Mitsumi CDROM, CMD640, EIDE Problem

Mitsumi CDROM, CMD640, EIDE Problem

Post by Andreas Kauffman » Thu, 08 Feb 1996 04:00:00



I have
   CMD 640 Chipset      (I think, because thats the name of its Dos-driver)
   Mitsumi FX400 CD-Rom as master on ide1
   Conner 1.2 GB on ide0 (forgot the name)
Kernel: aidecd from Slackware 3.0

The Problem:
Mounting the CD-Rom needs a looong time.
In that time it's possible to switch between consoles
with <Alt> <F*>, but commands will be done when mounting
is done, not before - so it's no real multitasking what happens.
Copying single files from CD-Rom seems to work fine.
Copying multiple files does maybe the same as mounting -
I don't know: I lost patience and did a hard reboot after
a few minutes.
Running scripts or binaries from CD-Rom gives the same behaviour.

Giving as boot-parameters 'hdc=cdrom' or 'hdc=serialize'
makes no visible difference.
I'm booting with LILO. Should I try LOADLIN from Dos ?
Should I use other kernels ?
Any other ideas about that problem ?

Hoping for hints
        Andreas Kauffmann

------------------------------------
P.S.:  That's what 'dmesg' says:

ide0: hda: cyls=2477,heads=16,sects=63
ide1: hdc: serialize
Console: 16 point font, 400 scans
Console: colour VGA+ 80x25, 1 virtual console (max 63)
pcibios_init : BIOS32 Service Directory structure at 0x000fb8c0
pcibios_init : BIOS32 Service Directory entry at 0xfbd50
pcibios_init : PCI BIOS revision 2.10 entry at 0xfbd80
Probing PCI hardware.
Warning : Unknown PCI device.  Please read include/linux/pci.h
Calibrating delay loop.. <snip>
...
hda: Conner Peripherals 1275MB - CFS1275A, 1219MB w/64KB Cache, LBA,
CHS=2477/16
/63, MaxMult=16
hdc: FX001DE, ATAPI, CDROM drive
ide1: secondary interface on irq 15 (single-threaded with ide0)
ide0: primary interface on irq 14
Floppy drive(s): fd0 is 1.44M
Floppy drive(s): fd0 is 1.44M
FDC 0 is a post-1991 82077
aha152x: porttest: failed
EATA0: address 0x1f0 in use, skipping probe.
EATA0: address 0x170 in use, skipping probe.
scsi : 0 hosts.
scsi : detected total.
... <snip>
...

03:38:24
PDT 1995
Partition check:
  hda: hda1 hda2 < hda5 hda6 hda7 hda8 hda9 >
VFS: Mounted root (ext2 filesystem) readonly.
Adding Swap: 102780k swap-space
hdc: media changed
hdc: code: 70  key: 6  asc: 29  ascq: 0
hdc: media changed
hdc: code: 70  key: 6  asc: 28  ascq: 0
VFS: Disk change detected on device 22/0
  hdc:ll_rw_block: only 2048-char blocks implemented (1024)
 unable to read partition table
Max size:340158   Log zone size:2048
First datazone:36   Root inode number 73728
ISO9660 Extensions: RRIP_1991A
found on /dev/hdc.
    Mounting unknown ISO9660 CD-ROM under /cdrom.
Running gpm...

The last 3 lines are not from 'dmesg', but are also written
when booting, just before the login-prompt.

 
 
 

Mitsumi CDROM, CMD640, EIDE Problem

Post by Andreas Kauffman » Fri, 09 Feb 1996 04:00:00



> I have
>    CMD 640 Chipset      (I think, because thats the name of its Dos-driver)
>    Mitsumi FX400 CD-Rom as master on ide1
>    Conner 1.2 GB on ide0 (forgot the name)
> Kernel: aidecd from Slackware 3.0

> The Problem:
> Mounting the CD-Rom needs a looong time.
> In that time it's possible to switch between consoles
> with <Alt> <F*>, but commands will be done when mounting
> is done, not before - so it's no real multitasking what happens.
> Copying single files from CD-Rom seems to work fine.
> Copying multiple files does maybe the same as mounting -
> I don't know: I lost patience and did a hard reboot after
> a few minutes.
> Running scripts or binaries from CD-Rom gives the same behaviour.

> Giving as boot-parameters 'hdc=cdrom' or 'hdc=serialize'
> makes no visible difference.
> I'm booting with LILO. Should I try LOADLIN from Dos ?
> Should I use other kernels ?
> Any other ideas about that problem ?

> Hoping for hints
>         Andreas Kauffmann

> ------------------------------------
> P.S.:  That's what 'dmesg' says:

> ide0: hda: cyls=2477,heads=16,sects=63
> ide1: hdc: serialize
> Console: 16 point font, 400 scans
> Console: colour VGA+ 80x25, 1 virtual console (max 63)
> pcibios_init : BIOS32 Service Directory structure at 0x000fb8c0
> pcibios_init : BIOS32 Service Directory entry at 0xfbd50
> pcibios_init : PCI BIOS revision 2.10 entry at 0xfbd80
> Probing PCI hardware.
> Warning : Unknown PCI device.  Please read include/linux/pci.h
> Calibrating delay loop.. <snip>
> ...
> hda: Conner Peripherals 1275MB - CFS1275A, 1219MB w/64KB Cache, LBA,
> CHS=2477/16
> /63, MaxMult=16
> hdc: FX001DE, ATAPI, CDROM drive
> ide1: secondary interface on irq 15 (single-threaded with ide0)
> ide0: primary interface on irq 14
> Floppy drive(s): fd0 is 1.44M
> Floppy drive(s): fd0 is 1.44M
> FDC 0 is a post-1991 82077
> aha152x: porttest: failed
> EATA0: address 0x1f0 in use, skipping probe.
> EATA0: address 0x170 in use, skipping probe.
> scsi : 0 hosts.
> scsi : detected total.
> ... <snip>
> ...

> 03:38:24
> PDT 1995
> Partition check:
>   hda: hda1 hda2 < hda5 hda6 hda7 hda8 hda9 >
> VFS: Mounted root (ext2 filesystem) readonly.
> Adding Swap: 102780k swap-space
> hdc: media changed
> hdc: code: 70  key: 6  asc: 29  ascq: 0
> hdc: media changed
> hdc: code: 70  key: 6  asc: 28  ascq: 0
> VFS: Disk change detected on device 22/0
>   hdc:ll_rw_block: only 2048-char blocks implemented (1024)
>  unable to read partition table
> Max size:340158   Log zone size:2048
> First datazone:36   Root inode number 73728
> ISO9660 Extensions: RRIP_1991A
> found on /dev/hdc.
>     Mounting unknown ISO9660 CD-ROM under /cdrom.
> Running gpm...

> The last 3 lines are not from 'dmesg', but are also written
> when booting, just before the login-prompt.

I think I solved my problem by using the kernel 1.3.50.
Now everything seems to work.    (Hurra !)

Good luck for all with similar problems -
        Andreas Kauffmann

---------------------------------------------------------------------
P.S.: (When compiling the kernel there occured an error in connection
with './drivers/char/consolemap.c' - something was not found.
I tried to use 'consolemap.c' & Co from other kernels -
of course that didn't work.
The I tried 'make' again and it worked - don't know why,
but won't complain about that.)

 
 
 

1. Mitsumi CDROM and CMD640 problems

I have a problem with my CDROM. I have realised that I probably
have the worst combination of hardware one can have.

What I have:
A buggy CMD640 chipset (on a PCI-card).
2 Quantum 730MB on ide0
1 Mitsumi FX400 as master on ide1

After hours of work I managed to get the CDROM online if I use
kernel version 1.2.13 and "hdc=serialize". Everything looks fine,
and I can read files with 'less' etc.

But, if I do something like 'cat * >/dev/null' I get an immediate
kernel lockup, no messages at all. I guess one explanation could be
the number of interrupts, another simultaneous interrupts on ide0 and
ide1. Anyhow, I havn't managed to solve the problem.

Is there perhaps a 1.3 kernel which works with this configuration ?
(1.3.59 doesn't). Or is there any other solutions ? Everything works
under things like DOS and Windows95.

/Per Fremrot

--
__________________________________________________________________________

School of Electrical Engineering       | WWW: http://www.efd.lth.se/~e92pf
Lund Institute of Technology           | Phone +46-(0)46-390637

2. PAM and Debian Slink Dist

3. KERNEL PANIC: Linux CDROM driver fails badly for Mitsumi FX400 with CMD640 IDEs

4. Printing X-Windows fonts?

5. CDROM / CMD640 EIDE controller / etc

6. Can't install LILO from RH 5.2

7. Problems with DCL 2278 EIDE and Mitsumi 4x cdrom drive

8. Mandrake download question

9. Problems with CMD640 EIDE Controler.

10. HP Vectra & Mitsumi EIDE FX400 CDROM

11. SB CDROM -ATAPI, PnP, or CMD640 problem???

12. EIDE & CMD CONTRLL & EIDE CDROM & KERN 1.2.9