2001 9:24 PM
To: File Systems Developers
Subject: [ntfsd] RE: Converting NT Driver to XP

> I have another argument in favor of the fees: It costs
Microsoft real
> money
> to develop the kit

Really?
Pick the FASTFAT/CDFS/RDBSS from the real world NT source and
distribute it? Costed money?

Max

You are currently subscribed to ntfsd as:
xxxxx@hollistech.com To unsubscribe send a blank email to
leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

33

> > Really?

> Pick the FASTFAT/CDFS/RDBSS from the real world NT source and distribute
> it?
> Costed money?
>
Don’t forget they had to remove all secret stuff from ntifs.h :wink:

I think they assemble the header from lots of private subsystem headers using some tool.
These “begin_ntddk” comments.

Max

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com

34

> ----------

From: xxxxx@storagecraft.com[SMTP:xxxxx@storagecraft.com]
Reply To: xxxxx@lists.osr.com
Sent: Saturday, December 01, 2001 8:43 PM
To: xxxxx@lists.osr.com
Subject: [ntfsd] RE: Converting NT Driver to XP

> > Really?
> > Pick the FASTFAT/CDFS/RDBSS from the real world NT source and
distribute
> > it?
> > Costed money?
> >
> Don’t forget they had to remove all secret stuff from ntifs.h :wink:

I think they assemble the header from lots of private subsystem headers
using some tool.
These “begin_ntddk” comments.

Probably yes but still had to select what to include and what to remove.

Best regards,

Michal Vodicka
STMicroelectronics Design and Application s.r.o.
[michal.vodicka@st.com, http:://www.st.com]

You are currently subscribed to ntfsd as: $subst(‘Recip.EmailAddr’)
To unsubscribe send a blank email to leave-ntfsd-$subst(‘Recip.MemberIDChar’)@lists.osr.com