xref: /netbsd/sys/dev/isa/README.seagate (revision bf9ec67e)
1#	$NetBSD: README.seagate,v 1.2 1998/01/05 07:31:07 perry Exp $
2
3The hardware:
4
5The ST01/02, and Future Domain 950 are very simple SCSI controllers. They are
6not busmastering, so the processor must do all transfers a la IDE. They support
7blind transfer by adding wait states (up to a certain limit). Interrupt is
8generated for reconnect and parity errors (maybe also for some other events).
9
10The card consists of one command port that writes to scsi control lines, reads
11status lines, and a data port that read/writes to the 8 data lines. The address
12decoding gives both data and control ports large memory areas to a single
13port. This is used by the code.
14
15The ST01/02 differs from the FD950 in memory address location and SCSI id.
16
17Probing for the card:
18
19A card is recognized by comparing the BIOS signature with known signatures. A
20new card may not be recognized if the BIOS signature has changed. Please send
21new signatures to me.
22
23Driver function:
24
25A scsi command is sent to scsi_cmd function. The command is either placed in
26the queue or an retryable message is returned. The routine may wait for
27completion of the command depending on the supplied flags. A timer is started
28for every command placed in the queue. The commands are added in the order they
29are received. There is a possiblity to make all REQUEST SENSE commands be
30queued before all other commands, but I dont think it is a good thing (Linux
31do however use this).
32
33The card is mostly controlled by the sea_main function. It is called by
34scsi_cmd, the interrupt routine, and the timeout routine. The sea_main routine
35runs as long there are something to do (transfer data, issue queued commands,
36and handle reconnected commands).
37
38The data transfers may be done in two different ways: Blind and polled
39transfers. They differ in the way the driver does it handshaking with the
40target. During a blind transfer, the driver code blindly transfers a block
41of data without checking for changed phase. During polled transfers, the
42phase is checked between every character transfered. The polled transfers
43are always used for status information and command transfers.
44
45Because the card does not use dma in any way, there is no need to handle
46physical addresses. There is no problem with the isa-bus address limit of
4716MB, making bounce-buffers unnecessary.
48
49The data structures:
50
51Every card has a sea_softc structure keeping the queues of commands waiting to
52be issued, and commands currently disconnected. The type of card (Seagate or
53Future Domain), data and control port addresses, scsi id, busy flags for all
54possible targets, and interrupt vector for the card.
55
56Every scsi command to be issued are stored in a sea_scb structure. It contains
57a flag describing status/error of the command, current data buffer position,
58and number of bytes remaining to be transfered.
59
60
61PROBLEMS
62
63I have had problems getting the ST02 boot using the boot floppies. I think it
64is some problem with BIOS calls not working. It is unfortunately impossible to
65disconnect the ST02 floppy controller.
66
67I have had problem to get the driver talk to a 40 MB Seagate disk. I dont have
68access to it any more, so I can't do any more checks on that.
69
70NOTE: The ST02 creates its own description of the disk attached. This is not
71the same as the disk says. This translation problem may cause problems when
72sharing a disk between both DOS and BSD. It is however not impossible.
73