Hello,
whenever I attach two /bin/wd sessions
(to different pids) I get a Proc32 crash.
Does anyone know the cause of this ?
thanks in advance
Below is crash dump from Proc32 and
output from sin ver.
Proc output to serial port 3f8 at 9600 baud
Path=0 - EIDE
target=0 lun=0 Direct-Access(0) - QUANTUM FIREBALL Rev:
Version: 425.O Aug 19 2002 Technical Support: +1 (613) 591-0941
Proc fault 1, ldt 100 /boot/sys/Proc32; fault e+0
cs:eip=5:7dd9 ss:esp=d:f7c0f78 efl=12287 ds=d es=8 fs=0 gs=0
eax/ff814fc ebx/ff814fc ecx/0 edx/ff81500 esi/0 edi/4ff1bd ebp/f7c0f7c
Stack (d:f7c0f78)
00000001 0f7c0fa0 00000001 0001667c 00007100 0000d2fc 80000089 f053f1bd
00010000 000000f8 0f7c0fb4 00000000 00023180 0000cd8b 00000099 0f7c0fe8
00000000 000059fd 0000000d 00003882 0f7c0fd0 00005768 00003880 0000000d
00000610 00003880 0000000d 00000016 00000001 000059e5 000059fd 0f7c0935
Process Entry (addr 5aa4)
00000000 00000001 00000000 00000001 00000000 00000000 30020207 00001e1e
000058d8 0100000d 00005b5c ffffffff 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000501 000d0005 00007118 00000000 00000001
00000010 00000000 00000182 0000c140 00000000 00000000 0001e1a0 00000000
00000000 00000000 00000000 ffff0001 00000000 00000000 00000000
PROGRAM NAME VERSION DATE
/boot/sys/Proc32 Proc 4.25O Aug 19 2002
/boot/sys/Proc32 Slib16 4.23G Oct 04 1996
/boot/sys/Slib32 Slib32 4.24B Aug 12 1997
/bin/Fsys Fsys32 4.24Y Apr 23 2002
/bin/Fsys.eide eide 4.25G Apr 15 2002
//1/bin/Dev32 Dev32 4.23G Oct 04 1996
//1/bin/Pipe Pipe 4.23A Feb 26 1996
//1/bin/Dev32.ser Dev.ser 4.25A Feb 14 2003
//1/bin/Dev32.ser Dev.ser 4.25A Feb 14 2003
//1/bin/Dev32.ansi Dev32.ansi 4.23H Nov 21 1996
//1/bin/Dev32.par Dev32.par 4.25A Jan 08 2001
//1/bin/Dev32.pty Dev32.pty 4.23G Oct 04 1996
//1/bin/Fsys.eide eide 4.25G Apr 15 2002
//1/bin/Iso9660fsys Iso9660fsys 4.23D Mar 20 2000
//1/bin/Net Net 4.25E Apr 24 2002
//1/bin/Net.ether905 Net.ether905 4.24N Apr 12 2002
//1//usr/ucb/Socket Socket 4.25H Jul 30 1999
//1/usr/bin/lpsrvr lpsrvr 4.24A Jun 26 1997
//1//photon/bin/Photon Photon 1.14B Sep 03 1999
//1/*/bin/phfontpfr Photon Font 1.14F Oct 27 1999
\
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F7BBFE0.1D25890C@carmenta.se…
whenever I attach two /bin/wd sessions
(to different pids) I get a Proc32 crash.
Does anyone know the cause of this ?
I have a serious suspect on the system itself or at the very least the
processes you’re debugging. I’ve never had problems with multiple wd
sessions (different pids). The crash you’re getting is in the physical page
selection code, which hasn’t changed in a dogs age, and to take a mapping
error exception while executing it seems rather bizare.
Does the crash only occur when debugging a specific pid (process), or can
you do it with any process (even a hello world program)?
-Adam
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F7BBFE0.1D25890C@carmenta.se> …
whenever I attach two /bin/wd sessions
(to different pids) I get a Proc32 crash.
Does anyone know the cause of this ?
I have a serious suspect on the system itself or at the very least the
processes you’re debugging. I’ve never had problems with multiple wd
sessions (different pids). The crash you’re getting is in the physical page
selection code, which hasn’t changed in a dogs age, and to take a mapping
error exception while executing it seems rather bizare.
Does the crash only occur when debugging a specific pid (process), or can
you do it with any process (even a hello world program)?
-Adam
I have simplified the steps for crashing Proc32:
cc -g -o hello hello.c
cc -g -o hallo hello.c
./hello &
./hallo &
nwd hello
nwd hallo
… and Proc32 crashes.
nwd is defined in my .kshrc as:
nwd()
{
wd ps | grep $1 | grep -v grep | awk '{print $1}'
}
alias wd=‘wd -Console=tty’
and hello.c:
main()
{
while (1)
printf(“hello \n”);
}
regards
–
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F7C4432.A9F31F5A@carmenta.se…
Adam Mallory wrote:
I have simplified the steps for crashing Proc32:
Aside from avoiding doing the awk stuff to get a pid, I can run multiple
copies of the hello.c and attach to them on the same console w/o issue.
Can you repeat this on a different computer with the same software?
-Adam
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F7C4432.A9F31F5A@carmenta.se> …
Adam Mallory wrote:
I have simplified the steps for crashing Proc32:
snip
Aside from avoiding doing the awk stuff to get a pid, I can run multiple
copies of the hello.c and attach to them on the same console w/o issue.
Can you repeat this on a different computer with the same software?
-Adam
Even simpler:
./hello &
ps
/bin/wd
./hallo &
…
and after a few seconds Proc32 crashes.
Note, just a single /bin/wd.
The hard disk has been moved to a Pentium 133, and
two different Pentium II with the same (original) behaviour.
If it is not an issue with Proc32, could it (really) be hard disk related
?
regards
–
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson wrote:
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F7C4432.A9F31F5A@carmenta.se> …
Adam Mallory wrote:
I have simplified the steps for crashing Proc32:
snip
Aside from avoiding doing the awk stuff to get a pid, I can run multiple
copies of the hello.c and attach to them on the same console w/o issue.
Can you repeat this on a different computer with the same software?
-Adam
Further information:
When I try the same on another bootable
partition, I don’t get a Proc32 crash, (different
version of PRoc32 though).
Is this, could it be, hard disk related ?
regards
\
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F7D1EF3.3BB5CC8A@carmenta.se…
When I try the same on another bootable
partition, I don’t get a Proc32 crash, (different
version of PRoc32 though).
Is this, could it be, hard disk related ?
I’m not sure how - unless at load time there was corruption. Perhaps check
the checksum of the Proc bin on your hd - mine is:
2035812 281314 ./Proc32.425O
(use the cksum utility).
-Adam
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F7D1EF3.3BB5CC8A@carmenta.se> …
When I try the same on another bootable
partition, I don’t get a Proc32 crash, (different
version of PRoc32 though).
Is this, could it be, hard disk related ?
I’m not sure how - unless at load time there was corruption. Perhaps check
the checksum of the Proc bin on your hd - mine is:
2035812 281314 ./Proc32.425O
(use the cksum utility).
-Adam
I matches mine:
cksum /boot/sys/Proc32
2035812 281314 /boot/sys/Proc32
Any other clues ?
brgds
\
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F7D8C91.3A9BB640@carmenta.se…
Adam Mallory wrote:
Any other clues ?
Well, you mentioned that going to a seperate partition allows things to work
(but with a different Proc), what if you use Proc ver O, does it still work.
If so, then perhaps you need to rebuild your OS image as it could be
corrupt.
Outside of that I’m out of ideas.
-Adam
<!doctype html public “-//w3c//dtd html 4.0 transitional//en”>
Adam Mallory wrote:
Per Akesson <> wrote in
message
news:<3F7D8C91.3A9BB640@carmenta.se>...
Adam Mallory wrote:
>Any other clues ?
Well, you mentioned that going to a seperate partition allows things
to work
(but with a different Proc), what if you use Proc ver O, does it still
work.
If so, then perhaps you need to rebuild your OS image as it could be
corrupt.
Outside of that I'm out of ideas.
-Adam
I tried that, I built a new boot image with Proc32 version O, that
mounted the OK partition (t77) as /, and Proc32 does NOT crash.
I seems that it is something strange with my default boot t79
partition,
doesn't it ? How could it crash Proc32 ?
chkfsys says the partition is OK.
below is output from fdisk /dev/hd0.0 show:
_____OS_____
Start End ______Number_____
Size Boot
name type
Cylinder Cylinder Cylinders Blocks
1. QNY ( 78)
0 1
2 16065 7 MB
2. QNX ( 77)
2 150
149 1201536 586 MB
3. ------ (---) -----
----- ----- -------- -----
4. QNZ ( 79)
151 400
250 2016000 984 MB *
Note, t79 is the partition that Proc32 crashes on and t77 is OK.
brgds
--
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F811770.8A266A58@carmenta.se…
I tried that, I built a new boot image with Proc32 version O, that
mounted the OK partition (t77) as /, and Proc32 does NOT crash.
I seems that it is something strange with my default boot t79 partition,
doesn’t it ? How could it crash Proc32 ?
chkfsys says the partition is OK.
below is output from fdisk /dev/hd0.0 show:
What if you rebuild your osimage on the 79 partition? Do things start
working for you?
-Adam
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F811770.8A266A58@carmenta.se> …
I tried that, I built a new boot image with Proc32 version O, that
mounted the OK partition (t77) as /, and Proc32 does NOT crash.
I seems that it is something strange with my default boot t79 partition,
doesn’t it ? How could it crash Proc32 ?
chkfsys says the partition is OK.
below is output from fdisk /dev/hd0.0 show:
What if you rebuild your osimage on the 79 partition? Do things start
working for you?
-Adam
I even copied (cp -Rcsp -Munix) the t79 partition to a new disk,
rebuilt the boot image, and booted from the new disk,
but still no change. Proc32 still crashes.
Is my /bin/wd corrupted ? since it is the same whenever
Proc32 crashes:
qnx1:root> cksum /usr/watcom/10.6/bin/wd
2201938911 598928 /usr/watcom/10.6/bin/wd
regards
–
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson wrote:
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F811770.8A266A58@carmenta.se> …
I tried that, I built a new boot image with Proc32 version O, that
mounted the OK partition (t77) as /, and Proc32 does NOT crash.
I seems that it is something strange with my default boot t79 partition,
doesn’t it ? How could it crash Proc32 ?
chkfsys says the partition is OK.
below is output from fdisk /dev/hd0.0 show:
What if you rebuild your osimage on the 79 partition? Do things start
working for you?
-Adam
I even copied (cp -Rcsp -Munix) the t79 partition to a new disk,
rebuilt the boot image, and booted from the new disk,
but still no change. Proc32 still crashes.
Is my /bin/wd corrupted ? since it is the same whenever
Proc32 crashes:
The /usr/watcom/10.6/bin/wd differs between the t79 and the t77
partitions. I copied the wd from the (good) t77 partition to the t79
partition, and booted from the t79 partition, but still no luck.
Proc32 still crashes.
\
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F82B847.BB29F580@carmenta.se…
Per Akesson wrote:
The /usr/watcom/10.6/bin/wd differs between the t79 and the t77
partitions. I copied the wd from the (good) t77 partition to the t79
partition, and booted from the t79 partition, but still no luck.
Proc32 still crashes.
I would suggest backing up your data, and reinstalling (not copying) onto a
different drive (not just a partition).
-Adam
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F82B847.BB29F580@carmenta.se> …
Per Akesson wrote:
The /usr/watcom/10.6/bin/wd differs between the t79 and the t77
partitions. I copied the wd from the (good) t77 partition to the t79
partition, and booted from the t79 partition, but still no luck.
Proc32 still crashes.
I would suggest backing up your data, and reinstalling (not copying) onto a
different drive (not just a partition).
-Adam
Well some further investigation shows that
if I move all source and binaries to the local harddisk (/tmp),
and start 2 wd’s, Proc32 does NOT crash. Neither the
hello.c example nor my application.
My development directories are NFS mounted
due to backup issues.
It seems if something, (binary or source), is
NFS mounted during the debug session, Proc32
crashes.
Comments ?
\
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F83B0E9.6ECD085D@carmenta.se…
Adam Mallory wrote:
Well some further investigation shows that
if I move all source and binaries to the local harddisk (/tmp),
and start 2 wd’s, Proc32 does NOT crash. Neither the
hello.c example nor my application.
My development directories are NFS mounted
due to backup issues.
Ahh, a pertinent piece of information.
It seems if something, (binary or source), is
NFS mounted during the debug session, Proc32
crashes.
I’ll take another peek and get back to you. Althrough I don’t see what the
difference is where the source files live… strange. Does the same issue
happen with the older Proc?
-Adam
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F83B0E9.6ECD085D@carmenta.se> …
Adam Mallory wrote:
Well some further investigation shows that
if I move all source and binaries to the local harddisk (/tmp),
and start 2 wd’s, Proc32 does NOT crash. Neither the
hello.c example nor my application.
My development directories are NFS mounted
due to backup issues.
Ahh, a pertinent piece of information.
It seems if something, (binary or source), is
NFS mounted during the debug session, Proc32
crashes.
I’ll take another peek and get back to you. Althrough I don’t see what the
difference is where the source files live… strange. Does the same issue
happen with the older Proc?
-Adam
At least with Proc32 version J. Then I tried
to upgrade to patch G (with Proc32 v O), to
see if the problem went away, but no …
I don’t recall that I had this problem before Proc32 v J
but then I don’t recall the version of Proc32.
BTW, on the “good” partition t77, where Proc32 v O didn’t crash.
the TCPIP runtime version seems to differs from the “bad” t79
partition.
mount /dev/hd0.0t77 /mnt # good partition
cksum /mnt/usr/tcprt/4.25/usr/ucb/Socket
2900286966 228688 /mnt/usr/tcprt/4.25/usr/ucb/Socket
cksum /usr/tcprt/4.25/usr/ucb/Socket
2086384102 228455 /usr/tcprt/4.25/usr/ucb/Socket
regards
–
Per Åkesson
Carmenta AB
SWEDEN
Adam Mallory wrote:
Per Akesson <> Per.Akesson@carmenta.se> > wrote in message
news:> 3F83B0E9.6ECD085D@carmenta.se> …
Adam Mallory wrote:
Well some further investigation shows that
if I move all source and binaries to the local harddisk (/tmp),
and start 2 wd’s, Proc32 does NOT crash. Neither the
hello.c example nor my application.
My development directories are NFS mounted
due to backup issues.
Ahh, a pertinent piece of information.
It seems if something, (binary or source), is
NFS mounted during the debug session, Proc32
crashes.
I’ll take another peek and get back to you.
Have you had a chance to look into it yet ?
BTW, I did a fresh install on a new harddisk from
the QNX Upgrade, July 2003 CD, but the problem
persists.
brgds
\
Per Åkesson
Carmenta AB
SWEDEN
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F8E9B47.3077F461@carmenta.se…
Adam Mallory wrote:
Have you had a chance to look into it yet ?
No, unfortunately not, it’s on the “list”. I’ll let you know as soon as I
do.
-Adam
Per Akesson <Per.Akesson@carmenta.se> wrote in message
news:3F8E9B47.3077F461@carmenta.se…
Adam Mallory wrote:
Have you had a chance to look into it yet ?
BTW, I did a fresh install on a new harddisk from
the QNX Upgrade, July 2003 CD, but the problem
persists.
I still cannot seem to reproduce this over an NFS mount (I’ve even tried
larger number of concurrent sessions). Seeing that you can see the issue
but only via NFS - have you tried eliminating/replacing your network card -
seems to me the only thing left.
-Adam