[x-posted from apps]: Re: fs-qnx filesystem questions

Operating_System_Tec · October 26, 2001, 7:20am

“Igor Kovalenko” <kovalenko@home.com> wrote in message
news:9r9eqq$rt8$1@inn.qnx.com…

So you share Rennie’s view. I already said that such definition is rather
useless because it does not provide for a formal way to tell what is
microkernel and what is not.

I agree, but much of the terminology in the real-time domain is similar in
“cloudy factor” (probably due to marketing ). I use the terms as a
guideline, and not a definitive view of exactly how a system should or
should not behave.

Perhaps. But then you should not have [sarcastically] asked why did not we
write a proper driver ourselves, especially when you know better that you
do
not provide a way to do it.

I appologize if sarcasm was when you read into it, but it was definately not
intended. It was meant to be only a question of curiosity.

-Adam

Armin_Steinhoff1 · October 26, 2001, 11:19am

Rennie Allen wrote:

I am getting sick of this kind of arguments. Reminds me very much the
old
days when large chorus of people in comp.os.qnx (except Armin) argued
to me
how unnecessary and unwanted would it be to have SMP and threads in
QNX.
Until Dan Dodge came up and told them otherwise, lol.

I recall this argument. IIRC you were arguing that real-time systems
need SMP, and that QNX4 could, and should support SMP, and that ASMP
would be totally useless for realtime.

IMHO … your recall is not correct.

You don’t honestly believe that
those working at QSSL didn’t know that Dan was working on SMP do you ?

So let us know why you have argued against SMP ??

Now it is 5 years later, and how many SMP QNX6 systems are there as a
percentage of total QNX6 systems ?

I don’t know how many SMP systems are running at Motorola … but is
definitely a fact that
the availability of SMP for Neutrino was one of the important reasons
for Motorola to make a decision for Neutrino. (Statement from a
Motorola manager …)

If SMP were so vitally important (as
you made it out to be at the time) don’t you think that more than (some
small number here)% of the (paying customer) QNX6 systems would be
shipping SMP systems ?

It is vitally important … look at the QSSL home page.

QNX6 was designed to be SMP from the very
beginning (which btw was a long time before your SMP tirade)

If it was really a tirade … then it was a tirade against SMP.

, so actually implementing it was relatively low cost.

Our industry has no requirement for SMP today.

I don’t know what industry your company owns … fact is if SMP is used
it is mostly
used in the industry.

You continuously
complain that QNX is “not advanced” yet it has many features that the
vast majority of customers still don’t use, and probably won’t use for
years. I don’t disagree with you about the technical coolness of SMP or
a HPFS, but where is the customer base ?

Well … I heard a lot of years ago similar arguements against Ethernet
and TCP/IP

Fact is that a HPFS is neccessary for multi media … a files system
which is 5-8 times slower than the standard LINUX file system is more or
less useless for performant multi media applications.

Yes, QNX has better realtime technology, but that’s not the only issue
considered when decisions are made. If general-purpose OSes are
getting more
realtime capabilities, QNX needs to counter that with more
general-purpose
features, especially as technology evolves and new opportunities
arise.

Or perhaps the M$ strategy might be to get QSSL to spend inordinate
amounts of it’s R&D effort trying to provide something that is
essentially useless to their market, and to bleed off their customer
base ?

sounds a little bit paranoid.

We use queueing into circular buffers to write logs and billing
information
(potentially coming from thousands of calls). Since those data are
produced
by realtime components, we did not want those components to be held by
undeterministic I/O operations.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

OK. Now let’s get back to the root of the discussion here. You want
QSSL to invest a lot of time and effort in to producing a HPFS,

Time? Efforts? There are a lof of sources in the internet … so what’s
the problem?

Armin

Jutta_Steinhoff1 · October 26, 2001, 12:57pm

Rennie Allen wrote:

[…]
Now it is 5 years later, and how many SMP QNX6 systems are there as a
percentage of total QNX6 systems ? If SMP were so vitally important (as
you made it out to be at the time) don’t you think that more than (some
small number here)% of the (paying customer) QNX6 systems would be
shipping SMP systems ? QNX6 was designed to be SMP from the very
beginning (which btw was a long time before your SMP tirade), so
actually implementing it was relatively low cost.

Our industry has no requirement for SMP today. …

Rennie, the requirements seem to be so low that QSSL is organizing
a seminar in the next month with the topic: “Taking the Pain out
of Building Distributed, Fault-tolerant, SMP Networking Equipment”

Attendees should be “Engineers and project managers involved in
the development of high-performance, scalable, next-generation
IP infrastructure, optical and storage networking equipment”

The requirements from your QNX segment are different from other QNX
users, but that’s no reason to ignore important technologies which
other QNXers will need…

Jutta

Mario_Charest1 · October 26, 2001, 4:00pm

“Miguel Simon” <simon@ou.edu> wrote in message
news:3BD84C71.332A3F39@ou.edu…

Hi…

Mario Charest wrote:

…
What does a decent performance file system has to do with “general
purpose
computing”?

…educate me if you will, please… What is so bad about the QNX6
file system?? I suppose that I have not pushed the system to its
limits, have I? In any case, if you enlighten me, then I would know
what is the complain all about.

Check one of Igor’s post, I don’t think there is a need to
add more.

I get the feeling that some people use QNX because it seems to be the
least upper bound of all the bad RTOS’s. Which is or what constitutes a
good RTOS, and why would QNX6 not be one of them? Because of the file
system performance? How relevant is a ‘decent performance file system’
to the real-time aspect of the os?

I’m confuse, no one said QNX6 isn’t a good real-time OS nor does
it directly have something to do with file system.

If you don’t need better file system performance then what is
currently available then good for you. But I’m baffle as to why
people claim that if it’s good enough for them that it’s good
enough for other. What’s wrong in asking for a filesystem that
perform (speed and feature wise) at least on par with other
OSes.

Many have said -and is indeed my own
experience- that the file system is irrelevant to some type of hard real
time systems and/or applications. Then, if you have an RTOS, would you
have to have a ‘decent performance file system’ for the RTOS to be
considered good?? I would suspect that this is a matter of opinion as
opposed to a matter of fact, would you not?

Again, I don’t see how you came with this observation, no one said
"would you have to have a ‘decent performance file system’ for the RTOS to
be
considered good?? " But what I understand is that the performance of the
filesystem
of QNX is below what other OS offer, no more no less.

As an example: what file system did the Apollo program folks utilized in
their embedded real-time computers?

I would guess the Apollo program didn’t use any file system at all,
but if they would it would probably be a flash file system .
QNX flash file system architecture is totaly different then HD file
system architecture… My observeration is that the
flash file system of QNX is as good, if not better, then
other OS flash file system…

If you want to list product that can happily run with the level
of performance QNX currenlty offer or that don’t even need
file system I can also name a bunch. But I can also come up
with a list of system that would greatly benefit from a better
performing FS.

Of course it’s up to QNX to decide what is their priority,
they probably have more customer that are not in need
of better FS which is why they haven’t improve it. But
I just don’t understand why it would be wrong to have a
FS that perform on par with other OS.

In fact some post are talking about HPFS (high performance FS).
I personnaly I’m not even talking about HPFS. Is FAT32
an on EIDE to be consider HPFS? I sure don’t think so…

Bill_Caroselli_Q-TPS · October 26, 2001, 6:20pm

Personally, I think that the QNX FS is fast enough for me. But like any
speed guru, faster is always better.

I commented earlier that I think that the short comings are in the drivers
not the file system. I would like to hear an official statement from QSSL.

Does QSSL see the short coming to be file system or drivers?
Is QSSL satisfied with the current file system/drivers?
Are there plans to develop a better/faster file system/drivers?

–
Bill Caroselli – 1(530) 510-7292
Q-TPS Consulting
QTPS@EarthLink.net

Igor_Kovalenko2 · October 26, 2001, 8:58pm

“Jutta Steinhoff” <j-steinhoff@web.de> wrote in message
news:3BD95DAC.D3EEE399@web.de…

Rennie Allen wrote:
[…]
Now it is 5 years later, and how many SMP QNX6 systems are there as a
percentage of total QNX6 systems ? If SMP were so vitally important (as
you made it out to be at the time) don’t you think that more than (some
small number here)% of the (paying customer) QNX6 systems would be
shipping SMP systems ? QNX6 was designed to be SMP from the very
beginning (which btw was a long time before your SMP tirade), so
actually implementing it was relatively low cost.

Our industry has no requirement for SMP today. …

Rennie, the requirements seem to be so low that QSSL is organizing
a seminar in the next month with the topic: “Taking the Pain out
of Building Distributed, Fault-tolerant, SMP Networking Equipment”

Attendees should be “Engineers and project managers involved in
the development of high-performance, scalable, next-generation
IP infrastructure, optical and storage networking equipment”

The requirements from your QNX segment are different from other QNX
users, but that’s no reason to ignore important technologies which
other QNXers will need…

Rennie is obviously thinking of SMP in the context of SMP servers/desktops.
But there are SMP capable PowerPC processors which are getting very popular
among developers of various high-performance telecom equipment. I think
that’s where QNX is targeting.

igor

Mario_Charest1 · October 28, 2001, 9:37pm

Can’t see my own post, reposting…