Hi,
we are having a problem when running Dev32.par on some new computers that we
never encountered before. Almost everytime (but not everytime 1 of 10 work)
Dev32.par (Dev.par &) runs it prints out this error and terminates.
//[node]/Dev32.par terminated (SIGFPE ) at 0005:000007AC
I have the exact same error on two machines whose only similarity is the
CPU, Pentium III 800mhz, the motherboards, video cards, and NIC’s are
different. I have come to the conclusion that the CPU might be the cause, as
I have tried to install QNX on several slower machines, (highest Pentium III
450mhz) with no error. Has anyone ever encountered this before? Is there a
fix/patch?
Chris Nasr <cnasr@mechtronix.ca> wrote:
Hi,
we are having a problem when running Dev32.par on some new computers that we
never encountered before. Almost everytime (but not everytime 1 of 10 work)
Dev32.par (Dev.par &) runs it prints out this error and terminates.
//[node]/Dev32.par terminated (SIGFPE ) at 0005:000007AC
I have the exact same error on two machines whose only similarity is the
CPU, Pentium III 800mhz, the motherboards, video cards, and NIC’s are
different. I have come to the conclusion that the CPU might be the cause, as
I have tried to install QNX on several slower machines, (highest Pentium III
450mhz) with no error.Has anyone ever encountered this before?
Yes… it’s a known issue with fast CPU’s
Is there a fix/patch?
There should be. If someone from QA doesn’t jump in here and tell you
where it is, ping this thread, and I’ll give you their private email
addresses.
Pete,
thanks for the reply, that takes a load off my mind to know that it is a
known issue. However, I must claim ignorance, what did you mean by ping this
thread? Or is that exactly what I’m doing right now 
Chris Nasr
cnasr[at]mechtronix[dot]ca
Mechtronix Systems Inc.
<pete@qnx.com> wrote in message news:8t79bp$do5$2@nntp.qnx.com…
Chris Nasr <> cnasr@mechtronix.ca> > wrote:
Hi,
we are having a problem when running Dev32.par on some new computers
that we
never encountered before. Almost everytime (but not everytime 1 of 10
work)
Dev32.par (Dev.par &) runs it prints out this error and terminates.
file://[node]/Dev32.par terminated (SIGFPE ) at 0005:000007AC
I have the exact same error on two machines whose only similarity is the
CPU, Pentium III 800mhz, the motherboards, video cards, and NIC’s are
different. I have come to the conclusion that the CPU might be the
cause, as
I have tried to install QNX on several slower machines, (highest Pentium
III
450mhz) with no error.Has anyone ever encountered this before?
Yes… it’s a known issue with fast CPU’s
Is there a fix/patch?
There should be. If someone from QA doesn’t jump in here and tell you
where it is, ping this thread, and I’ll give you their private email
addresses.
we are having a problem when running Dev32.par on some new computers that
we
never encountered before. Almost everytime (but not everytime 1 of 10
work)
Dev32.par (Dev.par &) runs it prints out this error and terminates.
file://[node]/Dev32.par terminated (SIGFPE ) at 0005:000007AC
I have the exact same error on two machines whose only similarity is the
CPU, Pentium III 800mhz, the motherboards, video cards, and NIC’s are
different. I have come to the conclusion that the CPU might be the cause,
as
I have tried to install QNX on several slower machines, (highest Pentium
III
450mhz) with no error. Has anyone ever encountered this before? Is there a
fix/patch?
I have this same problem…You are not alone…
I would also like the patch info…
Thanks…
~ Lee R. Copp
~ Project Engineer (EE/ME)
~ Michigan Scientific Corp.
~ 321 East Huron St.
~ Milford, MI 48381
~ 248-685-3939 x109 (V), 248-684-5406 (Fx)
~ http://www.michiganscientific.com
~ mailto:<Lee.R.Copp@MichiganScientific.com>
I think, it’s a common problem. We have the same difficulty on same
machines. As workaround we use the Dev32.par from patch C. But it can’t
be the solution.
Could you put the fixed version to an public place. I think there will be
other users with the same problem.
Thanks
Uwe
Ursprüngliche Nachricht
Am 25.10.00, 20:38:49, schrieb pete@qnx.com zum Thema Re: Dev32.par
error:
Chris Nasr <> cnasr@mechtronix.ca> > wrote:
Hi,
we are having a problem when running Dev32.par on some new computers that
we
never encountered before. Almost everytime (but not everytime 1 of 10
work)
Dev32.par (Dev.par &) runs it prints out this error and terminates.
//[node]/Dev32.par terminated (SIGFPE ) at 0005:000007AC
I have the exact same error on two machines whose only similarity is the
CPU, Pentium III 800mhz, the motherboards, video cards, and NIC’s are
different. I have come to the conclusion that the CPU might be the cause,
as
I have tried to install QNX on several slower machines, (highest Pentium
III
450mhz) with no error.Has anyone ever encountered this before?
Yes… it’s a known issue with fast CPU’s
Is there a fix/patch?
There should be. If someone from QA doesn’t jump in here and tell you
where it is, ping this thread, and I’ll give you their private email
addresses.
Uwe.Peuker@bitctrl.de wrote:
I think, it’s a common problem. We have the same difficulty on same
machines. As workaround we use the Dev32.par from patch C. But it can’t
be the solution.Could you put the fixed version to an public place. I think there will be
other users with the same problem.
I second this. I have the same problem on several machines. If fixed version
already exists (even beta) why not to make it available for download?
Greg Wrobel
Lee R. Copp <Lee.R.Copp@michiganscientific.com> wrote:
we are having a problem when running Dev32.par on some new computers that
we
never encountered before. Almost everytime (but not everytime 1 of 10
work)
Dev32.par (Dev.par &) runs it prints out this error and terminates.
file://[node]/Dev32.par terminated (SIGFPE ) at 0005:000007AC
I have the exact same error on two machines whose only similarity is the
CPU, Pentium III 800mhz, the motherboards, video cards, and NIC’s are
different. I have come to the conclusion that the CPU might be the cause,
as
I have tried to install QNX on several slower machines, (highest Pentium
III
450mhz) with no error. Has anyone ever encountered this before? Is there a
fix/patch?I have this same problem…You are not alone…>
I would also like the patch info…
There is already a patch for this problem. QA knows where it is, but
I don’t. Someone should be in here telling you all where to get it.
Sorry for the delay I will post the latest Dev32.par for Alpha today.
This will only be available for beta customers (for hopefully
obvious reasons).
I will post here as soon as this is available.
pete@qnx.com wrote:
: Lee R. Copp <Lee.R.Copp@michiganscientific.com> wrote:
:>> we are having a problem when running Dev32.par on some new computers that
:> we
:>> never encountered before. Almost everytime (but not everytime 1 of 10
:> work)
:>> Dev32.par (Dev.par &) runs it prints out this error and terminates.
:>> file://[node]/Dev32.par terminated (SIGFPE ) at 0005:000007AC
:>> I have the exact same error on two machines whose only similarity is the
:>> CPU, Pentium III 800mhz, the motherboards, video cards, and NIC’s are
:>> different. I have come to the conclusion that the CPU might be the cause,
:> as
:>> I have tried to install QNX on several slower machines, (highest Pentium
:> III
:>> 450mhz) with no error. Has anyone ever encountered this before? Is there a
:>> fix/patch?
:> I have this same problem…You are not alone…
:> I would also like the patch info…
: There is already a patch for this problem. QA knows where it is, but
: I don’t. Someone should be in here telling you all where to get it.
Michael Shane <mshane@qnx.com> wrote:
: Sorry for the delay I will post the latest Dev32.par for Alpha today.
: This will only be available for beta customers (for hopefully
: obvious reasons).
: I will post here as soon as this is available.
Ok this is now available. Check /updates/beta.qnx424/Alpha/Dev32.par
If you are having problems with this please email me directly.
Mike Shane
QA, QNX
mshane@qnx.com
“Michael Shane” <mshane@qnx.com> wrote in message
news:8t9ebq$mt1$1@nntp.qnx.com…
Sorry for the delay I will post the latest Dev32.par for Alpha today.
This will only be available for beta customers (for hopefully
obvious reasons).
Why?? What about I386??
I am wondering why a simple CPU speed increase (this was previously posted
as to what was
causing the problem) would crash a device driver for a device that has been
around since the begining…
I currently have no need for my printer port so I guess I’m just stirring up
some trouble…
Later…
~ Lee R. Copp
~ Project Engineer (EE/ME)
~ Michigan Scientific Corp.
~ 321 East Huron St.
~ Milford, MI 48381
~ 248-685-3939 x109 (V), 248-684-5406 (Fx)
~ http://www.michiganscientific.com
~ mailto:<Lee.R.Copp@MichiganScientific.com>
Previously, Lee R. Copp wrote in qdn.public.qnx4.devtools:
I am wondering why a simple CPU speed increase (this was previously posted
as to what was
causing the problem) would crash a device driver for a device that has been
around since the begining…
Note that this problem occured when doing an update. I was
one of, if not the first person to discover this problem. I
think that someone was trying to provide EPP support of some
kind. Standard parallel ports have no built in timing
features other than the period it takes to complete a port
I/O instructions. It is therefore common to have timing
loops in a parallel port driver of some sort. I’m guessing
that some new timing loop was under-run causing a divide
exception or something similar.
I currently have no need for my printer port so I guess I’m just stirring up
some trouble…>
So, do you want to take this outside? .
Mitchell Schoenbrun --------- maschoen@pobox.com
Note that this problem occured when doing an update. I was
I ran into it after applying the ‘D’ patch…
I/O instructions. It is therefore common to have timing
loops in a parallel port driver of some sort. I’m guessing
Why does this remind me of the ‘Turbo’ button on my 386
and all the software which broke simply because the button
was in the wrong position…
that some new timing loop was under-run causing a divide
exception or something similar.
That sounds about right…
So, do you want to take this outside? >
It is rather nice out today…
Later…
~ Lee R. Copp
~ Project Engineer (EE/ME)
~ Michigan Scientific Corp.
~ 321 East Huron St.
~ Milford, MI 48381
~ 248-685-3939 x109 (V), 248-684-5406 (Fx)
~ http://www.michiganscientific.com
~ mailto:<Lee.R.Copp@MichiganScientific.com>
The updated Dev32.par is now available for everyone in the
Released misc directory.
/updates/qnx42/Released/misc/Dev32.par
538087101 16903 Dev32.par
Mike Shane
QA, QNX
mshane@qnx.com
Previously, Michael Shane wrote in qdn.public.qnx4.devtools:
The updated Dev32.par is now available for everyone in the
Released misc directory.
We modified the released source for Dev32.par our puposes. Is there
something in the new driver we should include in our version? Is the
new source and lib available?
/updates/qnx42/Released/misc/Dev32.par
538087101 16903 Dev32.par
Mike Shane
QA, QNX
mshane@qnx.com
Mike,
thanks alot for the file.
Chris
“Michael Shane” <mshane@qnx.com> wrote in message
news:8tk1t9$sig$1@nntp.qnx.com…
The updated Dev32.par is now available for everyone in the
Released misc directory./updates/qnx42/Released/misc/Dev32.par
538087101 16903 Dev32.par
Mike Shane
QA, QNX
mshane@qnx.com