Bug 3755 - CDROM does not mount/ irq timeout/ only with ide-cdrom, not ide-scsi
Summary: CDROM does not mount/ irq timeout/ only with ide-cdrom, not ide-scsi
Status: REJECTED DUPLICATE of bug 3362
Alias: None
Product: IO/Storage
Classification: Unclassified
Component: IDE (show other bugs)
Hardware: i386 Linux
: P2 normal
Assignee: Bartlomiej Zolnierkiewicz
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2004-11-16 19:14 UTC by Paul Johnson
Modified: 2005-02-08 15:44 UTC (History)
0 users

See Also:
Kernel Version: 2.6.9-1.667
Subsystem:
Regression: ---
Bisected commit-id:


Attachments

Description Paul Johnson 2004-11-16 19:14:50 UTC
Distribution: Fedora Core 3
Hardware Environment: Dell Inspiron 8600 laptop, NEC ND-5100A cd-dvd+rw in a
"modular bay" (I *believe* it is a USB device, system sees as ide/ATAPI)
Software Environment:linux kernel

Problem Description: Can't mount CD-Rs that I write in Fedora Core 2 systems (or
any linux system, as far as I can see).  I can start the FC3 install from CD-R
disk and it runs OK, but after that install finishes, the same CD-R won't mount
with default kernel settings.  If I start with kernel option hdc=ide-scsi, I can
write CD's with ide-cdrom, but cannot write with ide-scsi.

I've been at this for so long that I'm dazed and confused, but I now know many
details about what does and does not work.

1. With default kernel settings or hdc=ide-cd (which I think are equivalent)
CD-R mounting fails. /var/log/messages says this, over and over

kernel: hdc: irq timeout: status=0xd0 { Busy }
Nov 16 20:18:00 pols111 kernel: hdc: irq timeout: error=0xd0LastFailedSense 0x0d 
Nov 16 20:18:01 pols111 kernel: hdc: ATAPI reset complete
Nov 16 20:18:21 pols111 kernel: hdc: irq timeout: status=0xd0 { Busy }
Nov 16 20:18:21 pols111 kernel: hdc: irq timeout: error=0xd0LastFailedSense 0x0d 
Nov 16 20:18:21 pols111 kernel: hdc: ATAPI reset complete

Note: The same disk will mount in MS windows XP on same machine

2. Factory-made CD's can be loaded with default settings or with hdc=ide-cd

3. If I boot with hdc=ide-scsi, then my CD-R's can be loaded, so can the
"factory written disks"

Steps to reproduce: 
1. Start system 
2. Insert disk
3. Watch output from tail -f /var/log/messages in horror.


If I start with no kernel option or hdc=ide-cd, the startup messages say:

Nov 16 19:40:04 pols111 kernel:     ide0: BM-DMA at 0xbfa0-0xbfa7, BIOS
settings: hda:DMA, hdb:p
io
Nov 16 19:40:04 pols111 kernel:     ide1: BM-DMA at 0xbfa8-0xbfaf, BIOS
settings: hdc:DMA, hdd:p
io
Nov 16 19:40:04 pols111 kernel: hda: IC25N060ATMR04-0, ATA DISK drive
Nov 16 19:40:04 pols111 kernel: Using cfq io scheduler
Nov 16 19:40:04 pols111 kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Nov 16 19:40:04 pols111 kernel: hdc: _NEC DVD+RW ND-5100A, ATAPI CD/DVD-ROM drive
Nov 16 19:40:05 pols111 mDNSResponder:  startup succeeded
Nov 16 19:40:05 pols111 kernel: ide1 at 0x170-0x177,0x376 on irq 15
Nov 16 19:40:05 pols111 smartd[2502]: smartd version 5.33
[i386-redhat-linux-gnu] Copyright (C)
2002-4 Bruce Allen
Nov 16 19:40:05 pols111 smartd[2502]: Home page is
http://smartmontools.sourceforge.net/
Nov 16 19:40:05 pols111 smartd[2502]: Opened configuration file /etc/smartd.conf
Nov 16 19:40:05 pols111 smartd[2502]: Configuration file /etc/smartd.conf parsed.
Nov 16 19:40:05 pols111 smartd[2502]: Device: /dev/hda, opened
Nov 16 19:40:05 pols111 smartd[2502]: Device: /dev/hda, found in smartd database.
Nov 16 19:40:05 pols111 kernel: hda: max request size: 1024KiB
Nov 16 19:40:05 pols111 kernel: hda: 117210240 sectors (60011 MB) w/7884KiB
Cache, CHS=16383/255
/63, UDMA(100)
Nov 16 19:40:05 pols111 smartd[2502]: Device: /dev/hda, is SMART capable. Adding
to "monitor" li
st.
Nov 16 19:40:05 pols111 kernel:  hda: hda1 hda2 hda3 hda4 < hda5 hda6 hda7 hda8
hda9 hda10 hda11
 hda12 hda13 hda14 >
Nov 16 19:40:05 pols111 smartd[2502]: Monitoring 1 ATA and 0 SCSI devices
Nov 16 19:40:05 pols111 kernel: hdc: ATAPI 24X DVD-ROM CD-R/RW drive, 2048kB
Cache, UDMA(33)
Nov 16 19:40:06 pols111 smartd[2504]: smartd has fork()ed into background mode.
New PID=2504.
Nov 16 19:40:06 pols111 kernel: Uniform CD-ROM driver Revision: 3.20

That looks right, doesn't it?  But it times out reading CD-R's.




If I mount with hdc=ide-scsi, (which allows CD-Rs to mount), I see this startup
info:

Nov 16 18:31:23 pols111 kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Nov 16 18:31:24 pols111 kernel: hdc: _NEC DVD+RW ND-5100A, ATAPI CD/DVD-ROM drive
Nov 16 18:31:24 pols111 kernel: ide1 at 0x170-0x177,0x376 on irq 15
Nov 16 18:31:24 pols111 kernel: hda: max request size: 1024KiB
Nov 16 18:31:24 pols111 kernel: hda: 117210240 sectors (60011 MB) w/7884KiB
Cache, CHS=16383/255
/63, UDMA(100)
Nov 16 18:31:24 pols111 kernel:  hda: hda1 hda2 hda3 hda4 < hda5 hda6 hda7 hda8
hda9 hda10 hda11
 hda12 hda13 hda14 >
Nov 16 18:31:24 pols111 kernel: ide-floppy driver 0.99.newide
Nov 16 18:31:24 pols111 kernel: usbcore: registered new driver hiddev
Nov 16 18:31:24 pols111 kernel: usbcore: registered new driver usbhid
Nov 16 18:31:24 pols111 kernel: drivers/usb/input/hid-core.c: v2.0:USB HID core
driver
Nov 16 18:31:24 pols111 kernel: mice: PS/2 mouse device common for all mice
Nov 16 18:31:24 pols111 kernel: input: AT Translated Set 2 keyboard on
isa0060/serio0
Nov 16 18:31:24 pols111 kernel: input: PS/2 Generic Mouse on isa0060/serio1
Nov 16 18:31:24 pols111 kernel: md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
Nov 16 18:31:24 pols111 kernel: NET: Registered protocol family 2
Nov 16 18:31:24 pols111 kernel: IP: routing cache hash table of 1024 buckets,
32Kbytes
Nov 16 18:31:24 pols111 kernel: TCP: Hash tables configured (established 32768
bind 9362)
Nov 16 18:31:24 pols111 kernel: Initializing IPsec netlink socket
Nov 16 18:31:24 pols111 kernel: NET: Registered protocol family 1
Nov 16 18:31:25 pols111 kernel: NET: Registered protocol family 17
Nov 16 18:31:25 pols111 kernel: ACPI: (supports S0 S1 S3 S4 S4bios S5)
Nov 16 18:31:25 pols111 kernel: ACPI wakeup devices:
Nov 16 18:31:25 pols111 kernel:  LID PBTN PCI0 USB0 USB1 USB2 USB3 MODM PCIE
Nov 16 18:31:25 pols111 kernel: Freeing unused kernel memory: 144k freed
Nov 16 18:31:25 pols111 kernel: kjournald starting.  Commit interval 5 seconds
Nov 16 18:31:25 pols111 kernel: EXT3-fs: mounted filesystem with ordered data mode.
Nov 16 18:31:25 pols111 kernel: SELinux:  Disabled at runtime.
Nov 16 18:31:25 pols111 kernel: SELinux:  Unregistering netfilter hooks
Nov 16 18:31:25 pols111 kernel: inserting floppy driver for 2.6.9-1.667

and after a bunch of stuff about ethernet and uhci, it redefines hdc
Nov 16 18:31:29 pols111 kernel: SCSI subsystem initialized
Nov 16 18:31:29 pols111 kernel: ide-scsi is deprecated for cd burning! Use
ide-cd and give dev=/
dev/hdX as device
Nov 16 18:31:29 pols111 kernel: scsi0 : SCSI host adapter emulation for IDE
ATAPI devices
Nov 16 18:31:29 pols111 kernel:   Vendor: _NEC      Model: DVD+RW ND-5100A  
Rev: 10AC
Nov 16 18:31:29 pols111 kernel:   Type:   CD-ROM                            
ANSI SCSI revision:
 02
Nov 16 18:31:29 pols111 kernel: sr0: scsi3-mmc drive: 24x/24x writer cd/rw
xa/form2 cdda tray
Nov 16 18:31:29 pols111 kernel: Uniform CD-ROM driver Revision: 3.20
Comment 1 Paul Johnson 2004-11-18 14:02:55 UTC
In fedora-test email list, I asked about this one and Alan Cox guessed correctly
the problem is in the kernel and/or the Fedora hald (media auto loader thingie).

Alan wrote: "The problem is that some of the other file system code looks
at the _end_ of the disk. Unfortunately CD-R doesn't have an "end" it 
sort of tails out into no-mans land. Most drives politely handle such
requests, some don't. The ide-scsi driver has the needed magic to work 
around some of this, ide-cd lacks it."  There's some discussion (over my head)
about how this can be addressed either in hal or in ide-cd code in the kernel.
Comment 2 Paul Johnson 2004-11-18 14:04:31 UTC
Submitted before I should have.  

The work-around for me is to stop haldaemon (if you are running it) and then mount 

#mount -t iso9660 /dev/cdrom /media/cdrecorder

Adding the type gives the system enough information to read the disk.
Comment 3 Bartlomiej Zolnierkiewicz 2005-02-08 15:44:06 UTC

*** This bug has been marked as a duplicate of 3362 ***

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