Just curious,
If I attempt to read a file using straight asm opposed to
ZwCreate/ZwRead/ZwClose in a filter, would that still result in a re-entry?
Since it is a driver and hardware can be written too directly, it is
possible to go around the IO manager and prevent a second IRP right?
Just curious,
where do you ppl get all those ideeas ? Is there an agency teaching this
kind of shit ?
Dan
----- Original Message -----
From: “Matt Martin”
To: “Windows File Systems Devs Interest List”
Sent: Wednesday, September 21, 2005 2:58 PM
Subject: [ntfsd] eliminate re-entry with asm?
> Just curious,
>
> If I attempt to read a file using straight asm opposed to
> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
> re-entry? Since it is a driver and hardware can be written too directly,
> it is possible to go around the IO manager and prevent a second IRP right?
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@rdsor.ro
> To unsubscribe send a blank email to xxxxx@lists.osr.com
> If I attempt to read a file using straight asm opposed to
ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
re-entry?
You mean with Assembler ?
If yes, that there are two rules related to using
assembler in drivers:
L.
> where do you ppl get all those ideeas ? Is there an agency teaching this
kind of shit ?
:-)))))))
L.
The answer to your question is, if I understand correctly that your
considering using no API functions and writing directly to the hardware,
yes as far as reentrancy is concerned.
That being said, you are, bluntly, you are out of your mind if you
attempt this. There is basically no chance of ever getting this to
work. There is no need to get in to details, such as basically assured
coruption, why.
I have a pet peve concerning people responding to questions on this
list with nothing more than criticism and, frequently, party line advice
that is most definitely not true in all cases, and, moreover, advice of
which they have no way of knowing the applicability in one’s specific
case, particularly based on a three line description. Don’t get me
wrong; advice such as “no assembler is drivers” is not only a good idea,
use of assembler in a driver these days is overwhelmingly unnecessary.
My question, however, is exactly what does mentioning things such as
this several time daily add to the discussion. In this particular case,
my apologies to author of the question, assembler is so far down on the
list of problems manifest in this idea, and, again, apologies, there is
basically zero chance of this idea ever working, that it really adds
nothing, and certainly does not merit its own, dedicated e-mail.
Finally, not only do real reasons exist for assembler’s use, and
sometimes these reasons may not legally be disclosed, but the world will
simply not going to end if assembler is used.
The problem, as I see it, is that this list, although very useful,
takes a lot of time to get through, mostly due to, in my opinion, people
going on and on about procedure, policy and the proverbial right thing
to do, suspect questions notwithstanding.
For example, what I am doing here.
My opinion is that a nice rule of thumb might be that, if you’re not
going to address the person’s question, simply do not respond. This, of
course, does not rule out adding anything else to the discussion.
MM
>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
Just curious,
If I attempt to read a file using straight asm opposed to
ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
re-entry?
Since it is a driver and hardware can be written too directly, it is
possible to go around the IO manager and prevent a second IRP right?
Questions? First check the IFS FAQ at
https://www.osronline.com/article.cfm?id=17
You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
To unsubscribe send a blank email to xxxxx@lists.osr.com
Dan,
Why be a dick about it? I teach myself, I learn threw trial and error. It
was just a question…
You might consider it dumb question, but hell, I’m learning. At least I
haven’t asked the same damn question 5 times this week about how to call w32
api in the kernel…
Ladislav, yes, I meant assembler. I interpret your answer (rule) to be that
you think if would be a debugging nightmare. At minimum, could the IO
manager be by passed to read data from the disk?
----- Original Message -----
From: “Dan Partelly”
To: “Windows File Systems Devs Interest List”
Sent: Wednesday, September 21, 2005 7:03 AM
Subject: Re: [ntfsd] eliminate re-entry with asm?
>
> Just curious,
>
> where do you ppl get all those ideeas ? Is there an agency teaching this
> kind of shit ?
>
> Dan
>
> ----- Original Message -----
> From: “Matt Martin”
> To: “Windows File Systems Devs Interest List”
> Sent: Wednesday, September 21, 2005 2:58 PM
> Subject: [ntfsd] eliminate re-entry with asm?
>
>
>> Just curious,
>>
>> If I attempt to read a file using straight asm opposed to
>> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
>> re-entry? Since it is a driver and hardware can be written too directly,
>> it is possible to go around the IO manager and prevent a second IRP
>> right?
>>
>> —
>> Questions? First check the IFS FAQ at
>> https://www.osronline.com/article.cfm?id=17
>>
>> You are currently subscribed to ntfsd as: xxxxx@rdsor.ro
>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@comcast.net
> To unsubscribe send a blank email to xxxxx@lists.osr.com
Martin,
I will take a different tack here. It is also the responsibility of
the people asking questions to attempt to ask informed questions. Now a
“How to get started with file system drivers?” can be considered a informed
question, since it is general and asking for help. But to look at the OP’s
question, before asking about assembler, well there is a heck of a lot
written in this forum and others about asm, a quick search might have caused
the question to either not be asked, or asked in a way that does not draw
the response you objected to.
Most of us who give data on these forum, do so on our own time, and
with the knowledge that everyone we help is potentially a competitor for the
next job or consult we try to get. It is reasonable to ask the people
comming to this forum to be prepared to ask intelligent questions and to
search the forums and the FAQ before asking. I have taught classes over the
years on various subjects, where if someone did do the basics, it was
reasonable to ask them to leave. There the teacher was getting paid,
shouldn’t the same rules apply when you do it for free.
–
Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting
Remove StopSpam from the email to reply
“Martin O’Brien” wrote in message
news:xxxxx@ntfsd…
> The answer to your question is, if I understand correctly that your
> considering using no API functions and writing directly to the hardware,
> yes as far as reentrancy is concerned.
>
> That being said, you are, bluntly, you are out of your mind if you
> attempt this. There is basically no chance of ever getting this to
> work. There is no need to get in to details, such as basically assured
> coruption, why.
>
> I have a pet peve concerning people responding to questions on this
> list with nothing more than criticism and, frequently, party line advice
> that is most definitely not true in all cases, and, moreover, advice of
> which they have no way of knowing the applicability in one’s specific
> case, particularly based on a three line description. Don’t get me
> wrong; advice such as “no assembler is drivers” is not only a good idea,
> use of assembler in a driver these days is overwhelmingly unnecessary.
> My question, however, is exactly what does mentioning things such as
> this several time daily add to the discussion. In this particular case,
> my apologies to author of the question, assembler is so far down on the
> list of problems manifest in this idea, and, again, apologies, there is
> basically zero chance of this idea ever working, that it really adds
> nothing, and certainly does not merit its own, dedicated e-mail.
> Finally, not only do real reasons exist for assembler’s use, and
> sometimes these reasons may not legally be disclosed, but the world will
> simply not going to end if assembler is used.
>
> The problem, as I see it, is that this list, although very useful,
> takes a lot of time to get through, mostly due to, in my opinion, people
> going on and on about procedure, policy and the proverbial right thing
> to do, suspect questions notwithstanding.
>
> For example, what I am doing here.
>
> My opinion is that a nice rule of thumb might be that, if you’re not
> going to address the person’s question, simply do not respond. This, of
> course, does not rule out adding anything else to the discussion.
>
> MM
>
>
>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
> Just curious,
>
> If I attempt to read a file using straight asm opposed to
> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
> re-entry?
> Since it is a driver and hardware can be written too directly, it is
> possible to go around the IO manager and prevent a second IRP right?
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
Sorry for the typo, the second to last sentence should obviously read “where
if someone did not do the basics”
–
Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting
Remove StopSpam from the email to reply
“Don Burn” wrote in message news:xxxxx@ntfsd…
> Martin,
>
> I will take a different tack here. It is also the responsibility of
> the people asking questions to attempt to ask informed questions. Now a
> “How to get started with file system drivers?” can be considered a
> informed question, since it is general and asking for help. But to look
> at the OP’s question, before asking about assembler, well there is a heck
> of a lot written in this forum and others about asm, a quick search might
> have caused the question to either not be asked, or asked in a way that
> does not draw the response you objected to.
>
> Most of us who give data on these forum, do so on our own time, and
> with the knowledge that everyone we help is potentially a competitor for
> the next job or consult we try to get. It is reasonable to ask the people
> comming to this forum to be prepared to ask intelligent questions and to
> search the forums and the FAQ before asking. I have taught classes over
> the years on various subjects, where if someone did do the basics, it was
> reasonable to ask them to leave. There the teacher was getting paid,
> shouldn’t the same rules apply when you do it for free.
>
>
> –
> Don Burn (MVP, Windows DDK)
> Windows 2k/XP/2k3 Filesystem and Driver Consulting
> Remove StopSpam from the email to reply
>
>
>
>
>
> “Martin O’Brien” wrote in message
> news:xxxxx@ntfsd…
>> The answer to your question is, if I understand correctly that your
>> considering using no API functions and writing directly to the hardware,
>> yes as far as reentrancy is concerned.
>>
>> That being said, you are, bluntly, you are out of your mind if you
>> attempt this. There is basically no chance of ever getting this to
>> work. There is no need to get in to details, such as basically assured
>> coruption, why.
>>
>> I have a pet peve concerning people responding to questions on this
>> list with nothing more than criticism and, frequently, party line advice
>> that is most definitely not true in all cases, and, moreover, advice of
>> which they have no way of knowing the applicability in one’s specific
>> case, particularly based on a three line description. Don’t get me
>> wrong; advice such as “no assembler is drivers” is not only a good idea,
>> use of assembler in a driver these days is overwhelmingly unnecessary.
>> My question, however, is exactly what does mentioning things such as
>> this several time daily add to the discussion. In this particular case,
>> my apologies to author of the question, assembler is so far down on the
>> list of problems manifest in this idea, and, again, apologies, there is
>> basically zero chance of this idea ever working, that it really adds
>> nothing, and certainly does not merit its own, dedicated e-mail.
>> Finally, not only do real reasons exist for assembler’s use, and
>> sometimes these reasons may not legally be disclosed, but the world will
>> simply not going to end if assembler is used.
>>
>> The problem, as I see it, is that this list, although very useful,
>> takes a lot of time to get through, mostly due to, in my opinion, people
>> going on and on about procedure, policy and the proverbial right thing
>> to do, suspect questions notwithstanding.
>>
>> For example, what I am doing here.
>>
>> My opinion is that a nice rule of thumb might be that, if you’re not
>> going to address the person’s question, simply do not respond. This, of
>> course, does not rule out adding anything else to the discussion.
>>
>> MM
>>
>>
>>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
>> Just curious,
>>
>> If I attempt to read a file using straight asm opposed to
>> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
>> re-entry?
>> Since it is a driver and hardware can be written too directly, it is
>> possible to go around the IO manager and prevent a second IRP right?
>>
>>
>> —
>> Questions? First check the IFS FAQ at
>> https://www.osronline.com/article.cfm?id=17
>>
>> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>>
>
>
>
Hello all
Am going thru the discussions taking place today and would just give two
comment.
Martin,
You stated, “you are out of your mind if you attempt this”. As I stated in a
previous reply, I learn threw trial and error. I’m attempting to learn
something new here. I’m not working in any sort of production environment,
so therefor, I will try this. And I’m willing to take the time to understand
the mistakes in this approach (I’m hard headed).
Don,
there isn’t a damn thing I could find in the archive regarding what I asked.
I wasn’t looking for your approval, simply whether or not it was plausible.
And further more, where do you get off on this whole potential competitor
crap, if my question is so lame, you should know that I can’t compete with
the majority of folks here in this news group. Why are you trying to spin it
now???
----- Original Message -----
From: “Martin O’Brien”
To: “Windows File Systems Devs Interest List”
Sent: Wednesday, September 21, 2005 7:58 AM
Subject: Re: [ntfsd] eliminate re-entry with asm?
> The answer to your question is, if I understand correctly that your
> considering using no API functions and writing directly to the hardware,
> yes as far as reentrancy is concerned.
>
> That being said, you are, bluntly, you are out of your mind if you
> attempt this. There is basically no chance of ever getting this to
> work. There is no need to get in to details, such as basically assured
> coruption, why.
>
> I have a pet peve concerning people responding to questions on this
> list with nothing more than criticism and, frequently, party line advice
> that is most definitely not true in all cases, and, moreover, advice of
> which they have no way of knowing the applicability in one’s specific
> case, particularly based on a three line description. Don’t get me
> wrong; advice such as “no assembler is drivers” is not only a good idea,
> use of assembler in a driver these days is overwhelmingly unnecessary.
> My question, however, is exactly what does mentioning things such as
> this several time daily add to the discussion. In this particular case,
> my apologies to author of the question, assembler is so far down on the
> list of problems manifest in this idea, and, again, apologies, there is
> basically zero chance of this idea ever working, that it really adds
> nothing, and certainly does not merit its own, dedicated e-mail.
> Finally, not only do real reasons exist for assembler’s use, and
> sometimes these reasons may not legally be disclosed, but the world will
> simply not going to end if assembler is used.
>
> The problem, as I see it, is that this list, although very useful,
> takes a lot of time to get through, mostly due to, in my opinion, people
> going on and on about procedure, policy and the proverbial right thing
> to do, suspect questions notwithstanding.
>
> For example, what I am doing here.
>
> My opinion is that a nice rule of thumb might be that, if you’re not
> going to address the person’s question, simply do not respond. This, of
> course, does not rule out adding anything else to the discussion.
>
> MM
>
>
>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
> Just curious,
>
> If I attempt to read a file using straight asm opposed to
> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
> re-entry?
> Since it is a driver and hardware can be written too directly, it is
> possible to go around the IO manager and prevent a second IRP right?
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@comcast.net
> To unsubscribe send a blank email to xxxxx@lists.osr.com
Martin,
There is a heck of a lot on why assembler is bad. Sorry, if you cannot
find this then you are not looking very hard. As far as the question:
“Is there a programitic way to avoid the re-entry?” The answer is no,
you would have to create a file system that bypasses all of OS beneath it
and lock the world (with the inevitable lack of performance) to stop this
type of behavior. So you need to learn to live with the re-entry.
–
Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting
Remove StopSpam from the email to reply
“Matt Martin” wrote in message news:xxxxx@ntfsd…
> Martin,
>
> You stated, “you are out of your mind if you attempt this”. As I stated in
> a previous reply, I learn threw trial and error. I’m attempting to learn
> something new here. I’m not working in any sort of production environment,
> so therefor, I will try this. And I’m willing to take the time to
> understand the mistakes in this approach (I’m hard headed).
>
> Don,
> there isn’t a damn thing I could find in the archive regarding what I
> asked. I wasn’t looking for your approval, simply whether or not it was
> plausible. And further more, where do you get off on this whole potential
> competitor crap, if my question is so lame, you should know that I can’t
> compete with the majority of folks here in this news group. Why are you
> trying to spin it now???
>
>
>
>
>
>
>
> ----- Original Message -----
> From: “Martin O’Brien”
> To: “Windows File Systems Devs Interest List”
> Sent: Wednesday, September 21, 2005 7:58 AM
> Subject: Re: [ntfsd] eliminate re-entry with asm?
>
>
>> The answer to your question is, if I understand correctly that your
>> considering using no API functions and writing directly to the hardware,
>> yes as far as reentrancy is concerned.
>>
>> That being said, you are, bluntly, you are out of your mind if you
>> attempt this. There is basically no chance of ever getting this to
>> work. There is no need to get in to details, such as basically assured
>> coruption, why.
>>
>> I have a pet peve concerning people responding to questions on this
>> list with nothing more than criticism and, frequently, party line advice
>> that is most definitely not true in all cases, and, moreover, advice of
>> which they have no way of knowing the applicability in one’s specific
>> case, particularly based on a three line description. Don’t get me
>> wrong; advice such as “no assembler is drivers” is not only a good idea,
>> use of assembler in a driver these days is overwhelmingly unnecessary.
>> My question, however, is exactly what does mentioning things such as
>> this several time daily add to the discussion. In this particular case,
>> my apologies to author of the question, assembler is so far down on the
>> list of problems manifest in this idea, and, again, apologies, there is
>> basically zero chance of this idea ever working, that it really adds
>> nothing, and certainly does not merit its own, dedicated e-mail.
>> Finally, not only do real reasons exist for assembler’s use, and
>> sometimes these reasons may not legally be disclosed, but the world will
>> simply not going to end if assembler is used.
>>
>> The problem, as I see it, is that this list, although very useful,
>> takes a lot of time to get through, mostly due to, in my opinion, people
>> going on and on about procedure, policy and the proverbial right thing
>> to do, suspect questions notwithstanding.
>>
>> For example, what I am doing here.
>>
>> My opinion is that a nice rule of thumb might be that, if you’re not
>> going to address the person’s question, simply do not respond. This, of
>> course, does not rule out adding anything else to the discussion.
>>
>> MM
>>
>>
>>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
>> Just curious,
>>
>> If I attempt to read a file using straight asm opposed to
>> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
>> re-entry?
>> Since it is a driver and hardware can be written too directly, it is
>> possible to go around the IO manager and prevent a second IRP right?
>>
>>
>> —
>> Questions? First check the IFS FAQ at
>> https://www.osronline.com/article.cfm?id=17
>>
>> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>>
>> —
>> Questions? First check the IFS FAQ at
>> https://www.osronline.com/article.cfm?id=17
>>
>> You are currently subscribed to ntfsd as: xxxxx@comcast.net
>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
DON:
Don, I hear what your saying, and I agree with you that it is the
responsibility of the requestor to be at least somewhat informed. But I
would ask the question who benefits from nothing but criticism? I mean,
take a look at how many questions that are either totally perposterous
(like the one which led us to this discussion), or have absolutely no
relation to any of these lists (like the Win32 API issue of which the
author wrote in his response to Dan); furthermore, how many times are
they asked? That is, personally, I would find no way to maintain that
the practice of criticing discourages these tpyes of questions.
Moreover, for any one of these questions, count up the number of
responses to it. It’s not even close. All of this misses the
fundamental problem, which is that there is no way prevent these
questions, because, unless someone is out there intentionally wasting
people’s time, he or she does not know that the question is
non-applicable, foolish, et. c. Insisting that people people have
minimum knowledge is a complete waste of time, and, as is telling people
to search the list first after they have already asked a bad question.
It is, by definition, unavoidably, impossible. The best and only thing
we can do is cut down on the volume. Fundamentally, responding 50 times
to 1 stupid question that we can not prevent some from asking, is sort
of cutting of your nose to spite your face type of deal, and every bit
as foolish as the question itself, and really pretty disingenious.
Just say no to responding.
MM
>> xxxxx@acm.org 09/21/05 9:13 AM >>>
Martin,
I will take a different tack here. It is also the responsibility
of
the people asking questions to attempt to ask informed questions. Now
a
“How to get started with file system drivers?” can be considered a
informed
question, since it is general and asking for help. But to look at the
OP’s
question, before asking about assembler, well there is a heck of a lot
written in this forum and others about asm, a quick search might have
caused
the question to either not be asked, or asked in a way that does not
draw
the response you objected to.
Most of us who give data on these forum, do so on our own time,
and
with the knowledge that everyone we help is potentially a competitor
for the
next job or consult we try to get. It is reasonable to ask the people
comming to this forum to be prepared to ask intelligent questions and
to
search the forums and the FAQ before asking. I have taught classes
over the
years on various subjects, where if someone did do the basics, it was
reasonable to ask them to leave. There the teacher was getting paid,
shouldn’t the same rules apply when you do it for free.
–
Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting
Remove StopSpam from the email to reply
“Martin O’Brien” wrote in message
news:xxxxx@ntfsd…
> The answer to your question is, if I understand correctly that your
> considering using no API functions and writing directly to the
hardware,
> yes as far as reentrancy is concerned.
>
> That being said, you are, bluntly, you are out of your mind if you
> attempt this. There is basically no chance of ever getting this to
> work. There is no need to get in to details, such as basically
assured
> coruption, why.
>
> I have a pet peve concerning people responding to questions on this
> list with nothing more than criticism and, frequently, party line
advice
> that is most definitely not true in all cases, and, moreover, advice
of
> which they have no way of knowing the applicability in one’s
specific
> case, particularly based on a three line description. Don’t get me
> wrong; advice such as “no assembler is drivers” is not only a good
idea,
> use of assembler in a driver these days is overwhelmingly
unnecessary.
> My question, however, is exactly what does mentioning things such as
> this several time daily add to the discussion. In this particular
case,
> my apologies to author of the question, assembler is so far down on
the
> list of problems manifest in this idea, and, again, apologies, there
is
> basically zero chance of this idea ever working, that it really adds
> nothing, and certainly does not merit its own, dedicated e-mail.
> Finally, not only do real reasons exist for assembler’s use, and
> sometimes these reasons may not legally be disclosed, but the world
will
> simply not going to end if assembler is used.
>
> The problem, as I see it, is that this list, although very useful,
> takes a lot of time to get through, mostly due to, in my opinion,
people
> going on and on about procedure, policy and the proverbial right
thing
> to do, suspect questions notwithstanding.
>
> For example, what I am doing here.
>
> My opinion is that a nice rule of thumb might be that, if you’re
not
> going to address the person’s question, simply do not respond. This,
of
> course, does not rule out adding anything else to the discussion.
>
> MM
>
>
>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
> Just curious,
>
> If I attempt to read a file using straight asm opposed to
> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
> re-entry?
> Since it is a driver and hardware can be written too directly, it is
> possible to go around the IO manager and prevent a second IRP right?
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
—
Questions? First check the IFS FAQ at
https://www.osronline.com/article.cfm?id=17
You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
To unsubscribe send a blank email to xxxxx@lists.osr.com
Oh, if it was so easy. For instance on another forum recently people did
just that on a question. Did the OP, get a clue, NO! at the end of the day
he flooded multiple newsgroups with multiple copies of the same question,
and complaining no one was helping him.
By the way having seen postings on the open source groups, where Linus, Eric
Raymond, or RMS have torn into someone with language that I have never seen
on a Windows group, I think you are romanticizing the world of open source.
–
Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting
Remove StopSpam from the email to reply
“Arijit Bhattacharyya” wrote in message
news:xxxxx@ntfsd…
Hello all
Am going thru the discussions taking place today and would just give two
comment.
1. DONT ANSWER TO A QUESTION IF YOU THINK THAT THE QUESTION IS A STUPID ONE.
2. IF U DO NOT RECEIVE AN ANSWER TO A QUESTION, CONSIDER IT WAS A STUPID
ONE.
Well…at least this is what I have seen in Linux groups, and ppl like
Linus can be considered “busy” and he “takes out time” to answer questions.
I think, keeping these 2 as tacit rules can avoid all kinds of
confrontations. Another point that I want to put here is that many ppl
coming from Linux background think that just like in Linux, we can do
nething with the windows Kernel. Well, this is unfortunately not so. So if u
think that windows takes away the freedom as a programmer (just as i do),
shift to Linux (just as I would do in some time). And then we would have
much more technical answers to questions like " How can I use asm in kernel"
rather than the kind of answers given 2day.
regards
On 9/21/05, Don Burn wrote:
>
> Martin,
>
> I will take a different tack here. It is also the responsibility of
> the people asking questions to attempt to ask informed questions. Now a
> “How to get started with file system drivers?” can be considered a
> informed
> question, since it is general and asking for help. But to look at the OP’s
> question, before asking about assembler, well there is a heck of a lot
> written in this forum and others about asm, a quick search might have
> caused
> the question to either not be asked, or asked in a way that does not draw
> the response you objected to.
>
> Most of us who give data on these forum, do so on our own time, and
> with the knowledge that everyone we help is potentially a competitor for
> the
> next job or consult we try to get. It is reasonable to ask the people
> comming to this forum to be prepared to ask intelligent questions and to
> search the forums and the FAQ before asking. I have taught classes over
> the
> years on various subjects, where if someone did do the basics, it was
> reasonable to ask them to leave. There the teacher was getting paid,
> shouldn’t the same rules apply when you do it for free.
>
>
> –
> Don Burn (MVP, Windows DDK)
> Windows 2k/XP/2k3 Filesystem and Driver Consulting
> Remove StopSpam from the email to reply
>
>
>
>
>
> “Martin O’Brien” wrote in message
> news:xxxxx@ntfsd…
> > The answer to your question is, if I understand correctly that your
> > considering using no API functions and writing directly to the hardware,
> > yes as far as reentrancy is concerned.
> >
> > That being said, you are, bluntly, you are out of your mind if you
> > attempt this. There is basically no chance of ever getting this to
> > work. There is no need to get in to details, such as basically assured
> > coruption, why.
> >
> > I have a pet peve concerning people responding to questions on this
> > list with nothing more than criticism and, frequently, party line advice
> > that is most definitely not true in all cases, and, moreover, advice of
> > which they have no way of knowing the applicability in one’s specific
> > case, particularly based on a three line description. Don’t get me
> > wrong; advice such as “no assembler is drivers” is not only a good idea,
> > use of assembler in a driver these days is overwhelmingly unnecessary.
> > My question, however, is exactly what does mentioning things such as
> > this several time daily add to the discussion. In this particular case,
> > my apologies to author of the question, assembler is so far down on the
> > list of problems manifest in this idea, and, again, apologies, there is
> > basically zero chance of this idea ever working, that it really adds
> > nothing, and certainly does not merit its own, dedicated e-mail.
> > Finally, not only do real reasons exist for assembler’s use, and
> > sometimes these reasons may not legally be disclosed, but the world will
> > simply not going to end if assembler is used.
> >
> > The problem, as I see it, is that this list, although very useful,
> > takes a lot of time to get through, mostly due to, in my opinion, people
> > going on and on about procedure, policy and the proverbial right thing
> > to do, suspect questions notwithstanding.
> >
> > For example, what I am doing here.
> >
> > My opinion is that a nice rule of thumb might be that, if you’re not
> > going to address the person’s question, simply do not respond. This, of
> > course, does not rule out adding anything else to the discussion.
> >
> > MM
> >
> >
> >>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
> > Just curious,
> >
> > If I attempt to read a file using straight asm opposed to
> > ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
> > re-entry?
> > Since it is a driver and hardware can be written too directly, it is
> > possible to go around the IO manager and prevent a second IRP right?
> >
> >
> > —
> > Questions? First check the IFS FAQ at
> > https://www.osronline.com/article.cfm?id=17
> >
> > You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
> > To unsubscribe send a blank email to xxxxx@lists.osr.com
> >
>
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@gmail.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
> At minimum, could the IO manager be by passed to read
data from the disk?
Yes, it could. You may build a read request IRP
and send it directly to a disk device, as file systems
do. You will then read disk sectors that may be, depends
on the read offset, a file, a directory entry, or whatever else
that has been written to disk by file system.
However, if ZwCreate/ZwRead/ZwClose has been mentioned
in the original post, this means that you think about attempting
to implement something like your own ZwCreate using
direct disk access. No, don’t even think about this, it is file
system’s work
You must either stay in file system level and work with files and
directories, or at disk level and work with sectors. It is just not
possible to mix these two layers.
Using of asm has nothing to do with this.
L.
MATT:
You are absolutely correct, and I apologize. I intentionally
overstated this point of the case, to, hopefully prempt and endless
stream of e-mail from people going on and on about the dangers inherent
in this et. c. My bias, as is I think most people’s on this list, is
that this will eventually end up in production. If this is going to end
up in production, I would reiterate my warning. Perhaps I shouldn’t,
but I’m in no way trying to tell you what to do. I think, for me only,
I couldn’t honestly answer this question without some sort of warning,
and I was trying to keep it short (the answer, not the diatribe).
Best of luck.
MM
>> xxxxx@comcast.net 09/21/05 9:31 AM >>>
Martin,
You stated, “you are out of your mind if you attempt this”. As I stated
in a
previous reply, I learn threw trial and error. I’m attempting to learn
something new here. I’m not working in any sort of production
environment,
so therefor, I will try this. And I’m willing to take the time to
understand
the mistakes in this approach (I’m hard headed).
Don,
there isn’t a damn thing I could find in the archive regarding what I
asked.
I wasn’t looking for your approval, simply whether or not it was
plausible.
And further more, where do you get off on this whole potential
competitor
crap, if my question is so lame, you should know that I can’t compete
with
the majority of folks here in this news group. Why are you trying to
spin it
now???
----- Original Message -----
From: “Martin O’Brien”
To: “Windows File Systems Devs Interest List”
Sent: Wednesday, September 21, 2005 7:58 AM
Subject: Re: [ntfsd] eliminate re-entry with asm?
> The answer to your question is, if I understand correctly that your
> considering using no API functions and writing directly to the
hardware,
> yes as far as reentrancy is concerned.
>
> That being said, you are, bluntly, you are out of your mind if you
> attempt this. There is basically no chance of ever getting this to
> work. There is no need to get in to details, such as basically
assured
> coruption, why.
>
> I have a pet peve concerning people responding to questions on this
> list with nothing more than criticism and, frequently, party line
advice
> that is most definitely not true in all cases, and, moreover, advice
of
> which they have no way of knowing the applicability in one’s
specific
> case, particularly based on a three line description. Don’t get me
> wrong; advice such as “no assembler is drivers” is not only a good
idea,
> use of assembler in a driver these days is overwhelmingly
unnecessary.
> My question, however, is exactly what does mentioning things such as
> this several time daily add to the discussion. In this particular
case,
> my apologies to author of the question, assembler is so far down on
the
> list of problems manifest in this idea, and, again, apologies, there
is
> basically zero chance of this idea ever working, that it really adds
> nothing, and certainly does not merit its own, dedicated e-mail.
> Finally, not only do real reasons exist for assembler’s use, and
> sometimes these reasons may not legally be disclosed, but the world
will
> simply not going to end if assembler is used.
>
> The problem, as I see it, is that this list, although very useful,
> takes a lot of time to get through, mostly due to, in my opinion,
people
> going on and on about procedure, policy and the proverbial right
thing
> to do, suspect questions notwithstanding.
>
> For example, what I am doing here.
>
> My opinion is that a nice rule of thumb might be that, if you’re
not
> going to address the person’s question, simply do not respond. This,
of
> course, does not rule out adding anything else to the discussion.
>
> MM
>
>
>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
> Just curious,
>
> If I attempt to read a file using straight asm opposed to
> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
> re-entry?
> Since it is a driver and hardware can be written too directly, it is
> possible to go around the IO manager and prevent a second IRP right?
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@comcast.net
> To unsubscribe send a blank email to xxxxx@lists.osr.com
—
Questions? First check the IFS FAQ at
https://www.osronline.com/article.cfm?id=17
You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
To unsubscribe send a blank email to xxxxx@lists.osr.com
Dan,
That is all I was asking, thanks for the response finally. What made me
curious is I’ve seen bootloaders that were written in assembler, and those
could all navigate to the start of the disk cyl 0, sect 1 and read. I was
thinking that it “might” be possible to get the physical location of a file
in the filter and then modify some bootloader code to read specific sectors
(or something like that) -thus, eliminating the re-entry.
I think I’ve seen products out there that could link assembler code into C
(thus eliminating C assembler limitations). However, I don’t know if those
products would work with a driver.
was just curious.
----- Original Message -----
From: “Don Burn”
Newsgroups: ntfsd
To: “Windows File Systems Devs Interest List”
Sent: Wednesday, September 21, 2005 8:37 AM
Subject: Re:[ntfsd] eliminate re-entry with asm?
> Martin,
>
> There is a heck of a lot on why assembler is bad. Sorry, if you
> cannot find this then you are not looking very hard. As far as the
> question:
>
> “Is there a programitic way to avoid the re-entry?” The answer is
> no, you would have to create a file system that bypasses all of OS beneath
> it and lock the world (with the inevitable lack of performance) to stop
> this type of behavior. So you need to learn to live with the re-entry.
>
>
> –
> Don Burn (MVP, Windows DDK)
> Windows 2k/XP/2k3 Filesystem and Driver Consulting
> Remove StopSpam from the email to reply
>
>
>
> “Matt Martin” wrote in message
> news:xxxxx@ntfsd…
>> Martin,
>>
>> You stated, “you are out of your mind if you attempt this”. As I stated
>> in a previous reply, I learn threw trial and error. I’m attempting to
>> learn something new here. I’m not working in any sort of production
>> environment, so therefor, I will try this. And I’m willing to take the
>> time to understand the mistakes in this approach (I’m hard headed).
>>
>> Don,
>> there isn’t a damn thing I could find in the archive regarding what I
>> asked. I wasn’t looking for your approval, simply whether or not it was
>> plausible. And further more, where do you get off on this whole potential
>> competitor crap, if my question is so lame, you should know that I can’t
>> compete with the majority of folks here in this news group. Why are you
>> trying to spin it now???
>>
>>
>>
>>
>>
>>
>>
>> ----- Original Message -----
>> From: “Martin O’Brien”
>> To: “Windows File Systems Devs Interest List”
>> Sent: Wednesday, September 21, 2005 7:58 AM
>> Subject: Re: [ntfsd] eliminate re-entry with asm?
>>
>>
>>> The answer to your question is, if I understand correctly that your
>>> considering using no API functions and writing directly to the hardware,
>>> yes as far as reentrancy is concerned.
>>>
>>> That being said, you are, bluntly, you are out of your mind if you
>>> attempt this. There is basically no chance of ever getting this to
>>> work. There is no need to get in to details, such as basically assured
>>> coruption, why.
>>>
>>> I have a pet peve concerning people responding to questions on this
>>> list with nothing more than criticism and, frequently, party line advice
>>> that is most definitely not true in all cases, and, moreover, advice of
>>> which they have no way of knowing the applicability in one’s specific
>>> case, particularly based on a three line description. Don’t get me
>>> wrong; advice such as “no assembler is drivers” is not only a good idea,
>>> use of assembler in a driver these days is overwhelmingly unnecessary.
>>> My question, however, is exactly what does mentioning things such as
>>> this several time daily add to the discussion. In this particular case,
>>> my apologies to author of the question, assembler is so far down on the
>>> list of problems manifest in this idea, and, again, apologies, there is
>>> basically zero chance of this idea ever working, that it really adds
>>> nothing, and certainly does not merit its own, dedicated e-mail.
>>> Finally, not only do real reasons exist for assembler’s use, and
>>> sometimes these reasons may not legally be disclosed, but the world will
>>> simply not going to end if assembler is used.
>>>
>>> The problem, as I see it, is that this list, although very useful,
>>> takes a lot of time to get through, mostly due to, in my opinion, people
>>> going on and on about procedure, policy and the proverbial right thing
>>> to do, suspect questions notwithstanding.
>>>
>>> For example, what I am doing here.
>>>
>>> My opinion is that a nice rule of thumb might be that, if you’re not
>>> going to address the person’s question, simply do not respond. This, of
>>> course, does not rule out adding anything else to the discussion.
>>>
>>> MM
>>>
>>>
>>>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
>>> Just curious,
>>>
>>> If I attempt to read a file using straight asm opposed to
>>> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
>>> re-entry?
>>> Since it is a driver and hardware can be written too directly, it is
>>> possible to go around the IO manager and prevent a second IRP right?
>>>
>>>
>>> —
>>> Questions? First check the IFS FAQ at
>>> https://www.osronline.com/article.cfm?id=17
>>>
>>> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
>>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>>>
>>> —
>>> Questions? First check the IFS FAQ at
>>> https://www.osronline.com/article.cfm?id=17
>>>
>>> You are currently subscribed to ntfsd as: xxxxx@comcast.net
>>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>>
>>
>
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@comcast.net
> To unsubscribe send a blank email to xxxxx@lists.osr.com
DON:
I have no idea of what you are talking about with regards to assembler.
I never said I couldn’t find any information; moreover, I never said I
was looking for it. I either I did not make myself clear, or you missed
my point. I agree both that it is a foolish question to which to expect
an answer to in a forum such as this, and that there is no reasonable
way to avoid reentrancy. The answer, however, as you point out, is that
yes, it is possible in a totally useless, basically meaningless way,
which is the reason I explicitly stated that there was no need to
explain why. Ignoring that, if had not wished to respond to the other
issues, I never would have responded to the technical aspects of this
question, when some is asking about writing directly to hardware, he
pretty clear is going to be bypassing the filesystem.
Yes, it is a terrible idea, and one which I would not, and did not (i.
e. - “you are absoultely out of your mind”) endorse. But the answer is
yes, as I see it.
>> xxxxx@acm.org 09/21/05 9:37 AM >>>
Martin,
There is a heck of a lot on why assembler is bad. Sorry, if you
cannot
find this then you are not looking very hard. As far as the question:
“Is there a programitic way to avoid the re-entry?” The answer
is no,
you would have to create a file system that bypasses all of OS beneath
it
and lock the world (with the inevitable lack of performance) to stop
this
type of behavior. So you need to learn to live with the re-entry.
–
Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting
Remove StopSpam from the email to reply
“Matt Martin” wrote in message
news:xxxxx@ntfsd…
> Martin,
>
> You stated, “you are out of your mind if you attempt this”. As I
stated in
> a previous reply, I learn threw trial and error. I’m attempting to
learn
> something new here. I’m not working in any sort of production
environment,
> so therefor, I will try this. And I’m willing to take the time to
> understand the mistakes in this approach (I’m hard headed).
>
> Don,
> there isn’t a damn thing I could find in the archive regarding what I
> asked. I wasn’t looking for your approval, simply whether or not it
was
> plausible. And further more, where do you get off on this whole
potential
> competitor crap, if my question is so lame, you should know that I
can’t
> compete with the majority of folks here in this news group. Why are
you
> trying to spin it now???
>
>
>
>
>
>
>
> ----- Original Message -----
> From: “Martin O’Brien”
> To: “Windows File Systems Devs Interest List”
> Sent: Wednesday, September 21, 2005 7:58 AM
> Subject: Re: [ntfsd] eliminate re-entry with asm?
>
>
>> The answer to your question is, if I understand correctly that your
>> considering using no API functions and writing directly to the
hardware,
>> yes as far as reentrancy is concerned.
>>
>> That being said, you are, bluntly, you are out of your mind if you
>> attempt this. There is basically no chance of ever getting this to
>> work. There is no need to get in to details, such as basically
assured
>> coruption, why.
>>
>> I have a pet peve concerning people responding to questions on this
>> list with nothing more than criticism and, frequently, party line
advice
>> that is most definitely not true in all cases, and, moreover, advice
of
>> which they have no way of knowing the applicability in one’s
specific
>> case, particularly based on a three line description. Don’t get me
>> wrong; advice such as “no assembler is drivers” is not only a good
idea,
>> use of assembler in a driver these days is overwhelmingly
unnecessary.
>> My question, however, is exactly what does mentioning things such
as
>> this several time daily add to the discussion. In this particular
case,
>> my apologies to author of the question, assembler is so far down on
the
>> list of problems manifest in this idea, and, again, apologies, there
is
>> basically zero chance of this idea ever working, that it really
adds
>> nothing, and certainly does not merit its own, dedicated e-mail.
>> Finally, not only do real reasons exist for assembler’s use, and
>> sometimes these reasons may not legally be disclosed, but the world
will
>> simply not going to end if assembler is used.
>>
>> The problem, as I see it, is that this list, although very useful,
>> takes a lot of time to get through, mostly due to, in my opinion,
people
>> going on and on about procedure, policy and the proverbial right
thing
>> to do, suspect questions notwithstanding.
>>
>> For example, what I am doing here.
>>
>> My opinion is that a nice rule of thumb might be that, if you’re
not
>> going to address the person’s question, simply do not respond.
This, of
>> course, does not rule out adding anything else to the discussion.
>>
>> MM
>>
>>
>>>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
>> Just curious,
>>
>> If I attempt to read a file using straight asm opposed to
>> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
>> re-entry?
>> Since it is a driver and hardware can be written too directly, it
is
>> possible to go around the IO manager and prevent a second IRP
right?
>>
>>
>> —
>> Questions? First check the IFS FAQ at
>> https://www.osronline.com/article.cfm?id=17
>>
>> You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
>> To unsubscribe send a blank email to
xxxxx@lists.osr.com
>>
>> —
>> Questions? First check the IFS FAQ at
>> https://www.osronline.com/article.cfm?id=17
>>
>> You are currently subscribed to ntfsd as: xxxxx@comcast.net
>> To unsubscribe send a blank email to
xxxxx@lists.osr.com
>
>
—
Questions? First check the IFS FAQ at
https://www.osronline.com/article.cfm?id=17
You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
To unsubscribe send a blank email to xxxxx@lists.osr.com
DON:
On this, you and I agree.
>> xxxxx@acm.org 09/21/05 9:42 AM >>>
Oh, if it was so easy. For instance on another forum recently people
did
just that on a question. Did the OP, get a clue, NO! at the end of
the day
he flooded multiple newsgroups with multiple copies of the same
question,
and complaining no one was helping him.
By the way having seen postings on the open source groups, where Linus,
Eric
Raymond, or RMS have torn into someone with language that I have never
seen
on a Windows group, I think you are romanticizing the world of open
source.
–
Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting
Remove StopSpam from the email to reply
“Arijit Bhattacharyya” wrote in message
news:xxxxx@ntfsd…
Hello all
Am going thru the discussions taking place today and would just give
two
comment.
1. DONT ANSWER TO A QUESTION IF YOU THINK THAT THE QUESTION IS A STUPID
ONE.
2. IF U DO NOT RECEIVE AN ANSWER TO A QUESTION, CONSIDER IT WAS A
STUPID
ONE.
Well…at least this is what I have seen in Linux groups, and ppl
like
Linus can be considered “busy” and he “takes out time” to answer
questions.
I think, keeping these 2 as tacit rules can avoid all kinds of
confrontations. Another point that I want to put here is that many ppl
coming from Linux background think that just like in Linux, we can do
nething with the windows Kernel. Well, this is unfortunately not so. So
if u
think that windows takes away the freedom as a programmer (just as i
do),
shift to Linux (just as I would do in some time). And then we would
have
much more technical answers to questions like " How can I use asm in
kernel"
rather than the kind of answers given 2day.
regards
On 9/21/05, Don Burn wrote:
>
> Martin,
>
> I will take a different tack here. It is also the responsibility of
> the people asking questions to attempt to ask informed questions. Now
a
> “How to get started with file system drivers?” can be considered a
> informed
> question, since it is general and asking for help. But to look at the
OP’s
> question, before asking about assembler, well there is a heck of a
lot
> written in this forum and others about asm, a quick search might
have
> caused
> the question to either not be asked, or asked in a way that does not
draw
> the response you objected to.
>
> Most of us who give data on these forum, do so on our own time, and
> with the knowledge that everyone we help is potentially a competitor
for
> the
> next job or consult we try to get. It is reasonable to ask the
people
> comming to this forum to be prepared to ask intelligent questions and
to
> search the forums and the FAQ before asking. I have taught classes
over
> the
> years on various subjects, where if someone did do the basics, it
was
> reasonable to ask them to leave. There the teacher was getting paid,
> shouldn’t the same rules apply when you do it for free.
>
>
> –
> Don Burn (MVP, Windows DDK)
> Windows 2k/XP/2k3 Filesystem and Driver Consulting
> Remove StopSpam from the email to reply
>
>
>
>
>
> “Martin O’Brien” wrote in message
> news:xxxxx@ntfsd…
> > The answer to your question is, if I understand correctly that
your
> > considering using no API functions and writing directly to the
hardware,
> > yes as far as reentrancy is concerned.
> >
> > That being said, you are, bluntly, you are out of your mind if you
> > attempt this. There is basically no chance of ever getting this to
> > work. There is no need to get in to details, such as basically
assured
> > coruption, why.
> >
> > I have a pet peve concerning people responding to questions on
this
> > list with nothing more than criticism and, frequently, party line
advice
> > that is most definitely not true in all cases, and, moreover,
advice of
> > which they have no way of knowing the applicability in one’s
specific
> > case, particularly based on a three line description. Don’t get me
> > wrong; advice such as “no assembler is drivers” is not only a good
idea,
> > use of assembler in a driver these days is overwhelmingly
unnecessary.
> > My question, however, is exactly what does mentioning things such
as
> > this several time daily add to the discussion. In this particular
case,
> > my apologies to author of the question, assembler is so far down on
the
> > list of problems manifest in this idea, and, again, apologies,
there is
> > basically zero chance of this idea ever working, that it really
adds
> > nothing, and certainly does not merit its own, dedicated e-mail.
> > Finally, not only do real reasons exist for assembler’s use, and
> > sometimes these reasons may not legally be disclosed, but the world
will
> > simply not going to end if assembler is used.
> >
> > The problem, as I see it, is that this list, although very useful,
> > takes a lot of time to get through, mostly due to, in my opinion,
people
> > going on and on about procedure, policy and the proverbial right
thing
> > to do, suspect questions notwithstanding.
> >
> > For example, what I am doing here.
> >
> > My opinion is that a nice rule of thumb might be that, if you’re
not
> > going to address the person’s question, simply do not respond.
This, of
> > course, does not rule out adding anything else to the discussion.
> >
> > MM
> >
> >
> >>>> xxxxx@comcast.net 09/21/05 7:58 AM >>>
> > Just curious,
> >
> > If I attempt to read a file using straight asm opposed to
> > ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
> > re-entry?
> > Since it is a driver and hardware can be written too directly, it
is
> > possible to go around the IO manager and prevent a second IRP
right?
> >
> >
> > —
> > Questions? First check the IFS FAQ at
> > https://www.osronline.com/article.cfm?id=17
> >
> > You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
> > To unsubscribe send a blank email to
xxxxx@lists.osr.com
> >
>
>
>
> —
> Questions? First check the IFS FAQ at
> https://www.osronline.com/article.cfm?id=17
>
> You are currently subscribed to ntfsd as: xxxxx@gmail.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
—
Questions? First check the IFS FAQ at
https://www.osronline.com/article.cfm?id=17
You are currently subscribed to ntfsd as: xxxxx@evitechnology.com
To unsubscribe send a blank email to xxxxx@lists.osr.com
With due respect to everybody, the experts, the newbies (like me) , I find
that in recent days a lot of heat is generated over trivial issues, as many
of us suggested, the best way is not to answer a question if you think it is
ridiculous/stupid/off tpoic, that way the poster of the question knows tht
it was irrelevant.
Arguing abt it, defending your own writing when some one else doesnt like u
critisizing just wastes more time.
Don, Matrin, Matt, Ladislav, Dan, we know *who* you are, and we know that
newbies like me should be grateful that you do answer us, I myself have
asked stupid questions many times, but then it is not intentional always.
I think it really comes from the wonderful world of bean counting and the
art of creative salesmanship. It all usually begins in a call to a customer
account with “We can have that multi-million dollar terahertz display system
to you by the end of next week Mr. Widget.”
–
The personal opinion of
Gary G. Little
“Dan Partelly” wrote in message news:xxxxx@ntfsd…
>
> Just curious,
>
> where do you ppl get all those ideeas ? Is there an agency teaching this
> kind of shit ?
>
> Dan
>
> ----- Original Message -----
> From: “Matt Martin”
> To: “Windows File Systems Devs Interest List”
> Sent: Wednesday, September 21, 2005 2:58 PM
> Subject: [ntfsd] eliminate re-entry with asm?
>
>
>> Just curious,
>>
>> If I attempt to read a file using straight asm opposed to
>> ZwCreate/ZwRead/ZwClose in a filter, would that still result in a
>> re-entry? Since it is a driver and hardware can be written too directly,
>> it is possible to go around the IO manager and prevent a second IRP
>> right?
>>
>> —
>> Questions? First check the IFS FAQ at
>> https://www.osronline.com/article.cfm?id=17
>>
>> You are currently subscribed to ntfsd as: xxxxx@rdsor.ro
>> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>