summaryrefslogtreecommitdiffstats
path: root/Documentation/cdrom/cdu31a
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 15:20:36 -0700
committerLinus Torvalds <torvalds@ppc970.osdl.org>2005-04-16 15:20:36 -0700
commit1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 (patch)
tree0bba044c4ce775e45a88a51686b5d9f90697ea9d /Documentation/cdrom/cdu31a
downloadtalos-obmc-linux-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.tar.gz
talos-obmc-linux-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.zip
Linux-2.6.12-rc2v2.6.12-rc2
Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
Diffstat (limited to 'Documentation/cdrom/cdu31a')
-rw-r--r--Documentation/cdrom/cdu31a196
1 files changed, 196 insertions, 0 deletions
diff --git a/Documentation/cdrom/cdu31a b/Documentation/cdrom/cdu31a
new file mode 100644
index 000000000000..c0667da09c00
--- /dev/null
+++ b/Documentation/cdrom/cdu31a
@@ -0,0 +1,196 @@
+
+ CDU31A/CDU33A Driver Info
+ -------------------------
+
+Information on the Sony CDU31A/CDU33A CDROM driver for the Linux
+kernel.
+
+ Corey Minyard (minyard@metronet.com)
+
+ Colossians 3:17
+
+Crude Table of Contents
+-----------------------
+
+ Setting Up the Hardware
+ Configuring the Kernel
+ Configuring as a Module
+ Driver Special Features
+
+
+This device driver handles Sony CDU31A/CDU33A CDROM drives and
+provides a complete block-level interface as well as an ioctl()
+interface as specified in include/linux/cdrom.h). With this
+interface, CDROMs can be accessed, standard audio CDs can be played
+back normally, and CD audio information can be read off the drive.
+
+Note that this will only work for CDU31A/CDU33A drives. Some vendors
+market their drives as CDU31A compatible. They lie. Their drives are
+really CDU31A hardware interface compatible (they can plug into the
+same card). They are not software compatible.
+
+Setting Up the Hardware
+-----------------------
+
+The CDU31A driver is unable to safely tell if an interface card is
+present that it can use because the interface card does not announce
+its presence in any way besides placing 4 I/O locations in memory. It
+used to just probe memory and attempt commands, but Linus wisely asked
+me to remove that because it could really screw up other hardware in
+the system.
+
+Because of this, you must tell the kernel where the drive interface
+is, what interrupts are used, and possibly if you are on a PAS-16
+soundcard.
+
+If you have the Sony CDU31A/CDU33A drive interface card, the following
+diagram will help you set it up. If you have another card, you are on
+your own. You need to make sure that the I/O address and interrupt is
+not used by another card in the system. You will need to know the I/O
+address and interrupt you have set. Note that use of interrupts is
+highly recommended, if possible, it really cuts down on CPU used.
+Unfortunately, most soundcards do not support interrupts for their
+CDROM interfaces. By default, the Sony interface card comes with
+interrupts disabled.
+
+ +----------+-----------------+----------------------+
+ | JP1 | 34 Pin Conn | |
+ | JP2 +-----------------+ |
+ | JP3 |
+ | JP4 |
+ | +--+
+ | | +-+
+ | | | | External
+ | | | | Connector
+ | | | |
+ | | +-+
+ | +--+
+ | |
+ | +--------+
+ | |
+ +------------------------------------------+
+
+ JP1 sets the Base Address, using the following settings:
+
+ Address Pin 1 Pin 2
+ ------- ----- -----
+ 0x320 Short Short
+ 0x330 Short Open
+ 0x340 Open Short
+ 0x360 Open Open
+
+ JP2 and JP3 configure the DMA channel; they must be set the same.
+
+ DMA Pin 1 Pin 2 Pin 3
+ --- ----- ----- -----
+ 1 On Off On
+ 2 Off On Off
+ 3 Off Off On
+
+ JP4 Configures the IRQ:
+
+ IRQ Pin 1 Pin 2 Pin 3 Pin 4
+ --- ----- ----- ----- -----
+ 3 Off Off On Off
+ 4 Off Off* Off On
+ 5 On Off Off Off
+ 6 Off On Off Off
+
+ The documentation states to set this for interrupt
+ 4, but I think that is a mistake.
+
+Note that if you have another interface card, you will need to look at
+the documentation to find the I/O base address. This is specified to
+the SLCD.SYS driver for DOS with the /B: parameter, so you can look at
+you DOS driver setup to find the address, if necessary.
+
+Configuring the Kernel
+----------------------
+
+You must tell the kernel where the drive is at boot time. This can be
+done at the Linux boot prompt, by using LILO, or by using Bootlin.
+Note that this is no substitute for HOWTOs and LILO documentation, if
+you are confused please read those for info on bootline configuration
+and LILO.
+
+At the linux boot prompt, press the ALT key and add the following line
+after the boot name (you can let the kernel boot, it will tell you the
+default boot name while booting):
+
+ cdu31a=<base address>,<interrupt>[,PAS]
+
+The base address needs to have "0x" in front of it, since it is in
+hex. For instance, to configure a drive at address 320 on interrupt 5,
+use the following:
+
+ cdu31a=0x320,5
+
+I use the following boot line:
+
+ cdu31a=0x1f88,0,PAS
+
+because I have a PAS-16 which does not support interrupt for the
+CDU31A interface.
+
+Adding this as an append line at the beginning of the /etc/lilo.conf
+file will set it for lilo configurations. I have the following as the
+first line in my lilo.conf file:
+
+ append="cdu31a=0x1f88,0"
+
+I'm not sure how to set up Bootlin (I have never used it), if someone
+would like to fill in this section please do.
+
+
+Configuring as a Module
+-----------------------
+
+The driver supports loading as a module. However, you must specify
+the boot address and interrupt on the boot line to insmod. You can't
+use modprobe to load it, since modprobe doesn't support setting
+variables.
+
+Anyway, I use the following line to load my driver as a module
+
+ /sbin/insmod /lib/modules/`uname -r`/misc/cdu31a.o cdu31a_port=0x1f88
+
+You can set the following variables in the driver:
+
+ cdu31a_port=<I/O address> - sets the base I/O. If hex, put 0x in
+ front of it. This must be specified.
+
+ cdu31a_irq=<interrupt> - Sets the interrupt number. Leaving this
+ off will turn interrupts off.
+
+
+Driver Special Features
+-----------------------
+
+This section describes features beyond the normal audio and CD-ROM
+functions of the drive.
+
+2048 byte buffer mode
+
+If a disk is mounted with -o block=2048, data is copied straight from
+the drive data port to the buffer. Otherwise, the readahead buffer
+must be involved to hold the other 1K of data when a 1K block
+operation is done. Note that with 2048 byte blocks you cannot execute
+files from the CD.
+
+XA compatibility
+
+The driver should support XA disks for both the CDU31A and CDU33A. It
+does this transparently, the using program doesn't need to set it.
+
+Multi-Session
+
+A multi-session disk looks just like a normal disk to the user. Just
+mount one normally, and all the data should be there. A special
+thanks to Koen for help with this!
+
+Raw sector I/O
+
+Using the CDROMREADAUDIO it is possible to read raw audio and data
+tracks. Both operations return 2352 bytes per sector. On the data
+tracks, the first 12 bytes is not returned by the drive and the value
+of that data is indeterminate.
OpenPOWER on IntegriCloud