| Borislav Petkov | 5ce78af | 2008-02-02 19:56:48 +0100 | [diff] [blame] | 1 | /* |
| 2 | * IDE ATAPI streaming tape driver. |
| 3 | * |
| 4 | * This driver is a part of the Linux ide driver. |
| 5 | * |
| 6 | * The driver, in co-operation with ide.c, basically traverses the |
| 7 | * request-list for the block device interface. The character device |
| 8 | * interface, on the other hand, creates new requests, adds them |
| 9 | * to the request-list of the block device, and waits for their completion. |
| 10 | * |
| Borislav Petkov | 5ce78af | 2008-02-02 19:56:48 +0100 | [diff] [blame] | 11 | * The block device major and minor numbers are determined from the |
| 12 | * tape's relative position in the ide interfaces, as explained in ide.c. |
| 13 | * |
| 14 | * The character device interface consists of the following devices: |
| 15 | * |
| 16 | * ht0 major 37, minor 0 first IDE tape, rewind on close. |
| 17 | * ht1 major 37, minor 1 second IDE tape, rewind on close. |
| 18 | * ... |
| 19 | * nht0 major 37, minor 128 first IDE tape, no rewind on close. |
| 20 | * nht1 major 37, minor 129 second IDE tape, no rewind on close. |
| 21 | * ... |
| 22 | * |
| 23 | * The general magnetic tape commands compatible interface, as defined by |
| 24 | * include/linux/mtio.h, is accessible through the character device. |
| 25 | * |
| 26 | * General ide driver configuration options, such as the interrupt-unmask |
| 27 | * flag, can be configured by issuing an ioctl to the block device interface, |
| 28 | * as any other ide device. |
| 29 | * |
| 30 | * Our own ide-tape ioctl's can be issued to either the block device or |
| 31 | * the character device interface. |
| 32 | * |
| 33 | * Maximal throughput with minimal bus load will usually be achieved in the |
| 34 | * following scenario: |
| 35 | * |
| 36 | * 1. ide-tape is operating in the pipelined operation mode. |
| 37 | * 2. No buffering is performed by the user backup program. |
| 38 | * |
| 39 | * Testing was done with a 2 GB CONNER CTMA 4000 IDE ATAPI Streaming Tape Drive. |
| 40 | * |
| 41 | * Here are some words from the first releases of hd.c, which are quoted |
| 42 | * in ide.c and apply here as well: |
| 43 | * |
| 44 | * | Special care is recommended. Have Fun! |
| 45 | * |
| Borislav Petkov | 5ce78af | 2008-02-02 19:56:48 +0100 | [diff] [blame] | 46 | * Possible improvements. |
| 47 | * |
| 48 | * 1. Support for the ATAPI overlap protocol. |
| 49 | * |
| 50 | * In order to maximize bus throughput, we currently use the DSC |
| 51 | * overlap method which enables ide.c to service requests from the |
| 52 | * other device while the tape is busy executing a command. The |
| 53 | * DSC overlap method involves polling the tape's status register |
| 54 | * for the DSC bit, and servicing the other device while the tape |
| 55 | * isn't ready. |
| 56 | * |
| 57 | * In the current QIC development standard (December 1995), |
| 58 | * it is recommended that new tape drives will *in addition* |
| 59 | * implement the ATAPI overlap protocol, which is used for the |
| 60 | * same purpose - efficient use of the IDE bus, but is interrupt |
| 61 | * driven and thus has much less CPU overhead. |
| 62 | * |
| 63 | * ATAPI overlap is likely to be supported in most new ATAPI |
| 64 | * devices, including new ATAPI cdroms, and thus provides us |
| 65 | * a method by which we can achieve higher throughput when |
| 66 | * sharing a (fast) ATA-2 disk with any (slow) new ATAPI device. |
| 67 | */ |