Bug 12092 - DVD-RW will not work with kernel 2.6.27
Summary: DVD-RW will not work with kernel 2.6.27
Status: RESOLVED CODE_FIX
Alias: None
Product: IO/Storage
Classification: Unclassified
Component: Serial ATA (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: Alan
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2008-11-23 23:30 UTC by Troy
Modified: 2009-04-03 09:23 UTC (History)
2 users (show)

See Also:
Kernel Version: 2.6.27.5-2
Subsystem:
Regression: No
Bisected commit-id:


Attachments
boot message (3.84 KB, text/plain)
2008-11-29 11:14 UTC, Troy
Details
Boot Message (20.44 KB, text/plain)
2008-11-29 16:07 UTC, Troy
Details
kernel log (72.72 KB, application/octet-stream)
2008-11-30 16:19 UTC, Troy
Details
New boot message (52.49 KB, text/plain)
2008-12-02 19:44 UTC, Troy
Details
lspci -vvxxx (37.74 KB, text/plain)
2008-12-04 18:01 UTC, Troy
Details

Description Troy 2008-11-23 23:30:32 UTC
Latest working kernel version: 2.6.27.5-2

Earliest failing kernel version: kernel-desktop586-2.6.27-0.rc8.2mnb

Distribution: Mandriva 2009 One

Hardware Environment: LG HL-DT-ST DVD-RAM GH22NP20 and Lite-on DVD-RW LH-20A1P

Software Environment: Mandriva 2009 One

Problem Description:

I have 2009 One and KDE 4 with all the latest updates

I'm using a LG HL-DT-ST DVD-RAM GH22NP20 and a Lite-on DVD-RW LH-20A1P for my DVD-RW needs

Ever since I upgraded to 2009 my DVD-RW player has not worked, the LG is the second DVD-RW I have tried the first one was a Lite-on so I figured that the new kernel did not work with a Lite-on DVD-RW so I bought the LG. But when I put in a CD-RW player it works just fine in 2009.

When I try to access a Data DVD-R, CD-R or CD-RW with Pictures that was burned using K3b in Mandriva 2008.1 the error is always "that it could not determine file system type and none was given" so after that the disc appears to be mounted but can not be accessed.

I also installed the Linux-Linus 2.6.27.5-1 kernel to no avail.

My DVD-RW worked just fine with no problems using the old kernel

Steps to reproduce: Install either the LG or Lite-on DVD-RW and try to use, they fail every time.

[root@localhost ~]# lspcidrake -v
Card:NVIDIA GeForce 3 to GeForce 4: nVidia Corporation|NV25 [GeForce4 Ti 4200] [DISPLAY_VGA] (vendor:10de device:0253)
ehci_hcd        : NEC Corporation|USB 2.0 [SERIAL_USB] (vendor:1033 device:00e0 subv:1631 subd:1600)
ohci_hcd        : NEC Corporation|USB [SERIAL_USB] (vendor:1033 device:0035 subv:1631 subd:0035)
ohci_hcd        : NEC Corporation|USB [SERIAL_USB] (vendor:1033 device:0035 subv:1631 subd:0035)
sis900          : Silicon Integrated Systems [SiS]|SiS900 PCI Fast Ethernet [NETWORK_ETHERNET] (vendor:1039 device:0900 subv:1019 subd:0a14)
snd_intel8x0    : Silicon Integrated Systems [SiS]|AC'97 Sound Controller [MULTIMEDIA_AUDIO] (vendor:1039 device:7012 subv:13f6 subd:0300)
pata_sis        : Silicon Integrated Systems [SiS]|5513 [IDE] [STORAGE_IDE] (vendor:1039 device:5513)
ohci_hcd        : Silicon Integrated Systems [SiS]|USB 1.1 Controller [SERIAL_USB] (vendor:1039 device:7001 subv:1019 subd:0a14)
ohci_hcd        : Silicon Integrated Systems [SiS]|USB 1.1 Controller [SERIAL_USB] (vendor:1039 device:7001 subv:1019 subd:0a14)
i2c_sis96x      : Silicon Integrated Systems [SiS]|SiS961/2 SMBus Controller [SERIAL_SMBUS] (vendor:1039 device:0016)
i2c_sis630      : Silicon Integrated Systems [SiS]|SiS85C503/5513 (LPC Bridge) [BRIDGE_ISA] (vendor:1039 device:0018)
shpchp          : Silicon Integrated Systems [SiS]|Virtual PCI-to-PCI bridge (AGP) [BRIDGE_PCI] (vendor:1039device:0001)
sis_agp         : Silicon Integrated Systems [SiS]|735 Host [BRIDGE_HOST] (vendor:1039 device:0735)
hub             : Linux 2.6.27.5-desktop586-2mnb ohci_hcd|OHCI Host Controller [Hub|Unused|Full speed (or root) hub] (vendor:1d6b device:0001)
hub             : Linux 2.6.27.5-desktop586-2mnb ohci_hcd|OHCI Host Controller [Hub|Unused|Full speed (or root) hub] (vendor:1d6b device:0001)
hub             : Linux 2.6.27.5-desktop586-2mnb ohci_hcd|OHCI Host Controller [Hub|Unused|Full speed (or root) hub] (vendor:1d6b device:0001)
hub             : Linux 2.6.27.5-desktop586-2mnb ohci_hcd|OHCI Host Controller [Hub|Unused|Full speed (or root) hub] (vendor:1d6b device:0001)
hub             : Linux 2.6.27.5-desktop586-2mnb ehci_hcd|EHCI Host Controller [Hub|Unused|Full speed (or root) hub] (vendor:1d6b device:0002)
Comment 1 Andrew Morton 2008-11-24 00:12:28 UTC
So which device driver is being used to drive that device? pata_sis?
Comment 2 Troy 2008-11-24 19:27:37 UTC
sorry about that

The driver is pata_sis
Comment 3 Tejun Heo 2008-11-25 21:46:34 UTC
Can you please post kernel log including the boot messages and, if any, the messages printed after cd/dvd access failure from the old working kernel and the new non-working one?  Thanks.
Comment 4 Troy 2008-11-28 12:52:26 UTC
Here is the info you requested

/var/log/syslog - Lite-on DVD-RW  report for kernel 2.6.27.5-2 

Nov 28 12:36:49 localhost kernel: sr 1:0:0:0: [sr0] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
Nov 28 12:36:49 localhost kernel: sr 1:0:0:0: [sr0] Sense Key : Hardware Error [current] 
Nov 28 12:36:49 localhost kernel: sr 1:0:0:0: [sr0] Add. Sense: Logical unit communication CRC error (Ultra-DMA/32)
Nov 28 12:36:49 localhost kernel: end_request: I/O error, dev sr0, sector 88
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 11
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 12
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 13
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 14
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 15
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 16
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 17
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 18
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 19
Nov 28 12:36:49 localhost kernel: Buffer I/O error on device sr0, logical block 20
Nov 28 12:36:54 localhost kernel: sr 1:0:0:0: [sr0] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
Nov 28 12:36:54 localhost kernel: sr 1:0:0:0: [sr0] Sense Key : Hardware Error [current] 
Nov 28 12:36:54 localhost kernel: sr 1:0:0:0: [sr0] Add. Sense: Logical unit communication CRC error (Ultra-DMA/32)
Nov 28 12:36:54 localhost kernel: end_request: I/O error, dev sr0, sector 8
Nov 28 12:36:54 localhost kernel: __ratelimit: 8 callbacks suppressed
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 1
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 2
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 3
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 4
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 5
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 6
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 7
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 8
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 9
Nov 28 12:36:54 localhost kernel: Buffer I/O error on device sr0, logical block 10
Nov 28 12:37:00 localhost kernel: sr 1:0:0:0: [sr0] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
Nov 28 12:37:00 localhost kernel: sr 1:0:0:0: [sr0] Sense Key : Hardware Error [current] 
Nov 28 12:37:00 localhost kernel: sr 1:0:0:0: [sr0] Add. Sense: Logical unit communication CRC error (Ultra-DMA/32)
Nov 28 12:37:00 localhost kernel: end_request: I/O error, dev sr0, sector 8
Nov 28 12:37:00 localhost kernel: __ratelimit: 6 callbacks suppressed
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 1
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 2
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 3
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 4
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 5
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 6
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 7
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 8
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 9
Nov 28 12:37:00 localhost kernel: Buffer I/O error on device sr0, logical block 10


/var/log/syslog - DVD-RW  report for kernel-desktop586 2.6.24.7-1

The Lite-on DVD-RW worked in this kernel but it did not post a message in the kernel log


/var/log/syslog – report from playing an audio CD using kernel 2.6.24.7-1 with Lite-on DVD-RW

Nov 28 12:01:01 localhost CROND[17982]: (root) CMD (nice -n 19 run-parts --report /etc/cron.hourly)
Nov 28 12:01:14 localhost kernel: Shorewall:net2fw:DROP:IN=eth0 OUT= MAC=00:07:95:36:b7:44:00:09:5b:0e:88:68:08:00 SRC=210.74.155.125 DST=192.168.0.3 LEN=48 TOS=0x00 PREC=0x00 TTL=106 ID=26742 PROTO=TCP SPT=48426 DPT=5900 WINDOW=16384 RES=0x00 SYN URGP=0 
Nov 28 12:01:17 localhost kernel: Shorewall:net2fw:DROP:IN=eth0 OUT= MAC=00:07:95:36:b7:44:00:09:5b:0e:88:68:08:00 SRC=210.74.155.125 DST=192.168.0.3 LEN=48 TOS=0x00 PREC=0x00 TTL=106 ID=28279 PROTO=TCP SPT=48426 DPT=5900 WINDOW=16384 RES=0x00 SYN URGP=0 


/var/log/syslog – error report from trying to play an audio CD using kernel 2.6.27.5-2 with Lite-on DVD-RW

Nov 28 12:25:29 localhost kernel: sr0: CDROM (ioctl) error, command: Volume set (in), Read cd be 00 00 00 00 29 00 00 01 f8 00 00
Nov 28 12:25:29 localhost kernel: sr: Sense Key : Hardware Error [current] 
Nov 28 12:25:29 localhost kernel: sr: Add. Sense: Logical unit communication CRC error (Ultra-DMA/32)



Boot message for kernel-desktop586 2.6.24.7-1

Starting irqbalance: #[65G[#[1;32m  OK  #[0;39m]
Starting acpi daemon: #[65G[#[1;32m  OK  #[0;39m]
Starting VirtualBox host to guest time synchronisation VirtualBox Additions module not loaded!
Starting atd: #[65G[#[1;32m  OK  #[0;39m]
ALSA driver is already running.Doing alsactl to restore mixer settings: #[65G[#[1;32m  OK  #[0;39m]
Starting numlock: #[65G[#[1;32m  OK  #[0;39m]
Starting system logger: #[65G[#[1;32m  OK  #[0;39m]
Starting kernel logger: #[65G[#[1;32m  OK  #[0;39m]
Starting jexec servicesChecking if partitions have enough free diskspace: #[65G[#[1;32m  OK  #[0;39m]
Starting crond: #[65G[#[1;32m  OK  #[0;39m]
Starting system message bus: #[65G[#[1;32m  OK  #[0;39m]
Starting display manager: #[65G[#[1;32m  OK  #[0;39m]
Starting resolvconf: #[65G[#[1;32m  OK  #[0;39m]
Bringing up loopback interface:  #[65G[#[1;32m  OK  #[0;39m]
Bringing up interface eth0:  #[65G[#[1;32m  OK  #[0;39m]
Starting HAL daemon: #[65G[#[1;32m  OK  #[0;39m]
Starting anacron: #[65G[#[1;32m  OK  #[0;39m]
Waiting for network to be up#[65G[#[1;32m  OK  #[0;39m]
Starting Avahi daemon: #[65G[#[1;32m  OK  #[0;39m]
Starting portreserve: #[65G[#[1;32m  OK  #[0;39m]
Mounting other filesystems:  #[65G[#[1;32m  OK  #[0;39m]
Loading keymap: us#[65G[#[1;32m  OK  #[0;39m]
The BackSpace key sends: ^?#[65G[#[1;32m  OK  #[0;39m]
Starting preload daemon: #[65G[#[1;32m  OK  #[0;39m]
Starting gnump3d:Unknown option: pid
  The server root directory you specified, /home/mp3, is missing.

  Please update your configuration file to specify the actual
 root directory you wish to serve media from.  

  You can fix this error by changing the line that currently
 reads:

   root = /home/mp3

#[65G[#[1;32m  OK  #[0;39m]
Adding loopback device to routing table ...
Loading parallel port printer kernel modules ...
Starting CUPS printing system: #[65G[#[1;32m  OK  #[0;39m]
Compiling...
Compiling /etc/shorewall/zones...
Compiling /etc/shorewall/interfaces...
Determining Hosts in Zones...
Preprocessing Action Files...
   Pre-processing /usr/share/shorewall/action.Drop...
   Pre-processing /usr/share/shorewall/action.Reject...
Compiling /etc/shorewall/policy...
Compiling Kernel Route Filtering...
Compiling Martian Logging...
Compiling MAC Filtration -- Phase 1...
Compiling /etc/shorewall/rules...
Generating Transitive Closure of Used-action List...
Processing /usr/share/shorewall/action.Reject for chain Reject...
Processing /usr/share/shorewall/action.Drop for chain Drop...
Compiling MAC Filtration -- Phase 2...
Applying Policies...
Generating Rule Matrix...
Creating iptables-restore input...
Shorewall configuration compiled to /var/lib/shorewall/.start
Processing /etc/shorewall/params ...
Starting Shorewall....
Initializing...
Processing /etc/shorewall/init ...
Setting up ARP filtering...
Setting up Route Filtering...
Setting up Martian Logging...
Setting up Accept Source Routing...
Setting up Proxy ARP...
Setting up Traffic Control...
Preparing iptables-restore input...
Running /sbin/iptables-restore...
Processing /etc/shorewall/start ...
Processing /etc/shorewall/started ...
done.
Starting mandi daemon: unable to open white list file
#[65G[#[1;32m  OK  #[0;39m]
Syncing time for ntpd: Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
#[65G[#[1;32m  OK  #[0;39m]
Starting ntpd: #[65G[#[1;32m  OK  #[0;39m]
Total time: 16 seconds


Boot message for kernel-desktop586 2.6.27.5-2

Starting irqbalance: #[65G[#[1;32m  OK  #[0;39m]
Starting acpi daemon: #[65G[#[1;32m  OK  #[0;39m]
Starting VirtualBox host to guest time synchronisation VirtualBox Additions module not loaded!
Starting atd: #[65G[#[1;32m  OK  #[0;39m]
ALSA driver is already running.Doing alsactl to restore mixer settings: #[65G[#[1;32m  OK  #[0;39m]
Starting numlock: #[65G[#[1;32m  OK  #[0;39m]
Starting system logger: #[65G[#[1;32m  OK  #[0;39m]
Starting kernel logger: #[65G[#[1;32m  OK  #[0;39m]
Checking if partitions have enough free diskspace: #[65G[#[1;32m  OK  #[0;39m]
Starting jexec servicesStarting system message bus: #[65G[#[1;32m  OK  #[0;39m]
Starting display manager: #[65G[#[1;32m  OK  #[0;39m]
Starting crond: #[65G[#[1;32m  OK  #[0;39m]
Starting resolvconf: #[65G[#[1;32m  OK  #[0;39m]
Bringing up loopback interface:  #[65G[#[1;32m  OK  #[0;39m]
Bringing up interface eth0:  #[65G[#[1;32m  OK  #[0;39m]
Starting HAL daemon: #[65G[#[1;32m  OK  #[0;39m]
Starting anacron: #[65G[#[1;32m  OK  #[0;39m]
Waiting for network to be up#[65G[#[1;32m  OK  #[0;39m]
Starting Avahi daemon: #[65G[#[1;32m  OK  #[0;39m]
Starting portreserve: #[65G[#[1;32m  OK  #[0;39m]
Mounting other filesystems:  #[65G[#[1;32m  OK  #[0;39m]
Starting preload daemon: #[65G[#[1;32m  OK  #[0;39m]
Loading keymap: us#[65G[#[1;32m  OK  #[0;39m]
The BackSpace key sends: ^?#[65G[#[1;32m  OK  #[0;39m]
Starting gnump3d:Unknown option: pid
  The server root directory you specified, /home/mp3, is missing.

  Please update your configuration file to specify the actual
 root directory you wish to serve media from.  

  You can fix this error by changing the line that currently
 reads:

   root = /home/mp3

#[65G[#[1;32m  OK  #[0;39m]
Adding loopback device to routing table ...
Loading parallel port printer kernel modules ...
Starting CUPS printing system: #[65G[#[1;32m  OK  #[0;39m]
Syncing time for ntpd: Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
Error : Name or service not known
#[65G[#[1;32m  OK  #[0;39m]
Starting ntpd: #[65G[#[1;32m  OK  #[0;39m]
Compiling...
Compiling /etc/shorewall/zones...
Compiling /etc/shorewall/interfaces...
Determining Hosts in Zones...
Preprocessing Action Files...
   Pre-processing /usr/share/shorewall/action.Drop...
   Pre-processing /usr/share/shorewall/action.Reject...
Compiling /etc/shorewall/policy...
Compiling Kernel Route Filtering...
Compiling Martian Logging...
Compiling MAC Filtration -- Phase 1...
Compiling /etc/shorewall/rules...
Generating Transitive Closure of Used-action List...
Processing /usr/share/shorewall/action.Reject for chain Reject...
Processing /usr/share/shorewall/action.Drop for chain Drop...
Compiling MAC Filtration -- Phase 2...
Applying Policies...
Generating Rule Matrix...
Creating iptables-restore input...
Shorewall configuration compiled to /var/lib/shorewall/.start
Processing /etc/shorewall/params ...
Starting Shorewall....
Initializing...
Processing /etc/shorewall/init ...
Setting up ARP filtering...
Setting up Route Filtering...
Setting up Martian Logging...
Setting up Accept Source Routing...
Setting up Proxy ARP...
Setting up Traffic Control...
Preparing iptables-restore input...
Running /sbin/iptables-restore...
Processing /etc/shorewall/start ...
Processing /etc/shorewall/started ...
done.
Starting mandi daemon: unable to open white list file
#[65G[#[1;32m  OK  #[0;39m]
Total time: 16 seconds
Comment 5 Tejun Heo 2008-11-29 02:20:26 UTC
Can you please attach (please don't post inline) /var/log/boot.msg.
Comment 6 Troy 2008-11-29 11:14:56 UTC
Created attachment 19074 [details]
boot message

Is this what your looking for?

This boot message is from kernel 2.6.27.5-2
Comment 7 Tejun Heo 2008-11-29 14:55:21 UTC
No, if you don't have boot.log on your distro, please do "dmesg > dmesg.out" right after boot and post the file.  Thanks.
Comment 8 Troy 2008-11-29 16:07:21 UTC
Created attachment 19076 [details]
Boot Message

Okay, this should be the one.
Comment 9 Tejun Heo 2008-11-30 02:04:49 UTC
1. Are you using 80c cable?

2. Can you please try "libata.force=2:udma33"?  Please note that if the driver is loaded from initrd, different distros use different methods to pass module parameters to initrd modules.  Please consult your distro for details.

3. Can you please post kernel log from the older (working) configuration?
Comment 10 Troy 2008-11-30 15:51:18 UTC
What cable are you talking about?

The IDE cable going from the DVD-RW to the mother board?
Comment 11 Troy 2008-11-30 16:19:42 UTC
Created attachment 19082 [details]
kernel log

This is the old kernel log right after reboot, the DVD-RW is working and was used to read a data DVD-R and it played an Audio CD
Comment 12 Tejun Heo 2008-11-30 16:38:54 UTC
Hmm... apparently something went wrong with cable detection.

 ata2.00: limited to UDMA/33 due to 40-wire cable

Alan, does anything ring a bell?  Older kernel is detecting 40c cable but newer kernel is detecting 80c.  The transmission errors are reported via sense data so libata never notices it. :-(
Comment 13 Alan 2008-12-01 02:01:23 UTC
The old cable detect logic was broken so we actually need to know what cable is *really* present.

The CRC reporting is an interesting twist 
Comment 14 Tejun Heo 2008-12-01 02:06:10 UTC
(In reply to comment #13)
> The old cable detect logic was broken so we actually need to know what cable
> is
> *really* present.

It's a laptop.  Maybe wrongly wired.  :-(

> The CRC reporting is an interesting twist 

Yeah, it's a long-standing FIXME in libata-eh logic.  libata eh currently doesn't interpret the sense data so if transport error is reported via SCSI sense code instead of ATA error or SError, it doesn't know that the failure is transport one and thus never gears down the transfer mode.
Comment 15 Troy 2008-12-01 11:02:46 UTC
(In reply to comment #14)
> It's a laptop.  Maybe wrongly wired.  :-(

If your talking about mine, it's a desktop that I built from scratch and it uses an IDE cable from the mother board to the internal DVD-RW.
Comment 16 Adam Williamson 2008-12-01 12:52:13 UTC
Troy, what they need to know is whether the cable is 40-conductor or 80-conductor. There's two types of IDE cable; both have 40 pins but old ones had only 40 conductors while newer ones have 80 conductors. 40 conductor cables can only manage up to 33MB/sec, while 80 conductor cables are needed for the (relatively) newer 66MB/sec, 100MB/sec and 133MB/sec UltraDMA modules. So that's why it's important, and that's why the module has to detect the cable. You should be able to tell which type it is from the number and diameter of actual individual wires which make up the cable - an IDE cable is a bunch of wires in grey (usually) shielding bonded together, an 80c cable will have 80 of these, a 40c cable will have 40.

The 'libata.force=2:udma33' thing is (presumably, right Tejun?) a module parameter Tejun wants you to try passing to pata_sis when it's loaded.

I might be wrong here - if it doesn't seem to change anything I'll check with the tools guys - but I believe if you just add this line to /etc/modprobe.conf:

options pata_sis libata.force=2:udma33

and then re-generate your initrd (with mkinitrd - if you just run 'mkinitrd' it'll give you an example command, you should either add -f to this command or delete the existing initrd file before running it), it should take effect at the next boot. Give that a shot.
Comment 17 Adam Williamson 2008-12-01 12:52:41 UTC
s/modules/modes/
Comment 18 Tejun Heo 2008-12-01 18:32:19 UTC
Somehow I read that the drive was directly connected to the motherboard and assumed that it was a laptop.  Adam, thanks a lot for clarifying what I tried to say.  Your explanation is correct in all accounts.  :-)
Comment 19 Troy 2008-12-01 22:42:02 UTC
Well, 

I have 40c cable

I also put "options pata_sis libata.force=2:udma33" into modprobe.conf, and the example command that mkinitrd provided was 
mkinitrd /boot/initrd-2.6.27.5-desktop586-2mnb.img 2.6.27.5-desktop586-2mnb -f

I then put -f after it and loaded it and then rebooted

but the boot failed right after loading pata_sis module and the error was

"FATAL: Error inserting pata_sis (/lib/modules/2.6.27.5-desktop586-2mnb/pata_sis.ko.gz): unknown symbol in module, or unknown parameter (see dmesg)

it tried to set up a file system setuproot but every /dev, /proc, /sys failed and said no such directory.

because of this error kernel 2.6.27.5-2 will not boot, I have to use the older kernel to use the computer.

it says to see dmesg, will it contain the info your looking for in the older kernel?
Comment 20 Tejun Heo 2008-12-02 00:17:29 UTC
So, the cable detection got broken somehow.  For the module loading error, "force" is not the module parameter of pata_sis, you need to specify "options libata force=2:udma33"
Comment 21 Troy 2008-12-02 10:34:07 UTC
I think I understand what your saying?

Are you saying that I need to add "options libata force=2:udma33" to the end of "mkinitrd /boot/initrd-2.6.27.5-desktop586-2mnb.img 2.6.27.5-desktop586-2mnb -f"

Or are you saying "-f" is not required?

Thanks again for your help
Comment 22 Tejun Heo 2008-12-02 18:26:49 UTC
The "options libata" line should go into modprobe.conf instead of the pata_sis line and I have no idea about the "-f" option as it depends on which distro you're using.
Comment 23 Troy 2008-12-02 19:44:24 UTC
Created attachment 19116 [details]
New boot message

I took out pata_sis from modprobe.conf and loaded mkinitrd and then rebooted.

kernel 2.6.27.5-2 works now but the DVD-RW still does not work.

So here is the new boot message
Comment 24 Tejun Heo 2008-12-02 19:54:53 UTC
Have you put the libata option line?  libata is not getting the force parameter.  When it does, it will whine loudly about it and configure the dvd drive to UDMA/33.
Comment 25 Troy 2008-12-02 22:40:10 UTC
I put this in modprobe.conf "options libata.force=2:udma33"

Then I loaded this in a console to re-generate initrd just like Adam said,

"mkinitrd /boot/initrd-2.6.27.5-desktop586-2mnb.img 2.6.27.5-desktop586-2mnb -f"
Comment 26 Tejun Heo 2008-12-03 00:55:07 UTC
The line should read "options libata force=2:udma33" without the dot.
Comment 27 Troy 2008-12-03 17:55:03 UTC
Thanks,

I took out the dot and it's now configured for udma33 and the DVD-RW works, I tried a DVD movie, data DVD-R and Audio CD every thing is good.

Thanks again for your help
Comment 28 Tejun Heo 2008-12-03 19:30:04 UTC
So, now we have to figure out what went wrong with cable detection.  Alan, do you want to take this from here?
Comment 29 Alan 2008-12-04 02:08:24 UTC
Yes.. will try and review that code next week
Comment 30 Alan 2008-12-04 03:40:09 UTC
Please attach an lspci -vvxxx
Comment 31 Troy 2008-12-04 18:01:23 UTC
Created attachment 19153 [details]
lspci -vvxxx

Here is the file you requested
Comment 32 Alan 2008-12-05 08:53:45 UTC
Thanks. So the hardware is reporting cable detect bits of 0x20 (Primary 80 wire, secondary 40 wire)

The PCI vendor identifiers match "Targa Visionary 1000' which is a laptop with known cable detect funnies so we ignored the BIOS.

Ok so the targa entry has to go and will need redoing some other way
Comment 33 Adam Williamson 2008-12-07 17:49:30 UTC
Alan, Tejun - thanks a bunch for your help with this issue.
Comment 34 Alan 2009-04-03 09:23:40 UTC
commit e33ba2c6c668a2be8f192b383186ee7cbb371da0

Note You need to log in before you can comment on or make changes to this bug.