I have the same problem.
And my guess is that it has to do something with the new south bridge over PCI
floppy controllers. It's the same problem with the onboard PCI ac97 sound
support.
I think that the hardware and kernel resources gets overloaded or miss
configured.
Maybe there are also BIOS
For the record, after replacing drives *and* cables, the problem
persists. I can only assume that it is the controller after all :-(
I think I will live without a floppy on that particular box.
ael
--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of
For what it is worth, here are the results from my debian testing box
under 2.6.32_exact-55846-gf405425
$ lsmod |grep floppy
floppy 45327 0
# setfdprm /dev/fd0 HD
# fdformat /dev/fd0
Double-sided, 80 tracks, 18 sec/track. Total capacity 1440 kB.
Formatting ... done
Verifying
On 15/12/09 15:59, ael wrote:
For what it is worth, here are the results from my debian testing box
under 2.6.32_exact-55846-gf405425
$ lsmod |grep floppy
floppy 45327 0
# setfdprm /dev/fd0 HD
# fdformat /dev/fd0
Double-sided, 80 tracks, 18 sec/track. Total capacity
ael wrote:
For what it is worth, here are the results from my debian testing box
under 2.6.32_exact-55846-gf405425
$ lsmod |grep floppy
floppy 45327 0
# setfdprm /dev/fd0 HD
# fdformat /dev/fd0
Double-sided, 80 tracks, 18 sec/track. Total capacity 1440 kB.
Formatting
Alain Knaff wrote:
Could you try the same with a higher repetition count:
fdrawcmd readid 0 repeat=40
Just to make sure that eventually all sectors show up
Also, on your case, the actual read error seems to be on track 0. Could you
give me also the headers of track 0?
Have to go out for an
On 12/15/2009 10:08 AM, Alain Knaff wrote:
This does indeed look like a different problem to me... No sector with a
bad track id, but lots of sectors skipped. But these skips look more
like they are happening during read (rather than being really missed), as
on the second pass, they are
ael wrote:
Mark Hounschell wrote:
On 12/15/2009 10:08 AM, Alain Knaff wrote:
I mentioned I had multiple machines with this problem. Some running
different versions of SuSE. Mainly 11.0, which is where all the info
I've provided came from thus far. This machine also has a SuSE-11.2
disk on
Alain Knaff wrote:
Could you try the same with a higher repetition count:
On same floppy (medium) as before:
# fdrawcmd readid 0 repeat=40
0: 0
1: 0
2: 0
3: 1
4: 0
5: 10
6: 2
no disk change
0: 0
1: 0
2: 0
3: 1
4: 0
5: 11
6: 2
no disk change
0: 0
1: 0
2: 0
3: 1
4: 0
5: 12
6: 2
no disk change
Mark Hounschell wrote:
On 12/15/2009 10:08 AM, Alain Knaff wrote:
I mentioned I had multiple machines with this problem. Some running
different versions of SuSE. Mainly 11.0, which is where all the info
I've provided came from thus far. This machine also has a SuSE-11.2
disk on it. When I do
ael wrote:
Alain Knaff wrote:
Could you try the same with a higher repetition count:
On same floppy (medium) as before:
[...]
All sector ids seem to be present (although occasionally they are
skipped during read...), and track is correct everywhere... but if I
remember correctly, you got
Alain Knaff wrote:
ael wrote:
Mark Hounschell wrote:
On 12/15/2009 10:08 AM, Alain Knaff wrote:
I mentioned I had multiple machines with this problem. Some running
different versions of SuSE. Mainly 11.0, which is where all the info
I've provided came from thus far. This machine also has a
Alain Knaff wrote:
Also, on your case, the actual read error seems to be on track 0. Could you
give me also the headers of track 0?
# fdrawcmd seek 0 0
0: 20
1: 0
no disk change
# fdrawcmd readid 0 repeat=40
0: 0
1: 0
2: 0
3: 0
4: 0
5: 9
6: 2
no disk change
0: 0
1: 0
2: 0
3: 0
4: 0
5: a
6:
ael wrote:
Is that what you wanted?
ael
Yes. All sectors are there, ... so I wonder why you are getting errors.
So, next round of tests: trying to read these sectors:
fdrawcmd recalibrate 0
fdrawcmd read 0 0 0 1 2 18 1 1 length=18432 /dev/null
Alain
--
To UNSUBSCRIBE, email to
Alain Knaff wrote:
ael wrote:
Is that what you wanted?
ael
Yes. All sectors are there, ... so I wonder why you are getting errors.
So, next round of tests: trying to read these sectors:
fdrawcmd recalibrate 0
fdrawcmd read 0 0 0 1 2 18 1 1 length=18432 /dev/null
# fdrawcmd
A.E.Lawrence wrote:
# fdrawcmd read 0 0 0 1 2 18 1 1 length=18432 /dev/null
remaining= 17920
0: 40 == So this is Abnormal termination?
1: 20 == CRC error? (id or data)
2: 20 == CRC error? (data)
Did I decode them correctly?
ael
--
To UNSUBSCRIBE, email to
A.E.Lawrence wrote:
Alain Knaff wrote:
ael wrote:
Is that what you wanted?
ael
Yes. All sectors are there, ... so I wonder why you are getting errors.
So, next round of tests: trying to read these sectors:
fdrawcmd recalibrate 0
fdrawcmd read 0 0 0 1 2 18 1 1 length=18432 /dev/null
ael wrote:
A.E.Lawrence wrote:
# fdrawcmd read 0 0 0 1 2 18 1 1 length=18432 /dev/null
remaining= 17920
0: 40 == So this is Abnormal termination?
1: 20 == CRC error? (id or data)
2: 20 == CRC error? (data)
Did I decode them correctly?
ael
Yes, that's correct.
Alain
--
To
Alain Knaff wrote:
A.E.Lawrence wrote:
Alain Knaff wrote:
ael wrote:
Is that what you wanted?
ael
Yes. All sectors are there, ... so I wonder why you are getting errors.
So, next round of tests: trying to read these sectors:
fdrawcmd recalibrate 0
fdrawcmd read 0 0 0 1 2 18 1 1
Alain Knaff wrote:
fdrawcmd read 0 0 0 1 2 18 1 1 length=512 /dev/null
fdrawcmd read 0 0 0 2 2 18 1 1 length=512 /dev/null
fdrawcmd read 0 0 0 3 2 18 1 1 length=512 /dev/null
...
fdrawcmd read 0 0 0 18 2 18 1 1 length=512 /dev/null
fdrawcmd read 4 0 1 1 2 18 1 1 length=512
Alain Knaff wrote:
Mark Hounschell wrote:
[...]
All kernels below 2.6.28 work on these boxes. All kernels 2.6.28 and higher do
NOT.
2.6.28 was when support for sector bases other than 0 or 1 were
introduced. So, rather than have sectors numbered from 1 to 18, you can
now have sectors
Alain Knaff wrote:
Mark Hounschell wrote:
[...]
All kernels below 2.6.28 work on these boxes. All kernels 2.6.28 and higher do
NOT.
2.6.28 was when support for sector bases other than 0 or 1 were
introduced. So, rather than have sectors numbered from 1 to 18, you can
now have sectors
ael wrote:
Alain Knaff wrote:
then do a getfdprm -o /dev/fd0u1440 (or getfdprm /dev/fd0)
I should have mentioned that my tests were done under the current git
kernel:
# uname -a
Linux exact 2.6.32_exact-55846-gf405425 #194 Sun Dec 13 16:30:46 GMT
2009 i686 GNU/Linux
ael
--
To
On 14/12/09 12:27, ael wrote:
# getfdprm -o /dev/fd0u1440
2880 18 2 80 0 0x1b 0x00 0xcf 0x6c
# fdrawcmd drive=/dev/fd0u1440 readid 0 repeat=18
raw cmd: Invalid argument
... and if you try with /dev/fd0 instead?
Alain
--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
ael wrote:
Alain Knaff wrote:
Mark Hounschell wrote:
[...]
All kernels below 2.6.28 work on these boxes. All kernels 2.6.28 and
higher do NOT.
2.6.28 was when support for sector bases other than 0 or 1 were
introduced. So, rather than have sectors numbered from 1 to 18, you can
now have
Alain Knaff wrote:
On 14/12/09 12:27, ael wrote:
# getfdprm -o /dev/fd0u1440
2880 18 2 80 0 0x1b 0x00 0xcf 0x6c
# fdrawcmd drive=/dev/fd0u1440 readid 0 repeat=18
raw cmd: Invalid argument
... and if you try with /dev/fd0 instead?
Yes. I tried all the obvious things. The man/info page for
On 14/12/09 15:58, ael wrote:
Any point in running under strace?
Yes, this would be useful, especially for analyzing the Invalid argument
issue.
Regards,
Alain
--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact
ael wrote:
Alain Knaff wrote:
On 14/12/09 12:27, ael wrote:
# getfdprm -o /dev/fd0u1440
2880 18 2 80 0 0x1b 0x00 0xcf 0x6c
# fdrawcmd drive=/dev/fd0u1440 readid 0 repeat=18
raw cmd: Invalid argument
... and if you try with /dev/fd0 instead?
Yes. I tried all the obvious things. The
Alain Knaff wrote:
On 14/12/09 15:58, ael wrote:
Any point in running under strace?
Yes, this would be useful, especially for analyzing the Invalid argument
issue.
Ok, Will try and fit it in later today. Meanwhile, what exactly is the
command line that should work? The one you suggested
Alain Knaff wrote:
On 14/12/09 15:58, ael wrote:
Any point in running under strace?
Yes, this would be useful, especially for analyzing the Invalid argument
issue.
Looks as if that was something to do with my command line. Below is the
strace giving the IO error which probably isn't much
Alain Knaff wrote:
On 14/12/09 16:24, ael wrote:
Alain Knaff wrote:
On 14/12/09 15:58, ael wrote:
Any point in running under strace?
Yes, this would be useful, especially for analyzing the Invalid
argument
issue.
Looks as if that was something to do with my command line. Below is the
On 14/12/09 16:24, ael wrote:
Alain Knaff wrote:
On 14/12/09 15:58, ael wrote:
Any point in running under strace?
Yes, this would be useful, especially for analyzing the Invalid
argument
issue.
Looks as if that was something to do with my command line. Below is the
strace giving the IO
32 matches
Mail list logo
