Skip to content

L
linux
Commit d5b4bb4d authored by Linus Torvalds's avatar Linus Torvalds
Browse files
Options

Merge branch 'delete-mca' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux 

Pull the MCA deletion branch from Paul Gortmaker:
 "It was good that we could support MCA machines back in the day, but
  realistically, nobody is using them anymore.  They were mostly limited
  to 386-sx 16MHz CPU and some 486 class machines and never more than
  64MB of RAM.  Even the enthusiast hobbyist community seems to have
  dried up close to ten years ago, based on what you can find searching
  various websites dedicated to the relatively short lived hardware.

  So lets remove the support relating to CONFIG_MCA.  There is no point
  carrying this forward, wasting cycles doing routine maintenance on it;
  wasting allyesconfig build time on validating it, wasting I/O on git
  grep'ping over it, and so on."

Let's see if anybody screams.  It generally has compiled, and James
Bottomley pointed out that there was a MCA extension from NCR that
allowed for up to 4GB of memory and PPro-class machines.  So in *theory*
there may be users out there.

But even James (technically listed as a maintainer) doesn't actually
have a system, and while Alan Cox claims to have a machine in his cellar
that he offered to anybody who wants to take it off his hands, he didn't
argue for keeping MCA support either.

So we could bring it back.  But somebody had better speak up and talk
about how they have actually been using said MCA hardware with modern
kernels for us to do that.  And David already took the patch to delete
all the networking driver code (commit a5e371f6: "drivers/net:
delete all code/drivers depending on CONFIG_MCA").

* 'delete-mca' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux:
  MCA: delete all remaining traces of microchannel bus support.
  scsi: delete the MCA specific drivers and driver code
  serial: delete the MCA specific 8250 support.
  arm: remove ability to select CONFIG_MCA
parents c80ddb52 bb8187d3
Show whitespace changes
Inline Side-by-side
Showing with 32 additions and 8033 deletions
Documentation/00-INDEX
View file @ d5b4bb4d
......@@ -218,8 +218,6 @@ m68k/
- directory with info about Linux on Motorola 68k architecture.
magic-number.txt
- list of magic numbers used to mark/protect kernel data structures.
mca.txt
- info on supporting Micro Channel Architecture (e.g. PS/2) systems.
md.txt
- info on boot arguments for the multiple devices driver.
memory-barriers.txt
......
Documentation/DocBook/Makefile
View file @ d5b4bb4d
......@@ -6,7 +6,7 @@
# To add a new book the only step required is to add the book to the
# list of DOCBOOKS.
DOCBOOKS := z8530book.xml mcabook.xml device-drivers.xml \
DOCBOOKS := z8530book.xml device-drivers.xml \
kernel-hacking.xml kernel-locking.xml deviceiobook.xml \
writing_usb_driver.xml networking.xml \
kernel-api.xml filesystems.xml lsm.xml usb.xml kgdb.xml \
......
Documentation/DocBook/kernel-api.tmpl
View file @ d5b4bb4d
......@@ -212,19 +212,6 @@ X!Edrivers/pci/hotplug.c
<sect1><title>PCI Hotplug Support Library</title>
!Edrivers/pci/hotplug/pci_hotplug_core.c
</sect1>
<sect1><title>MCA Architecture</title>
<sect2><title>MCA Device Functions</title>
<para>
Refer to the file arch/x86/kernel/mca_32.c for more information.
</para>
<!-- FIXME: Removed for now since no structured comments in source
X!Earch/x86/kernel/mca_32.c
-->
</sect2>
<sect2><title>MCA Bus DMA</title>
!Iarch/x86/include/asm/mca_dma.h
</sect2>
</sect1>
</chapter>
<chapter id="firmware">
......
Documentation/DocBook/mcabook.tmpl deleted 100644 → 0
View file @ c80ddb52
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
<book id="MCAGuide">
<bookinfo>
<title>MCA Driver Programming Interface</title>
<authorgroup>
<author>
<firstname>Alan</firstname>
<surname>Cox</surname>
<affiliation>
<address>
<email>alan@lxorguk.ukuu.org.uk</email>
</address>
</affiliation>
</author>
<author>
<firstname>David</firstname>
<surname>Weinehall</surname>
</author>
<author>
<firstname>Chris</firstname>
<surname>Beauregard</surname>
</author>
</authorgroup>
<copyright>
<year>2000</year>
<holder>Alan Cox</holder>
<holder>David Weinehall</holder>
<holder>Chris Beauregard</holder>
</copyright>
<legalnotice>
<para>
This documentation is free software; you can redistribute
it and/or modify it under the terms of the GNU General Public
License as published by the Free Software Foundation; either
version 2 of the License, or (at your option) any later
version.
</para>
<para>
This program is distributed in the hope that it will be
useful, but WITHOUT ANY WARRANTY; without even the implied
warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
See the GNU General Public License for more details.
</para>
<para>
You should have received a copy of the GNU General Public
License along with this program; if not, write to the Free
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston,
MA 02111-1307 USA
</para>
<para>
For more details see the file COPYING in the source
distribution of Linux.
</para>
</legalnotice>
</bookinfo>
<toc></toc>
<chapter id="intro">
<title>Introduction</title>
<para>
The MCA bus functions provide a generalised interface to find MCA
bus cards, to claim them for a driver, and to read and manipulate POS
registers without being aware of the motherboard internals or
certain deep magic specific to onboard devices.
</para>
<para>
The basic interface to the MCA bus devices is the slot. Each slot
is numbered and virtual slot numbers are assigned to the internal
devices. Using a pci_dev as other busses do does not really make
sense in the MCA context as the MCA bus resources require card
specific interpretation.
</para>
<para>
Finally the MCA bus functions provide a parallel set of DMA
functions mimicing the ISA bus DMA functions as closely as possible,
although also supporting the additional DMA functionality on the
MCA bus controllers.
</para>
</chapter>
<chapter id="bugs">
<title>Known Bugs And Assumptions</title>
<para>
None.
</para>
</chapter>
<chapter id="pubfunctions">
<title>Public Functions Provided</title>
!Edrivers/mca/mca-legacy.c
</chapter>
<chapter id="dmafunctions">
<title>DMA Functions Provided</title>
!Iarch/x86/include/asm/mca_dma.h
</chapter>
</book>
Documentation/devices.txt
View file @ d5b4bb4d
......@@ -847,13 +847,7 @@ Your cooperation is appreciated.
...
31 = /dev/tap15 16th Ethertap device
36 block MCA ESDI hard disk
0 = /dev/eda First ESDI disk whole disk
64 = /dev/edb Second ESDI disk whole disk
...
Partitions are handled in the same way as IDE disks
(see major number 3).
36 block OBSOLETE (was MCA ESDI hard disk)
37 char IDE tape
0 = /dev/ht0 First IDE tape
......
Documentation/eisa.txt
View file @ d5b4bb4d
......@@ -179,7 +179,7 @@ CONFIG_ALPHA_JENSEN or CONFIG_EISA_VLB_PRIMING are set.
Converting an EISA driver to the new API mostly involves *deleting*
code (since probing is now in the core EISA code). Unfortunately, most
drivers share their probing routine between ISA, MCA and EISA. Special
drivers share their probing routine between ISA, and EISA. Special
care must be taken when ripping out the EISA code, so other busses
won't suffer from these surgical strikes...
......
Documentation/kernel-parameters.txt
View file @ d5b4bb4d
......@@ -70,7 +70,6 @@ parameter is applicable:
M68k M68k architecture is enabled.
These options have more detailed description inside of
Documentation/m68k/kernel-options.txt.
MCA MCA bus support is enabled.
MDA MDA console support is enabled.
MIPS MIPS architecture is enabled.
MOUSE Appropriate mouse support is enabled.
......
Documentation/mca.txt deleted 100644 → 0
View file @ c80ddb52
i386 Micro Channel Architecture Support
=======================================
MCA support is enabled using the CONFIG_MCA define. A machine with a MCA
bus will have the kernel variable MCA_bus set, assuming the BIOS feature
bits are set properly (see arch/i386/boot/setup.S for information on
how this detection is done).
Adapter Detection
=================
The ideal MCA adapter detection is done through the use of the
Programmable Option Select registers. Generic functions for doing
this have been added in include/linux/mca.h and arch/x86/kernel/mca_32.c.
Everything needed to detect adapters and read (and write) configuration
information is there. A number of MCA-specific drivers already use
this. The typical probe code looks like the following:
#include <linux/mca.h>
unsigned char pos2, pos3, pos4, pos5;
struct net_device* dev;
int slot;
if( MCA_bus ) {
slot = mca_find_adapter( ADAPTER_ID, 0 );
if( slot == MCA_NOTFOUND ) {
return -ENODEV;
}
/* optional - see below */
mca_set_adapter_name( slot, "adapter name & description" );
mca_set_adapter_procfn( slot, dev_getinfo, dev );
/* read the POS registers. Most devices only use 2 and 3 */
pos2 = mca_read_stored_pos( slot, 2 );
pos3 = mca_read_stored_pos( slot, 3 );
pos4 = mca_read_stored_pos( slot, 4 );
pos5 = mca_read_stored_pos( slot, 5 );
} else {
return -ENODEV;
}
/* extract configuration from pos[2345] and set everything up */
Loadable modules should modify this to test that the specified IRQ and
IO ports (plus whatever other stuff) match. See 3c523.c for example
code (actually, smc-mca.c has a slightly more complex example that can
handle a list of adapter ids).
Keep in mind that devices should never directly access the POS registers
(via inb(), outb(), etc). While it's generally safe, there is a small
potential for blowing up hardware when it's done at the wrong time.
Furthermore, accessing a POS register disables a device temporarily.
This is usually okay during startup, but do _you_ want to rely on it?
During initial configuration, mca_init() reads all the POS registers
into memory. mca_read_stored_pos() accesses that data. mca_read_pos()
and mca_write_pos() are also available for (safer) direct POS access,
but their use is _highly_ discouraged. mca_write_pos() is particularly
dangerous, as it is possible for adapters to be put in inconsistent
states (i.e. sharing IO address, etc) and may result in crashes, toasted
hardware, and blindness.
User level drivers (such as the AGX X server) can use /proc/mca/pos to
find adapters (see below).
Some MCA adapters can also be detected via the usual ISA-style device
probing (many SCSI adapters, for example). This sort of thing is highly
discouraged. Perfectly good information is available telling you what's
there, so there's no excuse for messing with random IO ports. However,
we MCA people still appreciate any ISA-style driver that will work with
our hardware. You take what you can get...
Level-Triggered Interrupts
==========================
Because MCA uses level-triggered interrupts, a few problems arise with
what might best be described as the ISA mindset and its effects on
drivers. These sorts of problems are expected to become less common as
more people use shared IRQs on PCI machines.
In general, an interrupt must be acknowledged not only at the ICU (which
is done automagically by the kernel), but at the device level. In
particular, IRQ 0 must be reset after a timer interrupt (now done in
arch/x86/kernel/time.c) or the first timer interrupt hangs the system.
There were also problems with the 1.3.x floppy drivers, but that seems
to have been fixed.
IRQs are also shareable, and most MCA-specific devices should be coded
with shared IRQs in mind.
/proc/mca
=========
/proc/mca is a directory containing various files for adapters and
other stuff.
/proc/mca/pos Straight listing of POS registers
/proc/mca/slot[1-8] Information on adapter in specific slot
/proc/mca/video Same for integrated video
/proc/mca/scsi Same for integrated SCSI
/proc/mca/machine Machine information
See Appendix A for a sample.
Device drivers can easily add their own information function for
specific slots (including integrated ones) via the
mca_set_adapter_procfn() call. Drivers that support this are ESDI, IBM
SCSI, and 3c523. If a device is also a module, make sure that the proc
function is removed in the module cleanup. This will require storing
the slot information in a private structure somewhere. See the 3c523
driver for details.
Your typical proc function will look something like this:
static int
dev_getinfo( char* buf, int slot, void* d ) {
struct net_device* dev = (struct net_device*) d;
int len = 0;
len += sprintf( buf+len, "Device: %s\n", dev->name );
len += sprintf( buf+len, "IRQ: %d\n", dev->irq );
len += sprintf( buf+len, "IO Port: %#lx-%#lx\n", ... );
...
return len;
}
Some of the standard MCA information will already be printed, so don't
bother repeating it. Don't try putting in more than 3K of information.
Enable this function with:
mca_set_adapter_procfn( slot, dev_getinfo, dev );
Disable it with:
mca_set_adapter_procfn( slot, NULL, NULL );
It is also recommended that, even if you don't write a proc function, to
set the name of the adapter (i.e. "PS/2 ESDI Controller") via
mca_set_adapter_name( int slot, char* name ).
MCA Device Drivers
==================
Currently, there are a number of MCA-specific device drivers.
1) PS/2 SCSI
drivers/scsi/ibmmca.c
drivers/scsi/ibmmca.h
The driver for the IBM SCSI subsystem. Includes both integrated
controllers and adapter cards. May require command-line arg
"ibmmcascsi=io_port" to force detection of an adapter. If you have a
machine with a front-panel display (i.e. model 95), you can use
"ibmmcascsi=display" to enable a drive activity indicator.
2) 3c523
drivers/net/3c523.c
drivers/net/3c523.h
3Com 3c523 Etherlink/MC ethernet driver.
3) SMC Ultra/MCA and IBM Adapter/A
drivers/net/smc-mca.c
drivers/net/smc-mca.h
Driver for the MCA version of the SMC Ultra and various other
OEM'ed and work-alike cards (Elite, Adapter/A, etc).
4) NE/2
driver/net/ne2.c
driver/net/ne2.h
The NE/2 is the MCA version of the NE2000. This may not work
with clones that have a different adapter id than the original
NE/2.
5) Future Domain MCS-600/700, OEM'd IBM Fast SCSI Adapter/A and
Reply Sound Blaster/SCSI (SCSI part)
Better support for these cards than the driver for ISA.
Supports multiple cards with IRQ sharing.
Also added boot time option of scsi-probe, which can do reordering of
SCSI host adapters. This will direct the kernel on the order which
SCSI adapter should be detected. Example:
scsi-probe=ibmmca,fd_mcs,adaptec1542,buslogic
The serial drivers were modified to support the extended IO port range
of the typical MCA system (also #ifdef CONFIG_MCA).
The following devices work with existing drivers:
1) Token-ring
2) Future Domain SCSI (MCS-600, MCS-700, not MCS-350, OEM'ed IBM SCSI)
3) Adaptec 1640 SCSI (using the aha1542 driver)
4) Bustek/Buslogic SCSI (various)
5) Probably all Arcnet cards.
6) Some, possibly all, MCA IDE controllers.
7) 3Com 3c529 (MCA version of 3c509) (patched)
8) Intel EtherExpressMC (patched version)
You need to have CONFIG_MCA defined to have EtherExpressMC support.
9) Reply Sound Blaster/SCSI (SB part) (patched version)
Bugs & Other Weirdness
======================
NMIs tend to occur with MCA machines because of various hardware
weirdness, bus timeouts, and many other non-critical things. Some basic
code to handle them (inspired by the NetBSD MCA code) has been added to
detect the guilty device, but it's pretty incomplete. If NMIs are a
persistent problem (on some model 70 or 80s, they occur every couple
shell commands), the CONFIG_IGNORE_NMI flag will take care of that.
Various Pentium machines have had serious problems with the FPU test in
bugs.h. Basically, the machine hangs after the HLT test. This occurs,
as far as we know, on the Pentium-equipped 85s, 95s, and some PC Servers.
The PCI/MCA PC 750s are fine as far as I can tell. The ``mca-pentium''
boot-prompt flag will disable the FPU bug check if this is a problem
with your machine.
The model 80 has a raft of problems that are just too weird and unique
to get into here. Some people have no trouble while others have nothing
but problems. I'd suspect some problems are related to the age of the
average 80 and accompanying hardware deterioration, although others
are definitely design problems with the hardware. Among the problems
include SCSI controller problems, ESDI controller problems, and serious
screw-ups in the floppy controller. Oh, and the parallel port is also
pretty flaky. There were about 5 or 6 different model 80 motherboards
produced to fix various obscure problems. As far as I know, it's pretty
much impossible to tell which bugs a particular model 80 has (other than
triggering them, that is).
Drivers are required for some MCA memory adapters. If you're suddenly
short a few megs of RAM, this might be the reason. The (I think) Enhanced
Memory Adapter commonly found on the model 70 is one. There's a very
alpha driver floating around, but it's pretty ugly (disassembled from
the DOS driver, actually). See the MCA Linux web page (URL below)
for more current memory info.
The Thinkpad 700 and 720 will work, but various components are either
non-functional, flaky, or we don't know anything about them. The
graphics controller is supposed to be some WD, but we can't get things
working properly. The PCMCIA slots don't seem to work. Ditto for APM.
The serial ports work, but detection seems to be flaky.
Credits
=======
A whole pile of people have contributed to the MCA code. I'd include
their names here, but I don't have a list handy. Check the MCA Linux
home page (URL below) for a perpetually out-of-date list.
=====================================================================
MCA Linux Home Page: http://www.dgmicro.com/mca/
Christophe Beauregard
chrisb@truespectra.com
cpbeaure@calum.csclub.uwaterloo.ca
=====================================================================
Appendix A: Sample /proc/mca
This is from my model 8595. Slot 1 contains the standard IBM SCSI
adapter, slot 3 is an Adaptec AHA-1640, slot 5 is a XGA-1 video adapter,
and slot 7 is the 3c523 Etherlink/MC.
/proc/mca/machine:
Model Id: 0xf8
Submodel Id: 0x14
BIOS Revision: 0x5
/proc/mca/pos:
Slot 1: ff 8e f1 fc a0 ff ff ff IBM SCSI Adapter w/Cache
Slot 2: ff ff ff ff ff ff ff ff
Slot 3: 1f 0f 81 3b bf b6 ff ff
Slot 4: ff ff ff ff ff ff ff ff
Slot 5: db 8f 1d 5e fd c0 00 00
Slot 6: ff ff ff ff ff ff ff ff
Slot 7: 42 60 ff 08 ff ff ff ff 3Com 3c523 Etherlink/MC
Slot 8: ff ff ff ff ff ff ff ff
Video : ff ff ff ff ff ff ff ff
SCSI : ff ff ff ff ff ff ff ff
/proc/mca/slot1:
Slot: 1
Adapter Name: IBM SCSI Adapter w/Cache
Id: 8eff
Enabled: Yes
POS: ff 8e f1 fc a0 ff ff ff
Subsystem PUN: 7
Detected at boot: Yes
/proc/mca/slot3:
Slot: 3
Adapter Name: Unknown
Id: 0f1f
Enabled: Yes
POS: 1f 0f 81 3b bf b6 ff ff
/proc/mca/slot5:
Slot: 5
Adapter Name: Unknown
Id: 8fdb
Enabled: Yes
POS: db 8f 1d 5e fd c0 00 00
/proc/mca/slot7:
Slot: 7
Adapter Name: 3Com 3c523 Etherlink/MC
Id: 6042
Enabled: Yes
POS: 42 60 ff 08 ff ff ff ff
Revision: 0xe
IRQ: 9
IO Address: 0x3300-0x3308
Memory: 0xd8000-0xdbfff
Transceiver: External
Device: eth0
Hardware Address: 02 60 8c 45 c4 2a
Documentation/scsi/00-INDEX
View file @ d5b4bb4d
......@@ -56,8 +56,6 @@ g_NCR5380.txt
- info on driver for NCR5380 and NCR53c400 based adapters
hptiop.txt
- HIGHPOINT ROCKETRAID 3xxx RAID DRIVER
ibmmca.txt
- info on driver for IBM adapters with MCA bus
in2000.txt
- info on in2000 driver
libsas.txt
......
Documentation/scsi/ibmmca.txt deleted 100644 → 0
View file @ c80ddb52
-=< The IBM Microchannel SCSI-Subsystem >=-
for the IBM PS/2 series
Low Level Software-Driver for Linux
Copyright (c) 1995 Strom Systems, Inc. under the terms of the GNU
General Public License. Originally written by Martin Kolinek, December 1995.
Officially modified and maintained by Michael Lang since January 1999.
Version 4.0a
Last update: January 3, 2001
Before you Start
----------------
This is the common README.ibmmca file for all driver releases of the
IBM MCA SCSI driver for Linux. Please note, that driver releases 4.0
or newer do not work with kernel versions older than 2.4.0, while driver
versions older than 4.0 do not work with kernels 2.4.0 or later! If you
try to compile your kernel with the wrong driver source, the
compilation is aborted and you get a corresponding error message. This is
no bug in the driver; it prevents you from using the wrong source code
with the wrong kernel version.
Authors of this Driver
----------------------
- Chris Beauregard (improvement of the SCSI-device mapping by the driver)
- Martin Kolinek (origin, first release of this driver)
- Klaus Kudielka (multiple SCSI-host management/detection, adaption to
Linux Kernel 2.1.x, module support)
- Michael Lang (assigning original pun/lun mapping, dynamical ldn
assignment, rewritten adapter detection, this file,
patches, official driver maintenance and subsequent
debugging, related with the driver)
Table of Contents
-----------------
1 Abstract
2 Driver Description
2.1 IBM SCSI-Subsystem Detection
2.2 Physical Units, Logical Units, and Logical Devices
2.3 SCSI-Device Recognition and dynamical ldn Assignment
2.4 SCSI-Device Order
2.5 Regular SCSI-Command-Processing
2.6 Abort & Reset Commands
2.7 Disk Geometry
2.8 Kernel Boot Option
2.9 Driver Module Support
2.10 Multiple Hostadapter Support
2.11 /proc/scsi-Filesystem Information
2.12 /proc/mca-Filesystem Information
2.13 Supported IBM SCSI-Subsystems
2.14 Linux Kernel Versions
3 Code History
4 To do
5 Users' Manual
5.1 Commandline Parameters
5.2 Troubleshooting
5.3 Bug reports
5.4 Support WWW-page
6 References
7 Credits to
7.1 People
7.2 Sponsors & Supporters
8 Trademarks
9 Disclaimer
* * *
1 Abstract
----------
This README-file describes the IBM SCSI-subsystem low level driver for
Linux. The descriptions which were formerly kept in the source code have
been taken out of this file to simplify the codes readability. The driver
description has been updated, as most of the former description was already
quite outdated. The history of the driver development is also kept inside
here. Multiple historical developments have been summarized to shorten the
text size a bit. At the end of this file you can find a small manual for
this driver and hints to get it running on your machine.
2 Driver Description
--------------------
2.1 IBM SCSI-Subsystem Detection
--------------------------------
This is done in the ibmmca_detect() function. It first checks, if the
Microchannel-bus support is enabled, as the IBM SCSI-subsystem needs the
Microchannel. In a next step, a free interrupt is chosen and the main
interrupt handler is connected to it to handle answers of the SCSI-
subsystem(s). If the F/W SCSI-adapter is forced by the BIOS to use IRQ11
instead of IRQ14, IRQ11 is used for the IBM SCSI-2 F/W adapter. In a
further step it is checked, if the adapter gets detected by force from
the kernel commandline, where the I/O port and the SCSI-subsystem id can
be specified. The next step checks if there is an integrated SCSI-subsystem
installed. This register area is fixed through all IBM PS/2 MCA-machines
and appears as something like a virtual slot 10 of the MCA-bus. On most
PS/2 machines, the POS registers of slot 10 are set to 0xff or 0x00 if not
integrated SCSI-controller is available. But on certain PS/2s, like model
9595, this slot 10 is used to store other information which at earlier
stage confused the driver and resulted in the detection of some ghost-SCSI.
If POS-register 2 and 3 are not 0x00 and not 0xff, but all other POS
registers are either 0xff or 0x00, there must be an integrated SCSI-
subsystem present and it will be registered as IBM Integrated SCSI-
Subsystem. The next step checks, if there is a slot-adapter installed on
the MCA-bus. To get this, the first two POS-registers, that represent the
adapter ID are checked. If they fit to one of the ids, stored in the
adapter list, a SCSI-subsystem is assumed to be found in a slot and will be
registered. This check is done through all possible MCA-bus slots to allow
more than one SCSI-adapter to be present in the PS/2-system and this is
already the first point of problems. Looking into the technical reference
manual for the IBM PS/2 common interfaces, the POS2 register must have
different interpretation of its single bits to avoid overlapping I/O
regions. While one can assume, that the integrated subsystem has a fix
I/O-address at 0x3540 - 0x3547, further installed IBM SCSI-adapters must
use a different I/O-address. This is expressed by bit 1 to 3 of POS2
(multiplied by 8 + 0x3540). Bits 2 and 3 are reserved for the integrated
subsystem, but not for the adapters! The following list shows, how the
bits of POS2 and POS3 should be interpreted.
The POS2-register of all PS/2 models' integrated SCSI-subsystems has the
following interpretation of bits:
Bit 7 - 4 : Chip Revision ID (Release)
Bit 3 - 2 : Reserved
Bit 1 : 8k NVRAM Disabled
Bit 0 : Chip Enable (EN-Signal)
The POS3-register is interpreted as follows (for most IBM SCSI-subsys.):
Bit 7 - 5 : SCSI ID
Bit 4 - 0 : Reserved = 0
The slot-adapters have different interpretation of these bits. The IBM SCSI
adapter (w/Cache) and the IBM SCSI-2 F/W adapter use the following
interpretation of the POS2 register:
Bit 7 - 4 : ROM Segment Address Select
Bit 3 - 1 : Adapter I/O Address Select (*8+0x3540)
Bit 0 : Adapter Enable (EN-Signal)
and for the POS3 register:
Bit 7 - 5 : SCSI ID
Bit 4 : Fairness Enable (SCSI ID3 f. F/W)
Bit 3 - 0 : Arbitration Level
The most modern product of the series is the IBM SCSI-2 F/W adapter, it
allows dual-bus SCSI and SCSI-wide addressing, which means, PUNs may be
between 0 and 15. Here, Bit 4 is the high-order bit of the 4-bit wide
adapter PUN expression. In short words, this means, that IBM PS/2 machines
can only support 1 single integrated subsystem by default. Additional
slot-adapters get ports assigned by the automatic configuration tool.
One day I found a patch in ibmmca_detect(), forcing the I/O-address to be
0x3540 for integrated SCSI-subsystems, there was a remark placed, that on
integrated IBM SCSI-subsystems of model 56, the POS2 register was showing 5.
This means, that really for these models, POS2 has to be interpreted
sticking to the technical reference guide. In this case, the bit 2 (4) is
a reserved bit and may not be interpreted. These differences between the
adapters and the integrated controllers are taken into account by the
detection routine of the driver on from version >3.0g.
Every time, a SCSI-subsystem is discovered, the ibmmca_register() function
is called. This function checks first, if the requested area for the I/O-
address of this SCSI-subsystem is still available and assigns this I/O-
area to the SCSI-subsystem. There are always 8 sequential I/O-addresses
taken for each individual SCSI-subsystem found, which are:
Offset Type Permissions
0 Command Interface Register 1 Read/Write
1 Command Interface Register 2 Read/Write
2 Command Interface Register 3 Read/Write
3 Command Interface Register 4 Read/Write
4 Attention Register Read/Write
5 Basic Control Register Read/Write
6 Interrupt Status Register Read
7 Basic Status Register Read
After the I/O-address range is assigned, the host-adapter is assigned
to a local structure which keeps all adapter information needed for the
driver itself and the mid- and higher-level SCSI-drivers. The SCSI pun/lun
and the adapters' ldn tables are initialized and get probed afterwards by
the check_devices() function. If no further adapters are found,
ibmmca_detect() quits.
2.2 Physical Units, Logical Units, and Logical Devices
------------------------------------------------------
There can be up to 56 devices on the SCSI bus (besides the adapter):
there are up to 7 "physical units" (each identified by physical unit
number or pun, also called the scsi id, this is the number you select
with hardware jumpers), and each physical unit can have up to 8
"logical units" (each identified by logical unit number, or lun,
between 0 and 7). The IBM SCSI-2 F/W adapter offers this on up to two
busses and provides support for 30 logical devices at the same time, where
in wide-addressing mode you can have 16 puns with 32 luns on each device.
This section describes the handling of devices on non-F/W adapters.
Just imagine, that you can have 16 * 32 = 512 devices on a F/W adapter
which means a lot of possible devices for such a small machine.
Typically the adapter has pun=7, so puns of other physical units
are between 0 and 6(15). On a wide-adapter a pun higher than 7 is
possible, but is normally not used. Almost all physical units have only
one logical unit, with lun=0. A CD-ROM jukebox would be an example of a
physical unit with more than one logical unit.
The embedded microprocessor of the IBM SCSI-subsystem hides the complex
two-dimensional (pun,lun) organization from the operating system.
When the machine is powered-up (or rebooted), the embedded microprocessor
checks, on its own, all 56 possible (pun,lun) combinations, and the first
15 devices found are assigned into a one-dimensional array of so-called
"logical devices", identified by "logical device numbers" or ldn. The last
ldn=15 is reserved for the subsystem itself. Wide adapters may have
to check up to 15 * 8 = 120 pun/lun combinations.
2.3 SCSI-Device Recognition and Dynamical ldn Assignment
--------------------------------------------------------
One consequence of information hiding is that the real (pun,lun)
numbers are also hidden. The two possibilities to get around this problem
are to offer fake pun/lun combinations to the operating system or to
delete the whole mapping of the adapter and to reassign the ldns, using
the immediate assign command of the SCSI-subsystem for probing through
all possible pun/lun combinations. An ldn is a "logical device number"
which is used by IBM SCSI-subsystems to access some valid SCSI-device.
At the beginning of the development of this driver, the following approach
was used:
First, the driver checked the ldn's (0 to 6) to find out which ldn's
have devices assigned. This was done by the functions check_devices() and
device_exists(). The interrupt handler has a special paragraph of code
(see local_checking_phase_flag) to assist in the checking. Assume, for
example, that three logical devices were found assigned at ldn 0, 1, 2.
These are presented to the upper layer of Linux SCSI driver
as devices with bogus (pun, lun) equal to (0,0), (1,0), (2,0).
On the other hand, if the upper layer issues a command to device
say (4,0), this driver returns DID_NO_CONNECT error.
In a second step of the driver development, the following improvement has
been applied: The first approach limited the number of devices to 7, far
fewer than the 15 that it could use, then it just mapped ldn ->
(ldn/8,ldn%8) for pun,lun. We ended up with a real mishmash of puns
and luns, but it all seemed to work.
The latest development, which is implemented from the driver version 3.0
and later, realizes the device recognition in the following way:
The physical SCSI-devices on the SCSI-bus are probed via immediate_assign-
and device_inquiry-commands, that is all implemented in a completely new
made check_devices() subroutine. This delivers an exact map of the physical
SCSI-world that is now stored in the get_scsi[][]-array. This means,
that the once hidden pun,lun assignment is now known to this driver.
It no longer believes in default-settings of the subsystem and maps all
ldns to existing pun,lun "by foot". This assures full control of the ldn
mapping and allows dynamical remapping of ldns to different pun,lun, if
there are more SCSI-devices installed than ldns available (n>15). The
ldns from 0 to 6 get 'hardwired' by this driver to puns 0 to 7 at lun=0,
excluding the pun of the subsystem. This assures, that at least simple
SCSI-installations have optimum access-speed and are not touched by
dynamical remapping. The ldns 7 to 14 are put to existing devices with
lun>0 or to non-existing devices, in order to satisfy the subsystem, if
there are less than 15 SCSI-devices connected. In the case of more than 15
devices, the dynamical mapping goes active. If the get_scsi[][] reports a
device to be existent, but it has no ldn assigned, it gets an ldn out of 7
to 14. The numbers are assigned in cyclic order, therefore it takes 8
dynamical reassignments on the SCSI-devices until a certain device
loses its ldn again. This assures that dynamical remapping is avoided
during intense I/O between up to 15 SCSI-devices (means pun,lun
combinations). A further advantage of this method is that people who
build their kernel without probing on all luns will get what they expect,
because the driver just won't assign everything with lun>0 when
multiple lun probing is inactive.
2.4 SCSI-Device Order
---------------------
Because of the now correct recognition of physical pun,lun, and
their report to mid-level- and higher-level-drivers, the new reported puns
can be different from the old, faked puns. Therefore, Linux will eventually
change /dev/sdXXX assignments and prompt you for corrupted superblock
repair on boottime. In this case DO NOT PANIC, YOUR DISKS ARE STILL OK!!!
You have to reboot (CTRL-D) with an old kernel and set the /etc/fstab-file
entries right. After that, the system should come up as errorfree as before.
If your boot-partition is not coming up, also edit the /etc/lilo.conf-file
in a Linux session booted on old kernel and run lilo before reboot. Check
lilo.conf anyway to get boot on other partitions with foreign OSes right
again. But there exists a feature of this driver that allows you to change
the assignment order of the SCSI-devices by flipping the PUN-assignment.
See the next paragraph for a description.
The problem for this is, that Linux does not assign the SCSI-devices in the
way as described in the ANSI-SCSI-standard. Linux assigns /dev/sda to
the device with at minimum id 0. But the first drive should be at id 6,
because for historical reasons, drive at id 6 has, by hardware, the highest
priority and a drive at id 0 the lowest. IBM was one of the rare producers,
where the BIOS assigns drives belonging to the ANSI-SCSI-standard. Most
other producers' BIOS does not (I think even Adaptec-BIOS). The
IBMMCA_SCSI_ORDER_STANDARD flag, which you set while configuring the
kernel enables to choose the preferred way of SCSI-device-assignment.
Defining this flag would result in Linux determining the devices in the
same order as DOS and OS/2 does on your MCA-machine. This is also standard
on most industrial computers and OSes, like e.g. OS-9. Leaving this flag
undefined will get your devices ordered in the default way of Linux. See
also the remarks of Chris Beauregard from Dec 15, 1997 and the followups
in section 3.
2.5 Regular SCSI-Command-Processing
-----------------------------------
Only three functions get involved: ibmmca_queuecommand(), issue_cmd(),
and interrupt_handler().
The upper layer issues a scsi command by calling function
ibmmca_queuecommand(). This function fills a "subsystem control block"
(scb) and calls a local function issue_cmd(), which writes a scb
command into subsystem I/O ports. Once the scb command is carried out,
the interrupt_handler() is invoked. If a device is determined to be
existent and it has not assigned any ldn, it gets one dynamically.
For this, the whole stuff is done in ibmmca_queuecommand().
2.6 Abort & Reset Commands
--------------------------
These are implemented with busy waiting for interrupt to arrive.
ibmmca_reset() and ibmmca_abort() do not work sufficiently well
up to now and need still a lot of development work. This seems
to be a problem with other low-level SCSI drivers too, however
this should be no excuse.
2.7 Disk Geometry
-----------------
The ibmmca_biosparams() function should return the same disk geometry
as the bios. This is needed for fdisk, etc. The returned geometry is
certainly correct for disks smaller than 1 gigabyte. In the meantime,
it has been proved, that this works fine even with disks larger than
1 gigabyte.
2.8 Kernel Boot Option
----------------------
The function ibmmca_scsi_setup() is called if option ibmmcascsi=n
is passed to the kernel. See file linux/init/main.c for details.
2.9 Driver Module Support
-------------------------
Is implemented and tested by K. Kudielka. This could probably not work
on kernels <2.1.0.
2.10 Multiple Hostadapter Support
---------------------------------
This driver supports up to eight interfaces of type IBM-SCSI-Subsystem.
Integrated-, and MCA-adapters are automatically recognized. Unrecognizable
IBM-SCSI-Subsystem interfaces can be specified as kernel-parameters.
2.11 /proc/scsi-Filesystem Information
--------------------------------------
Information about the driver condition is given in
/proc/scsi/ibmmca/<host_no>. ibmmca_proc_info() provides this information.
This table is quite informative for interested users. It shows the load
of commands on the subsystem and whether you are running the bypassed
(software) or integrated (hardware) SCSI-command set (see below). The
amount of accesses is shown. Read, write, modeselect is shown separately
in order to help debugging problems with CD-ROMs or tapedrives.
The following table shows the list of 15 logical device numbers, that are
used by the SCSI-subsystem. The load on each ldn is shown in the table,
again, read and write commands are split. The last column shows the amount
of reassignments, that have been applied to the ldns, if you have more than
15 pun/lun combinations available on the SCSI-bus.
The last two tables show the pun/lun map and the positions of the ldns
on this pun/lun map. This may change during operation, when a ldn is
reassigned to another pun/lun combination. If the necessity for dynamical
assignments is set to 'no', the ldn structure keeps static.
2.12 /proc/mca-Filesystem Information
-------------------------------------
The slot-file contains all default entries and in addition chip and I/O-
address information of the SCSI-subsystem. This information is provided
by ibmmca_getinfo().
2.13 Supported IBM SCSI-Subsystems
----------------------------------
The following IBM SCSI-subsystems are supported by this driver:
- IBM Fast/Wide SCSI-2 Adapter
- IBM 7568 Industrial Computer SCSI Adapter w/Cache
- IBM Expansion Unit SCSI Controller
- IBM SCSI Adapter w/Cache
- IBM SCSI Adapter
- IBM Integrated SCSI Controller
- All clones, 100% compatible with the chipset and subsystem command
system of IBM SCSI-adapters (forced detection)
2.14 Linux Kernel Versions
--------------------------
The IBM SCSI-subsystem low level driver is prepared to be used with
all versions of Linux between 2.0.x and 2.4.x. The compatibility checks
are fully implemented up from version 3.1e of the driver. This means, that
you just need the latest ibmmca.h and ibmmca.c file and copy it in the
linux/drivers/scsi directory. The code is automatically adapted during
kernel compilation. This is different from kernel 2.4.0! Here version
4.0 or later of the driver must be used for kernel 2.4.0 or later. Version
4.0 or later does not work together with older kernels! Driver versions
older than 4.0 do not work together with kernel 2.4.0 or later. They work
on all older kernels.
3 Code History
--------------
Jan 15 1996: First public release.
- Martin Kolinek
Jan 23 1996: Scrapped code which reassigned scsi devices to logical
device numbers. Instead, the existing assignment (created
when the machine is powered-up or rebooted) is used.
A side effect is that the upper layer of Linux SCSI
device driver gets bogus scsi ids (this is benign),
and also the hard disks are ordered under Linux the
same way as they are under dos (i.e., C: disk is sda,
D: disk is sdb, etc.).
- Martin Kolinek
I think that the CD-ROM is now detected only if a CD is
inside CD_ROM while Linux boots. This can be fixed later,
once the driver works on all types of PS/2's.
- Martin Kolinek
Feb 7 1996: Modified biosparam function. Fixed the CD-ROM detection.
For now, devices other than harddisk and CD_ROM are
ignored. Temporarily modified abort() function
to behave like reset().
- Martin Kolinek
Mar 31 1996: The integrated scsi subsystem is correctly found
in PS/2 models 56,57, but not in model 76. Therefore
the ibmmca_scsi_setup() function has been added today.
This function allows the user to force detection of
scsi subsystem. The kernel option has format
ibmmcascsi=n
where n is the scsi_id (pun) of the subsystem. Most likely, n is 7.
- Martin Kolinek
Aug 21 1996: Modified the code which maps ldns to (pun,0). It was
insufficient for those of us with CD-ROM changers.
- Chris Beauregard
Dec 14 1996: More improvements to the ldn mapping. See check_devices
for details. Did more fiddling with the integrated SCSI detection,
but I think it's ultimately hopeless without actually testing the
model of the machine. The 56, 57, 76 and 95 (ultimedia) all have
different integrated SCSI register configurations. However, the 56
and 57 are the only ones that have problems with forced detection.
- Chris Beauregard
Mar 8-16 1997: Modified driver to run as a module and to support
multiple adapters. A structure, called ibmmca_hostdata, is now
present, containing all the variables, that were once only
available for one single adapter. The find_subsystem-routine has vanished.
The hardware recognition is now done in ibmmca_detect directly.
This routine checks for presence of MCA-bus, checks the interrupt
level and continues with checking the installed hardware.
Certain PS/2-models do not recognize a SCSI-subsystem automatically.
Hence, the setup defined by command-line-parameters is checked first.
Thereafter, the routine probes for an integrated SCSI-subsystem.
Finally, adapters are checked. This method has the advantage to cover all
possible combinations of multiple SCSI-subsystems on one MCA-board. Up to
eight SCSI-subsystems can be recognized and announced to the upper-level
drivers with this improvement. A set of defines made changes to other
routines as small as possible.
- Klaus Kudielka
May 30 1997: (v1.5b)
1) SCSI-command capability enlarged by the recognition of MODE_SELECT.
This needs the RD-Bit to be disabled on IM_OTHER_SCSI_CMD_CMD which
allows data to be written from the system to the device. It is a
necessary step to be allowed to set blocksize of SCSI-tape-drives and
the tape-speed, without confusing the SCSI-Subsystem.
2) The recognition of a tape is included in the check_devices routine.
This is done by checking for TYPE_TAPE, that is already defined in
the kernel-scsi-environment. The markup of a tape is done in the
global ldn_is_tape[] array. If the entry on index ldn
is 1, there is a tapedrive connected.
3) The ldn_is_tape[] array is necessary to distinguish between tape- and
other devices. Fixed blocklength devices should not cause a problem
with the SCB-command for read and write in the ibmmca_queuecommand
subroutine. Therefore, I only derivate the READ_XX, WRITE_XX for
the tape-devices, as recommended by IBM in this Technical Reference,
mentioned below. (IBM recommends to avoid using the read/write of the
subsystem, but the fact was, that read/write causes a command error from
the subsystem and this causes kernel-panic.)
4) In addition, I propose to use the ldn instead of a fix char for the
display of PS2_DISK_LED_ON(). On 95, one can distinguish between the
devices that are accessed. It shows activity and easyfies debugging.
The tape-support has been tested with a SONY SDT-5200 and a HP DDS-2
(I do not know yet the type). Optimization and CD-ROM audio-support,
I am working on ...
- Michael Lang
June 19 1997: (v1.6b)
1) Submitting the extra-array ldn_is_tape[] -> to the local ld[]
device-array.
2) CD-ROM Audio-Play seems to work now.
3) When using DDS-2 (120M) DAT-Tapes, mtst shows still density-code
0x13 for ordinary DDS (61000 BPM) instead 0x24 for DDS-2. This appears
also on Adaptec 2940 adaptor in a PCI-System. Therefore, I assume that
the problem is independent of the low-level-driver/bus-architecture.
4) Hexadecimal ldn on PS/2-95 LED-display.
5) Fixing of the PS/2-LED on/off that it works right with tapedrives and
does not confuse the disk_rw_in_progress counter.
- Michael Lang
June 21 1997: (v1.7b)
1) Adding of a proc_info routine to inform in /proc/scsi/ibmmca/<host> the
outer-world about operational load statistics on the different ldns,
seen by the driver. Everybody that has more than one IBM-SCSI should
test this, because I only have one and cannot see what happens with more
than one IBM-SCSI hosts.
2) Definition of a driver version-number to have a better recognition of
the source when there are existing too much releases that may confuse
the user, when reading about release-specific problems. Up to know,
I calculated the version-number to be 1.7. Because we are in BETA-test
yet, it is today 1.7b.
3) Sorry for the heavy bug I programmed on June 19 1997! After that, the
CD-ROM did not work any more! The C7-command was a fake impression
I got while programming. Now, the READ and WRITE commands for CD-ROM are
no longer running over the subsystem, but just over
IM_OTHER_SCSI_CMD_CMD. On my observations (PS/2-95), now CD-ROM mounts
much faster(!) and hopefully all fancy multimedia-functions, like direct
digital recording from audio-CDs also work. (I tried it with cdda2wav
from the cdwtools-package and it filled up the harddisk immediately :-).)
To easify boolean logics, a further local device-type in ld[], called
is_cdrom has been included.
4) If one uses a SCSI-device of unsupported type/commands, one
immediately runs into a kernel-panic caused by Command Error. To better
understand which SCSI-command caused the problem, I extended this
specific panic-message slightly.
- Michael Lang
June 25 1997: (v1.8b)
1) Some cosmetic changes for the handling of SCSI-device-types.
Now, also CD-Burners / WORMs and SCSI-scanners should work. For
MO-drives I have no experience, therefore not yet supported.
In logical_devices I changed from different type-variables to one
called 'device_type' where the values, corresponding to scsi.h,
of a SCSI-device are stored.
2) There existed a small bug, that maps a device, coming after a SCSI-tape
wrong. Therefore, e.g. a CD-ROM changer would have been mapped wrong
-> problem removed.
3) Extension of the logical_device structure. Now it contains also device,
vendor and revision-level of a SCSI-device for internal usage.
- Michael Lang
June 26-29 1997: (v2.0b)
1) The release number 2.0b is necessary because of the completely new done
recognition and handling of SCSI-devices with the adapter. As I got
from Chris the hint, that the subsystem can reassign ldns dynamically,
I remembered this immediate_assign-command, I found once in the handbook.
Now, the driver first kills all ldn assignments that are set by default
on the SCSI-subsystem. After that, it probes on all puns and luns for
devices by going through all combinations with immediate_assign and
probing for devices, using device_inquiry. The found physical(!) pun,lun
structure is stored in get_scsi[][] as device types. This is followed
by the assignment of all ldns to existing SCSI-devices. If more ldns
than devices are available, they are assigned to non existing pun,lun
combinations to satisfy the adapter. With this, the dynamical mapping
was possible to implement. (For further info see the text in the
source code and in the description below. Read the description
below BEFORE installing this driver on your system!)
2) Changed the name IBMMCA_DRIVER_VERSION to IBMMCA_SCSI_DRIVER_VERSION.
3) The LED-display shows on PS/2-95 no longer the ldn, but the SCSI-ID
(pun) of the accessed SCSI-device. This is now senseful, because the
pun known within the driver is exactly the pun of the physical device
and no longer a fake one.
4) The /proc/scsi/ibmmca/<host_no> consists now of the first part, where
hit-statistics of ldns is shown and a second part, where the maps of
physical and logical SCSI-devices are displayed. This could be very
interesting, when one is using more than 15 SCSI-devices in order to
follow the dynamical remapping of ldns.
- Michael Lang
June 26-29 1997: (v2.0b-1)
1) I forgot to switch the local_checking_phase_flag to 1 and back to 0
in the dynamical remapping part in ibmmca_queuecommand for the
device_exist routine. Sorry.
- Michael Lang
July 1-13 1997: (v3.0b,c)
1) Merging of the driver-developments of Klaus Kudielka and Michael Lang
in order to get a optimum and unified driver-release for the
IBM-SCSI-Subsystem-Adapter(s).
For people, using the Kernel-release >=2.1.0, module-support should
be no problem. For users, running under <2.1.0, module-support may not
work, because the methods have changed between 2.0.x and 2.1.x.
2) Added some more effective statistics for /proc-output.
3) Change typecasting at necessary points from (unsigned long) to
virt_to_bus().
4) Included #if... at special points to have specific adaption of the
driver to kernel 2.0.x and 2.1.x. It should therefore also run with
later releases.
5) Magneto-Optical drives and medium-changers are also recognized, now.
Therefore, we have a completely gapfree recognition of all SCSI-
device-types, that are known by Linux up to kernel 2.1.31.
6) The flag SCSI_IBMMCA_DEV_RESET has been inserted. If it is set within
the configuration, each connected SCSI-device will get a reset command
during boottime. This can be necessary for some special SCSI-devices.
This flag should be included in Config.in.
(See also the new Config.in file.)
Probable next improvement: bad disk handler.
- Michael Lang
Sept 14 1997: (v3.0c)
1) Some debugging and speed optimization applied.
- Michael Lang
Dec 15, 1997
- chrisb@truespectra.com
- made the front panel display thingy optional, specified from the
command-line via ibmmcascsi=display. Along the lines of the /LED
option for the OS/2 driver.
- fixed small bug in the LED display that would hang some machines.
- reversed ordering of the drives (using the
IBMMCA_SCSI_ORDER_STANDARD define). This is necessary for two main
reasons:
- users who've already installed Linux won't be screwed. Keep
in mind that not everyone is a kernel hacker.
- be consistent with the BIOS ordering of the drives. In the
BIOS, id 6 is C:, id 0 might be D:. With this scheme, they'd be
backwards. This confuses the crap out of those heathens who've
got a impure Linux installation (which, <wince>, I'm one of).
This whole problem arises because IBM is actually non-standard with
the id to BIOS mappings. You'll find, in fdomain.c, a similar
comment about a few FD BIOS revisions. The Linux (and apparently
industry) standard is that C: maps to scsi id (0,0). Let's stick
with that standard.
- Since this is technically a branch of my own, I changed the
version number to 3.0e-cpb.
Jan 17, 1998: (v3.0f)
1) Addition of some statistical info for /proc in proc_info.
2) Taking care of the SCSI-assignment problem, dealed by Chris at Dec 15
1997. In fact, IBM is right, concerning the assignment of SCSI-devices
to driveletters. It is conform to the ANSI-definition of the SCSI-
standard to assign drive C: to SCSI-id 6, because it is the highest
hardware priority after the hostadapter (that has still today by
default everywhere id 7). Also realtime-operating systems that I use,
like LynxOS and OS9, which are quite industrial systems use top-down
numbering of the harddisks, that is also starting at id 6. Now, one
sits a bit between two chairs. On one hand side, using the define
IBMMCA_SCSI_ORDER_STANDARD makes Linux assigning disks conform to
the IBM- and ANSI-SCSI-standard and keeps this driver downward
compatible to older releases, on the other hand side, people is quite
habituated in believing that C: is assigned to (0,0) and much other
SCSI-BIOS do so. Therefore, I moved the IBMMCA_SCSI_ORDER_STANDARD
define out of the driver and put it into Config.in as subitem of
'IBM SCSI support'. A help, added to Documentation/Configure.help
explains the differences between saying 'y' or 'n' to the user, when
IBMMCA_SCSI_ORDER_STANDARD prompts, so the ordinary user is enabled to
choose the way of assignment, depending on his own situation and gusto.
3) Adapted SCSI_IBMMCA_DEV_RESET to the local naming convention, so it is
now called IBMMCA_SCSI_DEV_RESET.
4) Optimization of proc_info and its subroutines.
5) Added more in-source-comments and extended the driver description by
some explanation about the SCSI-device-assignment problem.
- Michael Lang
Jan 18, 1998: (v3.0g)
1) Correcting names to be absolutely conform to the later 2.1.x releases.
This is necessary for
IBMMCA_SCSI_DEV_RESET -> CONFIG_IBMMCA_SCSI_DEV_RESET
IBMMCA_SCSI_ORDER_STANDARD -> CONFIG_IBMMCA_SCSI_ORDER_STANDARD
- Michael Lang
Jan 18, 1999: (v3.1 MCA-team internal)
1) The multiple hosts structure is accessed from every subroutine, so there
is no longer the address of the device structure passed from function
to function, but only the hostindex. A call by value, nothing more. This
should really be understood by the compiler and the subsystem should get
the right values and addresses.
2) The SCSI-subsystem detection was not complete and quite hugely buggy up
to now, compared to the technical manual. The interpretation of the pos2
register is not as assumed by people before, therefore, I dropped a note
in the ibmmca_detect function to show the registers' interpretation.
The pos-registers of integrated SCSI-subsystems do not contain any
information concerning the IO-port offset, really. Instead, they contain
some info about the adapter, the chip, the NVRAM .... The I/O-port is
fixed to 0x3540 - 0x3547. There can be more than one adapters in the
slots and they get an offset for the I/O area in order to get their own
I/O-address area. See chapter 2 for detailed description. At least, the
detection should now work right, even on models other than 95. The 95ers
came happily around the bug, as their pos2 register contains always 0
in the critical area. Reserved bits are not allowed to be interpreted,
therefore, IBM is allowed to set those bits as they like and they may
really vary between different PS/2 models. So, now, no interpretation
of reserved bits - hopefully no trouble here anymore.
3) The command error, which you may get on models 55, 56, 57, 70, 77 and
P70 may have been caused by the fact, that adapters of older design do
not like sending commands to non-existing SCSI-devices and will react
with a command error as a sign of protest. While this error is not
present on IBM SCSI Adapter w/cache, it appears on IBM Integrated SCSI
Adapters. Therefore, I implemented a workaround to forgive those
adapters their protests, but it is marked up in the statistics, so
after a successful boot, you can see in /proc/scsi/ibmmca/<host_number>
how often the command errors have been forgiven to the SCSI-subsystem.
If the number is bigger than 0, you have a SCSI subsystem of older
design, what should no longer matter.
4) ibmmca_getinfo() has been adapted very carefully, so it shows in the
slotn file really, what is senseful to be presented.
5) ibmmca_register() has been extended in its parameter list in order to
pass the right name of the SCSI-adapter to Linux.
- Michael Lang
Feb 6, 1999: (v3.1)
1) Finally, after some 3.1Beta-releases, the 3.1 release. Sorry, for
the delayed release, but it was not finished with the release of
Kernel 2.2.0.
- Michael Lang
Feb 10, 1999 (v3.1)
1) Added a new commandline parameter called 'bypass' in order to bypass
every integrated subsystem SCSI-command consequently in case of
troubles.
2) Concatenated read_capacity requests to the harddisks. It gave a lot
of troubles with some controllers and after I wanted to apply some
extensions, it jumped out in the same situation, on my w/cache, as like
on D. Weinehalls' Model 56, having integrated SCSI. This gave me the
decisive hint to move the code-part out and declare it global. Now
it seems to work far better and more stable. Let us see what
the world thinks of it...
3) By the way, only Sony DAT-drives seem to show density code 0x13. A
test with a HP drive gave right results, so the problem is vendor-
specific and not a problem of the OS or the driver.
- Michael Lang
Feb 18, 1999 (v3.1d)
1) The abort command and the reset function have been checked for
inconsistencies. From the logical point of thinking, they work
at their optimum, now, but as the subsystem does not answer with an
interrupt, abort never finishes, sigh...
2) Everything, that is accessed by a busmaster request from the adapter
is now declared as global variable, even the return-buffer in the
local checking phase. This assures, that no accesses to undefined memory
areas are performed.
3) In ibmmca.h, the line unchecked_isa_dma is added with 1 in order to
avoid memory-pointers for the areas higher than 16MByte in order to
be sure, it also works on 16-Bit Microchannel bus systems.
4) A lot of small things have been found, but nothing that endangered the
driver operations. Just it should be more stable, now.
- Michael Lang
Feb 20, 1999 (v3.1e)
1) I took the warning from the Linux Kernel Hackers Guide serious and
checked the cmd->result return value to the done-function very carefully.
It is obvious, that the IBM SCSI only delivers the tsb.dev_status, if
some error appeared, else it is undefined. Now, this is fixed. Before
any SCB command gets queued, the tsb.dev_status is set to 0, so the
cmd->result won't screw up Linux higher level drivers.
2) The reset-function has slightly improved. This is still planned for
abort. During the abort and the reset function, no interrupts are
allowed. This is however quite hard to cope with, so the INT-status
register is read. When the interrupt gets queued, one can find its
status immediately on that register and is enabled to continue in the
reset function. I had no chance to test this really, only in a bogus
situation, I got this function running, but the situation was too much
worse for Linux :-(, so tests will continue.
3) Buffers got now consistent. No open address mapping, as before and
therefore no further troubles with the unassigned memory segmentation
faults that scrambled probes on 95XX series and even on 85XX series,
when the kernel is done in a not so perfectly fitting way.
4) Spontaneous interrupts from the subsystem, appearing without any
command previously queued are answered with a DID_BAD_INTR result.
5) Taken into account ZP Gus' proposals to reverse the SCSI-device
scan order. As it does not work on Kernel 2.1.x or 2.2.x, as proposed
by him, I implemented it in a slightly derived way, which offers in
addition more flexibility.
- Michael Lang
Apr 23, 2000 (v3.2pre1)
1) During a very long time, I collected a huge amount of bug reports from
various people, trying really quite different things on their SCSI-
PS/2s. Today, all these bug reports are taken into account and should be
mostly solved. The major topics were:
- Driver crashes during boottime by no obvious reason.
- Driver panics while the midlevel-SCSI-driver is trying to inquire
the SCSI-device properties, even though hardware is in perfect state.
- Displayed info for the various slot-cards is interpreted wrong.
The main reasons for the crashes were two:
1) The commands to check for device information like INQUIRY,
TEST_UNIT_READY, REQUEST_SENSE and MODE_SENSE cause the devices
to deliver information of up to 255 bytes. Midlevel drivers offer
1024 bytes of space for the answer, but the IBM-SCSI-adapters do
not accept this, as they stick quite near to ANSI-SCSI and report
a COMMAND_ERROR message which causes the driver to panic. The main
problem was located around the INQUIRY command. Now, for all the
mentioned commands, the buffersize sent to the adapter is at
maximum 255 which seems to be a quite reasonable solution.
TEST_UNIT_READY gets a buffersize of 0 to make sure that no
data is transferred in order to avoid any possible command failure.
2) On unsuccessful TEST_UNIT_READY, the mid-level driver has to send
a REQUEST_SENSE in order to see where the problem is located. This
REQUEST_SENSE may have various length in its answer-buffer. IBM
SCSI-subsystems report a command failure if the returned buffersize
is different from the sent buffersize, but this can be suppressed by
a special bit, which is now done and problems seem to be solved.
2) Code adaption to all kernel-releases. Now, the 3.2 code compiles on
2.0.x, 2.1.x, 2.2.x and 2.3.x kernel releases without any code-changes.
3) Commandline-parameters are recognized again, even under Kernel 2.3.x or
higher.
- Michael Lang
April 27, 2000 (v3.2pre2)
1) Bypassed commands get read by the adapter by one cycle instead of two.
This increases SCSI-performance.
2) Synchronous datatransfer is provided for sure to be 5 MHz on older
SCSI and 10 MHz on internal F/W SCSI-adapter.
3) New commandline parameters allow to force the adapter to slow down while
in synchronous transfer. Could be helpful for very old devices.
- Michael Lang
June 2, 2000 (v3.2pre5)
1) Added Jim Shorney's contribution to make the activity indicator
flashing in addition to the LED-alphanumeric display-panel on
models 95A. To be enabled to choose this feature freely, a new
commandline parameter is added, called 'activity'.
2) Added the READ_CONTROL bit for test_unit_ready SCSI-command.
3) Added some suppress_exception bits to read_device_capacity and
all device_inquiry occurrences in the driver code.
4) Complaints about the various KERNEL_VERSION implementations are
taken into account. Every local_LinuxKernelVersion occurrence is
now replaced by KERNEL_VERSION, defined in linux/version.h.
Corresponding changes were applied to ibmmca.h, too. This was a
contribution to all kernel-parts by Philipp Hahn.
- Michael Lang
July 17, 2000 (v3.2pre8)
A long period of collecting bug reports from all corners of the world
now lead to the following corrections to the code:
1) SCSI-2 F/W support crashed with a COMMAND ERROR. The reason for this
was that it is possible to disable Fast-SCSI for the external bus.
The feature-control command, where this crash appeared regularly, tried
to set the maximum speed of 10MHz synchronous transfer speed and that
reports a COMMAND ERROR if external bus Fast-SCSI is disabled. Now,
the feature-command probes down from maximum speed until the adapter
stops to complain, which is at the same time the maximum possible
speed selected in the reference program. So, F/W external can run at
5 MHz (slow-) or 10 MHz (fast-SCSI). During feature probing, the
COMMAND ERROR message is used to detect if the adapter does not complain.
2) Up to now, only combined busmode is supported, if you use external
SCSI-devices, attached to the F/W-controller. If dual bus is selected,
only the internal SCSI-devices get accessed by Linux. For most
applications, this should do fine.
3) Wide-SCSI-addressing (16-Bit) is now possible for the internal F/W
bus on the F/W adapter. If F/W adapter is detected, the driver
automatically uses the extended PUN/LUN <-> LDN mapping tables, which
are now new from 3.2pre8. This allows PUNs between 0 and 15 and should
provide more fun with the F/W adapter.
4) Several machines use the SCSI: POS registers for internal/undocumented
storage of system relevant info. This confused the driver, mainly on
models 9595, as it expected no onboard SCSI only, if all POS in
the integrated SCSI-area are set to 0x00 or 0xff. Now, the mechanism
to check for integrated SCSI is much more restrictive and these problems
should be history.
- Michael Lang
July 18, 2000 (v3.2pre9)
This develop rather quickly at the moment. Two major things were still
missing in 3.2pre8:
1) The adapter PUN for F/W adapters has 4-bits, while all other adapters
have 3-bits. This is now taken into account for F/W.
2) When you select CONFIG_IBMMCA_SCSI_ORDER_STANDARD, you should
normally get the inverse probing order of your devices on the SCSI-bus.
The ANSI device order gets scrambled in version 3.2pre8!! Now, a new
and tested algorithm inverts the device-order on the SCSI-bus and
automatically avoids accidental access to whatever SCSI PUN the adapter
is set and works with SCSI- and Wide-SCSI-addressing.
- Michael Lang
July 23, 2000 (v3.2pre10 unpublished)
1) LED panel display supports wide-addressing in ibmmca=display mode.
2) Adapter-information and autoadaption to address-space is done.
3) Auto-probing for maximum synchronous SCSI transfer rate is working.
4) Optimization to some embedded function calls is applied.
5) Added some comment for the user to wait for SCSI-devices being probed.
6) Finished version 3.2 for Kernel 2.4.0. It least, I thought it is but...
- Michael Lang
July 26, 2000 (v3.2pre11)
1) I passed a horrible weekend getting mad with NMIs on kernel 2.2.14 and
a model 9595. Asking around in the community, nobody except of me has
seen such errors. Weird, but I am trying to recompile everything on
the model 9595. Maybe, as I use a specially modified gcc, that could
cause problems. But, it was not the reason. The true background was,
that the kernel was compiled for i386 and the 9595 has a 486DX-2.
Normally, no troubles should appear, but for this special machine,
only the right processor support is working fine!
2) Previous problems with synchronous speed, slowing down from one adapter
to the next during probing are corrected. Now, local variables store
the synchronous bitmask for every single adapter found on the MCA bus.
3) LED alphanumeric panel support for XX95 systems is now showing some
alive rotator during boottime. This makes sense, when no monitor is
connected to the system. You can get rid of all display activity, if
you do not use any parameter or just ibmmcascsi=activity, for the
harddrive activity LED, existent on all PS/2, except models 8595-XXX.
If no monitor is available, please use ibmmcascsi=display, which works
fine together with the linuxinfo utility for the LED-panel.
- Michael Lang
July 29, 2000 (v3.2)
1) Submission of this driver for kernel 2.4test-XX and 2.2.17.
- Michael Lang
December 28, 2000 (v3.2d / v4.0)
1) The interrupt handler had some wrong statement to wait for. This
was done due to experimental reasons during 3.2 development but it
has shown that this is not stable enough. Going back to wait for the
adapter to be not busy is best.
2) Inquiry requests can be shorter than 255 bytes of return buffer. Due
to a bug in the ibmmca_queuecommand routine, this buffer was forced
to 255 at minimum. If the memory address, this return buffer is pointing
to does not offer more space, invalid memory accesses destabilized the
kernel.
3) version 4.0 is only valid for kernel 2.4.0 or later. This is necessary
to remove old kernel version dependent waste from the driver. 3.2d is
only distributed with older kernels but keeps compatibility with older
kernel versions. 4.0 and higher versions cannot be used with older
kernels anymore!! You must have at least kernel 2.4.0!!
4) The commandline argument 'bypass' and all its functionality got removed
in version 4.0. This was never really necessary, as all troubles were
based on non-command related reasons up to now, so bypassing commands
did not help to avoid any bugs. It is kept in 3.2X for debugging reasons.
5) Dynamic reassignment of ldns was again verified and analyzed to be
completely inoperational. This is corrected and should work now.
6) All commands that get sent to the SCSI adapter were verified and
completed in such a way, that they are now completely conform to the
demands in the technical description of IBM. Main candidates were the
DEVICE_INQUIRY, REQUEST_SENSE and DEVICE_CAPACITY commands. They must
be transferred by bypassing the internal command buffer of the adapter
or else the response can be a random result. GET_POS_INFO would be more
safe in usage, if one could use the SUPRESS_EXCEPTION_SHORT, but this
is not allowed by the technical references of IBM. (Sorry, folks, the
model 80 problem is still a task to be solved in a different way.)
7) v3.2d is still hold back for some days for testing, while 4.0 is
released.
- Michael Lang
January 3, 2001 (v4.0a)
1) A lot of complains after the 2.4.0-prerelease kernel came in about
the impossibility to compile the driver as a module. This problem is
solved. In combination with that problem, some unprecise declaration
of the function option_setup() gave some warnings during compilation.
This is solved, too by a forward declaration in ibmmca.c.
2) #ifdef argument concerning CONFIG_SCSI_IBMMCA is no longer needed and
was entirely removed.
3) Some switch statements got optimized in code, as some minor variables
in internal SCSI-command handlers.
- Michael Lang
4 To do
-------
- IBM SCSI-2 F/W external SCSI bus support in separate mode!
- It seems that the handling of bad disks is really bad -
non-existent, in fact. However, a low-level driver cannot help
much, if such things happen.
5 Users' Manual
---------------
5.1 Commandline Parameters
--------------------------
There exist several features for the IBM SCSI-subsystem driver.
The commandline parameter format is:
ibmmcascsi=<command1>,<command2>,<command3>,...
where commandN can be one of the following:
display Owners of a model 95 or other PS/2 systems with an
alphanumeric LED display may set this to have their
display showing the following output of the 8 digits:
------DA
where '-' stays dark, 'D' shows the SCSI-device id
and 'A' shows the SCSI hostindex, being currently
accessed. During boottime, this will give the message
SCSIini*
on the LED-panel, where the * represents a rotator,
showing the activity during the probing phase of the
driver which can take up to two minutes per SCSI-adapter.
adisplay This works like display, but gives more optical overview
of the activities on the SCSI-bus. The display will have
the following output:
6543210A
where the numbers 0 to 6 light up at the shown position,
when the SCSI-device is accessed. 'A' shows again the SCSI
hostindex. If display nor adisplay is set, the internal
PS/2 harddisk LED is used for media-activities. So, if
you really do not have a system with a LED-display, you
should not set display or adisplay. Keep in mind, that
display and adisplay can only be used alternatively. It
is not recommended to use this option, if you have some
wide-addressed devices e.g. at the SCSI-2 F/W adapter in
your system. In addition, the usage of the display for
other tasks in parallel, like the linuxinfo-utility makes
no sense with this option.
activity This enables the PS/2 harddisk LED activity indicator.
Most PS/2 have no alphanumeric LED display, but some
indicator. So you should use this parameter to activate it.
If you own model 9595 (Server95), you can have both, the
LED panel and the activity indicator in parallel. However,
some PS/2s, like the 8595 do not have any harddisk LED
activity indicator, which means, that you must use the
alphanumeric LED display if you want to monitor SCSI-
activity.
bypass This is obsolete from driver version 4.0, as the adapters
got that far understood, that the selection between
integrated and bypassed commands should now work completely
correct! For historical reasons, the old description is
kept here:
This commandline parameter forces the driver never to use
SCSI-subsystems' integrated SCSI-command set. Except of
the immediate assign, which is of vital importance for
every IBM SCSI-subsystem to set its ldns right. Instead,
the ordinary ANSI-SCSI-commands are used and passed by the
controller to the SCSI-devices, therefore 'bypass'. The
effort, done by the subsystem is quite bogus and at a
minimum and therefore it should work everywhere. This
could maybe solve troubles with old or integrated SCSI-
controllers and nasty harddisks. Keep in mind, that using
this flag will slow-down SCSI-accesses slightly, as the
software generated commands are always slower than the
hardware. Non-harddisk devices always get read/write-
commands in bypass mode. On the most recent releases of
the Linux IBM-SCSI-driver, the bypass command should be
no longer a necessary thing, if you are sure about your
SCSI-hardware!
normal This is the parameter, introduced on the 2.0.x development
rail by ZP Gu. This parameter defines the SCSI-device
scan order in the new industry standard. This means, that
the first SCSI-device is the one with the lowest pun.
E.g. harddisk at pun=0 is scanned before harddisk at
pun=6, which means, that harddisk at pun=0 gets sda
and the one at pun=6 gets sdb.
ansi The ANSI-standard for the right scan order, as done by
IBM, Microware and Microsoft, scans SCSI-devices starting
at the highest pun, which means, that e.g. harddisk at
pun=6 gets sda and a harddisk at pun=0 gets sdb. If you
like to have the same SCSI-device order, as in DOS, OS-9
or OS/2, just use this parameter.
fast SCSI-I/O in synchronous mode is done at 5 MHz for IBM-
SCSI-devices. SCSI-2 Fast/Wide Adapter/A external bus
should then run at 10 MHz if Fast-SCSI is enabled,
and at 5 MHz if Fast-SCSI is disabled on the external
bus. This is the default setting when nothing is
specified here.
medium Synchronous rate is at 50% approximately, which means
2.5 MHz for IBM SCSI-adapters and 5.0 MHz for F/W ext.
SCSI-bus (when Fast-SCSI speed enabled on external bus).
slow The slowest possible synchronous transfer rate is set.
This means 1.82 MHz for IBM SCSI-adapters and 2.0 MHz
for F/W external bus at Fast-SCSI speed on the external
bus.
A further option is that you can force the SCSI-driver to accept a SCSI-
subsystem at a certain I/O-address with a predefined adapter PUN. This
is done by entering
commandN = I/O-base
commandN+1 = adapter PUN
e.g. ibmmcascsi=0x3540,7 will force the driver to detect a SCSI-subsystem
at I/O-address 0x3540 with adapter PUN 7. Please only use this method, if
the driver does really not recognize your SCSI-adapter! With driver version
3.2, this recognition of various adapters was hugely improved and you
should try first to remove your commandline arguments of such type with a
newer driver. I bet, it will be recognized correctly. Even multiple and
different types of IBM SCSI-adapters should be recognized correctly, too.
Use the forced detection method only as last solution!
Examples:
ibmmcascsi=adisplay
This will use the advanced display mode for the model 95 LED alphanumeric
display.
ibmmcascsi=display,0x3558,7
This will activate the default display mode for the model 95 LED display
and will force the driver to accept a SCSI-subsystem at I/O-base 0x3558
with adapter PUN 7.
5.2 Troubleshooting
-------------------
The following FAQs should help you to solve some major problems with this
driver.
Q: "Reset SCSI-devices at boottime" halts the system at boottime, why?
A: This is only tested with the IBM SCSI Adapter w/cache. It is not
yet proven to run on other adapters, however you may be lucky.
In version 3.1d this has been hugely improved and should work better,
now. Normally you really won't need to activate this flag in the
kernel configuration, as all post 1989 SCSI-devices should accept
the reset-signal, when the computer is switched on. The SCSI-
subsystem generates this reset while being initialized. This flag
is really reserved for users with very old, very strange or self-made
SCSI-devices.
Q: Why is the SCSI-order of my drives mirrored to the device-order
seen from OS/2 or DOS ?
A: It depends on the operating system, if it looks at the devices in
ANSI-SCSI-standard (starting from pun 6 and going down to pun 0) or
if it just starts at pun 0 and counts up. If you want to be conform
with OS/2 and DOS, you have to activate this flag in the kernel
configuration or you should set 'ansi' as parameter for the kernel.
The parameter 'normal' sets the new industry standard, starting
from pun 0, scanning up to pun 6. This allows you to change your
opinion still after having already compiled the kernel.
Q: Why can't I find IBM MCA SCSI support in the config menu?
A: You have to activate MCA bus support, first.
Q: Where can I find the latest info about this driver?
A: See the file MAINTAINERS for the current WWW-address, which offers
updates, info and Q/A lists. At this file's origin, the webaddress
was: http://www.staff.uni-mainz.de/mlang/linux.html
Q: My SCSI-adapter is not recognized by the driver, what can I do?
A: Just force it to be recognized by kernel parameters. See section 5.1.
If this really happens, do also send e-mail to the maintainer, as
forced detection should be never necessary. Forced detection is in
principal some flaw of the driver adapter detection and goes into
bug reports.
Q: The driver screws up, if it starts to probe SCSI-devices, is there
some way out of it?
A: Yes, that was some recognition problem of the correct SCSI-adapter
and its I/O base addresses. Upgrade your driver to the latest release
and it should be fine again.
Q: I get a message: panic IBM MCA SCSI: command error .... , what can
I do against this?
A: Previously, I followed the way by ignoring command errors by using
ibmmcascsi=forgiveall, but this command no longer exists and is
obsolete. If such a problem appears, it is caused by some segmentation
fault of the driver, which maps to some unallowed area. The latest
version of the driver should be ok, as most bugs have been solved.
Q: There are still kernel panics, even after having set
ibmmcascsi=forgiveall. Are there other possibilities to prevent
such panics?
A: No, get just the latest release of the driver and it should work
better and better with increasing version number. Forget about this
ibmmcascsi=forgiveall, as also ignorecmd are obsolete.!
Q: Linux panics or stops without any comment, but it is probable, that my
harddisk(s) have bad blocks.
A: Sorry, the bad-block handling is still a feeble point of this driver,
but is on the schedule for development in the near future.
Q: Linux panics while dynamically assigning SCSI-ids or ldns.
A: If you disconnect a SCSI-device from the machine, while Linux is up
and the driver uses dynamical reassignment of logical device numbers
(ldn), it really gets "angry" if it won't find devices, that were still
present at boottime and stops Linux.
Q: The system does not recover after an abort-command has been generated.
A: This is regrettably true, as it is not yet understood, why the
SCSI-adapter does really NOT generate any interrupt at the end of
the abort-command. As no interrupt is generated, the abort command
cannot get finished and the system hangs, sorry, but checks are
running to hunt down this problem. If there is a real pending command,
the interrupt MUST get generated after abort. In this case, it
should finish well.
Q: The system gets in bad shape after a SCSI-reset, is this known?
A: Yes, as there are a lot of prescriptions (see the Linux Hackers'
Guide) what has to be done for reset, we still share the bad shape of
the reset functions with all other low level SCSI-drivers.
Astonishingly, reset works in most cases quite ok, but the harddisks
won't run in synchronous mode anymore after a reset, until you reboot.
Q: Why does my XXX w/Cache adapter not use read-prefetch?
A: Ok, that is not completely possible. If a cache is present, the
adapter tries to use it internally. Explicitly, one can use the cache
with a read prefetch command, maybe in future, but this requires
some major overhead of SCSI-commands that risks the performance to
go down more than it gets improved. Tests with that are running.
Q: I have a IBM SCSI-2 Fast/Wide adapter, it boots in some way and hangs.
A: Yes, that is understood, as for sure, your SCSI-2 Fast/Wide adapter
was in such a case recognized as integrated SCSI-adapter or something
else, but not as the correct adapter. As the I/O-ports get assigned
wrongly by that reason, the system should crash in most cases. You
should upgrade to the latest release of the SCSI-driver. The
recommended version is 3.2 or later. Here, the F/W support is in
a stable and reliable condition. Wide-addressing is in addition
supported.
Q: I get an Oops message and something like "killing interrupt".
A: The reason for this is that the IBM SCSI-subsystem only sends a
termination status back, if some error appeared. In former releases
of the driver, it was not checked, if the termination status block
is NULL. From version 3.2, it is taken care of this.
Q: I have a F/W adapter and the driver sees my internal SCSI-devices,
but ignores the external ones.
A: Select combined busmode in the IBM config-program and check for that
no SCSI-id on the external devices appears on internal devices.
Reboot afterwards. Dual busmode is supported, but works only for the
internal bus, yet. External bus is still ignored. Take care for your
SCSI-ids. If combined bus-mode is activated, on some adapters,
the wide-addressing is not possible, so devices with ids between 8
and 15 get ignored by the driver & adapter!
Q: I have a 9595 and I get a NMI during heavy SCSI I/O e.g. during fsck.
A COMMAND ERROR is reported and characters on the screen are missing.
Warm reboot is not possible. Things look like quite weird.
A: Check the processor type of your 9595. If you have an 80486 or 486DX-2
processor complex on your mainboard and you compiled a kernel that
supports 80386 processors, it is possible, that the kernel cannot
keep track of the PS/2 interrupt handling and stops on an NMI. Just
compile a kernel for the correct processor type of your PS/2 and
everything should be fine. This is necessary even if one assumes,
that some 80486 system should be downward compatible to 80386
software.
Q: Some commands hang and interrupts block the machine. After some
timeout, the syslog reports that it tries to call abort, but the
machine is frozen.
A: This can be a busy wait bug in the interrupt handler of driver
version 3.2. You should at least upgrade to 3.2c if you use
kernel < 2.4.0 and driver version 4.0 if you use kernel 2.4.0 or
later (including all test releases).
Q: I have a PS/2 model 80 and more than 16 MBytes of RAM. The driver
completely refuses to work, reports NMIs, COMMAND ERRORs or other
ambiguous stuff. When reducing the RAM size down below 16 MB,
everything is running smoothly.
A: No real answer, yet. In any case, one should force the kernel to
present SCBs only below the 16 MBytes barrier. Maybe this solves the
problem. Not yet tried, but guessing that it could work. To get this,
set unchecked_isa_dma argument of ibmmca.h from 0 to 1.
5.3 Bug reports
--------------
If you really find bugs in the source code or the driver will successfully
refuse to work on your machine, you should send a bug report to me. The
best for this is to follow the instructions on the WWW-page for this
driver. Fill out the bug-report form, placed on the WWW-page and ship it,
so the bugs can be taken into account with maximum efforts. But, please
do not send bug reports about this driver to Linus Torvalds or Leonard
Zubkoff, as Linus is buried in E-Mail and Leonard is supervising all
SCSI-drivers and won't have the time left to look inside every single
driver to fix a bug and especially DO NOT send modified code to Linus
Torvalds or Alan J. Cox which has not been checked here!!! They are both
quite buried in E-mail (as me, sometimes, too) and one should first check
for problems on my local teststand. Recently, I got a lot of
bug reports for errors in the ibmmca.c code, which I could not imagine, but
a look inside some Linux-distribution showed me quite often some modified
code, which did no longer work on most other machines than the one of the
modifier. Ok, so now that there is maintenance service available for this
driver, please use this address first in order to keep the level of
confusion low. Thank you!
When you get a SCSI-error message that panics your system, a list of
register-entries of the SCSI-subsystem is shown (from Version 3.1d). With
this list, it is very easy for the maintainer to localize the problem in
the driver or in the configuration of the user. Please write down all the
values from this report and send them to the maintainer. This would really
help a lot and makes life easier concerning misunderstandings.
Use the bug-report form (see 5.4 for its address) to send all the bug-
stuff to the maintainer or write e-mail with the values from the table.
5.4 Support WWW-page
--------------------
The address of the IBM SCSI-subsystem supporting WWW-page is:
http://www.staff.uni-mainz.de/mlang/linux.html
Here you can find info about the background of this driver, patches,
troubleshooting support, news and a bugreport form. Please check that
WWW-page regularly for latest hints. If ever this URL changes, please
refer to the MAINTAINERS file in order to get the latest address.
For the bugreport, please fill out the formular on the corresponding
WWW-page. Read the dedicated instructions and write as much as you
know about your problem. If you do not like such formulars, please send
some e-mail directly, but at least with the same information as required by
the formular.
If you have extensive bug reports, including Oops messages and
screen-shots, please feel free to send it directly to the address
of the maintainer, too. The current address of the maintainer is:
Michael Lang <langa2@kph.uni-mainz.de>
6 References
------------
IBM Corp., "Update for the PS/2 Hardware Interface Technical Reference,
Common Interfaces", Armonk, September 1991, PN 04G3281,
(available in the U.S. for $21.75 at 1-800-IBM-PCTB or in Germany for
around 40,-DM at "Hallo IBM").
IBM Corp., "Personal System/2 Micro Channel SCSI
Adapter with Cache Technical Reference", Armonk, March 1990, PN 68X2365.
IBM Corp., "Personal System/2 Micro Channel SCSI
Adapter Technical Reference", Armonk, March 1990, PN 68X2397.
IBM Corp., "SCSI-2 Fast/Wide Adapter/A Technical Reference - Dual Bus",
Armonk, March 1994, PN 83G7545.
Friedhelm Schmidt, "SCSI-Bus und IDE-Schnittstelle - Moderne Peripherie-
Schnittstellen: Hardware, Protokollbeschreibung und Anwendung", 2. Aufl.
Addison Wesley, 1996.
Michael K. Johnson, "The Linux Kernel Hackers' Guide", Version 0.6, Chapel
Hill - North Carolina, 1995
Andreas Kaiser, "SCSI TAPE BACKUP for OS/2 2.0", Version 2.12, Stuttgart
1993
Helmut Rompel, "IBM Computerwelt GUIDE", What is what bei IBM., Systeme *
Programme * Begriffe, IWT-Verlag GmbH - Muenchen, 1988
7 Credits to
------------
7.1 People
----------
Klaus Grimm
who already a long time ago gave me the old code from the
SCSI-driver in order to get it running for some old machine
in our institute.
Martin Kolinek
who wrote the first release of the IBM SCSI-subsystem driver.
Chris Beauregard
who for a long time maintained MCA-Linux and the SCSI-driver
in the beginning. Chris, wherever you are: Cheers to you!
Klaus Kudielka
with whom in the 2.1.x times, I had a quite fruitful
cooperation to get the driver running as a module and to get
it running with multiple SCSI-adapters.
David Weinehall
for his excellent maintenance of the MCA-stuff and the quite
detailed bug reports and ideas for this driver (and his
patience ;-)).
Alan J. Cox
for his bug reports and his bold activities in cross-checking
the driver-code with his teststand.
7.2 Sponsors & Supporters
-------------------------
"Hallo IBM",
IBM-Deutschland GmbH
the service of IBM-Deutschland for customers. Their E-Mail
service is unbeatable. Whatever old stuff I asked for, I
always got some helpful answers.
Karl-Otto Reimers,
IBM Klub - Sparte IBM Geschichte, Sindelfingen
for sending me a copy of the w/Cache manual from the
IBM-Deutschland archives.
Harald Staiger
for his extensive hardware donations which allows me today
still to test the driver in various constellations.
Erich Fritscher
for his very kind sponsoring.
Louis Ohland,
Charles Lasitter
for support by shipping me an IBM SCSI-2 Fast/Wide manual.
In addition, the contribution of various hardware is quite
decessive and will make it possible to add FWSR (RAID)
adapter support to the driver in the near future! So,
complaints about no RAID support won't remain forever.
Yes, folks, that is no joke, RAID support is going to rise!
Erik Weber
for the great deal we made about a model 9595 and the nice
surrounding equipment and the cool trip to Mannheim
second-hand computer market. In addition, I would like
to thank him for his exhaustive SCSI-driver testing on his
95er PS/2 park.
Anthony Hogbin
for his direct shipment of a SCSI F/W adapter, which allowed
me immediately on the first stage to try it on model 8557
together with onboard SCSI adapter and some SCSI w/Cache.
Andreas Hotz
for his support by memory and an IBM SCSI-adapter. Collecting
all this together now allows me to try really things with
the driver at maximum load and variety on various models in
a very quick and efficient way.
Peter Jennewein
for his model 30, which serves me as part of my teststand
and his cool remark about how you make an ordinary diskette
drive working and how to connect it to an IBM-diskette port.
Johannes Gutenberg-Universitaet, Mainz &
Institut fuer Kernphysik, Mainz Microtron (MAMI)
for the offered space, the link, placed on the central
homepage and the space to store and offer the driver and
related material and the free working times, which allow
me to answer all your e-mail.
8 Trademarks
------------
IBM, PS/2, OS/2, Microchannel are registered trademarks of International
Business Machines Corporation
MS-DOS is a registered trademark of Microsoft Corporation
Microware, OS-9 are registered trademarks of Microware Systems
9 Disclaimer
------------
Beside the GNU General Public License and the dependent disclaimers and disclaimers
concerning the Linux-kernel in special, this SCSI-driver comes without any
warranty. Its functionality is tested as good as possible on certain
machines and combinations of computer hardware, which does not exclude,
that data loss or severe damage of hardware is possible while using this
part of software on some arbitrary computer hardware or in combination
with other software packages. It is highly recommended to make backup
copies of your data before using this software. Furthermore, personal
injuries by hardware defects, that could be caused by this SCSI-driver are
not excluded and it is highly recommended to handle this driver with a
maximum of carefulness.
This driver supports hardware, produced by International Business Machines
Corporation (IBM).
------
Michael Lang
(langa2@kph.uni-mainz.de)
Documentation/scsi/scsi-parameters.txt
View file @ d5b4bb4d
......@@ -37,9 +37,6 @@ parameters may be changed at runtime by the command
eata= [HW,SCSI]
fd_mcs= [HW,SCSI]
See header of drivers/scsi/fd_mcs.c.
fdomain= [HW,SCSI]
See header of drivers/scsi/fdomain.c.
......@@ -48,9 +45,6 @@ parameters may be changed at runtime by the command
gvp11= [HW,SCSI]
ibmmcascsi= [HW,MCA,SCSI] IBM MicroChannel SCSI adapter
See Documentation/mca.txt.
in2000= [HW,SCSI]
See header of drivers/scsi/in2000.c.
......
Documentation/scsi/scsi_mid_low_api.txt
View file @ d5b4bb4d
......@@ -30,7 +30,7 @@ the motherboard (or both). Some aic7xxx based HBAs are dual controllers
and thus represent two hosts. Like most modern HBAs, each aic7xxx host
has its own PCI device address. [The one-to-one correspondence between
a SCSI host and a PCI device is common but not required (e.g. with
ISA or MCA adapters).]
ISA adapters).]
The SCSI mid level isolates an LLD from other layers such as the SCSI
upper layer drivers and the block layer.
......
Documentation/serial/stallion.txt
View file @ d5b4bb4d
......@@ -20,10 +20,10 @@ There are two drivers that work with the different families of Stallion
multiport serial boards. One is for the Stallion smart boards - that is
EasyIO, EasyConnection 8/32 and EasyConnection 8/64-PCI, the other for
the true Stallion intelligent multiport boards - EasyConnection 8/64
(ISA, EISA, MCA), EasyConnection/RA-PCI, ONboard and Brumby.
(ISA, EISA), EasyConnection/RA-PCI, ONboard and Brumby.
If you are using any of the Stallion intelligent multiport boards (Brumby,
ONboard, EasyConnection 8/64 (ISA, EISA, MCA), EasyConnection/RA-PCI) with
ONboard, EasyConnection 8/64 (ISA, EISA), EasyConnection/RA-PCI) with
Linux you will need to get the driver utility package. This contains a
firmware loader and the firmware images necessary to make the devices operate.
......@@ -40,7 +40,7 @@ If you are using the EasyIO, EasyConnection 8/32 or EasyConnection 8/64-PCI
boards then you don't need this package, although it does have a serial stats
display program.
If you require DIP switch settings, EISA or MCA configuration files, or any
If you require DIP switch settings, or EISA configuration files, or any
other information related to Stallion boards then have a look at Stallion's
web pages at http://www.stallion.com.
......@@ -51,13 +51,13 @@ web pages at http://www.stallion.com.
The drivers can be used as loadable modules or compiled into the kernel.
You can choose which when doing a "config" on the kernel.
All ISA, EISA and MCA boards that you want to use need to be configured into
All ISA, and EISA boards that you want to use need to be configured into
the driver(s). All PCI boards will be automatically detected when you load
the driver - so they do not need to be entered into the driver(s)
configuration structure. Note that kernel PCI support is required to use PCI
boards.
There are two methods of configuring ISA, EISA and MCA boards into the drivers.
There are two methods of configuring ISA and EISA boards into the drivers.
If using the driver as a loadable module then the simplest method is to pass
the driver configuration as module arguments. The other method is to modify
the driver source to add configuration lines for each board in use.
......@@ -71,12 +71,12 @@ That makes things pretty simple to get going.
2.1 MODULE DRIVER CONFIGURATION:
The simplest configuration for modules is to use the module load arguments
to configure any ISA, EISA or MCA boards. PCI boards are automatically
to configure any ISA or EISA boards. PCI boards are automatically
detected, so do not need any additional configuration at all.
If using EasyIO, EasyConnection 8/32 ISA or MCA, or EasyConnection 8/63-PCI
If using EasyIO, EasyConnection 8/32 ISA, or EasyConnection 8/63-PCI
boards then use the "stallion" driver module, Otherwise if you are using
an EasyConnection 8/64 ISA, EISA or MCA, EasyConnection/RA-PCI, ONboard,
an EasyConnection 8/64 ISA or EISA, EasyConnection/RA-PCI, ONboard,
Brumby or original Stallion board then use the "istallion" driver module.
Typically to load up the smart board driver use:
......@@ -146,7 +146,7 @@ on each system boot. Typically configuration files are put in the
2.2 STATIC DRIVER CONFIGURATION:
For static driver configuration you need to modify the driver source code.
Entering ISA, EISA and MCA boards into the driver(s) configuration structure
Entering ISA and EISA boards into the driver(s) configuration structure
involves editing the driver(s) source file. It's pretty easy if you follow
the instructions below. Both drivers can support up to 4 boards. The smart
card driver (the stallion.c driver) supports any combination of EasyIO and
......@@ -157,7 +157,7 @@ supports any combination of ONboards, Brumbys, Stallions and EasyConnection
To set up the driver(s) for the boards that you want to use you need to
edit the appropriate driver file and add configuration entries.
If using EasyIO or EasyConnection 8/32 ISA or MCA boards,
If using EasyIO or EasyConnection 8/32 ISA boards,
In drivers/char/stallion.c:
- find the definition of the stl_brdconf array (of structures)
near the top of the file
......@@ -243,7 +243,7 @@ change it on the board.
On EasyIO and EasyConnection 8/32 boards the IRQ is software programmable, so
if there is a conflict you may need to change the IRQ used for a board. There
are no interrupts to worry about for ONboard, Brumby or EasyConnection 8/64
(ISA, EISA and MCA) boards. The memory region on EasyConnection 8/64 and
(ISA and EISA) boards. The memory region on EasyConnection 8/64 and
ONboard boards is software programmable, but not on the Brumby boards.
......
MAINTAINERS
View file @ d5b4bb4d
......@@ -3347,12 +3347,6 @@ T: git git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux.git
S: Maintained
F: arch/ia64/
IBM MCA SCSI SUBSYSTEM DRIVER
M: Michael Lang <langa2@kph.uni-mainz.de>
W: http://www.uni-mainz.de/~langm000/linux.html
S: Maintained
F: drivers/scsi/ibmmca.c
IBM Power Linux RAID adapter
M: Brian King <brking@us.ibm.com>
S: Supported
......@@ -4458,13 +4452,6 @@ T: git git://git.monstr.eu/linux-2.6-microblaze.git
S: Supported
F: arch/microblaze/
MICROCHANNEL ARCHITECTURE (MCA)
M: James Bottomley <James.Bottomley@HansenPartnership.com>
S: Maintained
F: Documentation/mca.txt
F: drivers/mca/
F: include/linux/mca*
MICROTEK X6 SCANNER
M: Oliver Neukum <oliver@neukum.name>
S: Maintained
......
arch/arm/Kconfig
View file @ d5b4bb4d
......@@ -107,14 +107,6 @@ config EISA
config SBUS
bool
config MCA
bool
help
MicroChannel Architecture is found in some IBM PS/2 machines and
laptops. It is a bus system similar to PCI or ISA. See
<file:Documentation/mca.txt> (and especially the web page given
there) before attempting to build an MCA bus kernel.
config STACKTRACE_SUPPORT
bool
default y
......
arch/frv/include/asm/processor.h
View file @ d5b4bb4d
......@@ -54,7 +54,6 @@ extern struct cpuinfo_frv __nongprelbss boot_cpu_data;
* Bus types
*/
#define EISA_bus 0
#define MCA_bus 0
struct thread_struct {
struct pt_regs *frame; /* [GR28] exception frame ptr for this thread */
......
arch/x86/Kconfig
View file @ d5b4bb4d
......@@ -2037,16 +2037,6 @@ config EISA
source "drivers/eisa/Kconfig"
config MCA
bool "MCA support"
---help---
MicroChannel Architecture is found in some IBM PS/2 machines and
laptops. It is a bus system similar to PCI or ISA. See
<file:Documentation/mca.txt> (and especially the web page given
there) before attempting to build an MCA bus kernel.
source "drivers/mca/Kconfig"
config SCx200
tristate "NatSemi SCx200 support"
---help---
......
arch/x86/include/asm/mca.h deleted 100644 → 0
View file @ c80ddb52
/* -*- mode: c; c-basic-offset: 8 -*- */
/* Platform specific MCA defines */
#ifndef _ASM_X86_MCA_H
#define _ASM_X86_MCA_H
/* Maximal number of MCA slots - actually, some machines have less, but
* they all have sufficient number of POS registers to cover 8.
*/
#define MCA_MAX_SLOT_NR 8
/* Most machines have only one MCA bus. The only multiple bus machines
* I know have at most two */
#define MAX_MCA_BUSSES 2
#define MCA_PRIMARY_BUS 0
#define MCA_SECONDARY_BUS 1
/* Dummy slot numbers on primary MCA for integrated functions */
#define MCA_INTEGSCSI (MCA_MAX_SLOT_NR)
#define MCA_INTEGVIDEO (MCA_MAX_SLOT_NR+1)
#define MCA_MOTHERBOARD (MCA_MAX_SLOT_NR+2)
/* Dummy POS values for integrated functions */
#define MCA_DUMMY_POS_START 0x10000
#define MCA_INTEGSCSI_POS (MCA_DUMMY_POS_START+1)
#define MCA_INTEGVIDEO_POS (MCA_DUMMY_POS_START+2)
#define MCA_MOTHERBOARD_POS (MCA_DUMMY_POS_START+3)
/* MCA registers */
#define MCA_MOTHERBOARD_SETUP_REG 0x94
#define MCA_ADAPTER_SETUP_REG 0x96
#define MCA_POS_REG(n) (0x100+(n))
#define MCA_ENABLED 0x01 /* POS 2, set if adapter enabled */
/* Max number of adapters, including both slots and various integrated
* things.
*/
#define MCA_NUMADAPTERS (MCA_MAX_SLOT_NR+3)
#endif /* _ASM_X86_MCA_H */
arch/x86/include/asm/mca_dma.h deleted 100644 → 0
View file @ c80ddb52
#ifndef _ASM_X86_MCA_DMA_H
#define _ASM_X86_MCA_DMA_H
#include <asm/io.h>
#include <linux/ioport.h>
/*
* Microchannel specific DMA stuff. DMA on an MCA machine is fairly similar to
* standard PC dma, but it certainly has its quirks. DMA register addresses
* are in a different place and there are some added functions. Most of this
* should be pretty obvious on inspection. Note that the user must divide
* count by 2 when using 16-bit dma; that is not handled by these functions.
*
* Ramen Noodles are yummy.
*
* 1998 Tymm Twillman <tymm@computer.org>
*/
/*
* Registers that are used by the DMA controller; FN is the function register
* (tell the controller what to do) and EXE is the execution register (how
* to do it)
*/
#define MCA_DMA_REG_FN 0x18
#define MCA_DMA_REG_EXE 0x1A
/*
* Functions that the DMA controller can do
*/
#define MCA_DMA_FN_SET_IO 0x00
#define MCA_DMA_FN_SET_ADDR 0x20
#define MCA_DMA_FN_GET_ADDR 0x30
#define MCA_DMA_FN_SET_COUNT 0x40
#define MCA_DMA_FN_GET_COUNT 0x50
#define MCA_DMA_FN_GET_STATUS 0x60
#define MCA_DMA_FN_SET_MODE 0x70
#define MCA_DMA_FN_SET_ARBUS 0x80
#define MCA_DMA_FN_MASK 0x90
#define MCA_DMA_FN_RESET_MASK 0xA0
#define MCA_DMA_FN_MASTER_CLEAR 0xD0
/*
* Modes (used by setting MCA_DMA_FN_MODE in the function register)
*
* Note that the MODE_READ is read from memory (write to device), and
* MODE_WRITE is vice-versa.
*/
#define MCA_DMA_MODE_XFER 0x04 /* read by default */
#define MCA_DMA_MODE_READ 0x04 /* same as XFER */
#define MCA_DMA_MODE_WRITE 0x08 /* OR with MODE_XFER to use */
#define MCA_DMA_MODE_IO 0x01 /* DMA from IO register */
#define MCA_DMA_MODE_16 0x40 /* 16 bit xfers */
/**
* mca_enable_dma - channel to enable DMA on
* @dmanr: DMA channel
*
* Enable the MCA bus DMA on a channel. This can be called from
* IRQ context.
*/
static inline void mca_enable_dma(unsigned int dmanr)
{
outb(MCA_DMA_FN_RESET_MASK | dmanr, MCA_DMA_REG_FN);
}
/**
* mca_disble_dma - channel to disable DMA on
* @dmanr: DMA channel
*
* Enable the MCA bus DMA on a channel. This can be called from
* IRQ context.
*/
static inline void mca_disable_dma(unsigned int dmanr)
{
outb(MCA_DMA_FN_MASK | dmanr, MCA_DMA_REG_FN);
}
/**
* mca_set_dma_addr - load a 24bit DMA address
* @dmanr: DMA channel
* @a: 24bit bus address
*
* Load the address register in the DMA controller. This has a 24bit
* limitation (16Mb).
*/
static inline void mca_set_dma_addr(unsigned int dmanr, unsigned int a)
{
outb(MCA_DMA_FN_SET_ADDR | dmanr, MCA_DMA_REG_FN);
outb(a & 0xff, MCA_DMA_REG_EXE);
outb((a >> 8) & 0xff, MCA_DMA_REG_EXE);
outb((a >> 16) & 0xff, MCA_DMA_REG_EXE);
}
/**
* mca_get_dma_addr - load a 24bit DMA address
* @dmanr: DMA channel
*
* Read the address register in the DMA controller. This has a 24bit
* limitation (16Mb). The return is a bus address.
*/
static inline unsigned int mca_get_dma_addr(unsigned int dmanr)
{
unsigned int addr;
outb(MCA_DMA_FN_GET_ADDR | dmanr, MCA_DMA_REG_FN);
addr = inb(MCA_DMA_REG_EXE);
addr |= inb(MCA_DMA_REG_EXE) << 8;
addr |= inb(MCA_DMA_REG_EXE) << 16;
return addr;
}
/**
* mca_set_dma_count - load a 16bit transfer count
* @dmanr: DMA channel
* @count: count
*
* Set the DMA count for this channel. This can be up to 64Kbytes.
* Setting a count of zero will not do what you expect.
*/
static inline void mca_set_dma_count(unsigned int dmanr, unsigned int count)
{
count--; /* transfers one more than count -- correct for this */
outb(MCA_DMA_FN_SET_COUNT | dmanr, MCA_DMA_REG_FN);
outb(count & 0xff, MCA_DMA_REG_EXE);
outb((count >> 8) & 0xff, MCA_DMA_REG_EXE);
}
/**
* mca_get_dma_residue - get the remaining bytes to transfer
* @dmanr: DMA channel
*
* This function returns the number of bytes left to transfer
* on this DMA channel.
*/
static inline unsigned int mca_get_dma_residue(unsigned int dmanr)
{
unsigned short count;
outb(MCA_DMA_FN_GET_COUNT | dmanr, MCA_DMA_REG_FN);
count = 1 + inb(MCA_DMA_REG_EXE);
count += inb(MCA_DMA_REG_EXE) << 8;
return count;
}
/**
* mca_set_dma_io - set the port for an I/O transfer
* @dmanr: DMA channel
* @io_addr: an I/O port number
*
* Unlike the ISA bus DMA controllers the DMA on MCA bus can transfer
* with an I/O port target.
*/
static inline void mca_set_dma_io(unsigned int dmanr, unsigned int io_addr)
{
/*
* DMA from a port address -- set the io address
*/
outb(MCA_DMA_FN_SET_IO | dmanr, MCA_DMA_REG_FN);
outb(io_addr & 0xff, MCA_DMA_REG_EXE);
outb((io_addr >> 8) & 0xff, MCA_DMA_REG_EXE);
}
/**
* mca_set_dma_mode - set the DMA mode
* @dmanr: DMA channel
* @mode: mode to set
*
* The DMA controller supports several modes. The mode values you can
* set are-
*
* %MCA_DMA_MODE_READ when reading from the DMA device.
*
* %MCA_DMA_MODE_WRITE to writing to the DMA device.
*
* %MCA_DMA_MODE_IO to do DMA to or from an I/O port.
*
* %MCA_DMA_MODE_16 to do 16bit transfers.
*/
static inline void mca_set_dma_mode(unsigned int dmanr, unsigned int mode)
{
outb(MCA_DMA_FN_SET_MODE | dmanr, MCA_DMA_REG_FN);
outb(mode, MCA_DMA_REG_EXE);
}
#endif /* _ASM_X86_MCA_DMA_H */
arch/x86/include/asm/mpspec.h
View file @ d5b4bb4d
......@@ -40,7 +40,7 @@ extern int quad_local_to_mp_bus_id [NR_CPUS/4][4];
#endif /* CONFIG_X86_64 */
#if defined(CONFIG_MCA) || defined(CONFIG_EISA)
#ifdef CONFIG_EISA
extern int mp_bus_id_to_type[MAX_MP_BUSSES];
#endif
......
arch/x86/include/asm/mpspec_def.h
View file @ d5b4bb4d
......@@ -84,7 +84,7 @@ struct mpc_bus {
#define BUSTYPE_EISA "EISA"
#define BUSTYPE_ISA "ISA"
#define BUSTYPE_INTERN "INTERN" /* Internal BUS */
#define BUSTYPE_MCA "MCA"
#define BUSTYPE_MCA "MCA" /* Obsolete */
#define BUSTYPE_VL "VL" /* Local bus */
#define BUSTYPE_PCI "PCI"
#define BUSTYPE_PCMCIA "PCMCIA"
......@@ -169,6 +169,5 @@ enum mp_bustype {
MP_BUS_ISA = 1,
MP_BUS_EISA,
MP_BUS_PCI,
MP_BUS_MCA,
};
#endif /* _ASM_X86_MPSPEC_DEF_H */
arch/x86/kernel/Makefile
View file @ d5b4bb4d
......@@ -49,7 +49,6 @@ obj-y += cpu/
obj-y += acpi/
obj-y += reboot.o
obj-$(CONFIG_X86_32) += reboot_32.o
obj-$(CONFIG_MCA) += mca_32.o
obj-$(CONFIG_X86_MSR) += msr.o
obj-$(CONFIG_X86_CPUID) += cpuid.o
obj-$(CONFIG_PCI) += early-quirks.o
......
arch/x86/kernel/acpi/boot.c
View file @ d5b4bb4d
......@@ -990,7 +990,7 @@ void __init mp_config_acpi_legacy_irqs(void)
int i;
struct mpc_intsrc mp_irq;
#if defined (CONFIG_MCA) || defined (CONFIG_EISA)
#ifdef CONFIG_EISA
/*
* Fabricate the legacy ISA bus (bus #31).
*/
......
arch/x86/kernel/apic/io_apic.c
View file @ d5b4bb4d
......@@ -140,7 +140,7 @@ int mp_irq_entries;
/* GSI interrupts */
static int nr_irqs_gsi = NR_IRQS_LEGACY;
#if defined (CONFIG_MCA) || defined (CONFIG_EISA)
#ifdef CONFIG_EISA
int mp_bus_id_to_type[MAX_MP_BUSSES];
#endif
......@@ -835,7 +835,7 @@ static int __init find_isa_irq_apic(int irq, int type)
return -1;
}
#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
#ifdef CONFIG_EISA
/*
* EISA Edge/Level control register, ELCR
*/
......@@ -872,12 +872,6 @@ static int EISA_ELCR(unsigned int irq)
#define default_PCI_trigger(idx) (1)
#define default_PCI_polarity(idx) (1)
/* MCA interrupts are always polarity zero level triggered,
* when listed as conforming in the MP table. */
#define default_MCA_trigger(idx) (1)
#define default_MCA_polarity(idx) default_ISA_polarity(idx)
static int irq_polarity(int idx)
{
int bus = mp_irqs[idx].srcbus;
......@@ -935,7 +929,7 @@ static int irq_trigger(int idx)
trigger = default_ISA_trigger(idx);
else
trigger = default_PCI_trigger(idx);
#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
#ifdef CONFIG_EISA
switch (mp_bus_id_to_type[bus]) {
case MP_BUS_ISA: /* ISA pin */
{
......@@ -952,11 +946,6 @@ static int irq_trigger(int idx)
/* set before the switch */
break;
}
case MP_BUS_MCA: /* MCA pin */
{
trigger = default_MCA_trigger(idx);
break;
}
default:
{
printk(KERN_WARNING "broken BIOS!!\n");
......
arch/x86/kernel/mca_32.c deleted 100644 → 0
View file @ c80ddb52
/*
* Written by Martin Kolinek, February 1996
*
* Changes:
*
* Chris Beauregard July 28th, 1996
* - Fixed up integrated SCSI detection
*
* Chris Beauregard August 3rd, 1996
* - Made mca_info local
* - Made integrated registers accessible through standard function calls
* - Added name field
* - More sanity checking
*
* Chris Beauregard August 9th, 1996
* - Rewrote /proc/mca
*
* Chris Beauregard January 7th, 1997
* - Added basic NMI-processing
* - Added more information to mca_info structure
*
* David Weinehall October 12th, 1998
* - Made a lot of cleaning up in the source
* - Added use of save_flags / restore_flags
* - Added the 'driver_loaded' flag in MCA_adapter
* - Added an alternative implemention of ZP Gu's mca_find_unused_adapter
*
* David Weinehall March 24th, 1999
* - Fixed the output of 'Driver Installed' in /proc/mca/pos
* - Made the Integrated Video & SCSI show up even if they have id 0000
*
* Alexander Viro November 9th, 1999
* - Switched to regular procfs methods
*
* Alfred Arnold & David Weinehall August 23rd, 2000
* - Added support for Planar POS-registers
*/
#include <linux/module.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/mca.h>
#include <linux/kprobes.h>
#include <linux/slab.h>
#include <asm/io.h>
#include <linux/proc_fs.h>
#include <linux/mman.h>
#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/ioport.h>
#include <asm/uaccess.h>
#include <linux/init.h>
static unsigned char which_scsi;
int MCA_bus;
EXPORT_SYMBOL(MCA_bus);
/*
* Motherboard register spinlock. Untested on SMP at the moment, but
* are there any MCA SMP boxes?
*
* Yes - Alan
*/
static DEFINE_SPINLOCK(mca_lock);
/* Build the status info for the adapter */
static void mca_configure_adapter_status(struct mca_device *mca_dev)
{
mca_dev->status = MCA_ADAPTER_NONE;
mca_dev->pos_id = mca_dev->pos[0]
+ (mca_dev->pos[1] << 8);
if (!mca_dev->pos_id && mca_dev->slot < MCA_MAX_SLOT_NR) {
/*
* id = 0x0000 usually indicates hardware failure,
* however, ZP Gu (zpg@castle.net> reports that his 9556
* has 0x0000 as id and everything still works. There
* also seem to be an adapter with id = 0x0000; the
* NCR Parallel Bus Memory Card. Until this is confirmed,
* however, this code will stay.
*/
mca_dev->status = MCA_ADAPTER_ERROR;
return;
} else if (mca_dev->pos_id != 0xffff) {
/*
* 0xffff usually indicates that there's no adapter,
* however, some integrated adapters may have 0xffff as
* their id and still be valid. Examples are on-board
* VGA of the 55sx, the integrated SCSI of the 56 & 57,
* and possibly also the 95 ULTIMEDIA.
*/
mca_dev->status = MCA_ADAPTER_NORMAL;
}
if ((mca_dev->pos_id == 0xffff ||
mca_dev->pos_id == 0x0000) && mca_dev->slot >= MCA_MAX_SLOT_NR) {
int j;
for (j = 2; j < 8; j++) {
if (mca_dev->pos[j] != 0xff) {
mca_dev->status = MCA_ADAPTER_NORMAL;
break;
}
}
}
if (!(mca_dev->pos[2] & MCA_ENABLED)) {
/* enabled bit is in POS 2 */
mca_dev->status = MCA_ADAPTER_DISABLED;
}
} /* mca_configure_adapter_status */
/*--------------------------------------------------------------------*/
static struct resource mca_standard_resources[] = {
{ .start = 0x60, .end = 0x60, .name = "system control port B (MCA)" },
{ .start = 0x90, .end = 0x90, .name = "arbitration (MCA)" },
{ .start = 0x91, .end = 0x91, .name = "card Select Feedback (MCA)" },
{ .start = 0x92, .end = 0x92, .name = "system Control port A (MCA)" },
{ .start = 0x94, .end = 0x94, .name = "system board setup (MCA)" },
{ .start = 0x96, .end = 0x97, .name = "POS (MCA)" },
{ .start = 0x100, .end = 0x107, .name = "POS (MCA)" }
};
#define MCA_STANDARD_RESOURCES ARRAY_SIZE(mca_standard_resources)
/*
* mca_read_and_store_pos - read the POS registers into a memory buffer
* @pos: a char pointer to 8 bytes, contains the POS register value on
* successful return
*
* Returns 1 if a card actually exists (i.e. the pos isn't
* all 0xff) or 0 otherwise
*/
static int mca_read_and_store_pos(unsigned char *pos)
{
int j;
int found = 0;
for (j = 0; j < 8; j++) {
pos[j] = inb_p(MCA_POS_REG(j));
if (pos[j] != 0xff) {
/* 0xff all across means no device. 0x00 means
* something's broken, but a device is
* probably there. However, if you get 0x00
* from a motherboard register it won't matter
* what we find. For the record, on the
* 57SLC, the integrated SCSI adapter has
* 0xffff for the adapter ID, but nonzero for
* other registers. */
found = 1;
}
}
return found;
}
static unsigned char mca_pc_read_pos(struct mca_device *mca_dev, int reg)
{
unsigned char byte;
unsigned long flags;
if (reg < 0 || reg >= 8)
return 0;
spin_lock_irqsave(&mca_lock, flags);
if (mca_dev->pos_register) {
/* Disable adapter setup, enable motherboard setup */
outb_p(0, MCA_ADAPTER_SETUP_REG);
outb_p(mca_dev->pos_register, MCA_MOTHERBOARD_SETUP_REG);
byte = inb_p(MCA_POS_REG(reg));
outb_p(0xff, MCA_MOTHERBOARD_SETUP_REG);
} else {
/* Make sure motherboard setup is off */
outb_p(0xff, MCA_MOTHERBOARD_SETUP_REG);
/* Read the appropriate register */
outb_p(0x8|(mca_dev->slot & 0xf), MCA_ADAPTER_SETUP_REG);
byte = inb_p(MCA_POS_REG(reg));
outb_p(0, MCA_ADAPTER_SETUP_REG);
}
spin_unlock_irqrestore(&mca_lock, flags);
mca_dev->pos[reg] = byte;
return byte;
}
static void mca_pc_write_pos(struct mca_device *mca_dev, int reg,
unsigned char byte)
{
unsigned long flags;
if (reg < 0 || reg >= 8)
return;
spin_lock_irqsave(&mca_lock, flags);
/* Make sure motherboard setup is off */
outb_p(0xff, MCA_MOTHERBOARD_SETUP_REG);
/* Read in the appropriate register */
outb_p(0x8|(mca_dev->slot&0xf), MCA_ADAPTER_SETUP_REG);
outb_p(byte, MCA_POS_REG(reg));
outb_p(0, MCA_ADAPTER_SETUP_REG);
spin_unlock_irqrestore(&mca_lock, flags);
/* Update the global register list, while we have the byte */
mca_dev->pos[reg] = byte;
}
/* for the primary MCA bus, we have identity transforms */
static int mca_dummy_transform_irq(struct mca_device *mca_dev, int irq)
{
return irq;
}
static int mca_dummy_transform_ioport(struct mca_device *mca_dev, int port)
{
return port;
}
static void *mca_dummy_transform_memory(struct mca_device *mca_dev, void *mem)
{
return mem;
}
static int __init mca_init(void)
{
unsigned int i, j;
struct mca_device *mca_dev;
unsigned char pos[8];
short mca_builtin_scsi_ports[] = {0xf7, 0xfd, 0x00};
struct mca_bus *bus;
/*
* WARNING: Be careful when making changes here. Putting an adapter
* and the motherboard simultaneously into setup mode may result in
* damage to chips (according to The Indispensable PC Hardware Book
* by Hans-Peter Messmer). Also, we disable system interrupts (so
* that we are not disturbed in the middle of this).
*/
/* Make sure the MCA bus is present */
if (mca_system_init()) {
printk(KERN_ERR "MCA bus system initialisation failed\n");
return -ENODEV;
}
if (!MCA_bus)
return -ENODEV;
printk(KERN_INFO "Micro Channel bus detected.\n");
/* All MCA systems have at least a primary bus */
bus = mca_attach_bus(MCA_PRIMARY_BUS);
if (!bus)
goto out_nomem;
bus->default_dma_mask = 0xffffffffLL;
bus->f.mca_write_pos = mca_pc_write_pos;
bus->f.mca_read_pos = mca_pc_read_pos;
bus->f.mca_transform_irq = mca_dummy_transform_irq;
bus->f.mca_transform_ioport = mca_dummy_transform_ioport;
bus->f.mca_transform_memory = mca_dummy_transform_memory;
/* get the motherboard device */
mca_dev = kzalloc(sizeof(struct mca_device), GFP_KERNEL);
if (unlikely(!mca_dev))
goto out_nomem;
/*
* We do not expect many MCA interrupts during initialization,
* but let us be safe:
*/
spin_lock_irq(&mca_lock);
/* Make sure adapter setup is off */
outb_p(0, MCA_ADAPTER_SETUP_REG);
/* Read motherboard POS registers */
mca_dev->pos_register = 0x7f;
outb_p(mca_dev->pos_register, MCA_MOTHERBOARD_SETUP_REG);
mca_dev->name[0] = 0;
mca_read_and_store_pos(mca_dev->pos);
mca_configure_adapter_status(mca_dev);
/* fake POS and slot for a motherboard */
mca_dev->pos_id = MCA_MOTHERBOARD_POS;
mca_dev->slot = MCA_MOTHERBOARD;
mca_register_device(MCA_PRIMARY_BUS, mca_dev);
mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC);
if (unlikely(!mca_dev))
goto out_unlock_nomem;
/* Put motherboard into video setup mode, read integrated video
* POS registers, and turn motherboard setup off.
*/
mca_dev->pos_register = 0xdf;
outb_p(mca_dev->pos_register, MCA_MOTHERBOARD_SETUP_REG);
mca_dev->name[0] = 0;
mca_read_and_store_pos(mca_dev->pos);
mca_configure_adapter_status(mca_dev);
/* fake POS and slot for the integrated video */
mca_dev->pos_id = MCA_INTEGVIDEO_POS;
mca_dev->slot = MCA_INTEGVIDEO;
mca_register_device(MCA_PRIMARY_BUS, mca_dev);
/*
* Put motherboard into scsi setup mode, read integrated scsi
* POS registers, and turn motherboard setup off.
*
* It seems there are two possible SCSI registers. Martin says that
* for the 56,57, 0xf7 is the one, but fails on the 76.
* Alfredo (apena@vnet.ibm.com) says
* 0xfd works on his machine. We'll try both of them. I figure it's
* a good bet that only one could be valid at a time. This could
* screw up though if one is used for something else on the other
* machine.
*/
for (i = 0; (which_scsi = mca_builtin_scsi_ports[i]) != 0; i++) {
outb_p(which_scsi, MCA_MOTHERBOARD_SETUP_REG);
if (mca_read_and_store_pos(pos))
break;
}
if (which_scsi) {
/* found a scsi card */
mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC);
if (unlikely(!mca_dev))
goto out_unlock_nomem;
for (j = 0; j < 8; j++)
mca_dev->pos[j] = pos[j];
mca_configure_adapter_status(mca_dev);
/* fake POS and slot for integrated SCSI controller */
mca_dev->pos_id = MCA_INTEGSCSI_POS;
mca_dev->slot = MCA_INTEGSCSI;
mca_dev->pos_register = which_scsi;
mca_register_device(MCA_PRIMARY_BUS, mca_dev);
}
/* Turn off motherboard setup */
outb_p(0xff, MCA_MOTHERBOARD_SETUP_REG);
/*
* Now loop over MCA slots: put each adapter into setup mode, and
* read its POS registers. Then put adapter setup off.
*/
for (i = 0; i < MCA_MAX_SLOT_NR; i++) {
outb_p(0x8|(i&0xf), MCA_ADAPTER_SETUP_REG);
if (!mca_read_and_store_pos(pos))
continue;
mca_dev = kzalloc(sizeof(struct mca_device), GFP_ATOMIC);
if (unlikely(!mca_dev))
goto out_unlock_nomem;
for (j = 0; j < 8; j++)
mca_dev->pos[j] = pos[j];
mca_dev->driver_loaded = 0;
mca_dev->slot = i;
mca_dev->pos_register = 0;
mca_configure_adapter_status(mca_dev);
mca_register_device(MCA_PRIMARY_BUS, mca_dev);
}
outb_p(0, MCA_ADAPTER_SETUP_REG);
/* Enable interrupts and return memory start */
spin_unlock_irq(&mca_lock);
for (i = 0; i < MCA_STANDARD_RESOURCES; i++)
request_resource(&ioport_resource, mca_standard_resources + i);
mca_do_proc_init();
return 0;
out_unlock_nomem:
spin_unlock_irq(&mca_lock);
out_nomem:
printk(KERN_EMERG "Failed memory allocation in MCA setup!\n");
return -ENOMEM;
}
subsys_initcall(mca_init);
/*--------------------------------------------------------------------*/
static __kprobes void
mca_handle_nmi_device(struct mca_device *mca_dev, int check_flag)
{
int slot = mca_dev->slot;
if (slot == MCA_INTEGSCSI) {
printk(KERN_CRIT "NMI: caused by MCA integrated SCSI adapter (%s)\n",
mca_dev->name);
} else if (slot == MCA_INTEGVIDEO) {
printk(KERN_CRIT "NMI: caused by MCA integrated video adapter (%s)\n",
mca_dev->name);
} else if (slot == MCA_MOTHERBOARD) {
printk(KERN_CRIT "NMI: caused by motherboard (%s)\n",
mca_dev->name);
}
/* More info available in POS 6 and 7? */
if (check_flag) {
unsigned char pos6, pos7;
pos6 = mca_device_read_pos(mca_dev, 6);
pos7 = mca_device_read_pos(mca_dev, 7);
printk(KERN_CRIT "NMI: POS 6 = 0x%x, POS 7 = 0x%x\n", pos6, pos7);
}
} /* mca_handle_nmi_slot */
/*--------------------------------------------------------------------*/
static int __kprobes mca_handle_nmi_callback(struct device *dev, void *data)
{
struct mca_device *mca_dev = to_mca_device(dev);
unsigned char pos5;
pos5 = mca_device_read_pos(mca_dev, 5);
if (!(pos5 & 0x80)) {
/*
* Bit 7 of POS 5 is reset when this adapter has a hardware
* error. Bit 7 it reset if there's error information
* available in POS 6 and 7.
*/
mca_handle_nmi_device(mca_dev, !(pos5 & 0x40));
return 1;
}
return 0;
}
void __kprobes mca_handle_nmi(void)
{
/*
* First try - scan the various adapters and see if a specific
* adapter was responsible for the error.
*/
bus_for_each_dev(&mca_bus_type, NULL, NULL, mca_handle_nmi_callback);
}
arch/x86/kernel/mpparse.c
View file @ d5b4bb4d
......@@ -97,7 +97,7 @@ static void __init MP_bus_info(struct mpc_bus *m)
set_bit(m->busid, mp_bus_not_pci);
if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA) - 1) == 0) {
#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
#ifdef CONFIG_EISA
mp_bus_id_to_type[m->busid] = MP_BUS_ISA;
#endif
} else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI) - 1) == 0) {
......@@ -105,12 +105,10 @@ static void __init MP_bus_info(struct mpc_bus *m)
x86_init.mpparse.mpc_oem_pci_bus(m);
clear_bit(m->busid, mp_bus_not_pci);
#if defined(CONFIG_EISA) || defined(CONFIG_MCA)
#ifdef CONFIG_EISA
mp_bus_id_to_type[m->busid] = MP_BUS_PCI;
} else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA) - 1) == 0) {
mp_bus_id_to_type[m->busid] = MP_BUS_EISA;
} else if (strncmp(str, BUSTYPE_MCA, sizeof(BUSTYPE_MCA) - 1) == 0) {
mp_bus_id_to_type[m->busid] = MP_BUS_MCA;
#endif
} else
printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str);
......@@ -368,9 +366,6 @@ static void __init construct_ioapic_table(int mpc_default_type)
case 3:
memcpy(bus.bustype, "EISA ", 6);
break;
case 4:
case 7:
memcpy(bus.bustype, "MCA ", 6);
}
MP_bus_info(&bus);
if (mpc_default_type > 4) {
......@@ -623,7 +618,7 @@ void __init default_find_smp_config(void)
return;
/*
* If it is an SMP machine we should know now, unless the
* configuration is in an EISA/MCA bus machine with an
* configuration is in an EISA bus machine with an
* extended bios data area.
*
* there is a real-mode segmented pointer pointing to the
......
arch/x86/kernel/nmi.c
View file @ d5b4bb4d
......@@ -19,8 +19,6 @@
#include <linux/slab.h>
#include <linux/export.h>
#include <linux/mca.h>
#if defined(CONFIG_EDAC)
#include <linux/edac.h>
#endif
......@@ -247,16 +245,6 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
__this_cpu_add(nmi_stats.unknown, 1);
#ifdef CONFIG_MCA
/*
* Might actually be able to figure out what the guilty party
* is:
*/
if (MCA_bus) {
mca_handle_nmi();
return;
}
#endif
pr_emerg("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
reason, smp_processor_id());
......
arch/x86/kernel/setup.c
View file @ d5b4bb4d
......@@ -34,7 +34,6 @@
#include <linux/memblock.h>
#include <linux/seq_file.h>
#include <linux/console.h>
#include <linux/mca.h>
#include <linux/root_dev.h>
#include <linux/highmem.h>
#include <linux/module.h>
......@@ -179,12 +178,6 @@ struct cpuinfo_x86 new_cpu_data __cpuinitdata = {0, 0, 0, 0, -1, 1, 0, 0, -1};
/* common cpu data for all cpus */
struct cpuinfo_x86 boot_cpu_data __read_mostly = {0, 0, 0, 0, -1, 1, 0, 0, -1};
EXPORT_SYMBOL(boot_cpu_data);
static void set_mca_bus(int x)
{
#ifdef CONFIG_MCA
MCA_bus = x;
#endif
}
unsigned int def_to_bigsmp;
......@@ -716,7 +709,6 @@ void __init setup_arch(char **cmdline_p)
apm_info.bios = boot_params.apm_bios_info;
ist_info = boot_params.ist_info;
if (boot_params.sys_desc_table.length != 0) {
set_mca_bus(boot_params.sys_desc_table.table[3] & 0x2);
machine_id = boot_params.sys_desc_table.table[0];
machine_submodel_id = boot_params.sys_desc_table.table[1];
BIOS_revision = boot_params.sys_desc_table.table[2];
......
arch/x86/kernel/time.c
View file @ d5b4bb4d
......@@ -14,7 +14,6 @@
#include <linux/i8253.h>
#include <linux/time.h>
#include <linux/export.h>
#include <linux/mca.h>
#include <asm/vsyscall.h>
#include <asm/x86_init.h>
......@@ -58,11 +57,6 @@ EXPORT_SYMBOL(profile_pc);
static irqreturn_t timer_interrupt(int irq, void *dev_id)
{
global_clock_event->event_handler(global_clock_event);
/* MCA bus quirk: Acknowledge irq0 by setting bit 7 in port 0x61 */
if (MCA_bus)
outb_p(inb_p(0x61)| 0x80, 0x61);
return IRQ_HANDLED;
}
......
arch/x86/kernel/traps.c
View file @ d5b4bb4d
......@@ -37,10 +37,6 @@
#include <linux/eisa.h>
#endif
#ifdef CONFIG_MCA
#include <linux/mca.h>
#endif
#if defined(CONFIG_EDAC)
#include <linux/edac.h>
#endif
......
drivers/Makefile
View file @ d5b4bb4d
......@@ -92,7 +92,6 @@ obj-$(CONFIG_BT) += bluetooth/
obj-$(CONFIG_ACCESSIBILITY) += accessibility/
obj-$(CONFIG_ISDN) += isdn/
obj-$(CONFIG_EDAC) += edac/
obj-$(CONFIG_MCA) += mca/
obj-$(CONFIG_EISA) += eisa/
obj-y += lguest/
obj-$(CONFIG_CPU_FREQ) += cpufreq/
......
drivers/mca/Kconfig deleted 100644 → 0
View file @ c80ddb52
config MCA_LEGACY
bool "Legacy MCA API Support"
depends on MCA
help
This compiles in support for the old slot based MCA API. If you
have an unconverted MCA driver, you will need to say Y here. It
is safe to say Y anyway.
config MCA_PROC_FS
bool "Support for the mca entry in /proc"
depends on MCA_LEGACY && PROC_FS
help
If you want the old style /proc/mca directory in addition to the
new style sysfs say Y here.
drivers/mca/Makefile deleted 100644 → 0
View file @ c80ddb52
# Makefile for the Linux MCA bus support
obj-y := mca-bus.o mca-device.o mca-driver.o
obj-$(CONFIG_MCA_PROC_FS) += mca-proc.o
obj-$(CONFIG_MCA_LEGACY) += mca-legacy.o
drivers/mca/mca-bus.c deleted 100644 → 0
View file @ c80ddb52
/* -*- mode: c; c-basic-offset: 8 -*- */
/*
* MCA bus support functions for sysfs.
*
* (C) 2002 James Bottomley <James.Bottomley@HansenPartnership.com>
*
**-----------------------------------------------------------------------------
**
** This program is free software; you can redistribute it and/or modify
** it under the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 2 of the License, or
** (at your option) any later version.
**
** This program is distributed in the hope that it will be useful,
** but WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
** GNU General Public License for more details.
**
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software
** Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
**
**-----------------------------------------------------------------------------
*/
#include <linux/kernel.h>
#include <linux/device.h>
#include <linux/mca.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/slab.h>
/* Very few machines have more than one MCA bus. However, there are
* those that do (Voyager 35xx/5xxx), so we do it this way for future
* expansion. None that I know have more than 2 */
static struct mca_bus *mca_root_busses[MAX_MCA_BUSSES];
#define MCA_DEVINFO(i,s) { .pos = i, .name = s }
struct mca_device_info {
short pos_id; /* the 2 byte pos id for this card */
char name[50];
};
static int mca_bus_match (struct device *dev, struct device_driver *drv)
{
struct mca_device *mca_dev = to_mca_device (dev);
struct mca_driver *mca_drv = to_mca_driver (drv);
const unsigned short *mca_ids = mca_drv->id_table;
int i = 0;
if (mca_ids) {
for(i = 0; mca_ids[i]; i++) {
if (mca_ids[i] == mca_dev->pos_id) {
mca_dev->index = i;
return 1;
}
}
}
/* If the integrated id is present, treat it as though it were an
* additional id in the id_table (it can't be because by definition,
* integrated id's overflow a short */
if (mca_drv->integrated_id && mca_dev->pos_id ==
mca_drv->integrated_id) {
mca_dev->index = i;
return 1;
}
return 0;
}
struct bus_type mca_bus_type = {
.name = "MCA",
.match = mca_bus_match,
};
EXPORT_SYMBOL (mca_bus_type);
static ssize_t mca_show_pos_id(struct device *dev, struct device_attribute *attr, char *buf)
{
/* four digits, \n and trailing \0 */
struct mca_device *mca_dev = to_mca_device(dev);
int len;
if(mca_dev->pos_id < MCA_DUMMY_POS_START)
len = sprintf(buf, "%04x\n", mca_dev->pos_id);
else
len = sprintf(buf, "none\n");
return len;
}
static ssize_t mca_show_pos(struct device *dev, struct device_attribute *attr, char *buf)
{
/* enough for 8 two byte hex chars plus space and new line */
int j, len=0;
struct mca_device *mca_dev = to_mca_device(dev);
for(j=0; j<8; j++)
len += sprintf(buf+len, "%02x ", mca_dev->pos[j]);
/* change last trailing space to new line */
buf[len-1] = '\n';
return len;
}
static DEVICE_ATTR(id, S_IRUGO, mca_show_pos_id, NULL);
static DEVICE_ATTR(pos, S_IRUGO, mca_show_pos, NULL);
int __init mca_register_device(int bus, struct mca_device *mca_dev)
{
struct mca_bus *mca_bus = mca_root_busses[bus];
int rc;
mca_dev->dev.parent = &mca_bus->dev;
mca_dev->dev.bus = &mca_bus_type;
dev_set_name(&mca_dev->dev, "%02d:%02X", bus, mca_dev->slot);
mca_dev->dma_mask = mca_bus->default_dma_mask;
mca_dev->dev.dma_mask = &mca_dev->dma_mask;
mca_dev->dev.coherent_dma_mask = mca_dev->dma_mask;
rc = device_register(&mca_dev->dev);
if (rc)
goto err_out;
rc = device_create_file(&mca_dev->dev, &dev_attr_id);
if (rc) goto err_out_devreg;
rc = device_create_file(&mca_dev->dev, &dev_attr_pos);
if (rc) goto err_out_id;
return 1;
err_out_id:
device_remove_file(&mca_dev->dev, &dev_attr_id);
err_out_devreg:
device_unregister(&mca_dev->dev);
err_out:
return 0;
}
/* */
struct mca_bus * __devinit mca_attach_bus(int bus)
{
struct mca_bus *mca_bus;
if (unlikely(mca_root_busses[bus] != NULL)) {
/* This should never happen, but just in case */
printk(KERN_EMERG "MCA tried to add already existing bus %d\n",
bus);
dump_stack();
return NULL;
}
mca_bus = kzalloc(sizeof(struct mca_bus), GFP_KERNEL);
if (!mca_bus)
return NULL;
dev_set_name(&mca_bus->dev, "mca%d", bus);
sprintf(mca_bus->name,"Host %s MCA Bridge", bus ? "Secondary" : "Primary");
if (device_register(&mca_bus->dev)) {
kfree(mca_bus);
return NULL;
}
mca_root_busses[bus] = mca_bus;
return mca_bus;
}
int __init mca_system_init (void)
{
return bus_register(&mca_bus_type);
}
drivers/mca/mca-device.c deleted 100644 → 0
View file @ c80ddb52
/* -*- mode: c; c-basic-offset: 8 -*- */
/*
* MCA device support functions
*
* These functions support the ongoing device access API.
*
* (C) 2002 James Bottomley <James.Bottomley@HansenPartnership.com>
*
**-----------------------------------------------------------------------------
**
** This program is free software; you can redistribute it and/or modify
** it under the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 2 of the License, or
** (at your option) any later version.
**
** This program is distributed in the hope that it will be useful,
** but WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
** GNU General Public License for more details.
**
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software
** Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
**
**-----------------------------------------------------------------------------
*/
#include <linux/module.h>
#include <linux/device.h>
#include <linux/mca.h>
#include <linux/string.h>
/**
* mca_device_read_stored_pos - read POS register from stored data
* @mca_dev: device to read from
* @reg: register to read from
*
* Fetch a POS value that was stored at boot time by the kernel
* when it scanned the MCA space. The register value is returned.
* Missing or invalid registers report 0.
*/
unsigned char mca_device_read_stored_pos(struct mca_device *mca_dev, int reg)
{
if(reg < 0 || reg >= 8)
return 0;
return mca_dev->pos[reg];
}
EXPORT_SYMBOL(mca_device_read_stored_pos);
/**
* mca_device_read_pos - read POS register from card
* @mca_dev: device to read from
* @reg: register to read from
*
* Fetch a POS value directly from the hardware to obtain the
* current value. This is much slower than
* mca_device_read_stored_pos and may not be invoked from
* interrupt context. It handles the deep magic required for
* onboard devices transparently.
*/
unsigned char mca_device_read_pos(struct mca_device *mca_dev, int reg)
{
struct mca_bus *mca_bus = to_mca_bus(mca_dev->dev.parent);
return mca_bus->f.mca_read_pos(mca_dev, reg);
return mca_dev->pos[reg];
}
EXPORT_SYMBOL(mca_device_read_pos);
/**
* mca_device_write_pos - read POS register from card
* @mca_dev: device to write pos register to
* @reg: register to write to
* @byte: byte to write to the POS registers
*
* Store a POS value directly to the hardware. You should not
* normally need to use this function and should have a very good
* knowledge of MCA bus before you do so. Doing this wrongly can
* damage the hardware.
*
* This function may not be used from interrupt context.
*
*/
void mca_device_write_pos(struct mca_device *mca_dev, int reg,
unsigned char byte)
{
struct mca_bus *mca_bus = to_mca_bus(mca_dev->dev.parent);
mca_bus->f.mca_write_pos(mca_dev, reg, byte);
}
EXPORT_SYMBOL(mca_device_write_pos);
/**
* mca_device_transform_irq - transform the ADF obtained IRQ
* @mca_device: device whose irq needs transforming
* @irq: input irq from ADF
*
* MCA Adapter Definition Files (ADF) contain irq, ioport, memory
* etc. definitions. In systems with more than one bus, these need
* to be transformed through bus mapping functions to get the real
* system global quantities.
*
* This function transforms the interrupt number and returns the
* transformed system global interrupt
*/
int mca_device_transform_irq(struct mca_device *mca_dev, int irq)
{
struct mca_bus *mca_bus = to_mca_bus(mca_dev->dev.parent);
return mca_bus->f.mca_transform_irq(mca_dev, irq);
}
EXPORT_SYMBOL(mca_device_transform_irq);
/**
* mca_device_transform_ioport - transform the ADF obtained I/O port
* @mca_device: device whose port needs transforming
* @ioport: input I/O port from ADF
*
* MCA Adapter Definition Files (ADF) contain irq, ioport, memory
* etc. definitions. In systems with more than one bus, these need
* to be transformed through bus mapping functions to get the real
* system global quantities.
*
* This function transforms the I/O port number and returns the
* transformed system global port number.
*
* This transformation can be assumed to be linear for port ranges.
*/
int mca_device_transform_ioport(struct mca_device *mca_dev, int port)
{
struct mca_bus *mca_bus = to_mca_bus(mca_dev->dev.parent);
return mca_bus->f.mca_transform_ioport(mca_dev, port);
}
EXPORT_SYMBOL(mca_device_transform_ioport);
/**
* mca_device_transform_memory - transform the ADF obtained memory
* @mca_device: device whose memory region needs transforming
* @mem: memory region start from ADF
*
* MCA Adapter Definition Files (ADF) contain irq, ioport, memory
* etc. definitions. In systems with more than one bus, these need
* to be transformed through bus mapping functions to get the real
* system global quantities.
*
* This function transforms the memory region start and returns the
* transformed system global memory region (physical).
*
* This transformation can be assumed to be linear for region ranges.
*/
void *mca_device_transform_memory(struct mca_device *mca_dev, void *mem)
{
struct mca_bus *mca_bus = to_mca_bus(mca_dev->dev.parent);
return mca_bus->f.mca_transform_memory(mca_dev, mem);
}
EXPORT_SYMBOL(mca_device_transform_memory);
/**
* mca_device_claimed - check if claimed by driver
* @mca_dev: device to check
*
* Returns 1 if the slot has been claimed by a driver
*/
int mca_device_claimed(struct mca_device *mca_dev)
{
return mca_dev->driver_loaded;
}
EXPORT_SYMBOL(mca_device_claimed);
/**
* mca_device_set_claim - set the claim value of the driver
* @mca_dev: device to set value for
* @val: claim value to set (1 claimed, 0 unclaimed)
*/
void mca_device_set_claim(struct mca_device *mca_dev, int val)
{
mca_dev->driver_loaded = val;
}
EXPORT_SYMBOL(mca_device_set_claim);
/**
* mca_device_status - get the status of the device
* @mca_device: device to get
*
* returns an enumeration of the device status:
*
* MCA_ADAPTER_NORMAL adapter is OK.
* MCA_ADAPTER_NONE no adapter at device (should never happen).
* MCA_ADAPTER_DISABLED adapter is disabled.
* MCA_ADAPTER_ERROR adapter cannot be initialised.
*/
enum MCA_AdapterStatus mca_device_status(struct mca_device *mca_dev)
{
return mca_dev->status;
}
EXPORT_SYMBOL(mca_device_status);
/**
* mca_device_set_name - set the name of the device
* @mca_device: device to set the name of
* @name: name to set
*/
void mca_device_set_name(struct mca_device *mca_dev, const char *name)
{
if(!mca_dev)
return;
strlcpy(mca_dev->name, name, sizeof(mca_dev->name));
}
EXPORT_SYMBOL(mca_device_set_name);
drivers/mca/mca-driver.c deleted 100644 → 0
View file @ c80ddb52
/* -*- mode: c; c-basic-offset: 8 -*- */
/*
* MCA driver support functions for sysfs.
*
* (C) 2002 James Bottomley <James.Bottomley@HansenPartnership.com>
*
**-----------------------------------------------------------------------------
**
** This program is free software; you can redistribute it and/or modify
** it under the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 2 of the License, or
** (at your option) any later version.
**
** This program is distributed in the hope that it will be useful,
** but WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
** GNU General Public License for more details.
**
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software
** Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
**
**-----------------------------------------------------------------------------
*/
#include <linux/device.h>
#include <linux/mca.h>
#include <linux/module.h>
int mca_register_driver(struct mca_driver *mca_drv)
{
int r;
if (MCA_bus) {
mca_drv->driver.bus = &mca_bus_type;
if ((r = driver_register(&mca_drv->driver)) < 0)
return r;
mca_drv->integrated_id = 0;
}
return 0;
}
EXPORT_SYMBOL(mca_register_driver);
int mca_register_driver_integrated(struct mca_driver *mca_driver,
int integrated_id)
{
int r = mca_register_driver(mca_driver);
if (!r)
mca_driver->integrated_id = integrated_id;
return r;
}
EXPORT_SYMBOL(mca_register_driver_integrated);
void mca_unregister_driver(struct mca_driver *mca_drv)
{
if (MCA_bus)
driver_unregister(&mca_drv->driver);
}
EXPORT_SYMBOL(mca_unregister_driver);
drivers/mca/mca-legacy.c deleted 100644 → 0
View file @ c80ddb52
/* -*- mode: c; c-basic-offset: 8 -*- */
/*
* MCA bus support functions for legacy (2.4) API.
*
* Legacy API means the API that operates in terms of MCA slot number
*
* (C) 2002 James Bottomley <James.Bottomley@HansenPartnership.com>
*
**-----------------------------------------------------------------------------
**
** This program is free software; you can redistribute it and/or modify
** it under the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 2 of the License, or
** (at your option) any later version.
**
** This program is distributed in the hope that it will be useful,
** but WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
** GNU General Public License for more details.
**
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software
** Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
**
**-----------------------------------------------------------------------------
*/
#include <linux/module.h>
#include <linux/device.h>
#include <linux/mca-legacy.h>
#include <asm/io.h>
/* NOTE: This structure is stack allocated */
struct mca_find_adapter_info {
int id;
int slot;
struct mca_device *mca_dev;
};
/* The purpose of this iterator is to loop over all the devices and
* find the one with the smallest slot number that's just greater than
* or equal to the required slot with a matching id */
static int mca_find_adapter_callback(struct device *dev, void *data)
{
struct mca_find_adapter_info *info = data;
struct mca_device *mca_dev = to_mca_device(dev);
if(mca_dev->pos_id != info->id)
return 0;
if(mca_dev->slot < info->slot)
return 0;
if(!info->mca_dev || info->mca_dev->slot >= mca_dev->slot)
info->mca_dev = mca_dev;
return 0;
}
/**
* mca_find_adapter - scan for adapters
* @id: MCA identification to search for
* @start: starting slot
*
* Search the MCA configuration for adapters matching the 16bit
* ID given. The first time it should be called with start as zero
* and then further calls made passing the return value of the
* previous call until %MCA_NOTFOUND is returned.
*
* Disabled adapters are not reported.
*/
int mca_find_adapter(int id, int start)
{
struct mca_find_adapter_info info;
if(id == 0xffff)
return MCA_NOTFOUND;
info.slot = start;
info.id = id;
info.mca_dev = NULL;
for(;;) {
bus_for_each_dev(&mca_bus_type, NULL, &info, mca_find_adapter_callback);
if(info.mca_dev == NULL)
return MCA_NOTFOUND;
if(info.mca_dev->status != MCA_ADAPTER_DISABLED)
break;
/* OK, found adapter but it was disabled. Go around
* again, excluding the slot we just found */
info.slot = info.mca_dev->slot + 1;
info.mca_dev = NULL;
}
return info.mca_dev->slot;
}
EXPORT_SYMBOL(mca_find_adapter);
/*--------------------------------------------------------------------*/
/**
* mca_find_unused_adapter - scan for unused adapters
* @id: MCA identification to search for
* @start: starting slot
*
* Search the MCA configuration for adapters matching the 16bit
* ID given. The first time it should be called with start as zero
* and then further calls made passing the return value of the
* previous call until %MCA_NOTFOUND is returned.
*
* Adapters that have been claimed by drivers and those that
* are disabled are not reported. This function thus allows a driver
* to scan for further cards when some may already be driven.
*/
int mca_find_unused_adapter(int id, int start)
{
struct mca_find_adapter_info info = { 0 };
if (!MCA_bus || id == 0xffff)
return MCA_NOTFOUND;
info.slot = start;
info.id = id;
info.mca_dev = NULL;
for(;;) {
bus_for_each_dev(&mca_bus_type, NULL, &info, mca_find_adapter_callback);
if(info.mca_dev == NULL)
return MCA_NOTFOUND;
if(info.mca_dev->status != MCA_ADAPTER_DISABLED
&& !info.mca_dev->driver_loaded)
break;
/* OK, found adapter but it was disabled or already in
* use. Go around again, excluding the slot we just
* found */
info.slot = info.mca_dev->slot + 1;
info.mca_dev = NULL;
}
return info.mca_dev->slot;
}
EXPORT_SYMBOL(mca_find_unused_adapter);
/* NOTE: stack allocated structure */
struct mca_find_device_by_slot_info {
int slot;
struct mca_device *mca_dev;
};
static int mca_find_device_by_slot_callback(struct device *dev, void *data)
{
struct mca_find_device_by_slot_info *info = data;
struct mca_device *mca_dev = to_mca_device(dev);
if(mca_dev->slot == info->slot)
info->mca_dev = mca_dev;
return 0;
}
struct mca_device *mca_find_device_by_slot(int slot)
{
struct mca_find_device_by_slot_info info;
info.slot = slot;
info.mca_dev = NULL;
bus_for_each_dev(&mca_bus_type, NULL, &info, mca_find_device_by_slot_callback);
return info.mca_dev;
}
/**
* mca_read_stored_pos - read POS register from boot data
* @slot: slot number to read from
* @reg: register to read from
*
* Fetch a POS value that was stored at boot time by the kernel
* when it scanned the MCA space. The register value is returned.
* Missing or invalid registers report 0.
*/
unsigned char mca_read_stored_pos(int slot, int reg)
{
struct mca_device *mca_dev = mca_find_device_by_slot(slot);
if(!mca_dev)
return 0;
return mca_device_read_stored_pos(mca_dev, reg);
}
EXPORT_SYMBOL(mca_read_stored_pos);
/**
* mca_read_pos - read POS register from card
* @slot: slot number to read from
* @reg: register to read from
*
* Fetch a POS value directly from the hardware to obtain the
* current value. This is much slower than mca_read_stored_pos and
* may not be invoked from interrupt context. It handles the
* deep magic required for onboard devices transparently.
*/
unsigned char mca_read_pos(int slot, int reg)
{
struct mca_device *mca_dev = mca_find_device_by_slot(slot);
if(!mca_dev)
return 0;
return mca_device_read_pos(mca_dev, reg);
}
EXPORT_SYMBOL(mca_read_pos);
/**
* mca_write_pos - read POS register from card
* @slot: slot number to read from
* @reg: register to read from
* @byte: byte to write to the POS registers
*
* Store a POS value directly from the hardware. You should not
* normally need to use this function and should have a very good
* knowledge of MCA bus before you do so. Doing this wrongly can
* damage the hardware.
*
* This function may not be used from interrupt context.
*
* Note that this a technically a Bad Thing, as IBM tech stuff says
* you should only set POS values through their utilities.
* However, some devices such as the 3c523 recommend that you write
* back some data to make sure the configuration is consistent.
* I'd say that IBM is right, but I like my drivers to work.
*
* This function can't do checks to see if multiple devices end up
* with the same resources, so you might see magic smoke if someone
* screws up.
*/
void mca_write_pos(int slot, int reg, unsigned char byte)
{
struct mca_device *mca_dev = mca_find_device_by_slot(slot);
if(!mca_dev)
return;
mca_device_write_pos(mca_dev, reg, byte);
}
EXPORT_SYMBOL(mca_write_pos);
/**
* mca_set_adapter_name - Set the description of the card
* @slot: slot to name
* @name: text string for the namen
*
* This function sets the name reported via /proc for this
* adapter slot. This is for user information only. Setting a
* name deletes any previous name.
*/
void mca_set_adapter_name(int slot, char* name)
{
struct mca_device *mca_dev = mca_find_device_by_slot(slot);
if(!mca_dev)
return;
mca_device_set_name(mca_dev, name);
}
EXPORT_SYMBOL(mca_set_adapter_name);
/**
* mca_mark_as_used - claim an MCA device
* @slot: slot to claim
* FIXME: should we make this threadsafe
*
* Claim an MCA slot for a device driver. If the
* slot is already taken the function returns 1,
* if it is not taken it is claimed and 0 is
* returned.
*/
int mca_mark_as_used(int slot)
{
struct mca_device *mca_dev = mca_find_device_by_slot(slot);
if(!mca_dev)
/* FIXME: this is actually a severe error */
return 1;
if(mca_device_claimed(mca_dev))
return 1;
mca_device_set_claim(mca_dev, 1);
return 0;
}
EXPORT_SYMBOL(mca_mark_as_used);
/**
* mca_mark_as_unused - release an MCA device
* @slot: slot to claim
*
* Release the slot for other drives to use.
*/
void mca_mark_as_unused(int slot)
{
struct mca_device *mca_dev = mca_find_device_by_slot(slot);
if(!mca_dev)
return;
mca_device_set_claim(mca_dev, 0);
}
EXPORT_SYMBOL(mca_mark_as_unused);
drivers/mca/mca-proc.c deleted 100644 → 0
View file @ c80ddb52
/* -*- mode: c; c-basic-offset: 8 -*- */
/*
* MCA bus support functions for the proc fs.
*
* NOTE: this code *requires* the legacy MCA api.
*
* Legacy API means the API that operates in terms of MCA slot number
*
* (C) 2002 James Bottomley <James.Bottomley@HansenPartnership.com>
*
**-----------------------------------------------------------------------------
**
** This program is free software; you can redistribute it and/or modify
** it under the terms of the GNU General Public License as published by
** the Free Software Foundation; either version 2 of the License, or
** (at your option) any later version.
**
** This program is distributed in the hope that it will be useful,
** but WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
** GNU General Public License for more details.
**
** You should have received a copy of the GNU General Public License
** along with this program; if not, write to the Free Software
** Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
**
**-----------------------------------------------------------------------------
*/
#include <linux/module.h>
#include <linux/init.h>
#include <linux/proc_fs.h>
#include <linux/mca.h>
static int get_mca_info_helper(struct mca_device *mca_dev, char *page, int len)
{
int j;
for(j=0; j<8; j++)
len += sprintf(page+len, "%02x ",
mca_dev ? mca_dev->pos[j] : 0xff);
len += sprintf(page+len, " %s\n", mca_dev ? mca_dev->name : "");
return len;
}
static int get_mca_info(char *page, char **start, off_t off,
int count, int *eof, void *data)
{
int i, len = 0;
if(MCA_bus) {
struct mca_device *mca_dev;
/* Format POS registers of eight MCA slots */
for(i=0; i<MCA_MAX_SLOT_NR; i++) {
mca_dev = mca_find_device_by_slot(i);
len += sprintf(page+len, "Slot %d: ", i+1);
len = get_mca_info_helper(mca_dev, page, len);
}
/* Format POS registers of integrated video subsystem */
mca_dev = mca_find_device_by_slot(MCA_INTEGVIDEO);
len += sprintf(page+len, "Video : ");
len = get_mca_info_helper(mca_dev, page, len);
/* Format POS registers of integrated SCSI subsystem */
mca_dev = mca_find_device_by_slot(MCA_INTEGSCSI);
len += sprintf(page+len, "SCSI : ");
len = get_mca_info_helper(mca_dev, page, len);
/* Format POS registers of motherboard */
mca_dev = mca_find_device_by_slot(MCA_MOTHERBOARD);
len += sprintf(page+len, "Planar: ");
len = get_mca_info_helper(mca_dev, page, len);
} else {
/* Leave it empty if MCA not detected - this should *never*
* happen!
*/
}
if (len <= off+count) *eof = 1;
*start = page + off;
len -= off;
if (len>count) len = count;
if (len<0) len = 0;
return len;
}
/*--------------------------------------------------------------------*/
static int mca_default_procfn(char* buf, struct mca_device *mca_dev)
{
int len = 0, i;
int slot = mca_dev->slot;
/* Print out the basic information */
if(slot < MCA_MAX_SLOT_NR) {
len += sprintf(buf+len, "Slot: %d\n", slot+1);
} else if(slot == MCA_INTEGSCSI) {
len += sprintf(buf+len, "Integrated SCSI Adapter\n");
} else if(slot == MCA_INTEGVIDEO) {
len += sprintf(buf+len, "Integrated Video Adapter\n");
} else if(slot == MCA_MOTHERBOARD) {
len += sprintf(buf+len, "Motherboard\n");
}
if (mca_dev->name[0]) {
/* Drivers might register a name without /proc handler... */
len += sprintf(buf+len, "Adapter Name: %s\n",
mca_dev->name);
} else {
len += sprintf(buf+len, "Adapter Name: Unknown\n");
}
len += sprintf(buf+len, "Id: %02x%02x\n",
mca_dev->pos[1], mca_dev->pos[0]);
len += sprintf(buf+len, "Enabled: %s\nPOS: ",
mca_device_status(mca_dev) == MCA_ADAPTER_NORMAL ?
"Yes" : "No");
for(i=0; i<8; i++) {
len += sprintf(buf+len, "%02x ", mca_dev->pos[i]);
}
len += sprintf(buf+len, "\nDriver Installed: %s",
mca_device_claimed(mca_dev) ? "Yes" : "No");
buf[len++] = '\n';
buf[len] = 0;
return len;
} /* mca_default_procfn() */
static int get_mca_machine_info(char* page, char **start, off_t off,
int count, int *eof, void *data)
{
int len = 0;
len += sprintf(page+len, "Model Id: 0x%x\n", machine_id);
len += sprintf(page+len, "Submodel Id: 0x%x\n", machine_submodel_id);
len += sprintf(page+len, "BIOS Revision: 0x%x\n", BIOS_revision);
if (len <= off+count) *eof = 1;
*start = page + off;
len -= off;
if (len>count) len = count;
if (len<0) len = 0;
return len;
}
static int mca_read_proc(char *page, char **start, off_t off,
int count, int *eof, void *data)
{
struct mca_device *mca_dev = (struct mca_device *)data;
int len = 0;
/* Get the standard info */
len = mca_default_procfn(page, mca_dev);
/* Do any device-specific processing, if there is any */
if(mca_dev->procfn) {
len += mca_dev->procfn(page+len, mca_dev->slot,
mca_dev->proc_dev);
}
if (len <= off+count) *eof = 1;
*start = page + off;
len -= off;
if (len>count) len = count;
if (len<0) len = 0;
return len;
} /* mca_read_proc() */
/*--------------------------------------------------------------------*/
void __init mca_do_proc_init(void)
{
int i;
struct proc_dir_entry *proc_mca;
struct proc_dir_entry* node = NULL;
struct mca_device *mca_dev;
proc_mca = proc_mkdir("mca", NULL);
create_proc_read_entry("pos",0,proc_mca,get_mca_info,NULL);
create_proc_read_entry("machine",0,proc_mca,get_mca_machine_info,NULL);
/* Initialize /proc/mca entries for existing adapters */
for(i = 0; i < MCA_NUMADAPTERS; i++) {
enum MCA_AdapterStatus status;
mca_dev = mca_find_device_by_slot(i);
if(!mca_dev)
continue;
mca_dev->procfn = NULL;
if(i < MCA_MAX_SLOT_NR) sprintf(mca_dev->procname,"slot%d", i+1);
else if(i == MCA_INTEGVIDEO) sprintf(mca_dev->procname,"video");
else if(i == MCA_INTEGSCSI) sprintf(mca_dev->procname,"scsi");
else if(i == MCA_MOTHERBOARD) sprintf(mca_dev->procname,"planar");
status = mca_device_status(mca_dev);
if (status != MCA_ADAPTER_NORMAL &&
status != MCA_ADAPTER_DISABLED)
continue;
node = create_proc_read_entry(mca_dev->procname, 0, proc_mca,
mca_read_proc, (void *)mca_dev);
if(node == NULL) {
printk("Failed to allocate memory for MCA proc-entries!");
return;
}
}
} /* mca_do_proc_init() */
/**
* mca_set_adapter_procfn - Set the /proc callback
* @slot: slot to configure
* @procfn: callback function to call for /proc
* @dev: device information passed to the callback
*
* This sets up an information callback for /proc/mca/slot?. The
* function is called with the buffer, slot, and device pointer (or
* some equally informative context information, or nothing, if you
* prefer), and is expected to put useful information into the
* buffer. The adapter name, ID, and POS registers get printed
* before this is called though, so don't do it again.
*
* This should be called with a %NULL @procfn when a module
* unregisters, thus preventing kernel crashes and other such
* nastiness.
*/
void mca_set_adapter_procfn(int slot, MCA_ProcFn procfn, void* proc_dev)
{
struct mca_device *mca_dev = mca_find_device_by_slot(slot);
if(!mca_dev)
return;
mca_dev->procfn = procfn;
mca_dev->proc_dev = proc_dev;
}
EXPORT_SYMBOL(mca_set_adapter_procfn);
drivers/message/i2o/i2o_proc.c
View file @ d5b4bb4d
......@@ -283,7 +283,6 @@ static char *bus_strings[] = {
"Local Bus",
"ISA",
"EISA",
"MCA",
"PCI",
"PCMCIA",
"NUBUS",
......@@ -351,18 +350,6 @@ static int i2o_seq_show_hrt(struct seq_file *seq, void *v)
EisaSlotNumber);
break;
case I2O_BUS_MCA:
seq_printf(seq, " IOBase: %0#6x,",
hrt->hrt_entry[i].bus.mca_bus.
McaBaseIOPort);
seq_printf(seq, " MemoryBase: %0#10x,",
hrt->hrt_entry[i].bus.mca_bus.
McaBaseMemoryAddress);
seq_printf(seq, " Slot: %0#4x,",
hrt->hrt_entry[i].bus.mca_bus.
McaSlotNumber);
break;
case I2O_BUS_PCI:
seq_printf(seq, " Bus: %0#4x",
hrt->hrt_entry[i].bus.pci_bus.
......
drivers/scsi/Kconfig
View file @ d5b4bb4d
......@@ -808,19 +808,6 @@ config SCSI_FUTURE_DOMAIN
To compile this driver as a module, choose M here: the
module will be called fdomain.
config SCSI_FD_MCS
tristate "Future Domain MCS-600/700 SCSI support"
depends on MCA_LEGACY && SCSI
---help---
This is support for Future Domain MCS 600/700 MCA SCSI adapters.
Some PS/2 computers are equipped with IBM Fast SCSI Adapter/A which
is identical to the MCS 700 and hence also supported by this driver.
This driver also supports the Reply SB16/SCSI card (the SCSI part).
It supports multiple adapters in the same system.
To compile this driver as a module, choose M here: the
module will be called fd_mcs.
config SCSI_GDTH
tristate "Intel/ICP (former GDT SCSI Disk Array) RAID Controller support"
depends on (ISA || EISA || PCI) && SCSI && ISA_DMA_API
......@@ -890,76 +877,6 @@ config SCSI_GENERIC_NCR53C400
not detect your card. See the file
<file:Documentation/scsi/g_NCR5380.txt> for details.
config SCSI_IBMMCA
tristate "IBMMCA SCSI support"
depends on MCA && SCSI
---help---
This is support for the IBM SCSI adapter found in many of the PS/2
series computers. These machines have an MCA bus, so you need to
answer Y to "MCA support" as well and read
<file:Documentation/mca.txt>.
If the adapter isn't found during boot (a common problem for models
56, 57, 76, and 77) you'll need to use the 'ibmmcascsi=<pun>' kernel
option, where <pun> is the id of the SCSI subsystem (usually 7, but
if that doesn't work check your reference diskette). Owners of
model 95 with a LED-matrix-display can in addition activate some
activity info like under OS/2, but more informative, by setting
'ibmmcascsi=display' as an additional kernel parameter. Try "man
bootparam" or see the documentation of your boot loader about how to
pass options to the kernel.
To compile this driver as a module, choose M here: the
module will be called ibmmca.
config IBMMCA_SCSI_ORDER_STANDARD
bool "Standard SCSI-order"
depends on SCSI_IBMMCA
---help---
In the PC-world and in most modern SCSI-BIOS-setups, SCSI-hard disks
are assigned to the drive letters, starting with the lowest SCSI-id
(physical number -- pun) to be drive C:, as seen from DOS and
similar operating systems. When looking into papers describing the
ANSI-SCSI-standard, this assignment of drives appears to be wrong.
The SCSI-standard follows a hardware-hierarchy which says that id 7
has the highest priority and id 0 the lowest. Therefore, the host
adapters are still today everywhere placed as SCSI-id 7 by default.
In the SCSI-standard, the drive letters express the priority of the
disk. C: should be the hard disk, or a partition on it, with the
highest priority. This must therefore be the disk with the highest
SCSI-id (e.g. 6) and not the one with the lowest! IBM-BIOS kept the
original definition of the SCSI-standard as also industrial- and
process-control-machines, like VME-CPUs running under realtime-OSes
(e.g. LynxOS, OS9) do.
If you like to run Linux on your MCA-machine with the same
assignment of hard disks as seen from e.g. DOS or OS/2 on your
machine, which is in addition conformant to the SCSI-standard, you
must say Y here. This is also necessary for MCA-Linux users who want
to keep downward compatibility to older releases of the
IBM-MCA-SCSI-driver (older than driver-release 2.00 and older than
June 1997).
If you like to have the lowest SCSI-id assigned as drive C:, as
modern SCSI-BIOSes do, which does not conform to the standard, but
is widespread and common in the PC-world of today, you must say N
here. If unsure, say Y.
config IBMMCA_SCSI_DEV_RESET
bool "Reset SCSI-devices at boottime"
depends on SCSI_IBMMCA
---help---
By default, SCSI-devices are reset when the machine is powered on.
However, some devices exist, like special-control-devices,
SCSI-CNC-machines, SCSI-printer or scanners of older type, that do
not reset when switched on. If you say Y here, each device connected
to your SCSI-bus will be issued a reset-command after it has been
probed, while the kernel is booting. This may cause problems with
more modern devices, like hard disks, which do not appreciate these
reset commands, and can cause your system to hang. So say Y only if
you know that one of your older devices needs it; N is the safe
answer.
config SCSI_IPS
tristate "IBM ServeRAID support"
depends on PCI && SCSI
......
drivers/scsi/Makefile
View file @ d5b4bb4d
......@@ -75,7 +75,6 @@ obj-$(CONFIG_SCSI_AIC94XX) += aic94xx/
obj-$(CONFIG_SCSI_PM8001) += pm8001/
obj-$(CONFIG_SCSI_ISCI) += isci/
obj-$(CONFIG_SCSI_IPS) += ips.o
obj-$(CONFIG_SCSI_FD_MCS) += fd_mcs.o
obj-$(CONFIG_SCSI_FUTURE_DOMAIN)+= fdomain.o
obj-$(CONFIG_SCSI_IN2000) += in2000.o
obj-$(CONFIG_SCSI_GENERIC_NCR5380) += g_NCR5380.o
......@@ -100,7 +99,6 @@ obj-$(CONFIG_SCSI_SYM53C8XX_2) += sym53c8xx_2/
obj-$(CONFIG_SCSI_ZALON) += zalon7xx.o
obj-$(CONFIG_SCSI_EATA_PIO) += eata_pio.o
obj-$(CONFIG_SCSI_7000FASST) += wd7000.o
obj-$(CONFIG_SCSI_IBMMCA) += ibmmca.o
obj-$(CONFIG_SCSI_EATA) += eata.o
obj-$(CONFIG_SCSI_DC395x) += dc395x.o
obj-$(CONFIG_SCSI_DC390T) += tmscsim.o
......
drivers/scsi/aha1542.c
View file @ d5b4bb4d
......@@ -22,7 +22,7 @@
* Added module command-line options
* 19-Jul-99
* Modified by Adam Fritzler
* Added proper detection of the AHA-1640 (MCA version of AHA-1540)
* Added proper detection of the AHA-1640 (MCA, now deleted)
*/
#include <linux/module.h>
......@@ -37,8 +37,6 @@
#include <linux/spinlock.h>
#include <linux/isapnp.h>
#include <linux/blkdev.h>
#include <linux/mca.h>
#include <linux/mca-legacy.h>
#include <linux/slab.h>
#include <asm/dma.h>
......@@ -71,7 +69,7 @@
#define MAXBOARDS 4 /* Increase this and the sizes of the
arrays below, if you need more.. */
/* Boards 3,4 slots are reserved for ISAPnP/MCA scans */
/* Boards 3,4 slots are reserved for ISAPnP scans */
static unsigned int bases[MAXBOARDS] __initdata = {0x330, 0x334, 0, 0};
......@@ -1008,66 +1006,6 @@ static int __init aha1542_detect(struct scsi_host_template * tpnt)
}
#endif
/*
* Find MicroChannel cards (AHA1640)
*/
#ifdef CONFIG_MCA_LEGACY
if(MCA_bus) {
int slot = 0;
int pos = 0;
for (indx = 0; (slot != MCA_NOTFOUND) && (indx < ARRAY_SIZE(bases)); indx++) {
if (bases[indx])
continue;
/* Detect only AHA-1640 cards -- MCA ID 0F1F */
slot = mca_find_unused_adapter(0x0f1f, slot);
if (slot == MCA_NOTFOUND)
break;
/* Found one */
pos = mca_read_stored_pos(slot, 3);
/* Decode address */
if (pos & 0x80) {
if (pos & 0x02) {
if (pos & 0x01)
bases[indx] = 0x334;
else
bases[indx] = 0x234;
} else {
if (pos & 0x01)
bases[indx] = 0x134;
}
} else {
if (pos & 0x02) {
if (pos & 0x01)
bases[indx] = 0x330;
else
bases[indx] = 0x230;
} else {
if (pos & 0x01)
bases[indx] = 0x130;
}
}
/* No need to decode IRQ and Arb level -- those are
* read off the card later.
*/
printk(KERN_INFO "Found an AHA-1640 in MCA slot %d, I/O 0x%04x\n", slot, bases[indx]);
mca_set_adapter_name(slot, "Adapter AHA-1640");
mca_set_adapter_procfn(slot, NULL, NULL);
mca_mark_as_used(slot);
/* Go on */
slot++;
}
}
#endif
/*
* Hunt for ISA Plug'n'Pray Adaptecs (AHA1535)
*/
......
drivers/scsi/fd_mcs.c deleted 100644 → 0
View file @ c80ddb52
/* fd_mcs.c -- Future Domain MCS 600/700 (or IBM OEM) driver
*
* FutureDomain MCS-600/700 v0.2 03/11/1998 by ZP Gu (zpg@castle.net)
*
* This driver is cloned from fdomain.* to specifically support
* the Future Domain MCS 600/700 MCA SCSI adapters. Some PS/2s
* also equipped with IBM Fast SCSI Adapter/A which is an OEM
* of MCS 700.
*
* This driver also supports Reply SB16/SCSI card (the SCSI part).
*
* What makes this driver different is that this driver is MCA only
* and it supports multiple adapters in the same system, IRQ
* sharing, some driver statistics, and maps highest SCSI id to sda.
* All cards are auto-detected.
*
* Assumptions: TMC-1800/18C50/18C30, BIOS >= 3.4
*
* LILO command-line options:
* fd_mcs=<FIFO_COUNT>[,<FIFO_SIZE>]
*
* ********************************************************
* Please see Copyrights/Comments in fdomain.* for credits.
* Following is from fdomain.c for acknowledgement:
*
* Created: Sun May 3 18:53:19 1992 by faith@cs.unc.edu
* Revised: Wed Oct 2 11:10:55 1996 by r.faith@ieee.org
* Author: Rickard E. Faith, faith@cs.unc.edu
* Copyright 1992, 1993, 1994, 1995, 1996 Rickard E. Faith
*
* $Id: fdomain.c,v 5.45 1996/10/02 15:13:06 root Exp $
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; either version 2, or (at your option) any
* later version.
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
* 675 Mass Ave, Cambridge, MA 02139, USA.
**************************************************************************
NOTES ON USER DEFINABLE OPTIONS:
DEBUG: This turns on the printing of various debug information.
ENABLE_PARITY: This turns on SCSI parity checking. With the current
driver, all attached devices must support SCSI parity. If none of your
devices support parity, then you can probably get the driver to work by
turning this option off. I have no way of testing this, however, and it
would appear that no one ever uses this option.
FIFO_COUNT: The host adapter has an 8K cache (host adapters based on the
18C30 chip have a 2k cache). When this many 512 byte blocks are filled by
the SCSI device, an interrupt will be raised. Therefore, this could be as
low as 0, or as high as 16. Note, however, that values which are too high
or too low seem to prevent any interrupts from occurring, and thereby lock
up the machine. I have found that 2 is a good number, but throughput may
be increased by changing this value to values which are close to 2.
Please let me know if you try any different values.
[*****Now a runtime option*****]
RESELECTION: This is no longer an option, since I gave up trying to
implement it in version 4.x of this driver. It did not improve
performance at all and made the driver unstable (because I never found one
of the two race conditions which were introduced by the multiple
outstanding command code). The instability seems a very high price to pay
just so that you don't have to wait for the tape to rewind. If you want
this feature implemented, send me patches. I'll be happy to send a copy
of my (broken) driver to anyone who would like to see a copy.
**************************************************************************/
#include <linux/module.h>
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/blkdev.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/ioport.h>
#include <linux/proc_fs.h>
#include <linux/delay.h>
#include <linux/mca.h>
#include <linux/spinlock.h>
#include <linux/slab.h>
#include <scsi/scsicam.h>
#include <linux/mca-legacy.h>
#include <asm/io.h>
#include "scsi.h"
#include <scsi/scsi_host.h>
#define DRIVER_VERSION "v0.2 by ZP Gu<zpg@castle.net>"
/* START OF USER DEFINABLE OPTIONS */
#define DEBUG 0 /* Enable debugging output */
#define ENABLE_PARITY 1 /* Enable SCSI Parity */
/* END OF USER DEFINABLE OPTIONS */
#if DEBUG
#define EVERY_ACCESS 0 /* Write a line on every scsi access */
#define ERRORS_ONLY 1 /* Only write a line if there is an error */
#define DEBUG_MESSAGES 1 /* Debug MESSAGE IN phase */
#define DEBUG_ABORT 1 /* Debug abort() routine */
#define DEBUG_RESET 1 /* Debug reset() routine */
#define DEBUG_RACE 1 /* Debug interrupt-driven race condition */
#else
#define EVERY_ACCESS 0 /* LEAVE THESE ALONE--CHANGE THE ONES ABOVE */
#define ERRORS_ONLY 0
#define DEBUG_MESSAGES 0
#define DEBUG_ABORT 0
#define DEBUG_RESET 0
#define DEBUG_RACE 0
#endif
/* Errors are reported on the line, so we don't need to report them again */
#if EVERY_ACCESS
#undef ERRORS_ONLY
#define ERRORS_ONLY 0
#endif
#if ENABLE_PARITY
#define PARITY_MASK 0x08
#else
#define PARITY_MASK 0x00
#endif
enum chip_type {
unknown = 0x00,
tmc1800 = 0x01,
tmc18c50 = 0x02,
tmc18c30 = 0x03,
};
enum {
in_arbitration = 0x02,
in_selection = 0x04,
in_other = 0x08,
disconnect = 0x10,
aborted = 0x20,
sent_ident = 0x40,
};
enum in_port_type {
Read_SCSI_Data = 0,
SCSI_Status = 1,
TMC_Status = 2,
FIFO_Status = 3, /* tmc18c50/tmc18c30 only */
Interrupt_Cond = 4, /* tmc18c50/tmc18c30 only */
LSB_ID_Code = 5,
MSB_ID_Code = 6,
Read_Loopback = 7,
SCSI_Data_NoACK = 8,
Interrupt_Status = 9,
Configuration1 = 10,
Configuration2 = 11, /* tmc18c50/tmc18c30 only */
Read_FIFO = 12,
FIFO_Data_Count = 14
};
enum out_port_type {
Write_SCSI_Data = 0,
SCSI_Cntl = 1,
Interrupt_Cntl = 2,
SCSI_Mode_Cntl = 3,
TMC_Cntl = 4,
Memory_Cntl = 5, /* tmc18c50/tmc18c30 only */
Write_Loopback = 7,
IO_Control = 11, /* tmc18c30 only */
Write_FIFO = 12
};
struct fd_hostdata {
unsigned long _bios_base;
int _bios_major;
int _bios_minor;
volatile int _in_command;
Scsi_Cmnd *_current_SC;
enum chip_type _chip;
int _adapter_mask;
int _fifo_count; /* Number of 512 byte blocks before INTR */
char _adapter_name[64];
#if DEBUG_RACE
volatile int _in_interrupt_flag;
#endif
int _SCSI_Mode_Cntl_port;
int _FIFO_Data_Count_port;
int _Interrupt_Cntl_port;
int _Interrupt_Status_port;
int _Interrupt_Cond_port;
int _Read_FIFO_port;
int _Read_SCSI_Data_port;
int _SCSI_Cntl_port;
int _SCSI_Data_NoACK_port;
int _SCSI_Status_port;
int _TMC_Cntl_port;
int _TMC_Status_port;
int _Write_FIFO_port;
int _Write_SCSI_Data_port;
int _FIFO_Size; /* = 0x2000; 8k FIFO for
pre-tmc18c30 chips */
/* simple stats */
int _Bytes_Read;
int _Bytes_Written;
int _INTR_Processed;
};
#define FD_MAX_HOSTS 3 /* enough? */
#define HOSTDATA(shpnt) ((struct fd_hostdata *) shpnt->hostdata)
#define bios_base (HOSTDATA(shpnt)->_bios_base)
#define bios_major (HOSTDATA(shpnt)->_bios_major)
#define bios_minor (HOSTDATA(shpnt)->_bios_minor)
#define in_command (HOSTDATA(shpnt)->_in_command)
#define current_SC (HOSTDATA(shpnt)->_current_SC)
#define chip (HOSTDATA(shpnt)->_chip)
#define adapter_mask (HOSTDATA(shpnt)->_adapter_mask)
#define FIFO_COUNT (HOSTDATA(shpnt)->_fifo_count)
#define adapter_name (HOSTDATA(shpnt)->_adapter_name)
#if DEBUG_RACE
#define in_interrupt_flag (HOSTDATA(shpnt)->_in_interrupt_flag)
#endif
#define SCSI_Mode_Cntl_port (HOSTDATA(shpnt)->_SCSI_Mode_Cntl_port)
#define FIFO_Data_Count_port (HOSTDATA(shpnt)->_FIFO_Data_Count_port)
#define Interrupt_Cntl_port (HOSTDATA(shpnt)->_Interrupt_Cntl_port)
#define Interrupt_Status_port (HOSTDATA(shpnt)->_Interrupt_Status_port)
#define Interrupt_Cond_port (HOSTDATA(shpnt)->_Interrupt_Cond_port)
#define Read_FIFO_port (HOSTDATA(shpnt)->_Read_FIFO_port)
#define Read_SCSI_Data_port (HOSTDATA(shpnt)->_Read_SCSI_Data_port)
#define SCSI_Cntl_port (HOSTDATA(shpnt)->_SCSI_Cntl_port)
#define SCSI_Data_NoACK_port (HOSTDATA(shpnt)->_SCSI_Data_NoACK_port)
#define SCSI_Status_port (HOSTDATA(shpnt)->_SCSI_Status_port)
#define TMC_Cntl_port (HOSTDATA(shpnt)->_TMC_Cntl_port)
#define TMC_Status_port (HOSTDATA(shpnt)->_TMC_Status_port)
#define Write_FIFO_port (HOSTDATA(shpnt)->_Write_FIFO_port)
#define Write_SCSI_Data_port (HOSTDATA(shpnt)->_Write_SCSI_Data_port)
#define FIFO_Size (HOSTDATA(shpnt)->_FIFO_Size)
#define Bytes_Read (HOSTDATA(shpnt)->_Bytes_Read)
#define Bytes_Written (HOSTDATA(shpnt)->_Bytes_Written)
#define INTR_Processed (HOSTDATA(shpnt)->_INTR_Processed)
struct fd_mcs_adapters_struct {
char *name;
int id;
enum chip_type fd_chip;
int fifo_size;
int fifo_count;
};
#define REPLY_ID 0x5137
static struct fd_mcs_adapters_struct fd_mcs_adapters[] = {
{"Future Domain SCSI Adapter MCS-700(18C50)",
0x60e9,
tmc18c50,
0x2000,
4},
{"Future Domain SCSI Adapter MCS-600/700(TMC-1800)",
0x6127,
tmc1800,
0x2000,
4},
{"Reply Sound Blaster/SCSI Adapter",
REPLY_ID,
tmc18c30,
0x800,
2},
};
#define FD_BRDS ARRAY_SIZE(fd_mcs_adapters)
static irqreturn_t fd_mcs_intr(int irq, void *dev_id);
static unsigned long addresses[] = { 0xc8000, 0xca000, 0xce000, 0xde000 };
static unsigned short ports[] = { 0x140, 0x150, 0x160, 0x170 };
static unsigned short interrupts[] = { 3, 5, 10, 11, 12, 14, 15, 0 };
/* host information */
static int found = 0;
static struct Scsi_Host *hosts[FD_MAX_HOSTS + 1] = { NULL };
static int user_fifo_count = 0;
static int user_fifo_size = 0;
#ifndef MODULE
static int __init fd_mcs_setup(char *str)
{
static int done_setup = 0;
int ints[3];
get_options(str, 3, ints);
if (done_setup++ || ints[0] < 1 || ints[0] > 2 || ints[1] < 1 || ints[1] > 16) {
printk("fd_mcs: usage: fd_mcs=FIFO_COUNT, FIFO_SIZE\n");
return 0;
}
user_fifo_count = ints[0] >= 1 ? ints[1] : 0;
user_fifo_size = ints[0] >= 2 ? ints[2] : 0;
return 1;
}
__setup("fd_mcs=", fd_mcs_setup);
#endif /* !MODULE */
static void print_banner(struct Scsi_Host *shpnt)
{
printk("scsi%d <fd_mcs>: ", shpnt->host_no);
if (bios_base) {
printk("BIOS at 0x%lX", bios_base);
} else {
printk("No BIOS");
}
printk(", HostID %d, %s Chip, IRQ %d, IO 0x%lX\n", shpnt->this_id, chip == tmc18c50 ? "TMC-18C50" : (chip == tmc18c30 ? "TMC-18C30" : (chip == tmc1800 ? "TMC-1800" : "Unknown")), shpnt->irq, shpnt->io_port);
}
static void do_pause(unsigned amount)
{ /* Pause for amount*10 milliseconds */
do {
mdelay(10);
} while (--amount);
}
static void fd_mcs_make_bus_idle(struct Scsi_Host *shpnt)
{
outb(0, SCSI_Cntl_port);
outb(0, SCSI_Mode_Cntl_port);
if (chip == tmc18c50 || chip == tmc18c30)
outb(0x21 | PARITY_MASK, TMC_Cntl_port); /* Clear forced intr. */
else
outb(0x01 | PARITY_MASK, TMC_Cntl_port);
}
static int fd_mcs_detect(struct scsi_host_template * tpnt)
{
int loop;
struct Scsi_Host *shpnt;
/* get id, port, bios, irq */
int slot;
u_char pos2, pos3, pos4;
int id, port, irq;
unsigned long bios;
/* if not MCA machine, return */
if (!MCA_bus)
return 0;
/* changeable? */
id = 7;
for (loop = 0; loop < FD_BRDS; loop++) {
slot = 0;
while (MCA_NOTFOUND != (slot = mca_find_adapter(fd_mcs_adapters[loop].id, slot))) {
/* if we get this far, an adapter has been detected and is
enabled */
printk(KERN_INFO "scsi <fd_mcs>: %s at slot %d\n", fd_mcs_adapters[loop].name, slot + 1);
pos2 = mca_read_stored_pos(slot, 2);
pos3 = mca_read_stored_pos(slot, 3);
pos4 = mca_read_stored_pos(slot, 4);
/* ready for next probe */
slot++;
if (fd_mcs_adapters[loop].id == REPLY_ID) { /* reply card */
static int reply_irq[] = { 10, 11, 14, 15 };
bios = 0; /* no bios */
if (pos2 & 0x2)
port = ports[pos4 & 0x3];
else
continue;
/* can't really disable it, same as irq=10 */
irq = reply_irq[((pos4 >> 2) & 0x1) + 2 * ((pos4 >> 4) & 0x1)];
} else {
bios = addresses[pos2 >> 6];
port = ports[(pos2 >> 4) & 0x03];
irq = interrupts[(pos2 >> 1) & 0x07];
}
if (irq) {
/* claim the slot */
mca_set_adapter_name(slot - 1, fd_mcs_adapters[loop].name);
/* check irq/region */
if (request_irq(irq, fd_mcs_intr, IRQF_SHARED, "fd_mcs", hosts)) {
printk(KERN_ERR "fd_mcs: interrupt is not available, skipping...\n");
continue;
}
/* request I/O region */
if (request_region(port, 0x10, "fd_mcs")) {
printk(KERN_ERR "fd_mcs: I/O region is already in use, skipping...\n");
continue;
}
/* register */
if (!(shpnt = scsi_register(tpnt, sizeof(struct fd_hostdata)))) {
printk(KERN_ERR "fd_mcs: scsi_register() failed\n");
release_region(port, 0x10);
free_irq(irq, hosts);
continue;
}
/* save name */
strcpy(adapter_name, fd_mcs_adapters[loop].name);
/* chip/fifo */
chip = fd_mcs_adapters[loop].fd_chip;
/* use boot time value if available */
FIFO_COUNT = user_fifo_count ? user_fifo_count : fd_mcs_adapters[loop].fifo_count;
FIFO_Size = user_fifo_size ? user_fifo_size : fd_mcs_adapters[loop].fifo_size;
/* FIXME: Do we need to keep this bit of code inside NOT_USED around at all? */
#ifdef NOT_USED
/* *************************************************** */
/* Try to toggle 32-bit mode. This only
works on an 18c30 chip. (User reports
say this works, so we should switch to
it in the near future.) */
outb(0x80, port + IO_Control);
if ((inb(port + Configuration2) & 0x80) == 0x80) {
outb(0x00, port + IO_Control);
if ((inb(port + Configuration2) & 0x80) == 0x00) {
chip = tmc18c30;
FIFO_Size = 0x800; /* 2k FIFO */
printk("FIRST: chip=%s, fifo_size=0x%x\n", (chip == tmc18c30) ? "tmc18c30" : "tmc18c50", FIFO_Size);
}
}
/* That should have worked, but appears to
have problems. Let's assume it is an
18c30 if the RAM is disabled. */
if (inb(port + Configuration2) & 0x02) {
chip = tmc18c30;
FIFO_Size = 0x800; /* 2k FIFO */
printk("SECOND: chip=%s, fifo_size=0x%x\n", (chip == tmc18c30) ? "tmc18c30" : "tmc18c50", FIFO_Size);
}
/* *************************************************** */
#endif
/* IBM/ANSI scsi scan ordering */
/* Stick this back in when the scsi.c changes are there */
shpnt->reverse_ordering = 1;
/* saving info */
hosts[found++] = shpnt;
shpnt->this_id = id;
shpnt->irq = irq;
shpnt->io_port = port;
shpnt->n_io_port = 0x10;
/* save */
bios_base = bios;
adapter_mask = (1 << id);
/* save more */
SCSI_Mode_Cntl_port = port + SCSI_Mode_Cntl;
FIFO_Data_Count_port = port + FIFO_Data_Count;
Interrupt_Cntl_port = port + Interrupt_Cntl;
Interrupt_Status_port = port + Interrupt_Status;
Interrupt_Cond_port = port + Interrupt_Cond;
Read_FIFO_port = port + Read_FIFO;
Read_SCSI_Data_port = port + Read_SCSI_Data;
SCSI_Cntl_port = port + SCSI_Cntl;
SCSI_Data_NoACK_port = port + SCSI_Data_NoACK;
SCSI_Status_port = port + SCSI_Status;
TMC_Cntl_port = port + TMC_Cntl;
TMC_Status_port = port + TMC_Status;
Write_FIFO_port = port + Write_FIFO;
Write_SCSI_Data_port = port + Write_SCSI_Data;
Bytes_Read = 0;
Bytes_Written = 0;
INTR_Processed = 0;
/* say something */
print_banner(shpnt);
/* reset */
outb(1, SCSI_Cntl_port);
do_pause(2);
outb(0, SCSI_Cntl_port);
do_pause(115);
outb(0, SCSI_Mode_Cntl_port);
outb(PARITY_MASK, TMC_Cntl_port);
/* done reset */
}
}
if (found == FD_MAX_HOSTS) {
printk("fd_mcs: detecting reached max=%d host adapters.\n", FD_MAX_HOSTS);
break;
}
}
return found;
}
static const char *fd_mcs_info(struct Scsi_Host *shpnt)
{
return adapter_name;
}
static int TOTAL_INTR = 0;
/*
* inout : decides on the direction of the dataflow and the meaning of the
* variables
* buffer: If inout==FALSE data is being written to it else read from it
* *start: If inout==FALSE start of the valid data in the buffer
* offset: If inout==FALSE offset from the beginning of the imaginary file
* from which we start writing into the buffer
* length: If inout==FALSE max number of bytes to be written into the buffer
* else number of bytes in the buffer
*/
static int fd_mcs_proc_info(struct Scsi_Host *shpnt, char *buffer, char **start, off_t offset, int length, int inout)
{
int len = 0;
if (inout)
return (-ENOSYS);
*start = buffer + offset;
len += sprintf(buffer + len, "Future Domain MCS-600/700 Driver %s\n", DRIVER_VERSION);
len += sprintf(buffer + len, "HOST #%d: %s\n", shpnt->host_no, adapter_name);
len += sprintf(buffer + len, "FIFO Size=0x%x, FIFO Count=%d\n", FIFO_Size, FIFO_COUNT);
len += sprintf(buffer + len, "DriverCalls=%d, Interrupts=%d, BytesRead=%d, BytesWrite=%d\n\n", TOTAL_INTR, INTR_Processed, Bytes_Read, Bytes_Written);
if ((len -= offset) <= 0)
return 0;
if (len > length)
len = length;
return len;
}
static int fd_mcs_select(struct Scsi_Host *shpnt, int target)
{
int status;
unsigned long timeout;
outb(0x82, SCSI_Cntl_port); /* Bus Enable + Select */
outb(adapter_mask | (1 << target), SCSI_Data_NoACK_port);
/* Stop arbitration and enable parity */
outb(PARITY_MASK, TMC_Cntl_port);
timeout = 350; /* 350mS -- because of timeouts
(was 250mS) */
do {
status = inb(SCSI_Status_port); /* Read adapter status */
if (status & 1) { /* Busy asserted */
/* Enable SCSI Bus (on error, should make bus idle with 0) */
outb(0x80, SCSI_Cntl_port);
return 0;
}
udelay(1000); /* wait one msec */
} while (--timeout);
/* Make bus idle */
fd_mcs_make_bus_idle(shpnt);
#if EVERY_ACCESS
if (!target)
printk("Selection failed\n");
#endif
#if ERRORS_ONLY
if (!target) {
static int flag = 0;
if (!flag) /* Skip first failure for all chips. */
++flag;
else
printk("fd_mcs: Selection failed\n");
}
#endif
return 1;
}
static void my_done(struct Scsi_Host *shpnt, int error)
{
if (in_command) {
in_command = 0;
outb(0x00, Interrupt_Cntl_port);
fd_mcs_make_bus_idle(shpnt);
current_SC->result = error;
current_SC->scsi_done(current_SC);
} else {
panic("fd_mcs: my_done() called outside of command\n");
}
#if DEBUG_RACE
in_interrupt_flag = 0;
#endif
}
/* only my_done needs to be protected */
static irqreturn_t fd_mcs_intr(int irq, void *dev_id)
{
unsigned long flags;
int status;
int done = 0;
unsigned data_count, tmp_count;
int i = 0;
struct Scsi_Host *shpnt;
TOTAL_INTR++;
/* search for one adapter-response on shared interrupt */
while ((shpnt = hosts[i++])) {
if ((inb(TMC_Status_port)) & 1)
break;
}
/* return if some other device on this IRQ caused the interrupt */
if (!shpnt) {
return IRQ_NONE;
}
INTR_Processed++;
outb(0x00, Interrupt_Cntl_port);
/* Abort calls my_done, so we do nothing here. */
if (current_SC->SCp.phase & aborted) {
#if DEBUG_ABORT
printk("Interrupt after abort, ignoring\n");
#endif
/* return IRQ_HANDLED; */
}
#if DEBUG_RACE
++in_interrupt_flag;
#endif
if (current_SC->SCp.phase & in_arbitration) {
status = inb(TMC_Status_port); /* Read adapter status */
if (!(status & 0x02)) {
#if EVERY_ACCESS
printk(" AFAIL ");
#endif
spin_lock_irqsave(shpnt->host_lock, flags);
my_done(shpnt, DID_BUS_BUSY << 16);
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
}
current_SC->SCp.phase = in_selection;
outb(0x40 | FIFO_COUNT, Interrupt_Cntl_port);
outb(0x82, SCSI_Cntl_port); /* Bus Enable + Select */
outb(adapter_mask | (1 << scmd_id(current_SC)), SCSI_Data_NoACK_port);
/* Stop arbitration and enable parity */
outb(0x10 | PARITY_MASK, TMC_Cntl_port);
#if DEBUG_RACE
in_interrupt_flag = 0;
#endif
return IRQ_HANDLED;
} else if (current_SC->SCp.phase & in_selection) {
status = inb(SCSI_Status_port);
if (!(status & 0x01)) {
/* Try again, for slow devices */
if (fd_mcs_select(shpnt, scmd_id(current_SC))) {
#if EVERY_ACCESS
printk(" SFAIL ");
#endif
spin_lock_irqsave(shpnt->host_lock, flags);
my_done(shpnt, DID_NO_CONNECT << 16);
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
} else {
#if EVERY_ACCESS
printk(" AltSel ");
#endif
/* Stop arbitration and enable parity */
outb(0x10 | PARITY_MASK, TMC_Cntl_port);
}
}
current_SC->SCp.phase = in_other;
outb(0x90 | FIFO_COUNT, Interrupt_Cntl_port);
outb(0x80, SCSI_Cntl_port);
#if DEBUG_RACE
in_interrupt_flag = 0;
#endif
return IRQ_HANDLED;
}
/* current_SC->SCp.phase == in_other: this is the body of the routine */
status = inb(SCSI_Status_port);
if (status & 0x10) { /* REQ */
switch (status & 0x0e) {
case 0x08: /* COMMAND OUT */
outb(current_SC->cmnd[current_SC->SCp.sent_command++], Write_SCSI_Data_port);
#if EVERY_ACCESS
printk("CMD = %x,", current_SC->cmnd[current_SC->SCp.sent_command - 1]);
#endif
break;
case 0x00: /* DATA OUT -- tmc18c50/tmc18c30 only */
if (chip != tmc1800 && !current_SC->SCp.have_data_in) {
current_SC->SCp.have_data_in = -1;
outb(0xd0 | PARITY_MASK, TMC_Cntl_port);
}
break;
case 0x04: /* DATA IN -- tmc18c50/tmc18c30 only */
if (chip != tmc1800 && !current_SC->SCp.have_data_in) {
current_SC->SCp.have_data_in = 1;
outb(0x90 | PARITY_MASK, TMC_Cntl_port);
}
break;
case 0x0c: /* STATUS IN */
current_SC->SCp.Status = inb(Read_SCSI_Data_port);
#if EVERY_ACCESS
printk("Status = %x, ", current_SC->SCp.Status);
#endif
#if ERRORS_ONLY
if (current_SC->SCp.Status && current_SC->SCp.Status != 2 && current_SC->SCp.Status != 8) {
printk("ERROR fd_mcs: target = %d, command = %x, status = %x\n", current_SC->device->id, current_SC->cmnd[0], current_SC->SCp.Status);
}
#endif
break;
case 0x0a: /* MESSAGE OUT */
outb(MESSAGE_REJECT, Write_SCSI_Data_port); /* Reject */
break;
case 0x0e: /* MESSAGE IN */
current_SC->SCp.Message = inb(Read_SCSI_Data_port);
#if EVERY_ACCESS
printk("Message = %x, ", current_SC->SCp.Message);
#endif
if (!current_SC->SCp.Message)
++done;
#if DEBUG_MESSAGES || EVERY_ACCESS
if (current_SC->SCp.Message) {
printk("fd_mcs: message = %x\n", current_SC->SCp.Message);
}
#endif
break;
}
}
if (chip == tmc1800 && !current_SC->SCp.have_data_in && (current_SC->SCp.sent_command >= current_SC->cmd_len)) {
/* We have to get the FIFO direction
correct, so I've made a table based
on the SCSI Standard of which commands
appear to require a DATA OUT phase.
*/
/*
p. 94: Command for all device types
CHANGE DEFINITION 40 DATA OUT
COMPARE 39 DATA OUT
COPY 18 DATA OUT
COPY AND VERIFY 3a DATA OUT
INQUIRY 12
LOG SELECT 4c DATA OUT
LOG SENSE 4d
MODE SELECT (6) 15 DATA OUT
MODE SELECT (10) 55 DATA OUT
MODE SENSE (6) 1a
MODE SENSE (10) 5a
READ BUFFER 3c
RECEIVE DIAGNOSTIC RESULTS 1c
REQUEST SENSE 03
SEND DIAGNOSTIC 1d DATA OUT
TEST UNIT READY 00
WRITE BUFFER 3b DATA OUT
p.178: Commands for direct-access devices (not listed on p. 94)
FORMAT UNIT 04 DATA OUT
LOCK-UNLOCK CACHE 36
PRE-FETCH 34
PREVENT-ALLOW MEDIUM REMOVAL 1e
READ (6)/RECEIVE 08
READ (10) 3c
READ CAPACITY 25
READ DEFECT DATA (10) 37
READ LONG 3e
REASSIGN BLOCKS 07 DATA OUT
RELEASE 17
RESERVE 16 DATA OUT
REZERO UNIT/REWIND 01
SEARCH DATA EQUAL (10) 31 DATA OUT
SEARCH DATA HIGH (10) 30 DATA OUT
SEARCH DATA LOW (10) 32 DATA OUT
SEEK (6) 0b
SEEK (10) 2b
SET LIMITS (10) 33
START STOP UNIT 1b
SYNCHRONIZE CACHE 35
VERIFY (10) 2f
WRITE (6)/PRINT/SEND 0a DATA OUT
WRITE (10)/SEND 2a DATA OUT
WRITE AND VERIFY (10) 2e DATA OUT
WRITE LONG 3f DATA OUT
WRITE SAME 41 DATA OUT ?
p. 261: Commands for sequential-access devices (not previously listed)
ERASE 19
LOAD UNLOAD 1b
LOCATE 2b
READ BLOCK LIMITS 05
READ POSITION 34
READ REVERSE 0f
RECOVER BUFFERED DATA 14
SPACE 11
WRITE FILEMARKS 10 ?
p. 298: Commands for printer devices (not previously listed)
****** NOT SUPPORTED BY THIS DRIVER, since 0b is SEEK (6) *****
SLEW AND PRINT 0b DATA OUT -- same as seek
STOP PRINT 1b
SYNCHRONIZE BUFFER 10
p. 315: Commands for processor devices (not previously listed)
p. 321: Commands for write-once devices (not previously listed)
MEDIUM SCAN 38
READ (12) a8
SEARCH DATA EQUAL (12) b1 DATA OUT
SEARCH DATA HIGH (12) b0 DATA OUT
SEARCH DATA LOW (12) b2 DATA OUT
SET LIMITS (12) b3
VERIFY (12) af
WRITE (12) aa DATA OUT
WRITE AND VERIFY (12) ae DATA OUT
p. 332: Commands for CD-ROM devices (not previously listed)
PAUSE/RESUME 4b
PLAY AUDIO (10) 45
PLAY AUDIO (12) a5
PLAY AUDIO MSF 47
PLAY TRACK RELATIVE (10) 49
PLAY TRACK RELATIVE (12) a9
READ HEADER 44
READ SUB-CHANNEL 42
READ TOC 43
p. 370: Commands for scanner devices (not previously listed)
GET DATA BUFFER STATUS 34
GET WINDOW 25
OBJECT POSITION 31
SCAN 1b
SET WINDOW 24 DATA OUT
p. 391: Commands for optical memory devices (not listed)
ERASE (10) 2c
ERASE (12) ac
MEDIUM SCAN 38 DATA OUT
READ DEFECT DATA (12) b7
READ GENERATION 29
READ UPDATED BLOCK 2d
UPDATE BLOCK 3d DATA OUT
p. 419: Commands for medium changer devices (not listed)
EXCHANGE MEDIUM 46
INITIALIZE ELEMENT STATUS 07
MOVE MEDIUM a5
POSITION TO ELEMENT 2b
READ ELEMENT STATUS b8
REQUEST VOL. ELEMENT ADDRESS b5
SEND VOLUME TAG b6 DATA OUT
p. 454: Commands for communications devices (not listed previously)
GET MESSAGE (6) 08
GET MESSAGE (10) 28
GET MESSAGE (12) a8
*/
switch (current_SC->cmnd[0]) {
case CHANGE_DEFINITION:
case COMPARE:
case COPY:
case COPY_VERIFY:
case LOG_SELECT:
case MODE_SELECT:
case MODE_SELECT_10:
case SEND_DIAGNOSTIC:
case WRITE_BUFFER:
case FORMAT_UNIT:
case REASSIGN_BLOCKS:
case RESERVE:
case SEARCH_EQUAL:
case SEARCH_HIGH:
case SEARCH_LOW:
case WRITE_6:
case WRITE_10:
case WRITE_VERIFY:
case 0x3f:
case 0x41:
case 0xb1:
case 0xb0:
case 0xb2:
case 0xaa:
case 0xae:
case 0x24:
case 0x38:
case 0x3d:
case 0xb6:
case 0xea: /* alternate number for WRITE LONG */
current_SC->SCp.have_data_in = -1;
outb(0xd0 | PARITY_MASK, TMC_Cntl_port);
break;
case 0x00:
default:
current_SC->SCp.have_data_in = 1;
outb(0x90 | PARITY_MASK, TMC_Cntl_port);
break;
}
}
if (current_SC->SCp.have_data_in == -1) { /* DATA OUT */
while ((data_count = FIFO_Size - inw(FIFO_Data_Count_port)) > 512) {
#if EVERY_ACCESS
printk("DC=%d, ", data_count);
#endif
if (data_count > current_SC->SCp.this_residual)
data_count = current_SC->SCp.this_residual;
if (data_count > 0) {
#if EVERY_ACCESS
printk("%d OUT, ", data_count);
#endif
if (data_count == 1) {
Bytes_Written++;
outb(*current_SC->SCp.ptr++, Write_FIFO_port);
--current_SC->SCp.this_residual;
} else {
data_count >>= 1;
tmp_count = data_count << 1;
outsw(Write_FIFO_port, current_SC->SCp.ptr, data_count);
current_SC->SCp.ptr += tmp_count;
Bytes_Written += tmp_count;
current_SC->SCp.this_residual -= tmp_count;
}
}
if (!current_SC->SCp.this_residual) {
if (current_SC->SCp.buffers_residual) {
--current_SC->SCp.buffers_residual;
++current_SC->SCp.buffer;
current_SC->SCp.ptr = sg_virt(current_SC->SCp.buffer);
current_SC->SCp.this_residual = current_SC->SCp.buffer->length;
} else
break;
}
}
} else if (current_SC->SCp.have_data_in == 1) { /* DATA IN */
while ((data_count = inw(FIFO_Data_Count_port)) > 0) {
#if EVERY_ACCESS
printk("DC=%d, ", data_count);
#endif
if (data_count > current_SC->SCp.this_residual)
data_count = current_SC->SCp.this_residual;
if (data_count) {
#if EVERY_ACCESS
printk("%d IN, ", data_count);
#endif
if (data_count == 1) {
Bytes_Read++;
*current_SC->SCp.ptr++ = inb(Read_FIFO_port);
--current_SC->SCp.this_residual;
} else {
data_count >>= 1; /* Number of words */
tmp_count = data_count << 1;
insw(Read_FIFO_port, current_SC->SCp.ptr, data_count);
current_SC->SCp.ptr += tmp_count;
Bytes_Read += tmp_count;
current_SC->SCp.this_residual -= tmp_count;
}
}
if (!current_SC->SCp.this_residual && current_SC->SCp.buffers_residual) {
--current_SC->SCp.buffers_residual;
++current_SC->SCp.buffer;
current_SC->SCp.ptr = sg_virt(current_SC->SCp.buffer);
current_SC->SCp.this_residual = current_SC->SCp.buffer->length;
}
}
}
if (done) {
#if EVERY_ACCESS
printk(" ** IN DONE %d ** ", current_SC->SCp.have_data_in);
#endif
#if EVERY_ACCESS
printk("BEFORE MY_DONE. . .");
#endif
spin_lock_irqsave(shpnt->host_lock, flags);
my_done(shpnt, (current_SC->SCp.Status & 0xff)
| ((current_SC->SCp.Message & 0xff) << 8) | (DID_OK << 16));
spin_unlock_irqrestore(shpnt->host_lock, flags);
#if EVERY_ACCESS
printk("RETURNING.\n");
#endif
} else {
if (current_SC->SCp.phase & disconnect) {
outb(0xd0 | FIFO_COUNT, Interrupt_Cntl_port);
outb(0x00, SCSI_Cntl_port);
} else {
outb(0x90 | FIFO_COUNT, Interrupt_Cntl_port);
}
}
#if DEBUG_RACE
in_interrupt_flag = 0;
#endif
return IRQ_HANDLED;
}
static int fd_mcs_release(struct Scsi_Host *shpnt)
{
int i, this_host, irq_usage;
release_region(shpnt->io_port, shpnt->n_io_port);
this_host = -1;
irq_usage = 0;
for (i = 0; i < found; i++) {
if (shpnt == hosts[i])
this_host = i;
if (shpnt->irq == hosts[i]->irq)
irq_usage++;
}
/* only for the last one */
if (1 == irq_usage)
free_irq(shpnt->irq, hosts);
found--;
for (i = this_host; i < found; i++)
hosts[i] = hosts[i + 1];
hosts[found] = NULL;
return 0;
}
static int fd_mcs_queue_lck(Scsi_Cmnd * SCpnt, void (*done) (Scsi_Cmnd *))
{
struct Scsi_Host *shpnt = SCpnt->device->host;
if (in_command) {
panic("fd_mcs: fd_mcs_queue() NOT REENTRANT!\n");
}
#if EVERY_ACCESS
printk("queue: target = %d cmnd = 0x%02x pieces = %d size = %u\n",
SCpnt->target, *(unsigned char *) SCpnt->cmnd,
scsi_sg_count(SCpnt), scsi_bufflen(SCpnt));
#endif
fd_mcs_make_bus_idle(shpnt);
SCpnt->scsi_done = done; /* Save this for the done function */
current_SC = SCpnt;
/* Initialize static data */
if (scsi_bufflen(current_SC)) {
current_SC->SCp.buffer = scsi_sglist(current_SC);
current_SC->SCp.ptr = sg_virt(current_SC->SCp.buffer);
current_SC->SCp.this_residual = current_SC->SCp.buffer->length;
current_SC->SCp.buffers_residual = scsi_sg_count(current_SC) - 1;
} else {
current_SC->SCp.ptr = NULL;
current_SC->SCp.this_residual = 0;
current_SC->SCp.buffer = NULL;
current_SC->SCp.buffers_residual = 0;
}
current_SC->SCp.Status = 0;
current_SC->SCp.Message = 0;
current_SC->SCp.have_data_in = 0;
current_SC->SCp.sent_command = 0;
current_SC->SCp.phase = in_arbitration;
/* Start arbitration */
outb(0x00, Interrupt_Cntl_port);
outb(0x00, SCSI_Cntl_port); /* Disable data drivers */
outb(adapter_mask, SCSI_Data_NoACK_port); /* Set our id bit */
in_command = 1;
outb(0x20, Interrupt_Cntl_port);
outb(0x14 | PARITY_MASK, TMC_Cntl_port); /* Start arbitration */
return 0;
}
static DEF_SCSI_QCMD(fd_mcs_queue)
#if DEBUG_ABORT || DEBUG_RESET
static void fd_mcs_print_info(Scsi_Cmnd * SCpnt)
{
unsigned int imr;
unsigned int irr;
unsigned int isr;
struct Scsi_Host *shpnt = SCpnt->host;
if (!SCpnt || !SCpnt->host) {
printk("fd_mcs: cannot provide detailed information\n");
}
printk("%s\n", fd_mcs_info(SCpnt->host));
print_banner(SCpnt->host);
switch (SCpnt->SCp.phase) {
case in_arbitration:
printk("arbitration ");
break;
case in_selection:
printk("selection ");
break;
case in_other:
printk("other ");
break;
default:
printk("unknown ");
break;
}
printk("(%d), target = %d cmnd = 0x%02x pieces = %d size = %u\n",
SCpnt->SCp.phase, SCpnt->device->id, *(unsigned char *) SCpnt->cmnd,
scsi_sg_count(SCpnt), scsi_bufflen(SCpnt));
printk("sent_command = %d, have_data_in = %d, timeout = %d\n", SCpnt->SCp.sent_command, SCpnt->SCp.have_data_in, SCpnt->timeout);
#if DEBUG_RACE
printk("in_interrupt_flag = %d\n", in_interrupt_flag);
#endif
imr = (inb(0x0a1) << 8) + inb(0x21);
outb(0x0a, 0xa0);
irr = inb(0xa0) << 8;
outb(0x0a, 0x20);
irr += inb(0x20);
outb(0x0b, 0xa0);
isr = inb(0xa0) << 8;
outb(0x0b, 0x20);
isr += inb(0x20);
/* Print out interesting information */
printk("IMR = 0x%04x", imr);
if (imr & (1 << shpnt->irq))
printk(" (masked)");
printk(", IRR = 0x%04x, ISR = 0x%04x\n", irr, isr);
printk("SCSI Status = 0x%02x\n", inb(SCSI_Status_port));
printk("TMC Status = 0x%02x", inb(TMC_Status_port));
if (inb(TMC_Status_port) & 1)
printk(" (interrupt)");
printk("\n");
printk("Interrupt Status = 0x%02x", inb(Interrupt_Status_port));
if (inb(Interrupt_Status_port) & 0x08)
printk(" (enabled)");
printk("\n");
if (chip == tmc18c50 || chip == tmc18c30) {
printk("FIFO Status = 0x%02x\n", inb(shpnt->io_port + FIFO_Status));
printk("Int. Condition = 0x%02x\n", inb(shpnt->io_port + Interrupt_Cond));
}
printk("Configuration 1 = 0x%02x\n", inb(shpnt->io_port + Configuration1));
if (chip == tmc18c50 || chip == tmc18c30)
printk("Configuration 2 = 0x%02x\n", inb(shpnt->io_port + Configuration2));
}
#endif
static int fd_mcs_abort(Scsi_Cmnd * SCpnt)
{
struct Scsi_Host *shpnt = SCpnt->device->host;
unsigned long flags;
#if EVERY_ACCESS || ERRORS_ONLY || DEBUG_ABORT
printk("fd_mcs: abort ");
#endif
spin_lock_irqsave(shpnt->host_lock, flags);
if (!in_command) {
#if EVERY_ACCESS || ERRORS_ONLY
printk(" (not in command)\n");
#endif
spin_unlock_irqrestore(shpnt->host_lock, flags);
return FAILED;
} else
printk("\n");
#if DEBUG_ABORT
fd_mcs_print_info(SCpnt);
#endif
fd_mcs_make_bus_idle(shpnt);
current_SC->SCp.phase |= aborted;
current_SC->result = DID_ABORT << 16;
/* Aborts are not done well. . . */
my_done(shpnt, DID_ABORT << 16);
spin_unlock_irqrestore(shpnt->host_lock, flags);
return SUCCESS;
}
static int fd_mcs_bus_reset(Scsi_Cmnd * SCpnt) {
struct Scsi_Host *shpnt = SCpnt->device->host;
unsigned long flags;
#if DEBUG_RESET
static int called_once = 0;
#endif
#if ERRORS_ONLY
if (SCpnt)
printk("fd_mcs: SCSI Bus Reset\n");
#endif
#if DEBUG_RESET
if (called_once)
fd_mcs_print_info(current_SC);
called_once = 1;
#endif
spin_lock_irqsave(shpnt->host_lock, flags);
outb(1, SCSI_Cntl_port);
do_pause(2);
outb(0, SCSI_Cntl_port);
do_pause(115);
outb(0, SCSI_Mode_Cntl_port);
outb(PARITY_MASK, TMC_Cntl_port);
spin_unlock_irqrestore(shpnt->host_lock, flags);
/* Unless this is the very first call (i.e., SCPnt == NULL), everything
is probably hosed at this point. We will, however, try to keep
things going by informing the high-level code that we need help. */
return SUCCESS;
}
#include <scsi/scsi_ioctl.h>
static int fd_mcs_biosparam(struct scsi_device * disk, struct block_device *bdev,
sector_t capacity, int *info_array)
{
unsigned char *p = scsi_bios_ptable(bdev);
int size = capacity;
/* BIOS >= 3.4 for MCA cards */
/* This algorithm was provided by Future Domain (much thanks!). */
if (p && p[65] == 0xaa && p[64] == 0x55 /* Partition table valid */
&& p[4]) { /* Partition type */
/* The partition table layout is as follows:
Start: 0x1b3h
Offset: 0 = partition status
1 = starting head
2 = starting sector and cylinder (word, encoded)
4 = partition type
5 = ending head
6 = ending sector and cylinder (word, encoded)
8 = starting absolute sector (double word)
c = number of sectors (double word)
Signature: 0x1fe = 0x55aa
So, this algorithm assumes:
1) the first partition table is in use,
2) the data in the first entry is correct, and
3) partitions never divide cylinders
Note that (1) may be FALSE for NetBSD (and other BSD flavors),
as well as for Linux. Note also, that Linux doesn't pay any
attention to the fields that are used by this algorithm -- it
only uses the absolute sector data. Recent versions of Linux's
fdisk(1) will fill this data in correctly, and forthcoming
versions will check for consistency.
Checking for a non-zero partition type is not part of the
Future Domain algorithm, but it seemed to be a reasonable thing
to do, especially in the Linux and BSD worlds. */
info_array[0] = p[5] + 1; /* heads */
info_array[1] = p[6] & 0x3f; /* sectors */
} else {
/* Note that this new method guarantees that there will always be
less than 1024 cylinders on a platter. This is good for drives
up to approximately 7.85GB (where 1GB = 1024 * 1024 kB). */
if ((unsigned int) size >= 0x7e0000U)
{
info_array[0] = 0xff; /* heads = 255 */
info_array[1] = 0x3f; /* sectors = 63 */
} else if ((unsigned int) size >= 0x200000U) {
info_array[0] = 0x80; /* heads = 128 */
info_array[1] = 0x3f; /* sectors = 63 */
} else {
info_array[0] = 0x40; /* heads = 64 */
info_array[1] = 0x20; /* sectors = 32 */
}
}
/* For both methods, compute the cylinders */
info_array[2] = (unsigned int) size / (info_array[0] * info_array[1]);
kfree(p);
return 0;
}
static struct scsi_host_template driver_template = {
.proc_name = "fd_mcs",
.proc_info = fd_mcs_proc_info,
.detect = fd_mcs_detect,
.release = fd_mcs_release,
.info = fd_mcs_info,
.queuecommand = fd_mcs_queue,
.eh_abort_handler = fd_mcs_abort,
.eh_bus_reset_handler = fd_mcs_bus_reset,
.bios_param = fd_mcs_biosparam,
.can_queue = 1,
.this_id = 7,
.sg_tablesize = 64,
.cmd_per_lun = 1,
.use_clustering = DISABLE_CLUSTERING,
};
#include "scsi_module.c"
MODULE_LICENSE("GPL");
drivers/scsi/ibmmca.c deleted 100644 → 0
View file @ c80ddb52
/*
Low Level Linux Driver for the IBM Microchannel SCSI Subsystem for
Linux Kernel >= 2.4.0.
Copyright (c) 1995 Strom Systems, Inc. under the terms of the GNU
General Public License. Written by Martin Kolinek, December 1995.
Further development by: Chris Beauregard, Klaus Kudielka, Michael Lang
See the file Documentation/scsi/ibmmca.txt for a detailed description
of this driver, the commandline arguments and the history of its
development.
See the WWW-page: http://www.uni-mainz.de/~langm000/linux.html for latest
updates, info and ADF-files for adapters supported by this driver.
Alan Cox <alan@lxorguk.ukuu.org.uk>
Updated for Linux 2.5.45 to use the new error handler, cleaned up the
lock macros and did a few unavoidable locking tweaks, plus one locking
fix in the irq and completion path.
*/
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/ctype.h>
#include <linux/string.h>
#include <linux/interrupt.h>
#include <linux/ioport.h>
#include <linux/delay.h>
#include <linux/blkdev.h>
#include <linux/proc_fs.h>
#include <linux/stat.h>
#include <linux/mca.h>
#include <linux/spinlock.h>
#include <linux/init.h>
#include <asm/io.h>
#include "scsi.h"
#include <scsi/scsi_host.h>
/* Common forward declarations for all Linux-versions: */
static int ibmmca_queuecommand (struct Scsi_Host *, struct scsi_cmnd *);
static int ibmmca_abort (Scsi_Cmnd *);
static int ibmmca_host_reset (Scsi_Cmnd *);
static int ibmmca_biosparam (struct scsi_device *, struct block_device *, sector_t, int *);
static int ibmmca_proc_info(struct Scsi_Host *shpnt, char *buffer, char **start, off_t offset, int length, int inout);
/* current version of this driver-source: */
#define IBMMCA_SCSI_DRIVER_VERSION "4.0b-ac"
/* driver configuration */
#define IM_MAX_HOSTS 8 /* maximum number of host adapters */
#define IM_RESET_DELAY 60 /* seconds allowed for a reset */
/* driver debugging - #undef all for normal operation */
/* if defined: count interrupts and ignore this special one: */
#undef IM_DEBUG_TIMEOUT //50
#define TIMEOUT_PUN 0
#define TIMEOUT_LUN 0
/* verbose interrupt: */
#undef IM_DEBUG_INT
/* verbose queuecommand: */
#undef IM_DEBUG_CMD
/* verbose queucommand for specific SCSI-device type: */
#undef IM_DEBUG_CMD_SPEC_DEV
/* verbose device probing */
#undef IM_DEBUG_PROBE
/* device type that shall be displayed on syslog (only during debugging): */
#define IM_DEBUG_CMD_DEVICE TYPE_TAPE
/* relative addresses of hardware registers on a subsystem */
#define IM_CMD_REG(h) ((h)->io_port) /*Command Interface, (4 bytes long) */
#define IM_ATTN_REG(h) ((h)->io_port+4) /*Attention (1 byte) */
#define IM_CTR_REG(h) ((h)->io_port+5) /*Basic Control (1 byte) */
#define IM_INTR_REG(h) ((h)->io_port+6) /*Interrupt Status (1 byte, r/o) */
#define IM_STAT_REG(h) ((h)->io_port+7) /*Basic Status (1 byte, read only) */
/* basic I/O-port of first adapter */
#define IM_IO_PORT 0x3540
/* maximum number of hosts that can be found */
#define IM_N_IO_PORT 8
/*requests going into the upper nibble of the Attention register */
/*note: the lower nibble specifies the device(0-14), or subsystem(15) */
#define IM_IMM_CMD 0x10 /*immediate command */
#define IM_SCB 0x30 /*Subsystem Control Block command */
#define IM_LONG_SCB 0x40 /*long Subsystem Control Block command */
#define IM_EOI 0xe0 /*end-of-interrupt request */
/*values for bits 7,1,0 of Basic Control reg. (bits 6-2 reserved) */
#define IM_HW_RESET 0x80 /*hardware reset */
#define IM_ENABLE_DMA 0x02 /*enable subsystem's busmaster DMA */
#define IM_ENABLE_INTR 0x01 /*enable interrupts to the system */
/*to interpret the upper nibble of Interrupt Status register */
/*note: the lower nibble specifies the device(0-14), or subsystem(15) */
#define IM_SCB_CMD_COMPLETED 0x10
#define IM_SCB_CMD_COMPLETED_WITH_RETRIES 0x50
#define IM_LOOP_SCATTER_BUFFER_FULL 0x60
#define IM_ADAPTER_HW_FAILURE 0x70
#define IM_IMMEDIATE_CMD_COMPLETED 0xa0
#define IM_CMD_COMPLETED_WITH_FAILURE 0xc0
#define IM_CMD_ERROR 0xe0
#define IM_SOFTWARE_SEQUENCING_ERROR 0xf0
/*to interpret bits 3-0 of Basic Status register (bits 7-4 reserved) */
#define IM_CMD_REG_FULL 0x08
#define IM_CMD_REG_EMPTY 0x04
#define IM_INTR_REQUEST 0x02
#define IM_BUSY 0x01
/*immediate commands (word written into low 2 bytes of command reg) */
#define IM_RESET_IMM_CMD 0x0400
#define IM_FEATURE_CTR_IMM_CMD 0x040c
#define IM_DMA_PACING_IMM_CMD 0x040d
#define IM_ASSIGN_IMM_CMD 0x040e
#define IM_ABORT_IMM_CMD 0x040f
#define IM_FORMAT_PREP_IMM_CMD 0x0417
/*SCB (Subsystem Control Block) structure */
struct im_scb {
unsigned short command; /*command word (read, etc.) */
unsigned short enable; /*enable word, modifies cmd */
union {
unsigned long log_blk_adr; /*block address on SCSI device */
unsigned char scsi_cmd_length; /*6,10,12, for other scsi cmd */
} u1;
unsigned long sys_buf_adr; /*physical system memory adr */
unsigned long sys_buf_length; /*size of sys mem buffer */
unsigned long tsb_adr; /*Termination Status Block adr */
unsigned long scb_chain_adr; /*optional SCB chain address */
union {
struct {
unsigned short count; /*block count, on SCSI device */
unsigned short length; /*block length, on SCSI device */
} blk;
unsigned char scsi_command[12]; /*other scsi command */
} u2;
};
/*structure scatter-gather element (for list of system memory areas) */
struct im_sge {
void *address;
unsigned long byte_length;
};
/*structure returned by a get_pos_info command: */
struct im_pos_info {
unsigned short pos_id; /* adapter id */
unsigned char pos_3a; /* pos 3 (if pos 6 = 0) */
unsigned char pos_2; /* pos 2 */
unsigned char int_level; /* interrupt level IRQ 11 or 14 */
unsigned char pos_4a; /* pos 4 (if pos 6 = 0) */
unsigned short connector_size; /* MCA connector size: 16 or 32 Bit */
unsigned char num_luns; /* number of supported luns per device */
unsigned char num_puns; /* number of supported puns */
unsigned char pacing_factor; /* pacing factor */
unsigned char num_ldns; /* number of ldns available */
unsigned char eoi_off; /* time EOI and interrupt inactive */
unsigned char max_busy; /* time between reset and busy on */
unsigned short cache_stat; /* ldn cachestat. Bit=1 = not cached */
unsigned short retry_stat; /* retry status of ldns. Bit=1=disabled */
unsigned char pos_4b; /* pos 4 (if pos 6 = 1) */
unsigned char pos_3b; /* pos 3 (if pos 6 = 1) */
unsigned char pos_6; /* pos 6 */
unsigned char pos_5; /* pos 5 */
unsigned short max_overlap; /* maximum overlapping requests */
unsigned short num_bus; /* number of SCSI-busses */
};
/*values for SCB command word */
#define IM_NO_SYNCHRONOUS 0x0040 /*flag for any command */
#define IM_NO_DISCONNECT 0x0080 /*flag for any command */
#define IM_READ_DATA_CMD 0x1c01
#define IM_WRITE_DATA_CMD 0x1c02
#define IM_READ_VERIFY_CMD 0x1c03
#define IM_WRITE_VERIFY_CMD 0x1c04
#define IM_REQUEST_SENSE_CMD 0x1c08
#define IM_READ_CAPACITY_CMD 0x1c09
#define IM_DEVICE_INQUIRY_CMD 0x1c0b
#define IM_READ_LOGICAL_CMD 0x1c2a
#define IM_OTHER_SCSI_CMD_CMD 0x241f
/* unused, but supported, SCB commands */
#define IM_GET_COMMAND_COMPLETE_STATUS_CMD 0x1c07 /* command status */
#define IM_GET_POS_INFO_CMD 0x1c0a /* returns neat stuff */
#define IM_READ_PREFETCH_CMD 0x1c31 /* caching controller only */
#define IM_FOMAT_UNIT_CMD 0x1c16 /* format unit */
#define IM_REASSIGN_BLOCK_CMD 0x1c18 /* in case of error */
/*values to set bits in the enable word of SCB */
#define IM_READ_CONTROL 0x8000
#define IM_REPORT_TSB_ONLY_ON_ERROR 0x4000
#define IM_RETRY_ENABLE 0x2000
#define IM_POINTER_TO_LIST 0x1000
#define IM_SUPRESS_EXCEPTION_SHORT 0x0400
#define IM_BYPASS_BUFFER 0x0200
#define IM_CHAIN_ON_NO_ERROR 0x0001
/*TSB (Termination Status Block) structure */
struct im_tsb {
unsigned short end_status;
unsigned short reserved1;
unsigned long residual_byte_count;
unsigned long sg_list_element_adr;
unsigned short status_length;
unsigned char dev_status;
unsigned char cmd_status;
unsigned char dev_error;
unsigned char cmd_error;
unsigned short reserved2;
unsigned short reserved3;
unsigned short low_of_last_scb_adr;
unsigned short high_of_last_scb_adr;
};
/*subsystem uses interrupt request level 14 */
#define IM_IRQ 14
/*SCSI-2 F/W may evade to interrupt 11 */
#define IM_IRQ_FW 11
/* Model 95 has an additional alphanumeric display, which can be used
to display SCSI-activities. 8595 models do not have any disk led, which
makes this feature quite useful.
The regular PS/2 disk led is turned on/off by bits 6,7 of system
control port. */
/* LED display-port (actually, last LED on display) */
#define MOD95_LED_PORT 0x108
/* system-control-register of PS/2s with diskindicator */
#define PS2_SYS_CTR 0x92
/* activity displaying methods */
#define LED_DISP 1
#define LED_ADISP 2
#define LED_ACTIVITY 4
/* failed intr */
#define CMD_FAIL 255
/* The SCSI-ID(!) of the accessed SCSI-device is shown on PS/2-95 machines' LED
displays. ldn is no longer displayed here, because the ldn mapping is now
done dynamically and the ldn <-> pun,lun maps can be looked-up at boottime
or during uptime in /proc/scsi/ibmmca/<host_no> in case of trouble,
interest, debugging or just for having fun. The left number gives the
host-adapter number and the right shows the accessed SCSI-ID. */
/* display_mode is set by the ibmmcascsi= command line arg */
static int display_mode = 0;
/* set default adapter timeout */
static unsigned int adapter_timeout = 45;
/* for probing on feature-command: */
static unsigned int global_command_error_excuse = 0;
/* global setting by command line for adapter_speed */
static int global_adapter_speed = 0; /* full speed by default */
/* Panel / LED on, do it right for F/W addressin, too. adisplay will
* just ignore ids>7, as the panel has only 7 digits available */
#define PS2_DISK_LED_ON(ad,id) { if (display_mode & LED_DISP) { if (id>9) \
outw((ad+48)|((id+55)<<8), MOD95_LED_PORT ); else \
outw((ad+48)|((id+48)<<8), MOD95_LED_PORT ); } else \
if (display_mode & LED_ADISP) { if (id<7) outb((char)(id+48),MOD95_LED_PORT+1+id); \
outb((char)(ad+48), MOD95_LED_PORT); } \
if ((display_mode & LED_ACTIVITY)||(!display_mode)) \
outb(inb(PS2_SYS_CTR) | 0xc0, PS2_SYS_CTR); }
/* Panel / LED off */
/* bug fixed, Dec 15, 1997, where | was replaced by & here */
#define PS2_DISK_LED_OFF() { if (display_mode & LED_DISP) \
outw(0x2020, MOD95_LED_PORT ); else if (display_mode & LED_ADISP) { \
outl(0x20202020,MOD95_LED_PORT); outl(0x20202020,MOD95_LED_PORT+4); } \
if ((display_mode & LED_ACTIVITY)||(!display_mode)) \
outb(inb(PS2_SYS_CTR) & 0x3f, PS2_SYS_CTR); }
/* types of different supported hardware that goes to hostdata special */
#define IBM_SCSI2_FW 0
#define IBM_7568_WCACHE 1
#define IBM_EXP_UNIT 2
#define IBM_SCSI_WCACHE 3
#define IBM_SCSI 4
#define IBM_INTEGSCSI 5
/* other special flags for hostdata structure */
#define FORCED_DETECTION 100
#define INTEGRATED_SCSI 101
/* List of possible IBM-SCSI-adapters */
static short ibmmca_id_table[] = {
0x8efc,
0x8efd,
0x8ef8,
0x8eff,
0x8efe,
/* No entry for integrated SCSI, that's part of the register */
0
};
static const char *ibmmca_description[] = {
"IBM SCSI-2 F/W Adapter", /* special = 0 */
"IBM 7568 Industrial Computer SCSI Adapter w/Cache", /* special = 1 */
"IBM Expansion Unit SCSI Controller", /* special = 2 */
"IBM SCSI Adapter w/Cache", /* special = 3 */
"IBM SCSI Adapter", /* special = 4 */
"IBM Integrated SCSI Controller", /* special = 5 */
};
/* Max number of logical devices (can be up from 0 to 14). 15 is the address
of the adapter itself. */
#define MAX_LOG_DEV 15
/*local data for a logical device */
struct logical_device {
struct im_scb scb; /* SCSI-subsystem-control-block structure */
struct im_tsb tsb; /* SCSI command complete status block structure */
struct im_sge sge[16]; /* scatter gather list structure */
unsigned char buf[256]; /* SCSI command return data buffer */
Scsi_Cmnd *cmd; /* SCSI-command that is currently in progress */
int device_type; /* type of the SCSI-device. See include/scsi/scsi.h
for interpretation of the possible values */
int block_length; /* blocksize of a particular logical SCSI-device */
int cache_flag; /* 1 if this is uncached, 0 if cache is present for ldn */
int retry_flag; /* 1 if adapter retry is disabled, 0 if enabled */
};
/* statistics of the driver during operations (for proc_info) */
struct Driver_Statistics {
/* SCSI statistics on the adapter */
int ldn_access[MAX_LOG_DEV + 1]; /* total accesses on a ldn */
int ldn_read_access[MAX_LOG_DEV + 1]; /* total read-access on a ldn */
int ldn_write_access[MAX_LOG_DEV + 1]; /* total write-access on a ldn */
int ldn_inquiry_access[MAX_LOG_DEV + 1]; /* total inquiries on a ldn */
int ldn_modeselect_access[MAX_LOG_DEV + 1]; /* total mode selects on ldn */
int scbs; /* short SCBs queued */
int long_scbs; /* long SCBs queued */
int total_accesses; /* total accesses on all ldns */
int total_interrupts; /* total interrupts (should be
same as total_accesses) */
int total_errors; /* command completed with error */
/* dynamical assignment statistics */
int total_scsi_devices; /* number of physical pun,lun */
int dyn_flag; /* flag showing dynamical mode */
int dynamical_assignments; /* number of remappings of ldns */
int ldn_assignments[MAX_LOG_DEV + 1]; /* number of remappings of each
ldn */
};
/* data structure for each host adapter */
struct ibmmca_hostdata {
/* array of logical devices: */
struct logical_device _ld[MAX_LOG_DEV + 1];
/* array to convert (pun, lun) into logical device number: */
unsigned char _get_ldn[16][8];
/*array that contains the information about the physical SCSI-devices
attached to this host adapter: */
unsigned char _get_scsi[16][8];
/* used only when checking logical devices: */
int _local_checking_phase_flag;
/* report received interrupt: */
int _got_interrupt;
/* report termination-status of SCSI-command: */
int _stat_result;
/* reset status (used only when doing reset): */
int _reset_status;
/* code of the last SCSI command (needed for panic info): */
int _last_scsi_command[MAX_LOG_DEV + 1];
/* identifier of the last SCSI-command type */
int _last_scsi_type[MAX_LOG_DEV + 1];
/* last blockcount */
int _last_scsi_blockcount[MAX_LOG_DEV + 1];
/* last locgical block address */
unsigned long _last_scsi_logical_block[MAX_LOG_DEV + 1];
/* Counter that points on the next reassignable ldn for dynamical
remapping. The default value is 7, that is the first reassignable
number in the list at boottime: */
int _next_ldn;
/* Statistics-structure for this IBM-SCSI-host: */
struct Driver_Statistics _IBM_DS;
/* This hostadapters pos-registers pos2 until pos6 */
unsigned int _pos[8];
/* assign a special variable, that contains dedicated info about the
adaptertype */
int _special;
/* connector size on the MCA bus */
int _connector_size;
/* synchronous SCSI transfer rate bitpattern */
int _adapter_speed;
};
/* macros to access host data structure */
#define subsystem_pun(h) ((h)->this_id)
#define subsystem_maxid(h) ((h)->max_id)
#define ld(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_ld)
#define get_ldn(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_get_ldn)
#define get_scsi(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_get_scsi)
#define local_checking_phase_flag(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_local_checking_phase_flag)
#define got_interrupt(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_got_interrupt)
#define stat_result(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_stat_result)
#define reset_status(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_reset_status)
#define last_scsi_command(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_last_scsi_command)
#define last_scsi_type(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_last_scsi_type)
#define last_scsi_blockcount(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_last_scsi_blockcount)
#define last_scsi_logical_block(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_last_scsi_logical_block)
#define last_scsi_type(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_last_scsi_type)
#define next_ldn(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_next_ldn)
#define IBM_DS(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_IBM_DS)
#define special(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_special)
#define subsystem_connector_size(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_connector_size)
#define adapter_speed(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_adapter_speed)
#define pos2(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_pos[2])
#define pos3(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_pos[3])
#define pos4(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_pos[4])
#define pos5(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_pos[5])
#define pos6(h) (((struct ibmmca_hostdata *) (h)->hostdata)->_pos[6])
/* Define a arbitrary number as subsystem-marker-type. This number is, as
described in the ANSI-SCSI-standard, not occupied by other device-types. */
#define TYPE_IBM_SCSI_ADAPTER 0x2F
/* Define 0xFF for no device type, because this type is not defined within
the ANSI-SCSI-standard, therefore, it can be used and should not cause any
harm. */
#define TYPE_NO_DEVICE 0xFF
/* define medium-changer. If this is not defined previously, e.g. Linux
2.0.x, define this type here. */
#ifndef TYPE_MEDIUM_CHANGER
#define TYPE_MEDIUM_CHANGER 0x08
#endif
/* define possible operations for the immediate_assign command */
#define SET_LDN 0
#define REMOVE_LDN 1
/* ldn which is used to probe the SCSI devices */
#define PROBE_LDN 0
/* reset status flag contents */
#define IM_RESET_NOT_IN_PROGRESS 0
#define IM_RESET_IN_PROGRESS 1
#define IM_RESET_FINISHED_OK 2
#define IM_RESET_FINISHED_FAIL 3
#define IM_RESET_NOT_IN_PROGRESS_NO_INT 4
#define IM_RESET_FINISHED_OK_NO_INT 5
/* define undefined SCSI-command */
#define NO_SCSI 0xffff
/*-----------------------------------------------------------------------*/
/* if this is nonzero, ibmmcascsi option has been passed to the kernel */
static int io_port[IM_MAX_HOSTS] = { 0, 0, 0, 0, 0, 0, 0, 0 };
static int scsi_id[IM_MAX_HOSTS] = { 7, 7, 7, 7, 7, 7, 7, 7 };
/* fill module-parameters only, when this define is present.
(that is kernel version 2.1.x) */
#if defined(MODULE)
static char *boot_options = NULL;
module_param(boot_options, charp, 0);
module_param_array(io_port, int, NULL, 0);
module_param_array(scsi_id, int, NULL, 0);
MODULE_LICENSE("GPL");
#endif
/*counter of concurrent disk read/writes, to turn on/off disk led */
static int disk_rw_in_progress = 0;
static unsigned int pos[8]; /* whole pos register-line for diagnosis */
/* Taking into account the additions, made by ZP Gu.
* This selects now the preset value from the configfile and
* offers the 'normal' commandline option to be accepted */
#ifdef CONFIG_IBMMCA_SCSI_ORDER_STANDARD
static char ibm_ansi_order = 1;
#else
static char ibm_ansi_order = 0;
#endif
static void issue_cmd(struct Scsi_Host *, unsigned long, unsigned char);
static void internal_done(Scsi_Cmnd * cmd);
static void check_devices(struct Scsi_Host *, int);
static int immediate_assign(struct Scsi_Host *, unsigned int, unsigned int, unsigned int, unsigned int);
static int immediate_feature(struct Scsi_Host *, unsigned int, unsigned int);
#ifdef CONFIG_IBMMCA_SCSI_DEV_RESET
static int immediate_reset(struct Scsi_Host *, unsigned int);
#endif
static int device_inquiry(struct Scsi_Host *, int);
static int read_capacity(struct Scsi_Host *, int);
static int get_pos_info(struct Scsi_Host *);
static char *ti_p(int);
static char *ti_l(int);
static char *ibmrate(unsigned int, int);
static int probe_display(int);
static int probe_bus_mode(struct Scsi_Host *);
static int device_exists(struct Scsi_Host *, int, int *, int *);
static int option_setup(char *);
/* local functions needed for proc_info */
static int ldn_access_load(struct Scsi_Host *, int);
static int ldn_access_total_read_write(struct Scsi_Host *);
static irqreturn_t interrupt_handler(int irq, void *dev_id)
{
unsigned int intr_reg;
unsigned int cmd_result;
unsigned int ldn;
unsigned long flags;
Scsi_Cmnd *cmd;
int lastSCSI;
struct device *dev = dev_id;
struct Scsi_Host *shpnt = dev_get_drvdata(dev);
spin_lock_irqsave(shpnt->host_lock, flags);
if(!(inb(IM_STAT_REG(shpnt)) & IM_INTR_REQUEST)) {
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_NONE;
}
/* the reset-function already did all the job, even ints got
renabled on the subsystem, so just return */
if ((reset_status(shpnt) == IM_RESET_NOT_IN_PROGRESS_NO_INT) || (reset_status(shpnt) == IM_RESET_FINISHED_OK_NO_INT)) {
reset_status(shpnt) = IM_RESET_NOT_IN_PROGRESS;
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
}
/*must wait for attention reg not busy, then send EOI to subsystem */
while (1) {
if (!(inb(IM_STAT_REG(shpnt)) & IM_BUSY))
break;
cpu_relax();
}
/*get command result and logical device */
intr_reg = (unsigned char) (inb(IM_INTR_REG(shpnt)));
cmd_result = intr_reg & 0xf0;
ldn = intr_reg & 0x0f;
/* get the last_scsi_command here */
lastSCSI = last_scsi_command(shpnt)[ldn];
outb(IM_EOI | ldn, IM_ATTN_REG(shpnt));
/*these should never happen (hw fails, or a local programming bug) */
if (!global_command_error_excuse) {
switch (cmd_result) {
/* Prevent from Ooopsing on error to show the real reason */
case IM_ADAPTER_HW_FAILURE:
case IM_SOFTWARE_SEQUENCING_ERROR:
case IM_CMD_ERROR:
printk(KERN_ERR "IBM MCA SCSI: Fatal Subsystem ERROR!\n");
printk(KERN_ERR " Last cmd=0x%x, ena=%x, len=", lastSCSI, ld(shpnt)[ldn].scb.enable);
if (ld(shpnt)[ldn].cmd)
printk("%ld/%ld,", (long) (scsi_bufflen(ld(shpnt)[ldn].cmd)), (long) (ld(shpnt)[ldn].scb.sys_buf_length));
else
printk("none,");
if (ld(shpnt)[ldn].cmd)
printk("Blocksize=%d", ld(shpnt)[ldn].scb.u2.blk.length);
else
printk("Blocksize=none");
printk(", host=%p, ldn=0x%x\n", shpnt, ldn);
if (ld(shpnt)[ldn].cmd) {
printk(KERN_ERR "Blockcount=%d/%d\n", last_scsi_blockcount(shpnt)[ldn], ld(shpnt)[ldn].scb.u2.blk.count);
printk(KERN_ERR "Logical block=%lx/%lx\n", last_scsi_logical_block(shpnt)[ldn], ld(shpnt)[ldn].scb.u1.log_blk_adr);
}
printk(KERN_ERR "Reason given: %s\n", (cmd_result == IM_ADAPTER_HW_FAILURE) ? "HARDWARE FAILURE" : (cmd_result == IM_SOFTWARE_SEQUENCING_ERROR) ? "SOFTWARE SEQUENCING ERROR" : (cmd_result == IM_CMD_ERROR) ? "COMMAND ERROR" : "UNKNOWN");
/* if errors appear, enter this section to give detailed info */
printk(KERN_ERR "IBM MCA SCSI: Subsystem Error-Status follows:\n");
printk(KERN_ERR " Command Type................: %x\n", last_scsi_type(shpnt)[ldn]);
printk(KERN_ERR " Attention Register..........: %x\n", inb(IM_ATTN_REG(shpnt)));
printk(KERN_ERR " Basic Control Register......: %x\n", inb(IM_CTR_REG(shpnt)));
printk(KERN_ERR " Interrupt Status Register...: %x\n", intr_reg);
printk(KERN_ERR " Basic Status Register.......: %x\n", inb(IM_STAT_REG(shpnt)));
if ((last_scsi_type(shpnt)[ldn] == IM_SCB) || (last_scsi_type(shpnt)[ldn] == IM_LONG_SCB)) {
printk(KERN_ERR " SCB-Command.................: %x\n", ld(shpnt)[ldn].scb.command);
printk(KERN_ERR " SCB-Enable..................: %x\n", ld(shpnt)[ldn].scb.enable);
printk(KERN_ERR " SCB-logical block address...: %lx\n", ld(shpnt)[ldn].scb.u1.log_blk_adr);
printk(KERN_ERR " SCB-system buffer address...: %lx\n", ld(shpnt)[ldn].scb.sys_buf_adr);
printk(KERN_ERR " SCB-system buffer length....: %lx\n", ld(shpnt)[ldn].scb.sys_buf_length);
printk(KERN_ERR " SCB-tsb address.............: %lx\n", ld(shpnt)[ldn].scb.tsb_adr);
printk(KERN_ERR " SCB-Chain address...........: %lx\n", ld(shpnt)[ldn].scb.scb_chain_adr);
printk(KERN_ERR " SCB-block count.............: %x\n", ld(shpnt)[ldn].scb.u2.blk.count);
printk(KERN_ERR " SCB-block length............: %x\n", ld(shpnt)[ldn].scb.u2.blk.length);
}
printk(KERN_ERR " Send this report to the maintainer.\n");
panic("IBM MCA SCSI: Fatal error message from the subsystem (0x%X,0x%X)!\n", lastSCSI, cmd_result);
break;
}
} else {
/* The command error handling is made silent, but we tell the
* calling function, that there is a reported error from the
* adapter. */
switch (cmd_result) {
case IM_ADAPTER_HW_FAILURE:
case IM_SOFTWARE_SEQUENCING_ERROR:
case IM_CMD_ERROR:
global_command_error_excuse = CMD_FAIL;
break;
default:
global_command_error_excuse = 0;
break;
}
}
/* if no panic appeared, increase the interrupt-counter */
IBM_DS(shpnt).total_interrupts++;
/*only for local checking phase */
if (local_checking_phase_flag(shpnt)) {
stat_result(shpnt) = cmd_result;
got_interrupt(shpnt) = 1;
reset_status(shpnt) = IM_RESET_FINISHED_OK;
last_scsi_command(shpnt)[ldn] = NO_SCSI;
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
}
/* handling of commands coming from upper level of scsi driver */
if (last_scsi_type(shpnt)[ldn] == IM_IMM_CMD) {
/* verify ldn, and may handle rare reset immediate command */
if ((reset_status(shpnt) == IM_RESET_IN_PROGRESS) && (last_scsi_command(shpnt)[ldn] == IM_RESET_IMM_CMD)) {
if (cmd_result == IM_CMD_COMPLETED_WITH_FAILURE) {
disk_rw_in_progress = 0;
PS2_DISK_LED_OFF();
reset_status(shpnt) = IM_RESET_FINISHED_FAIL;
} else {
/*reset disk led counter, turn off disk led */
disk_rw_in_progress = 0;
PS2_DISK_LED_OFF();
reset_status(shpnt) = IM_RESET_FINISHED_OK;
}
stat_result(shpnt) = cmd_result;
last_scsi_command(shpnt)[ldn] = NO_SCSI;
last_scsi_type(shpnt)[ldn] = 0;
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
} else if (last_scsi_command(shpnt)[ldn] == IM_ABORT_IMM_CMD) {
/* react on SCSI abort command */
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: Interrupt from SCSI-abort.\n");
#endif
disk_rw_in_progress = 0;
PS2_DISK_LED_OFF();
cmd = ld(shpnt)[ldn].cmd;
ld(shpnt)[ldn].cmd = NULL;
if (cmd_result == IM_CMD_COMPLETED_WITH_FAILURE)
cmd->result = DID_NO_CONNECT << 16;
else
cmd->result = DID_ABORT << 16;
stat_result(shpnt) = cmd_result;
last_scsi_command(shpnt)[ldn] = NO_SCSI;
last_scsi_type(shpnt)[ldn] = 0;
if (cmd->scsi_done)
(cmd->scsi_done) (cmd); /* should be the internal_done */
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
} else {
disk_rw_in_progress = 0;
PS2_DISK_LED_OFF();
reset_status(shpnt) = IM_RESET_FINISHED_OK;
stat_result(shpnt) = cmd_result;
last_scsi_command(shpnt)[ldn] = NO_SCSI;
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
}
}
last_scsi_command(shpnt)[ldn] = NO_SCSI;
last_scsi_type(shpnt)[ldn] = 0;
cmd = ld(shpnt)[ldn].cmd;
ld(shpnt)[ldn].cmd = NULL;
#ifdef IM_DEBUG_TIMEOUT
if (cmd) {
if ((cmd->target == TIMEOUT_PUN) && (cmd->device->lun == TIMEOUT_LUN)) {
spin_unlock_irqsave(shpnt->host_lock, flags);
printk("IBM MCA SCSI: Ignoring interrupt from pun=%x, lun=%x.\n", cmd->target, cmd->device->lun);
return IRQ_HANDLED;
}
}
#endif
/*if no command structure, just return, else clear cmd */
if (!cmd)
{
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
}
#ifdef IM_DEBUG_INT
printk("cmd=%02x ireg=%02x ds=%02x cs=%02x de=%02x ce=%02x\n", cmd->cmnd[0], intr_reg, ld(shpnt)[ldn].tsb.dev_status, ld(shpnt)[ldn].tsb.cmd_status, ld(shpnt)[ldn].tsb.dev_error, ld(shpnt)[ldn].tsb.cmd_error);
#endif
/*if this is end of media read/write, may turn off PS/2 disk led */
if ((ld(shpnt)[ldn].device_type != TYPE_NO_LUN) && (ld(shpnt)[ldn].device_type != TYPE_NO_DEVICE)) {
/* only access this, if there was a valid device addressed */
if (--disk_rw_in_progress == 0)
PS2_DISK_LED_OFF();
}
/* IBM describes the status-mask to be 0x1e, but this is not conform
* with SCSI-definition, I suppose, the reason for it is that IBM
* adapters do not support CMD_TERMINATED, TASK_SET_FULL and
* ACA_ACTIVE as returning statusbyte information. (ML) */
if (cmd_result == IM_CMD_COMPLETED_WITH_FAILURE) {
cmd->result = (unsigned char) (ld(shpnt)[ldn].tsb.dev_status & 0x1e);
IBM_DS(shpnt).total_errors++;
} else
cmd->result = 0;
/* write device status into cmd->result, and call done function */
if (lastSCSI == NO_SCSI) { /* unexpected interrupt :-( */
cmd->result |= DID_BAD_INTR << 16;
printk("IBM MCA SCSI: WARNING - Interrupt from non-pending SCSI-command!\n");
} else /* things went right :-) */
cmd->result |= DID_OK << 16;
if (cmd->scsi_done)
(cmd->scsi_done) (cmd);
spin_unlock_irqrestore(shpnt->host_lock, flags);
return IRQ_HANDLED;
}
static void issue_cmd(struct Scsi_Host *shpnt, unsigned long cmd_reg,
unsigned char attn_reg)
{
unsigned long flags;
/* must wait for attention reg not busy */
while (1) {
spin_lock_irqsave(shpnt->host_lock, flags);
if (!(inb(IM_STAT_REG(shpnt)) & IM_BUSY))
break;
spin_unlock_irqrestore(shpnt->host_lock, flags);
}
/* write registers and enable system interrupts */
outl(cmd_reg, IM_CMD_REG(shpnt));
outb(attn_reg, IM_ATTN_REG(shpnt));
spin_unlock_irqrestore(shpnt->host_lock, flags);
}
static void internal_done(Scsi_Cmnd * cmd)
{
cmd->SCp.Status++;
return;
}
/* SCSI-SCB-command for device_inquiry */
static int device_inquiry(struct Scsi_Host *shpnt, int ldn)
{
int retr;
struct im_scb *scb;
struct im_tsb *tsb;
unsigned char *buf;
scb = &(ld(shpnt)[ldn].scb);
tsb = &(ld(shpnt)[ldn].tsb);
buf = (unsigned char *) (&(ld(shpnt)[ldn].buf));
ld(shpnt)[ldn].tsb.dev_status = 0; /* prepare statusblock */
for (retr = 0; retr < 3; retr++) {
/* fill scb with inquiry command */
scb->command = IM_DEVICE_INQUIRY_CMD | IM_NO_DISCONNECT;
scb->enable = IM_REPORT_TSB_ONLY_ON_ERROR | IM_READ_CONTROL | IM_SUPRESS_EXCEPTION_SHORT | IM_RETRY_ENABLE | IM_BYPASS_BUFFER;
last_scsi_command(shpnt)[ldn] = IM_DEVICE_INQUIRY_CMD;
last_scsi_type(shpnt)[ldn] = IM_SCB;
scb->sys_buf_adr = isa_virt_to_bus(buf);
scb->sys_buf_length = 255; /* maximum bufferlength gives max info */
scb->tsb_adr = isa_virt_to_bus(tsb);
/* issue scb to passed ldn, and busy wait for interrupt */
got_interrupt(shpnt) = 0;
issue_cmd(shpnt, isa_virt_to_bus(scb), IM_SCB | ldn);
while (!got_interrupt(shpnt))
barrier();
/*if command successful, break */
if ((stat_result(shpnt) == IM_SCB_CMD_COMPLETED) || (stat_result(shpnt) == IM_SCB_CMD_COMPLETED_WITH_RETRIES))
return 1;
}
/*if all three retries failed, return "no device at this ldn" */
if (retr >= 3)
return 0;
else
return 1;
}
static int read_capacity(struct Scsi_Host *shpnt, int ldn)
{
int retr;
struct im_scb *scb;
struct im_tsb *tsb;
unsigned char *buf;
scb = &(ld(shpnt)[ldn].scb);
tsb = &(ld(shpnt)[ldn].tsb);
buf = (unsigned char *) (&(ld(shpnt)[ldn].buf));
ld(shpnt)[ldn].tsb.dev_status = 0;
for (retr = 0; retr < 3; retr++) {
/*fill scb with read capacity command */
scb->command = IM_READ_CAPACITY_CMD;
scb->enable = IM_REPORT_TSB_ONLY_ON_ERROR | IM_READ_CONTROL | IM_RETRY_ENABLE | IM_BYPASS_BUFFER;
last_scsi_command(shpnt)[ldn] = IM_READ_CAPACITY_CMD;
last_scsi_type(shpnt)[ldn] = IM_SCB;
scb->sys_buf_adr = isa_virt_to_bus(buf);
scb->sys_buf_length = 8;
scb->tsb_adr = isa_virt_to_bus(tsb);
/*issue scb to passed ldn, and busy wait for interrupt */
got_interrupt(shpnt) = 0;
issue_cmd(shpnt, isa_virt_to_bus(scb), IM_SCB | ldn);
while (!got_interrupt(shpnt))
barrier();
/*if got capacity, get block length and return one device found */
if ((stat_result(shpnt) == IM_SCB_CMD_COMPLETED) || (stat_result(shpnt) == IM_SCB_CMD_COMPLETED_WITH_RETRIES))
return 1;
}
/*if all three retries failed, return "no device at this ldn" */
if (retr >= 3)
return 0;
else
return 1;
}
static int get_pos_info(struct Scsi_Host *shpnt)
{
int retr;
struct im_scb *scb;
struct im_tsb *tsb;
unsigned char *buf;
scb = &(ld(shpnt)[MAX_LOG_DEV].scb);
tsb = &(ld(shpnt)[MAX_LOG_DEV].tsb);
buf = (unsigned char *) (&(ld(shpnt)[MAX_LOG_DEV].buf));
ld(shpnt)[MAX_LOG_DEV].tsb.dev_status = 0;
for (retr = 0; retr < 3; retr++) {
/*fill scb with get_pos_info command */
scb->command = IM_GET_POS_INFO_CMD;
scb->enable = IM_READ_CONTROL | IM_REPORT_TSB_ONLY_ON_ERROR | IM_RETRY_ENABLE | IM_BYPASS_BUFFER;
last_scsi_command(shpnt)[MAX_LOG_DEV] = IM_GET_POS_INFO_CMD;
last_scsi_type(shpnt)[MAX_LOG_DEV] = IM_SCB;
scb->sys_buf_adr = isa_virt_to_bus(buf);
if (special(shpnt) == IBM_SCSI2_FW)
scb->sys_buf_length = 256; /* get all info from F/W adapter */
else
scb->sys_buf_length = 18; /* get exactly 18 bytes for other SCSI */
scb->tsb_adr = isa_virt_to_bus(tsb);
/*issue scb to ldn=15, and busy wait for interrupt */
got_interrupt(shpnt) = 0;
issue_cmd(shpnt, isa_virt_to_bus(scb), IM_SCB | MAX_LOG_DEV);
/* FIXME: timeout */
while (!got_interrupt(shpnt))
barrier();
/*if got POS-stuff, get block length and return one device found */
if ((stat_result(shpnt) == IM_SCB_CMD_COMPLETED) || (stat_result(shpnt) == IM_SCB_CMD_COMPLETED_WITH_RETRIES))
return 1;
}
/* if all three retries failed, return "no device at this ldn" */
if (retr >= 3)
return 0;
else
return 1;
}
/* SCSI-immediate-command for assign. This functions maps/unmaps specific
ldn-numbers on SCSI (PUN,LUN). It is needed for presetting of the
subsystem and for dynamical remapping od ldns. */
static int immediate_assign(struct Scsi_Host *shpnt, unsigned int pun,
unsigned int lun, unsigned int ldn,
unsigned int operation)
{
int retr;
unsigned long imm_cmd;
for (retr = 0; retr < 3; retr++) {
/* select mutation level of the SCSI-adapter */
switch (special(shpnt)) {
case IBM_SCSI2_FW:
imm_cmd = (unsigned long) (IM_ASSIGN_IMM_CMD);
imm_cmd |= (unsigned long) ((lun & 7) << 24);
imm_cmd |= (unsigned long) ((operation & 1) << 23);
imm_cmd |= (unsigned long) ((pun & 7) << 20) | ((pun & 8) << 24);
imm_cmd |= (unsigned long) ((ldn & 15) << 16);
break;
default:
imm_cmd = inl(IM_CMD_REG(shpnt));
imm_cmd &= (unsigned long) (0xF8000000); /* keep reserved bits */
imm_cmd |= (unsigned long) (IM_ASSIGN_IMM_CMD);
imm_cmd |= (unsigned long) ((lun & 7) << 24);
imm_cmd |= (unsigned long) ((operation & 1) << 23);
imm_cmd |= (unsigned long) ((pun & 7) << 20);
imm_cmd |= (unsigned long) ((ldn & 15) << 16);
break;
}
last_scsi_command(shpnt)[MAX_LOG_DEV] = IM_ASSIGN_IMM_CMD;
last_scsi_type(shpnt)[MAX_LOG_DEV] = IM_IMM_CMD;
got_interrupt(shpnt) = 0;
issue_cmd(shpnt, (unsigned long) (imm_cmd), IM_IMM_CMD | MAX_LOG_DEV);
while (!got_interrupt(shpnt))
barrier();
/*if command successful, break */
if (stat_result(shpnt) == IM_IMMEDIATE_CMD_COMPLETED)
return 1;
}
if (retr >= 3)
return 0;
else
return 1;
}
static int immediate_feature(struct Scsi_Host *shpnt, unsigned int speed, unsigned int timeout)
{
int retr;
unsigned long imm_cmd;
for (retr = 0; retr < 3; retr++) {
/* select mutation level of the SCSI-adapter */
imm_cmd = IM_FEATURE_CTR_IMM_CMD;
imm_cmd |= (unsigned long) ((speed & 0x7) << 29);
imm_cmd |= (unsigned long) ((timeout & 0x1fff) << 16);
last_scsi_command(shpnt)[MAX_LOG_DEV] = IM_FEATURE_CTR_IMM_CMD;
last_scsi_type(shpnt)[MAX_LOG_DEV] = IM_IMM_CMD;
got_interrupt(shpnt) = 0;
/* we need to run into command errors in order to probe for the
* right speed! */
global_command_error_excuse = 1;
issue_cmd(shpnt, (unsigned long) (imm_cmd), IM_IMM_CMD | MAX_LOG_DEV);
/* FIXME: timeout */
while (!got_interrupt(shpnt))
barrier();
if (global_command_error_excuse == CMD_FAIL) {
global_command_error_excuse = 0;
return 2;
} else
global_command_error_excuse = 0;
/*if command successful, break */
if (stat_result(shpnt) == IM_IMMEDIATE_CMD_COMPLETED)
return 1;
}
if (retr >= 3)
return 0;
else
return 1;
}
#ifdef CONFIG_IBMMCA_SCSI_DEV_RESET
static int immediate_reset(struct Scsi_Host *shpnt, unsigned int ldn)
{
int retries;
int ticks;
unsigned long imm_command;
for (retries = 0; retries < 3; retries++) {
imm_command = inl(IM_CMD_REG(shpnt));
imm_command &= (unsigned long) (0xFFFF0000); /* keep reserved bits */
imm_command |= (unsigned long) (IM_RESET_IMM_CMD);
last_scsi_command(shpnt)[ldn] = IM_RESET_IMM_CMD;
last_scsi_type(shpnt)[ldn] = IM_IMM_CMD;
got_interrupt(shpnt) = 0;
reset_status(shpnt) = IM_RESET_IN_PROGRESS;
issue_cmd(shpnt, (unsigned long) (imm_command), IM_IMM_CMD | ldn);
ticks = IM_RESET_DELAY * HZ;
while (reset_status(shpnt) == IM_RESET_IN_PROGRESS && --ticks) {
udelay((1 + 999 / HZ) * 1000);
barrier();
}
/* if reset did not complete, just complain */
if (!ticks) {
printk(KERN_ERR "IBM MCA SCSI: reset did not complete within %d seconds.\n", IM_RESET_DELAY);
reset_status(shpnt) = IM_RESET_FINISHED_OK;
/* did not work, finish */
return 1;
}
/*if command successful, break */
if (stat_result(shpnt) == IM_IMMEDIATE_CMD_COMPLETED)
return 1;
}
if (retries >= 3)
return 0;
else
return 1;
}
#endif
/* type-interpreter for physical device numbers */
static char *ti_p(int dev)
{
switch (dev) {
case TYPE_IBM_SCSI_ADAPTER:
return ("A");
case TYPE_DISK:
return ("D");
case TYPE_TAPE:
return ("T");
case TYPE_PROCESSOR:
return ("P");
case TYPE_WORM:
return ("W");
case TYPE_ROM:
return ("R");
case TYPE_SCANNER:
return ("S");
case TYPE_MOD:
return ("M");
case TYPE_MEDIUM_CHANGER:
return ("C");
case TYPE_NO_LUN:
return ("+"); /* show NO_LUN */
}
return ("-"); /* TYPE_NO_DEVICE and others */
}
/* interpreter for logical device numbers (ldn) */
static char *ti_l(int val)
{
const char hex[16] = "0123456789abcdef";
static char answer[2];
answer[1] = (char) (0x0);
if (val <= MAX_LOG_DEV)
answer[0] = hex[val];
else
answer[0] = '-';
return (char *) &answer;
}
/* transfers bitpattern of the feature command to values in MHz */
static char *ibmrate(unsigned int speed, int i)
{
switch (speed) {
case 0:
return i ? "5.00" : "10.00";
case 1:
return i ? "4.00" : "8.00";
case 2:
return i ? "3.33" : "6.66";
case 3:
return i ? "2.86" : "5.00";
case 4:
return i ? "2.50" : "4.00";
case 5:
return i ? "2.22" : "3.10";
case 6:
return i ? "2.00" : "2.50";
case 7:
return i ? "1.82" : "2.00";
}
return "---";
}
static int probe_display(int what)
{
static int rotator = 0;
const char rotor[] = "|/-\\";
if (!(display_mode & LED_DISP))
return 0;
if (!what) {
outl(0x20202020, MOD95_LED_PORT);
outl(0x20202020, MOD95_LED_PORT + 4);
} else {
outb('S', MOD95_LED_PORT + 7);
outb('C', MOD95_LED_PORT + 6);
outb('S', MOD95_LED_PORT + 5);
outb('I', MOD95_LED_PORT + 4);
outb('i', MOD95_LED_PORT + 3);
outb('n', MOD95_LED_PORT + 2);
outb('i', MOD95_LED_PORT + 1);
outb((char) (rotor[rotator]), MOD95_LED_PORT);
rotator++;
if (rotator > 3)
rotator = 0;
}
return 0;
}
static int probe_bus_mode(struct Scsi_Host *shpnt)
{
struct im_pos_info *info;
int num_bus = 0;
int ldn;
info = (struct im_pos_info *) (&(ld(shpnt)[MAX_LOG_DEV].buf));
if (get_pos_info(shpnt)) {
if (info->connector_size & 0xf000)
subsystem_connector_size(shpnt) = 16;
else
subsystem_connector_size(shpnt) = 32;
num_bus |= (info->pos_4b & 8) >> 3;
for (ldn = 0; ldn <= MAX_LOG_DEV; ldn++) {
if ((special(shpnt) == IBM_SCSI_WCACHE) || (special(shpnt) == IBM_7568_WCACHE)) {
if (!((info->cache_stat >> ldn) & 1))
ld(shpnt)[ldn].cache_flag = 0;
}
if (!((info->retry_stat >> ldn) & 1))
ld(shpnt)[ldn].retry_flag = 0;
}
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: SCSI-Cache bits: ");
for (ldn = 0; ldn <= MAX_LOG_DEV; ldn++) {
printk("%d", ld(shpnt)[ldn].cache_flag);
}
printk("\nIBM MCA SCSI: SCSI-Retry bits: ");
for (ldn = 0; ldn <= MAX_LOG_DEV; ldn++) {
printk("%d", ld(shpnt)[ldn].retry_flag);
}
printk("\n");
#endif
}
return num_bus;
}
/* probing scsi devices */
static void check_devices(struct Scsi_Host *shpnt, int adaptertype)
{
int id, lun, ldn, ticks;
int count_devices; /* local counter for connected device */
int max_pun;
int num_bus;
int speedrun; /* local adapter_speed check variable */
/* assign default values to certain variables */
ticks = 0;
count_devices = 0;
IBM_DS(shpnt).dyn_flag = 0; /* normally no need for dynamical ldn management */
IBM_DS(shpnt).total_errors = 0; /* set errorcounter to 0 */
next_ldn(shpnt) = 7; /* next ldn to be assigned is 7, because 0-6 is 'hardwired' */
/* initialize the very important driver-informational arrays/structs */
memset(ld(shpnt), 0, sizeof(ld(shpnt)));
for (ldn = 0; ldn <= MAX_LOG_DEV; ldn++) {
last_scsi_command(shpnt)[ldn] = NO_SCSI; /* emptify last SCSI-command storage */
last_scsi_type(shpnt)[ldn] = 0;
ld(shpnt)[ldn].cache_flag = 1;
ld(shpnt)[ldn].retry_flag = 1;
}
memset(get_ldn(shpnt), TYPE_NO_DEVICE, sizeof(get_ldn(shpnt))); /* this is essential ! */
memset(get_scsi(shpnt), TYPE_NO_DEVICE, sizeof(get_scsi(shpnt))); /* this is essential ! */
for (lun = 0; lun < 8; lun++) {
/* mark the adapter at its pun on all luns */
get_scsi(shpnt)[subsystem_pun(shpnt)][lun] = TYPE_IBM_SCSI_ADAPTER;
get_ldn(shpnt)[subsystem_pun(shpnt)][lun] = MAX_LOG_DEV; /* make sure, the subsystem
ldn is active for all
luns. */
}
probe_display(0); /* Supercool display usage during SCSI-probing. */
/* This makes sense, when booting without any */
/* monitor connected on model XX95. */
/* STEP 1: */
adapter_speed(shpnt) = global_adapter_speed;
speedrun = adapter_speed(shpnt);
while (immediate_feature(shpnt, speedrun, adapter_timeout) == 2) {
probe_display(1);
if (speedrun == 7)
panic("IBM MCA SCSI: Cannot set Synchronous-Transfer-Rate!\n");
speedrun++;
if (speedrun > 7)
speedrun = 7;
}
adapter_speed(shpnt) = speedrun;
/* Get detailed information about the current adapter, necessary for
* device operations: */
num_bus = probe_bus_mode(shpnt);
/* num_bus contains only valid data for the F/W adapter! */
if (adaptertype == IBM_SCSI2_FW) { /* F/W SCSI adapter: */
/* F/W adapter PUN-space extension evaluation: */
if (num_bus) {
printk(KERN_INFO "IBM MCA SCSI: Separate bus mode (wide-addressing enabled)\n");
subsystem_maxid(shpnt) = 16;
} else {
printk(KERN_INFO "IBM MCA SCSI: Combined bus mode (wide-addressing disabled)\n");
subsystem_maxid(shpnt) = 8;
}
printk(KERN_INFO "IBM MCA SCSI: Sync.-Rate (F/W: 20, Int.: 10, Ext.: %s) MBytes/s\n", ibmrate(speedrun, adaptertype));
} else /* all other IBM SCSI adapters: */
printk(KERN_INFO "IBM MCA SCSI: Synchronous-SCSI-Transfer-Rate: %s MBytes/s\n", ibmrate(speedrun, adaptertype));
/* assign correct PUN device space */
max_pun = subsystem_maxid(shpnt);
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: Current SCSI-host index: %d\n", shpnt);
printk("IBM MCA SCSI: Removing default logical SCSI-device mapping.");
#else
printk(KERN_INFO "IBM MCA SCSI: Dev. Order: %s, Mapping (takes <2min): ", (ibm_ansi_order) ? "ANSI" : "New");
#endif
for (ldn = 0; ldn < MAX_LOG_DEV; ldn++) {
probe_display(1);
#ifdef IM_DEBUG_PROBE
printk(".");
#endif
immediate_assign(shpnt, 0, 0, ldn, REMOVE_LDN); /* remove ldn (wherever) */
}
lun = 0; /* default lun is 0 */
#ifndef IM_DEBUG_PROBE
printk("cleared,");
#endif
/* STEP 2: */
#ifdef IM_DEBUG_PROBE
printk("\nIBM MCA SCSI: Scanning SCSI-devices.");
#endif
for (id = 0; id < max_pun; id++)
#ifdef CONFIG_SCSI_MULTI_LUN
for (lun = 0; lun < 8; lun++)
#endif
{
probe_display(1);
#ifdef IM_DEBUG_PROBE
printk(".");
#endif
if (id != subsystem_pun(shpnt)) {
/* if pun is not the adapter: */
/* set ldn=0 to pun,lun */
immediate_assign(shpnt, id, lun, PROBE_LDN, SET_LDN);
if (device_inquiry(shpnt, PROBE_LDN)) { /* probe device */
get_scsi(shpnt)[id][lun] = (unsigned char) (ld(shpnt)[PROBE_LDN].buf[0]);
/* entry, even for NO_LUN */
if (ld(shpnt)[PROBE_LDN].buf[0] != TYPE_NO_LUN)
count_devices++; /* a existing device is found */
}
/* remove ldn */
immediate_assign(shpnt, id, lun, PROBE_LDN, REMOVE_LDN);
}
}
#ifndef IM_DEBUG_PROBE
printk("scanned,");
#endif
/* STEP 3: */
#ifdef IM_DEBUG_PROBE
printk("\nIBM MCA SCSI: Mapping SCSI-devices.");
#endif
ldn = 0;
lun = 0;
#ifdef CONFIG_SCSI_MULTI_LUN
for (lun = 0; lun < 8 && ldn < MAX_LOG_DEV; lun++)
#endif
for (id = 0; id < max_pun && ldn < MAX_LOG_DEV; id++) {
probe_display(1);
#ifdef IM_DEBUG_PROBE
printk(".");
#endif
if (id != subsystem_pun(shpnt)) {
if (get_scsi(shpnt)[id][lun] != TYPE_NO_LUN && get_scsi(shpnt)[id][lun] != TYPE_NO_DEVICE) {
/* Only map if accepted type. Always enter for
lun == 0 to get no gaps into ldn-mapping for ldn<7. */
immediate_assign(shpnt, id, lun, ldn, SET_LDN);
get_ldn(shpnt)[id][lun] = ldn; /* map ldn */
if (device_exists(shpnt, ldn, &ld(shpnt)[ldn].block_length, &ld(shpnt)[ldn].device_type)) {
#ifdef CONFIG_IBMMCA_SCSI_DEV_RESET
printk("resetting device at ldn=%x ... ", ldn);
immediate_reset(shpnt, ldn);
#endif
ldn++;
} else {
/* device vanished, probably because we don't know how to
* handle it or because it has problems */
if (lun > 0) {
/* remove mapping */
get_ldn(shpnt)[id][lun] = TYPE_NO_DEVICE;
immediate_assign(shpnt, 0, 0, ldn, REMOVE_LDN);
} else
ldn++;
}
} else if (lun == 0) {
/* map lun == 0, even if no device exists */
immediate_assign(shpnt, id, lun, ldn, SET_LDN);
get_ldn(shpnt)[id][lun] = ldn; /* map ldn */
ldn++;
}
}
}
/* STEP 4: */
/* map remaining ldns to non-existing devices */
for (lun = 1; lun < 8 && ldn < MAX_LOG_DEV; lun++)
for (id = 0; id < max_pun && ldn < MAX_LOG_DEV; id++) {
if (get_scsi(shpnt)[id][lun] == TYPE_NO_LUN || get_scsi(shpnt)[id][lun] == TYPE_NO_DEVICE) {
probe_display(1);
/* Map remaining ldns only to NON-existing pun,lun
combinations to make sure an inquiry will fail.
For MULTI_LUN, it is needed to avoid adapter autonome
SCSI-remapping. */
immediate_assign(shpnt, id, lun, ldn, SET_LDN);
get_ldn(shpnt)[id][lun] = ldn;
ldn++;
}
}
#ifndef IM_DEBUG_PROBE
printk("mapped.");
#endif
printk("\n");
#ifdef IM_DEBUG_PROBE
if (ibm_ansi_order)
printk("IBM MCA SCSI: Device order: IBM/ANSI (pun=7 is first).\n");
else
printk("IBM MCA SCSI: Device order: New Industry Standard (pun=0 is first).\n");
#endif
#ifdef IM_DEBUG_PROBE
/* Show the physical and logical mapping during boot. */
printk("IBM MCA SCSI: Determined SCSI-device-mapping:\n");
printk(" Physical SCSI-Device Map Logical SCSI-Device Map\n");
printk("ID\\LUN 0 1 2 3 4 5 6 7 ID\\LUN 0 1 2 3 4 5 6 7\n");
for (id = 0; id < max_pun; id++) {
printk("%2d ", id);
for (lun = 0; lun < 8; lun++)
printk("%2s ", ti_p(get_scsi(shpnt)[id][lun]));
printk(" %2d ", id);
for (lun = 0; lun < 8; lun++)
printk("%2s ", ti_l(get_ldn(shpnt)[id][lun]));
printk("\n");
}
#endif
/* assign total number of found SCSI-devices to the statistics struct */
IBM_DS(shpnt).total_scsi_devices = count_devices;
/* decide for output in /proc-filesystem, if the configuration of
SCSI-devices makes dynamical reassignment of devices necessary */
if (count_devices >= MAX_LOG_DEV)
IBM_DS(shpnt).dyn_flag = 1; /* dynamical assignment is necessary */
else
IBM_DS(shpnt).dyn_flag = 0; /* dynamical assignment is not necessary */
/* If no SCSI-devices are assigned, return 1 in order to cause message. */
if (ldn == 0)
printk("IBM MCA SCSI: Warning: No SCSI-devices found/assigned!\n");
/* reset the counters for statistics on the current adapter */
IBM_DS(shpnt).scbs = 0;
IBM_DS(shpnt).long_scbs = 0;
IBM_DS(shpnt).total_accesses = 0;
IBM_DS(shpnt).total_interrupts = 0;
IBM_DS(shpnt).dynamical_assignments = 0;
memset(IBM_DS(shpnt).ldn_access, 0x0, sizeof(IBM_DS(shpnt).ldn_access));
memset(IBM_DS(shpnt).ldn_read_access, 0x0, sizeof(IBM_DS(shpnt).ldn_read_access));
memset(IBM_DS(shpnt).ldn_write_access, 0x0, sizeof(IBM_DS(shpnt).ldn_write_access));
memset(IBM_DS(shpnt).ldn_inquiry_access, 0x0, sizeof(IBM_DS(shpnt).ldn_inquiry_access));
memset(IBM_DS(shpnt).ldn_modeselect_access, 0x0, sizeof(IBM_DS(shpnt).ldn_modeselect_access));
memset(IBM_DS(shpnt).ldn_assignments, 0x0, sizeof(IBM_DS(shpnt).ldn_assignments));
probe_display(0);
return;
}
static int device_exists(struct Scsi_Host *shpnt, int ldn, int *block_length, int *device_type)
{
unsigned char *buf;
/* if no valid device found, return immediately with 0 */
if (!(device_inquiry(shpnt, ldn)))
return 0;
buf = (unsigned char *) (&(ld(shpnt)[ldn].buf));
if (*buf == TYPE_ROM) {
*device_type = TYPE_ROM;
*block_length = 2048; /* (standard blocksize for yellow-/red-book) */
return 1;
}
if (*buf == TYPE_WORM) {
*device_type = TYPE_WORM;
*block_length = 2048;
return 1;
}
if (*buf == TYPE_DISK) {
*device_type = TYPE_DISK;
if (read_capacity(shpnt, ldn)) {
*block_length = *(buf + 7) + (*(buf + 6) << 8) + (*(buf + 5) << 16) + (*(buf + 4) << 24);
return 1;
} else
return 0;
}
if (*buf == TYPE_MOD) {
*device_type = TYPE_MOD;
if (read_capacity(shpnt, ldn)) {
*block_length = *(buf + 7) + (*(buf + 6) << 8) + (*(buf + 5) << 16) + (*(buf + 4) << 24);
return 1;
} else
return 0;
}
if (*buf == TYPE_TAPE) {
*device_type = TYPE_TAPE;
*block_length = 0; /* not in use (setting by mt and mtst in op.) */
return 1;
}
if (*buf == TYPE_PROCESSOR) {
*device_type = TYPE_PROCESSOR;
*block_length = 0; /* they set their stuff on drivers */
return 1;
}
if (*buf == TYPE_SCANNER) {
*device_type = TYPE_SCANNER;
*block_length = 0; /* they set their stuff on drivers */
return 1;
}
if (*buf == TYPE_MEDIUM_CHANGER) {
*device_type = TYPE_MEDIUM_CHANGER;
*block_length = 0; /* One never knows, what to expect on a medium
changer device. */
return 1;
}
return 0;
}
static void internal_ibmmca_scsi_setup(char *str, int *ints)
{
int i, j, io_base, id_base;
char *token;
io_base = 0;
id_base = 0;
if (str) {
j = 0;
while ((token = strsep(&str, ",")) != NULL) {
if (!strcmp(token, "activity"))
display_mode |= LED_ACTIVITY;
if (!strcmp(token, "display"))
display_mode |= LED_DISP;
if (!strcmp(token, "adisplay"))
display_mode |= LED_ADISP;
if (!strcmp(token, "normal"))
ibm_ansi_order = 0;
if (!strcmp(token, "ansi"))
ibm_ansi_order = 1;
if (!strcmp(token, "fast"))
global_adapter_speed = 0;
if (!strcmp(token, "medium"))
global_adapter_speed = 4;
if (!strcmp(token, "slow"))
global_adapter_speed = 7;
if ((*token == '-') || (isdigit(*token))) {
if (!(j % 2) && (io_base < IM_MAX_HOSTS))
io_port[io_base++] = simple_strtoul(token, NULL, 0);
if ((j % 2) && (id_base < IM_MAX_HOSTS))
scsi_id[id_base++] = simple_strtoul(token, NULL, 0);
j++;
}
}
} else if (ints) {
for (i = 0; i < IM_MAX_HOSTS && 2 * i + 2 < ints[0]; i++) {
io_port[i] = ints[2 * i + 2];
scsi_id[i] = ints[2 * i + 2];
}
}
return;
}
#if 0
FIXME NEED TO MOVE TO SYSFS
static int ibmmca_getinfo(char *buf, int slot, void *dev_id)
{
struct Scsi_Host *shpnt;
int len, speciale, connectore, k;
unsigned int pos[8];
unsigned long flags;
struct Scsi_Host *dev = dev_id;
spin_lock_irqsave(dev->host_lock, flags);
shpnt = dev; /* assign host-structure to local pointer */
len = 0; /* set filled text-buffer index to 0 */
/* get the _special contents of the hostdata structure */
speciale = ((struct ibmmca_hostdata *) shpnt->hostdata)->_special;
connectore = ((struct ibmmca_hostdata *) shpnt->hostdata)->_connector_size;
for (k = 2; k < 4; k++)
pos[k] = ((struct ibmmca_hostdata *) shpnt->hostdata)->_pos[k];
if (speciale == FORCED_DETECTION) { /* forced detection */
len += sprintf(buf + len,
"Adapter category: forced detected\n" "***************************************\n" "*** Forced detected SCSI Adapter ***\n" "*** No chip-information available ***\n" "***************************************\n");
} else if (speciale == INTEGRATED_SCSI) {
/* if the integrated subsystem has been found automatically: */
len += sprintf(buf + len,
"Adapter category: integrated\n" "Chip revision level: %d\n" "Chip status: %s\n" "8 kByte NVRAM status: %s\n", ((pos[2] & 0xf0) >> 4), (pos[2] & 1) ? "enabled" : "disabled", (pos[2] & 2) ? "locked" : "accessible");
} else if ((speciale >= 0) && (speciale < ARRAY_SIZE(subsys_list))) {
/* if the subsystem is a slot adapter */
len += sprintf(buf + len, "Adapter category: slot-card\n" "ROM Segment Address: ");
if ((pos[2] & 0xf0) == 0xf0)
len += sprintf(buf + len, "off\n");
else
len += sprintf(buf + len, "0x%x\n", ((pos[2] & 0xf0) << 13) + 0xc0000);
len += sprintf(buf + len, "Chip status: %s\n", (pos[2] & 1) ? "enabled" : "disabled");
len += sprintf(buf + len, "Adapter I/O Offset: 0x%x\n", ((pos[2] & 0x0e) << 2));
} else {
len += sprintf(buf + len, "Adapter category: unknown\n");
}
/* common subsystem information to write to the slotn file */
len += sprintf(buf + len, "Subsystem PUN: %d\n", shpnt->this_id);
len += sprintf(buf + len, "I/O base address range: 0x%x-0x%x\n", (unsigned int) (shpnt->io_port), (unsigned int) (shpnt->io_port + 7));
len += sprintf(buf + len, "MCA-slot size: %d bits", connectore);
/* Now make sure, the bufferlength is devidable by 4 to avoid
* paging problems of the buffer. */
while (len % sizeof(int) != (sizeof(int) - 1))
len += sprintf(buf + len, " ");
len += sprintf(buf + len, "\n");
spin_unlock_irqrestore(shpnt->host_lock, flags);
return len;
}
#endif
static struct scsi_host_template ibmmca_driver_template = {
.proc_name = "ibmmca",
.proc_info = ibmmca_proc_info,
.name = "IBM SCSI-Subsystem",
.queuecommand = ibmmca_queuecommand,
.eh_abort_handler = ibmmca_abort,
.eh_host_reset_handler = ibmmca_host_reset,
.bios_param = ibmmca_biosparam,
.can_queue = 16,
.this_id = 7,
.sg_tablesize = 16,
.cmd_per_lun = 1,
.use_clustering = ENABLE_CLUSTERING,
};
static int ibmmca_probe(struct device *dev)
{
struct Scsi_Host *shpnt;
int port, id, i, j, k, irq, enabled, ret = -EINVAL;
struct mca_device *mca_dev = to_mca_device(dev);
const char *description = ibmmca_description[mca_dev->index];
/* First of all, print the version number of the driver. This is
* important to allow better user bugreports in case of already
* having problems with the MCA_bus probing. */
printk(KERN_INFO "IBM MCA SCSI: Version %s\n", IBMMCA_SCSI_DRIVER_VERSION);
/* The POS2-register of all PS/2 model SCSI-subsystems has the following
* interpretation of bits:
* Bit 7 - 4 : Chip Revision ID (Release)
* Bit 3 - 2 : Reserved
* Bit 1 : 8k NVRAM Disabled
* Bit 0 : Chip Enable (EN-Signal)
* The POS3-register is interpreted as follows:
* Bit 7 - 5 : SCSI ID
* Bit 4 : Reserved = 0
* Bit 3 - 0 : Reserved = 0
* (taken from "IBM, PS/2 Hardware Interface Technical Reference, Common
* Interfaces (1991)").
* In short words, this means, that IBM PS/2 machines only support
* 1 single subsystem by default. The slot-adapters must have another
* configuration on pos2. Here, one has to assume the following
* things for POS2-register:
* Bit 7 - 4 : Chip Revision ID (Release)
* Bit 3 - 1 : port offset factor
* Bit 0 : Chip Enable (EN-Signal)
* As I found a patch here, setting the IO-registers to 0x3540 forced,
* as there was a 0x05 in POS2 on a model 56, I assume, that the
* port 0x3540 must be fix for integrated SCSI-controllers.
* Ok, this discovery leads to the following implementation: (M.Lang) */
/* first look for the IBM SCSI integrated subsystem on the motherboard */
for (j = 0; j < 8; j++) /* read the pos-information */
pos[j] = mca_device_read_pos(mca_dev, j);
id = (pos[3] & 0xe0) >> 5; /* this is correct and represents the PUN */
enabled = (pos[2] &0x01);
if (!enabled) {
printk(KERN_WARNING "IBM MCA SCSI: WARNING - Your SCSI-subsystem is disabled!\n");
printk(KERN_WARNING " SCSI-operations may not work.\n");
}
/* pos2 = pos3 = 0xff if there is no integrated SCSI-subsystem present, but
* if we ignore the settings of all surrounding pos registers, it is not
* completely sufficient to only check pos2 and pos3. */
/* Therefore, now the following if statement is used to
* make sure, we see a real integrated onboard SCSI-interface and no
* internal system information, which gets mapped to some pos registers
* on models 95xx. */
if (mca_dev->slot == MCA_INTEGSCSI &&
((!pos[0] && !pos[1] && pos[2] > 0 &&
pos[3] > 0 && !pos[4] && !pos[5] &&
!pos[6] && !pos[7]) ||
(pos[0] == 0xff && pos[1] == 0xff &&
pos[2] < 0xff && pos[3] < 0xff &&
pos[4] == 0xff && pos[5] == 0xff &&
pos[6] == 0xff && pos[7] == 0xff))) {
irq = IM_IRQ;
port = IM_IO_PORT;
} else {
irq = IM_IRQ;
port = IM_IO_PORT + ((pos[2] &0x0e) << 2);
if ((mca_dev->index == IBM_SCSI2_FW) && (pos[6] != 0)) {
printk(KERN_ERR "IBM MCA SCSI: ERROR - Wrong POS(6)-register setting!\n");
printk(KERN_ERR " Impossible to determine adapter PUN!\n");
printk(KERN_ERR " Guessing adapter PUN = 7.\n");
id = 7;
} else {
id = (pos[3] & 0xe0) >> 5; /* get subsystem PUN */
if (mca_dev->index == IBM_SCSI2_FW) {
id |= (pos[3] & 0x10) >> 1; /* get subsystem PUN high-bit
* for F/W adapters */
}
}
if ((mca_dev->index == IBM_SCSI2_FW) &&
(pos[4] & 0x01) && (pos[6] == 0)) {
/* IRQ11 is used by SCSI-2 F/W Adapter/A */
printk(KERN_DEBUG "IBM MCA SCSI: SCSI-2 F/W adapter needs IRQ 11.\n");
irq = IM_IRQ_FW;
}
}
/* give detailed information on the subsystem. This helps me
* additionally during debugging and analyzing bug-reports. */
printk(KERN_INFO "IBM MCA SCSI: %s found, io=0x%x, scsi id=%d,\n",
description, port, id);
if (mca_dev->slot == MCA_INTEGSCSI)
printk(KERN_INFO " chip rev.=%d, 8K NVRAM=%s, subsystem=%s\n", ((pos[2] & 0xf0) >> 4), (pos[2] & 2) ? "locked" : "accessible", (pos[2] & 1) ? "enabled." : "disabled.");
else {
if ((pos[2] & 0xf0) == 0xf0)
printk(KERN_DEBUG " ROM Addr.=off,");
else
printk(KERN_DEBUG " ROM Addr.=0x%x,", ((pos[2] & 0xf0) << 13) + 0xc0000);
printk(KERN_DEBUG " port-offset=0x%x, subsystem=%s\n", ((pos[2] & 0x0e) << 2), (pos[2] & 1) ? "enabled." : "disabled.");
}
/* check I/O region */
if (!request_region(port, IM_N_IO_PORT, description)) {
printk(KERN_ERR "IBM MCA SCSI: Unable to get I/O region 0x%x-0x%x (%d ports).\n", port, port + IM_N_IO_PORT - 1, IM_N_IO_PORT);
goto out_fail;
}
/* register host */
shpnt = scsi_host_alloc(&ibmmca_driver_template,
sizeof(struct ibmmca_hostdata));
if (!shpnt) {
printk(KERN_ERR "IBM MCA SCSI: Unable to register host.\n");
goto out_release;
}
dev_set_drvdata(dev, shpnt);
if(request_irq(irq, interrupt_handler, IRQF_SHARED, description, dev)) {
printk(KERN_ERR "IBM MCA SCSI: failed to request interrupt %d\n", irq);
goto out_free_host;
}
/* request I/O region */
special(shpnt) = mca_dev->index; /* important assignment or else crash! */
subsystem_connector_size(shpnt) = 0; /* preset slot-size */
shpnt->irq = irq; /* assign necessary stuff for the adapter */
shpnt->io_port = port;
shpnt->n_io_port = IM_N_IO_PORT;
shpnt->this_id = id;
shpnt->max_id = 8; /* 8 PUNs are default */
/* now, the SCSI-subsystem is connected to Linux */
#ifdef IM_DEBUG_PROBE
ctrl = (unsigned int) (inb(IM_CTR_REG(found))); /* get control-register status */
printk("IBM MCA SCSI: Control Register contents: %x, status: %x\n", ctrl, inb(IM_STAT_REG(found)));
printk("IBM MCA SCSI: This adapters' POS-registers: ");
for (i = 0; i < 8; i++)
printk("%x ", pos[i]);
printk("\n");
#endif
reset_status(shpnt) = IM_RESET_NOT_IN_PROGRESS;
for (i = 0; i < 16; i++) /* reset the tables */
for (j = 0; j < 8; j++)
get_ldn(shpnt)[i][j] = MAX_LOG_DEV;
/* check which logical devices exist */
/* after this line, local interrupting is possible: */
local_checking_phase_flag(shpnt) = 1;
check_devices(shpnt, mca_dev->index); /* call by value, using the global variable hosts */
local_checking_phase_flag(shpnt) = 0;
/* an ibm mca subsystem has been detected */
for (k = 2; k < 7; k++)
((struct ibmmca_hostdata *) shpnt->hostdata)->_pos[k] = pos[k];
((struct ibmmca_hostdata *) shpnt->hostdata)->_special = INTEGRATED_SCSI;
mca_device_set_name(mca_dev, description);
/* FIXME: NEED TO REPLUMB TO SYSFS
mca_set_adapter_procfn(MCA_INTEGSCSI, (MCA_ProcFn) ibmmca_getinfo, shpnt);
*/
mca_device_set_claim(mca_dev, 1);
if (scsi_add_host(shpnt, dev)) {
dev_printk(KERN_ERR, dev, "IBM MCA SCSI: scsi_add_host failed\n");
goto out_free_host;
}
scsi_scan_host(shpnt);
return 0;
out_free_host:
scsi_host_put(shpnt);
out_release:
release_region(port, IM_N_IO_PORT);
out_fail:
return ret;
}
static int __devexit ibmmca_remove(struct device *dev)
{
struct Scsi_Host *shpnt = dev_get_drvdata(dev);
scsi_remove_host(shpnt);
release_region(shpnt->io_port, shpnt->n_io_port);
free_irq(shpnt->irq, dev);
scsi_host_put(shpnt);
return 0;
}
/* The following routine is the SCSI command queue for the midlevel driver */
static int ibmmca_queuecommand_lck(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *))
{
unsigned int ldn;
unsigned int scsi_cmd;
struct im_scb *scb;
struct Scsi_Host *shpnt;
int current_ldn;
int id, lun;
int target;
int max_pun;
int i;
struct scatterlist *sg;
shpnt = cmd->device->host;
max_pun = subsystem_maxid(shpnt);
if (ibm_ansi_order) {
target = max_pun - 1 - cmd->device->id;
if ((target <= subsystem_pun(shpnt)) && (cmd->device->id <= subsystem_pun(shpnt)))
target--;
else if ((target >= subsystem_pun(shpnt)) && (cmd->device->id >= subsystem_pun(shpnt)))
target++;
} else
target = cmd->device->id;
/* if (target,lun) is NO LUN or not existing at all, return error */
if ((get_scsi(shpnt)[target][cmd->device->lun] == TYPE_NO_LUN) || (get_scsi(shpnt)[target][cmd->device->lun] == TYPE_NO_DEVICE)) {
cmd->result = DID_NO_CONNECT << 16;
if (done)
done(cmd);
return 0;
}
/*if (target,lun) unassigned, do further checks... */
ldn = get_ldn(shpnt)[target][cmd->device->lun];
if (ldn >= MAX_LOG_DEV) { /* on invalid ldn do special stuff */
if (ldn > MAX_LOG_DEV) { /* dynamical remapping if ldn unassigned */
current_ldn = next_ldn(shpnt); /* stop-value for one circle */
while (ld(shpnt)[next_ldn(shpnt)].cmd) { /* search for a occupied, but not in */
/* command-processing ldn. */
next_ldn(shpnt)++;
if (next_ldn(shpnt) >= MAX_LOG_DEV)
next_ldn(shpnt) = 7;
if (current_ldn == next_ldn(shpnt)) { /* One circle done ? */
/* no non-processing ldn found */
scmd_printk(KERN_WARNING, cmd,
"IBM MCA SCSI: Cannot assign SCSI-device dynamically!\n"
" On ldn 7-14 SCSI-commands everywhere in progress.\n"
" Reporting DID_NO_CONNECT for device.\n");
cmd->result = DID_NO_CONNECT << 16; /* return no connect */
if (done)
done(cmd);
return 0;
}
}
/* unmap non-processing ldn */
for (id = 0; id < max_pun; id++)
for (lun = 0; lun < 8; lun++) {
if (get_ldn(shpnt)[id][lun] == next_ldn(shpnt)) {
get_ldn(shpnt)[id][lun] = TYPE_NO_DEVICE;
get_scsi(shpnt)[id][lun] = TYPE_NO_DEVICE;
/* unmap entry */
}
}
/* set reduced interrupt_handler-mode for checking */
local_checking_phase_flag(shpnt) = 1;
/* map found ldn to pun,lun */
get_ldn(shpnt)[target][cmd->device->lun] = next_ldn(shpnt);
/* change ldn to the right value, that is now next_ldn */
ldn = next_ldn(shpnt);
/* unassign all ldns (pun,lun,ldn does not matter for remove) */
immediate_assign(shpnt, 0, 0, 0, REMOVE_LDN);
/* set only LDN for remapped device */
immediate_assign(shpnt, target, cmd->device->lun, ldn, SET_LDN);
/* get device information for ld[ldn] */
if (device_exists(shpnt, ldn, &ld(shpnt)[ldn].block_length, &ld(shpnt)[ldn].device_type)) {
ld(shpnt)[ldn].cmd = NULL; /* To prevent panic set 0, because
devices that were not assigned,
should have nothing in progress. */
get_scsi(shpnt)[target][cmd->device->lun] = ld(shpnt)[ldn].device_type;
/* increase assignment counters for statistics in /proc */
IBM_DS(shpnt).dynamical_assignments++;
IBM_DS(shpnt).ldn_assignments[ldn]++;
} else
/* panic here, because a device, found at boottime has
vanished */
panic("IBM MCA SCSI: ldn=0x%x, SCSI-device on (%d,%d) vanished!\n", ldn, target, cmd->device->lun);
/* unassign again all ldns (pun,lun,ldn does not matter for remove) */
immediate_assign(shpnt, 0, 0, 0, REMOVE_LDN);
/* remap all ldns, as written in the pun/lun table */
lun = 0;
#ifdef CONFIG_SCSI_MULTI_LUN
for (lun = 0; lun < 8; lun++)
#endif
for (id = 0; id < max_pun; id++) {
if (get_ldn(shpnt)[id][lun] <= MAX_LOG_DEV)
immediate_assign(shpnt, id, lun, get_ldn(shpnt)[id][lun], SET_LDN);
}
/* set back to normal interrupt_handling */
local_checking_phase_flag(shpnt) = 0;
#ifdef IM_DEBUG_PROBE
/* Information on syslog terminal */
printk("IBM MCA SCSI: ldn=0x%x dynamically reassigned to (%d,%d).\n", ldn, target, cmd->device->lun);
#endif
/* increase next_ldn for next dynamical assignment */
next_ldn(shpnt)++;
if (next_ldn(shpnt) >= MAX_LOG_DEV)
next_ldn(shpnt) = 7;
} else { /* wall against Linux accesses to the subsystem adapter */
cmd->result = DID_BAD_TARGET << 16;
if (done)
done(cmd);
return 0;
}
}
/*verify there is no command already in progress for this log dev */
if (ld(shpnt)[ldn].cmd)
panic("IBM MCA SCSI: cmd already in progress for this ldn.\n");
/*save done in cmd, and save cmd for the interrupt handler */
cmd->scsi_done = done;
ld(shpnt)[ldn].cmd = cmd;
/*fill scb information independent of the scsi command */
scb = &(ld(shpnt)[ldn].scb);
ld(shpnt)[ldn].tsb.dev_status = 0;
scb->enable = IM_REPORT_TSB_ONLY_ON_ERROR | IM_RETRY_ENABLE;
scb->tsb_adr = isa_virt_to_bus(&(ld(shpnt)[ldn].tsb));
scsi_cmd = cmd->cmnd[0];
if (scsi_sg_count(cmd)) {
BUG_ON(scsi_sg_count(cmd) > 16);
scsi_for_each_sg(cmd, sg, scsi_sg_count(cmd), i) {
ld(shpnt)[ldn].sge[i].address = (void *) (isa_page_to_bus(sg_page(sg)) + sg->offset);
ld(shpnt)[ldn].sge[i].byte_length = sg->length;
}
scb->enable |= IM_POINTER_TO_LIST;
scb->sys_buf_adr = isa_virt_to_bus(&(ld(shpnt)[ldn].sge[0]));
scb->sys_buf_length = scsi_sg_count(cmd) * sizeof(struct im_sge);
} else {
scb->sys_buf_adr = isa_virt_to_bus(scsi_sglist(cmd));
/* recent Linux midlevel SCSI places 1024 byte for inquiry
* command. Far too much for old PS/2 hardware. */
switch (scsi_cmd) {
/* avoid command errors by setting bufferlengths to
* ANSI-standard. Beware of forcing it to 255,
* this could SEGV the kernel!!! */
case INQUIRY:
case REQUEST_SENSE:
case MODE_SENSE:
case MODE_SELECT:
if (scsi_bufflen(cmd) > 255)
scb->sys_buf_length = 255;
else
scb->sys_buf_length = scsi_bufflen(cmd);
break;
case TEST_UNIT_READY:
scb->sys_buf_length = 0;
break;
default:
scb->sys_buf_length = scsi_bufflen(cmd);
break;
}
}
/*fill scb information dependent on scsi command */
#ifdef IM_DEBUG_CMD
printk("issue scsi cmd=%02x to ldn=%d\n", scsi_cmd, ldn);
#endif
/* for specific device-type debugging: */
#ifdef IM_DEBUG_CMD_SPEC_DEV
if (ld(shpnt)[ldn].device_type == IM_DEBUG_CMD_DEVICE)
printk("(SCSI-device-type=0x%x) issue scsi cmd=%02x to ldn=%d\n", ld(shpnt)[ldn].device_type, scsi_cmd, ldn);
#endif
/* for possible panics store current command */
last_scsi_command(shpnt)[ldn] = scsi_cmd;
last_scsi_type(shpnt)[ldn] = IM_SCB;
/* update statistical info */
IBM_DS(shpnt).total_accesses++;
IBM_DS(shpnt).ldn_access[ldn]++;
switch (scsi_cmd) {
case READ_6:
case WRITE_6:
case READ_10:
case WRITE_10:
case READ_12:
case WRITE_12:
/* Distinguish between disk and other devices. Only disks (that are the
most frequently accessed devices) should be supported by the
IBM-SCSI-Subsystem commands. */
switch (ld(shpnt)[ldn].device_type) {
case TYPE_DISK: /* for harddisks enter here ... */
case TYPE_MOD: /* ... try it also for MO-drives (send flames as */
/* you like, if this won't work.) */
if (scsi_cmd == READ_6 || scsi_cmd == READ_10 || scsi_cmd == READ_12) {
/* read command preparations */
scb->enable |= IM_READ_CONTROL;
IBM_DS(shpnt).ldn_read_access[ldn]++; /* increase READ-access on ldn stat. */
scb->command = IM_READ_DATA_CMD | IM_NO_DISCONNECT;
} else { /* write command preparations */
IBM_DS(shpnt).ldn_write_access[ldn]++; /* increase write-count on ldn stat. */
scb->command = IM_WRITE_DATA_CMD | IM_NO_DISCONNECT;
}
if (scsi_cmd == READ_6 || scsi_cmd == WRITE_6) {
scb->u1.log_blk_adr = (((unsigned) cmd->cmnd[3]) << 0) | (((unsigned) cmd->cmnd[2]) << 8) | ((((unsigned) cmd->cmnd[1]) & 0x1f) << 16);
scb->u2.blk.count = (unsigned) cmd->cmnd[4];
} else {
scb->u1.log_blk_adr = (((unsigned) cmd->cmnd[5]) << 0) | (((unsigned) cmd->cmnd[4]) << 8) | (((unsigned) cmd->cmnd[3]) << 16) | (((unsigned) cmd->cmnd[2]) << 24);
scb->u2.blk.count = (((unsigned) cmd->cmnd[8]) << 0) | (((unsigned) cmd->cmnd[7]) << 8);
}
last_scsi_logical_block(shpnt)[ldn] = scb->u1.log_blk_adr;
last_scsi_blockcount(shpnt)[ldn] = scb->u2.blk.count;
scb->u2.blk.length = ld(shpnt)[ldn].block_length;
break;
/* for other devices, enter here. Other types are not known by
Linux! TYPE_NO_LUN is forbidden as valid device. */
case TYPE_ROM:
case TYPE_TAPE:
case TYPE_PROCESSOR:
case TYPE_WORM:
case TYPE_SCANNER:
case TYPE_MEDIUM_CHANGER:
/* If there is a sequential-device, IBM recommends to use
IM_OTHER_SCSI_CMD_CMD instead of subsystem READ/WRITE.
This includes CD-ROM devices, too, due to the partial sequential
read capabilities. */
scb->command = IM_OTHER_SCSI_CMD_CMD;
if (scsi_cmd == READ_6 || scsi_cmd == READ_10 || scsi_cmd == READ_12)
/* enable READ */
scb->enable |= IM_READ_CONTROL;
scb->enable |= IM_BYPASS_BUFFER;
scb->u1.scsi_cmd_length = cmd->cmd_len;
memcpy(scb->u2.scsi_command, cmd->cmnd, cmd->cmd_len);
last_scsi_type(shpnt)[ldn] = IM_LONG_SCB;
/* Read/write on this non-disk devices is also displayworthy,
so flash-up the LED/display. */
break;
}
break;
case INQUIRY:
IBM_DS(shpnt).ldn_inquiry_access[ldn]++;
scb->command = IM_DEVICE_INQUIRY_CMD;
scb->enable |= IM_READ_CONTROL | IM_SUPRESS_EXCEPTION_SHORT | IM_BYPASS_BUFFER;
scb->u1.log_blk_adr = 0;
break;
case TEST_UNIT_READY:
scb->command = IM_OTHER_SCSI_CMD_CMD;
scb->enable |= IM_READ_CONTROL | IM_SUPRESS_EXCEPTION_SHORT | IM_BYPASS_BUFFER;
scb->u1.log_blk_adr = 0;
scb->u1.scsi_cmd_length = 6;
memcpy(scb->u2.scsi_command, cmd->cmnd, 6);
last_scsi_type(shpnt)[ldn] = IM_LONG_SCB;
break;
case READ_CAPACITY:
/* the length of system memory buffer must be exactly 8 bytes */
scb->command = IM_READ_CAPACITY_CMD;
scb->enable |= IM_READ_CONTROL | IM_BYPASS_BUFFER;
if (scb->sys_buf_length > 8)
scb->sys_buf_length = 8;
break;
/* Commands that need read-only-mode (system <- device): */
case REQUEST_SENSE:
scb->command = IM_REQUEST_SENSE_CMD;
scb->enable |= IM_READ_CONTROL | IM_SUPRESS_EXCEPTION_SHORT | IM_BYPASS_BUFFER;
break;
/* Commands that need write-only-mode (system -> device): */
case MODE_SELECT:
case MODE_SELECT_10:
IBM_DS(shpnt).ldn_modeselect_access[ldn]++;
scb->command = IM_OTHER_SCSI_CMD_CMD;
scb->enable |= IM_SUPRESS_EXCEPTION_SHORT | IM_BYPASS_BUFFER; /*Select needs WRITE-enabled */
scb->u1.scsi_cmd_length = cmd->cmd_len;
memcpy(scb->u2.scsi_command, cmd->cmnd, cmd->cmd_len);
last_scsi_type(shpnt)[ldn] = IM_LONG_SCB;
break;
/* For other commands, read-only is useful. Most other commands are
running without an input-data-block. */
default:
scb->command = IM_OTHER_SCSI_CMD_CMD;
scb->enable |= IM_READ_CONTROL | IM_SUPRESS_EXCEPTION_SHORT | IM_BYPASS_BUFFER;
scb->u1.scsi_cmd_length = cmd->cmd_len;
memcpy(scb->u2.scsi_command, cmd->cmnd, cmd->cmd_len);
last_scsi_type(shpnt)[ldn] = IM_LONG_SCB;
break;
}
/*issue scb command, and return */
if (++disk_rw_in_progress == 1)
PS2_DISK_LED_ON(shpnt->host_no, target);
if (last_scsi_type(shpnt)[ldn] == IM_LONG_SCB) {
issue_cmd(shpnt, isa_virt_to_bus(scb), IM_LONG_SCB | ldn);
IBM_DS(shpnt).long_scbs++;
} else {
issue_cmd(shpnt, isa_virt_to_bus(scb), IM_SCB | ldn);
IBM_DS(shpnt).scbs++;
}
return 0;
}
static DEF_SCSI_QCMD(ibmmca_queuecommand)
static int __ibmmca_abort(Scsi_Cmnd * cmd)
{
/* Abort does not work, as the adapter never generates an interrupt on
* whatever situation is simulated, even when really pending commands
* are running on the adapters' hardware ! */
struct Scsi_Host *shpnt;
unsigned int ldn;
void (*saved_done) (Scsi_Cmnd *);
int target;
int max_pun;
unsigned long imm_command;
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: Abort subroutine called...\n");
#endif
shpnt = cmd->device->host;
max_pun = subsystem_maxid(shpnt);
if (ibm_ansi_order) {
target = max_pun - 1 - cmd->device->id;
if ((target <= subsystem_pun(shpnt)) && (cmd->device->id <= subsystem_pun(shpnt)))
target--;
else if ((target >= subsystem_pun(shpnt)) && (cmd->device->id >= subsystem_pun(shpnt)))
target++;
} else
target = cmd->device->id;
/* get logical device number, and disable system interrupts */
printk(KERN_WARNING "IBM MCA SCSI: Sending abort to device pun=%d, lun=%d.\n", target, cmd->device->lun);
ldn = get_ldn(shpnt)[target][cmd->device->lun];
/*if cmd for this ldn has already finished, no need to abort */
if (!ld(shpnt)[ldn].cmd) {
return SUCCESS;
}
/* Clear ld.cmd, save done function, install internal done,
* send abort immediate command (this enables sys. interrupts),
* and wait until the interrupt arrives.
*/
saved_done = cmd->scsi_done;
cmd->scsi_done = internal_done;
cmd->SCp.Status = 0;
last_scsi_command(shpnt)[ldn] = IM_ABORT_IMM_CMD;
last_scsi_type(shpnt)[ldn] = IM_IMM_CMD;
imm_command = inl(IM_CMD_REG(shpnt));
imm_command &= (unsigned long) (0xffff0000); /* mask reserved stuff */
imm_command |= (unsigned long) (IM_ABORT_IMM_CMD);
/* must wait for attention reg not busy */
/* FIXME - timeout, politeness */
while (1) {
if (!(inb(IM_STAT_REG(shpnt)) & IM_BUSY))
break;
}
/* write registers and enable system interrupts */
outl(imm_command, IM_CMD_REG(shpnt));
outb(IM_IMM_CMD | ldn, IM_ATTN_REG(shpnt));
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: Abort queued to adapter...\n");
#endif
spin_unlock_irq(shpnt->host_lock);
while (!cmd->SCp.Status)
yield();
spin_lock_irq(shpnt->host_lock);
cmd->scsi_done = saved_done;
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: Abort returned with adapter response...\n");
#endif
/*if abort went well, call saved done, then return success or error */
if (cmd->result == (DID_ABORT << 16))
{
cmd->result |= DID_ABORT << 16;
if (cmd->scsi_done)
(cmd->scsi_done) (cmd);
ld(shpnt)[ldn].cmd = NULL;
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: Abort finished with success.\n");
#endif
return SUCCESS;
} else {
cmd->result |= DID_NO_CONNECT << 16;
if (cmd->scsi_done)
(cmd->scsi_done) (cmd);
ld(shpnt)[ldn].cmd = NULL;
#ifdef IM_DEBUG_PROBE
printk("IBM MCA SCSI: Abort failed.\n");
#endif
return FAILED;
}
}
static int ibmmca_abort(Scsi_Cmnd * cmd)
{
struct Scsi_Host *shpnt = cmd->device->host;
int rc;
spin_lock_irq(shpnt->host_lock);
rc = __ibmmca_abort(cmd);
spin_unlock_irq(shpnt->host_lock);
return rc;
}
static int __ibmmca_host_reset(Scsi_Cmnd * cmd)
{
struct Scsi_Host *shpnt;
Scsi_Cmnd *cmd_aid;
int ticks, i;
unsigned long imm_command;
BUG_ON(cmd == NULL);
ticks = IM_RESET_DELAY * HZ;
shpnt = cmd->device->host;
if (local_checking_phase_flag(shpnt)) {
printk(KERN_WARNING "IBM MCA SCSI: unable to reset while checking devices.\n");
return FAILED;
}
/* issue reset immediate command to subsystem, and wait for interrupt */
printk("IBM MCA SCSI: resetting all devices.\n");
reset_status(shpnt) = IM_RESET_IN_PROGRESS;
last_scsi_command(shpnt)[0xf] = IM_RESET_IMM_CMD;
last_scsi_type(shpnt)[0xf] = IM_IMM_CMD;
imm_command = inl(IM_CMD_REG(shpnt));
imm_command &= (unsigned long) (0xffff0000); /* mask reserved stuff */
imm_command |= (unsigned long) (IM_RESET_IMM_CMD);
/* must wait for attention reg not busy */
while (1) {
if (!(inb(IM_STAT_REG(shpnt)) & IM_BUSY))
break;
spin_unlock_irq(shpnt->host_lock);
yield();
spin_lock_irq(shpnt->host_lock);
}
/*write registers and enable system interrupts */
outl(imm_command, IM_CMD_REG(shpnt));
outb(IM_IMM_CMD | 0xf, IM_ATTN_REG(shpnt));
/* wait for interrupt finished or intr_stat register to be set, as the
* interrupt will not be executed, while we are in here! */
/* FIXME: This is really really icky we so want a sleeping version of this ! */
while (reset_status(shpnt) == IM_RESET_IN_PROGRESS && --ticks && ((inb(IM_INTR_REG(shpnt)) & 0x8f) != 0x8f)) {
udelay((1 + 999 / HZ) * 1000);
barrier();
}
/* if reset did not complete, just return an error */
if (!ticks) {
printk(KERN_ERR "IBM MCA SCSI: reset did not complete within %d seconds.\n", IM_RESET_DELAY);
reset_status(shpnt) = IM_RESET_FINISHED_FAIL;
return FAILED;
}
if ((inb(IM_INTR_REG(shpnt)) & 0x8f) == 0x8f) {
/* analysis done by this routine and not by the intr-routine */
if (inb(IM_INTR_REG(shpnt)) == 0xaf)
reset_status(shpnt) = IM_RESET_FINISHED_OK_NO_INT;
else if (inb(IM_INTR_REG(shpnt)) == 0xcf)
reset_status(shpnt) = IM_RESET_FINISHED_FAIL;
else /* failed, 4get it */
reset_status(shpnt) = IM_RESET_NOT_IN_PROGRESS_NO_INT;
outb(IM_EOI | 0xf, IM_ATTN_REG(shpnt));
}
/* if reset failed, just return an error */
if (reset_status(shpnt) == IM_RESET_FINISHED_FAIL) {
printk(KERN_ERR "IBM MCA SCSI: reset failed.\n");
return FAILED;
}
/* so reset finished ok - call outstanding done's, and return success */
printk(KERN_INFO "IBM MCA SCSI: Reset successfully completed.\n");
for (i = 0; i < MAX_LOG_DEV; i++) {
cmd_aid = ld(shpnt)[i].cmd;
if (cmd_aid && cmd_aid->scsi_done) {
ld(shpnt)[i].cmd = NULL;
cmd_aid->result = DID_RESET << 16;
}
}
return SUCCESS;
}
static int ibmmca_host_reset(Scsi_Cmnd * cmd)
{
struct Scsi_Host *shpnt = cmd->device->host;
int rc;
spin_lock_irq(shpnt->host_lock);
rc = __ibmmca_host_reset(cmd);
spin_unlock_irq(shpnt->host_lock);
return rc;
}
static int ibmmca_biosparam(struct scsi_device *sdev, struct block_device *bdev, sector_t capacity, int *info)
{
int size = capacity;
info[0] = 64;
info[1] = 32;
info[2] = size / (info[0] * info[1]);
if (info[2] >= 1024) {
info[0] = 128;
info[1] = 63;
info[2] = size / (info[0] * info[1]);
if (info[2] >= 1024) {
info[0] = 255;
info[1] = 63;
info[2] = size / (info[0] * info[1]);
if (info[2] >= 1024)
info[2] = 1023;
}
}
return 0;
}
/* calculate percentage of total accesses on a ldn */
static int ldn_access_load(struct Scsi_Host *shpnt, int ldn)
{
if (IBM_DS(shpnt).total_accesses == 0)
return (0);
if (IBM_DS(shpnt).ldn_access[ldn] == 0)
return (0);
return (IBM_DS(shpnt).ldn_access[ldn] * 100) / IBM_DS(shpnt).total_accesses;
}
/* calculate total amount of r/w-accesses */
static int ldn_access_total_read_write(struct Scsi_Host *shpnt)
{
int a;
int i;
a = 0;
for (i = 0; i <= MAX_LOG_DEV; i++)
a += IBM_DS(shpnt).ldn_read_access[i] + IBM_DS(shpnt).ldn_write_access[i];
return (a);
}
static int ldn_access_total_inquiry(struct Scsi_Host *shpnt)
{
int a;
int i;
a = 0;
for (i = 0; i <= MAX_LOG_DEV; i++)
a += IBM_DS(shpnt).ldn_inquiry_access[i];
return (a);
}
static int ldn_access_total_modeselect(struct Scsi_Host *shpnt)
{
int a;
int i;
a = 0;
for (i = 0; i <= MAX_LOG_DEV; i++)
a += IBM_DS(shpnt).ldn_modeselect_access[i];
return (a);
}
/* routine to display info in the proc-fs-structure (a deluxe feature) */
static int ibmmca_proc_info(struct Scsi_Host *shpnt, char *buffer, char **start, off_t offset, int length, int inout)
{
int len = 0;
int i, id, lun;
unsigned long flags;
int max_pun;
spin_lock_irqsave(shpnt->host_lock, flags); /* Check it */
max_pun = subsystem_maxid(shpnt);
len += sprintf(buffer + len, "\n IBM-SCSI-Subsystem-Linux-Driver, Version %s\n\n\n", IBMMCA_SCSI_DRIVER_VERSION);
len += sprintf(buffer + len, " SCSI Access-Statistics:\n");
len += sprintf(buffer + len, " Device Scanning Order....: %s\n", (ibm_ansi_order) ? "IBM/ANSI" : "New Industry Standard");
#ifdef CONFIG_SCSI_MULTI_LUN
len += sprintf(buffer + len, " Multiple LUN probing.....: Yes\n");
#else
len += sprintf(buffer + len, " Multiple LUN probing.....: No\n");
#endif
len += sprintf(buffer + len, " This Hostnumber..........: %d\n", shpnt->host_no);
len += sprintf(buffer + len, " Base I/O-Port............: 0x%x\n", (unsigned int) (IM_CMD_REG(shpnt)));
len += sprintf(buffer + len, " (Shared) IRQ.............: %d\n", IM_IRQ);
len += sprintf(buffer + len, " Total Interrupts.........: %d\n", IBM_DS(shpnt).total_interrupts);
len += sprintf(buffer + len, " Total SCSI Accesses......: %d\n", IBM_DS(shpnt).total_accesses);
len += sprintf(buffer + len, " Total short SCBs.........: %d\n", IBM_DS(shpnt).scbs);
len += sprintf(buffer + len, " Total long SCBs..........: %d\n", IBM_DS(shpnt).long_scbs);
len += sprintf(buffer + len, " Total SCSI READ/WRITE..: %d\n", ldn_access_total_read_write(shpnt));
len += sprintf(buffer + len, " Total SCSI Inquiries...: %d\n", ldn_access_total_inquiry(shpnt));
len += sprintf(buffer + len, " Total SCSI Modeselects.: %d\n", ldn_access_total_modeselect(shpnt));
len += sprintf(buffer + len, " Total SCSI other cmds..: %d\n", IBM_DS(shpnt).total_accesses - ldn_access_total_read_write(shpnt)
- ldn_access_total_modeselect(shpnt)
- ldn_access_total_inquiry(shpnt));
len += sprintf(buffer + len, " Total SCSI command fails.: %d\n\n", IBM_DS(shpnt).total_errors);
len += sprintf(buffer + len, " Logical-Device-Number (LDN) Access-Statistics:\n");
len += sprintf(buffer + len, " LDN | Accesses [%%] | READ | WRITE | ASSIGNMENTS\n");
len += sprintf(buffer + len, " -----|--------------|-----------|-----------|--------------\n");
for (i = 0; i <= MAX_LOG_DEV; i++)
len += sprintf(buffer + len, " %2X | %3d | %8d | %8d | %8d\n", i, ldn_access_load(shpnt, i), IBM_DS(shpnt).ldn_read_access[i], IBM_DS(shpnt).ldn_write_access[i], IBM_DS(shpnt).ldn_assignments[i]);
len += sprintf(buffer + len, " -----------------------------------------------------------\n\n");
len += sprintf(buffer + len, " Dynamical-LDN-Assignment-Statistics:\n");
len += sprintf(buffer + len, " Number of physical SCSI-devices..: %d (+ Adapter)\n", IBM_DS(shpnt).total_scsi_devices);
len += sprintf(buffer + len, " Dynamical Assignment necessary...: %s\n", IBM_DS(shpnt).dyn_flag ? "Yes" : "No ");
len += sprintf(buffer + len, " Next LDN to be assigned..........: 0x%x\n", next_ldn(shpnt));
len += sprintf(buffer + len, " Dynamical assignments done yet...: %d\n", IBM_DS(shpnt).dynamical_assignments);
len += sprintf(buffer + len, "\n Current SCSI-Device-Mapping:\n");
len += sprintf(buffer + len, " Physical SCSI-Device Map Logical SCSI-Device Map\n");
len += sprintf(buffer + len, " ID\\LUN 0 1 2 3 4 5 6 7 ID\\LUN 0 1 2 3 4 5 6 7\n");
for (id = 0; id < max_pun; id++) {
len += sprintf(buffer + len, " %2d ", id);
for (lun = 0; lun < 8; lun++)
len += sprintf(buffer + len, "%2s ", ti_p(get_scsi(shpnt)[id][lun]));
len += sprintf(buffer + len, " %2d ", id);
for (lun = 0; lun < 8; lun++)
len += sprintf(buffer + len, "%2s ", ti_l(get_ldn(shpnt)[id][lun]));
len += sprintf(buffer + len, "\n");
}
len += sprintf(buffer + len, "(A = IBM-Subsystem, D = Harddisk, T = Tapedrive, P = Processor, W = WORM,\n");
len += sprintf(buffer + len, " R = CD-ROM, S = Scanner, M = MO-Drive, C = Medium-Changer, + = unprovided LUN,\n");
len += sprintf(buffer + len, " - = nothing found, nothing assigned or unprobed LUN)\n\n");
*start = buffer + offset;
len -= offset;
if (len > length)
len = length;
spin_unlock_irqrestore(shpnt->host_lock, flags);
return len;
}
static int option_setup(char *str)
{
int ints[IM_MAX_HOSTS];
char *cur = str;
int i = 1;
while (cur && isdigit(*cur) && i < IM_MAX_HOSTS) {
ints[i++] = simple_strtoul(cur, NULL, 0);
if ((cur = strchr(cur, ',')) != NULL)
cur++;
}
ints[0] = i - 1;
internal_ibmmca_scsi_setup(cur, ints);
return 1;
}
__setup("ibmmcascsi=", option_setup);
static struct mca_driver ibmmca_driver = {
.id_table = ibmmca_id_table,
.driver = {
.name = "ibmmca",
.bus = &mca_bus_type,
.probe = ibmmca_probe,
.remove = __devexit_p(ibmmca_remove),
},
};
static int __init ibmmca_init(void)
{
#ifdef MODULE
/* If the driver is run as module, read from conf.modules or cmd-line */
if (boot_options)
option_setup(boot_options);
#endif
return mca_register_driver_integrated(&ibmmca_driver, MCA_INTEGSCSI);
}
static void __exit ibmmca_exit(void)
{
mca_unregister_driver(&ibmmca_driver);
}
module_init(ibmmca_init);
module_exit(ibmmca_exit);
drivers/scsi/sim710.c
View file @ d5b4bb4d
......@@ -17,7 +17,7 @@
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*----------------------------------------------------------------------------
*
* MCA card detection code by Trent McNair.
* MCA card detection code by Trent McNair. (now deleted)
* Fixes to not explicitly nul bss data from Xavier Bestel.
* Some multiboard fixes from Rolf Eike Beer.
* Auto probing of EISA config space from Trevor Hemsley.
......@@ -32,7 +32,6 @@
#include <linux/blkdev.h>
#include <linux/device.h>
#include <linux/init.h>
#include <linux/mca.h>
#include <linux/eisa.h>
#include <linux/interrupt.h>
#include <scsi/scsi_host.h>
......@@ -43,7 +42,7 @@
#include "53c700.h"
/* Must be enough for both EISA and MCA */
/* Must be enough for EISA */
#define MAX_SLOTS 8
static __u8 __initdata id_array[MAX_SLOTS] = { [0 ... MAX_SLOTS-1] = 7 };
......@@ -89,7 +88,7 @@ param_setup(char *str)
__setup("sim710=", param_setup);
static struct scsi_host_template sim710_driver_template = {
.name = "LSI (Symbios) 710 MCA/EISA",
.name = "LSI (Symbios) 710 EISA",
.proc_name = "sim710",
.this_id = 7,
.module = THIS_MODULE,
......@@ -169,114 +168,6 @@ sim710_device_remove(struct device *dev)
return 0;
}
#ifdef CONFIG_MCA
/* CARD ID 01BB and 01BA use the same pos values */
#define MCA_01BB_IO_PORTS { 0x0000, 0x0000, 0x0800, 0x0C00, 0x1000, 0x1400, \
0x1800, 0x1C00, 0x2000, 0x2400, 0x2800, \
0x2C00, 0x3000, 0x3400, 0x3800, 0x3C00, \
0x4000, 0x4400, 0x4800, 0x4C00, 0x5000 }
#define MCA_01BB_IRQS { 3, 5, 11, 14 }
/* CARD ID 004f */
#define MCA_004F_IO_PORTS { 0x0000, 0x0200, 0x0300, 0x0400, 0x0500, 0x0600 }
#define MCA_004F_IRQS { 5, 9, 14 }
static short sim710_mca_id_table[] = { 0x01bb, 0x01ba, 0x004f, 0};
static __init int
sim710_mca_probe(struct device *dev)
{
struct mca_device *mca_dev = to_mca_device(dev);
int slot = mca_dev->slot;
int pos[3];
unsigned int base;
int irq_vector;
short id = sim710_mca_id_table[mca_dev->index];
static int io_004f_by_pos[] = MCA_004F_IO_PORTS;
static int irq_004f_by_pos[] = MCA_004F_IRQS;
static int io_01bb_by_pos[] = MCA_01BB_IO_PORTS;
static int irq_01bb_by_pos[] = MCA_01BB_IRQS;
char *name;
int clock;
pos[0] = mca_device_read_stored_pos(mca_dev, 2);
pos[1] = mca_device_read_stored_pos(mca_dev, 3);
pos[2] = mca_device_read_stored_pos(mca_dev, 4);
/*
* 01BB & 01BA port base by bits 7,6,5,4,3,2 in pos[2]
*
* 000000 <disabled> 001010 0x2800
* 000001 <invalid> 001011 0x2C00
* 000010 0x0800 001100 0x3000
* 000011 0x0C00 001101 0x3400
* 000100 0x1000 001110 0x3800
* 000101 0x1400 001111 0x3C00
* 000110 0x1800 010000 0x4000
* 000111 0x1C00 010001 0x4400
* 001000 0x2000 010010 0x4800
* 001001 0x2400 010011 0x4C00
* 010100 0x5000
*
* 00F4 port base by bits 3,2,1 in pos[0]
*
* 000 <disabled> 001 0x200
* 010 0x300 011 0x400
* 100 0x500 101 0x600
*
* 01BB & 01BA IRQ is specified in pos[0] bits 7 and 6:
*
* 00 3 10 11
* 01 5 11 14
*
* 00F4 IRQ specified by bits 6,5,4 in pos[0]
*
* 100 5 101 9
* 110 14
*/
if (id == 0x01bb || id == 0x01ba) {
base = io_01bb_by_pos[(pos[2] & 0xFC) >> 2];
irq_vector =
irq_01bb_by_pos[((pos[0] & 0xC0) >> 6)];
clock = 50;
if (id == 0x01bb)
name = "NCR 3360/3430 SCSI SubSystem";
else
name = "NCR Dual SIOP SCSI Host Adapter Board";
} else if ( id == 0x004f ) {
base = io_004f_by_pos[((pos[0] & 0x0E) >> 1)];
irq_vector =
irq_004f_by_pos[((pos[0] & 0x70) >> 4) - 4];
clock = 50;
name = "NCR 53c710 SCSI Host Adapter Board";
} else {
return -ENODEV;
}
mca_device_set_name(mca_dev, name);
mca_device_set_claim(mca_dev, 1);
base = mca_device_transform_ioport(mca_dev, base);
irq_vector = mca_device_transform_irq(mca_dev, irq_vector);
return sim710_probe_common(dev, base, irq_vector, clock,
0, id_array[slot]);
}
static struct mca_driver sim710_mca_driver = {
.id_table = sim710_mca_id_table,
.driver = {
.name = "sim710",
.bus = &mca_bus_type,
.probe = sim710_mca_probe,
.remove = __devexit_p(sim710_device_remove),
},
};
#endif /* CONFIG_MCA */
#ifdef CONFIG_EISA
static struct eisa_device_id sim710_eisa_ids[] = {
{ "CPQ4410" },
......@@ -344,10 +235,6 @@ static int __init sim710_init(void)
param_setup(sim710);
#endif
#ifdef CONFIG_MCA
err = mca_register_driver(&sim710_mca_driver);
#endif
#ifdef CONFIG_EISA
err = eisa_driver_register(&sim710_eisa_driver);
#endif
......@@ -361,11 +248,6 @@ static int __init sim710_init(void)
static void __exit sim710_exit(void)
{
#ifdef CONFIG_MCA
if (MCA_bus)
mca_unregister_driver(&sim710_mca_driver);
#endif
#ifdef CONFIG_EISA
eisa_driver_unregister(&sim710_eisa_driver);
#endif
......
drivers/tty/serial/8250/8250_mca.c deleted 100644 → 0
View file @ c80ddb52
/*
* Copyright (C) 2005 Russell King.
* Data taken from include/asm-i386/serial.h
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
#include <linux/module.h>
#include <linux/init.h>
#include <linux/mca.h>
#include <linux/serial_8250.h>
/*
* FIXME: Should we be doing AUTO_IRQ here?
*/
#ifdef CONFIG_SERIAL_8250_DETECT_IRQ
#define MCA_FLAGS UPF_BOOT_AUTOCONF | UPF_SKIP_TEST | UPF_AUTO_IRQ
#else
#define MCA_FLAGS UPF_BOOT_AUTOCONF | UPF_SKIP_TEST
#endif
#define PORT(_base,_irq) \
{ \
.iobase = _base, \
.irq = _irq, \
.uartclk = 1843200, \
.iotype = UPIO_PORT, \
.flags = MCA_FLAGS, \
}
static struct plat_serial8250_port mca_data[] = {
PORT(0x3220, 3),
PORT(0x3228, 3),
PORT(0x4220, 3),
PORT(0x4228, 3),
PORT(0x5220, 3),
PORT(0x5228, 3),
{ },
};
static struct platform_device mca_device = {
.name = "serial8250",
.id = PLAT8250_DEV_MCA,
.dev = {
.platform_data = mca_data,
},
};
static int __init mca_init(void)
{
if (!MCA_bus)
return -ENODEV;
return platform_device_register(&mca_device);
}
module_init(mca_init);
MODULE_AUTHOR("Russell King");
MODULE_DESCRIPTION("8250 serial probe module for MCA ports");
MODULE_LICENSE("GPL");
drivers/tty/serial/8250/Kconfig
View file @ d5b4bb4d
......@@ -241,15 +241,6 @@ config SERIAL_8250_RSA
help
::: To be written :::
config SERIAL_8250_MCA
tristate "Support 8250-type ports on MCA buses"
depends on SERIAL_8250 != n && MCA
help
Say Y here if you have a MCA serial ports.
To compile this driver as a module, choose M here: the module
will be called 8250_mca.
config SERIAL_8250_ACORN
tristate "Acorn expansion card serial port support"
depends on ARCH_ACORN && SERIAL_8250
......
drivers/tty/serial/8250/Makefile
View file @ d5b4bb4d
......@@ -15,7 +15,6 @@ obj-$(CONFIG_SERIAL_8250_ACCENT) += 8250_accent.o
obj-$(CONFIG_SERIAL_8250_BOCA) += 8250_boca.o
obj-$(CONFIG_SERIAL_8250_EXAR_ST16C554) += 8250_exar_st16c554.o
obj-$(CONFIG_SERIAL_8250_HUB6) += 8250_hub6.o
obj-$(CONFIG_SERIAL_8250_MCA) += 8250_mca.o
obj-$(CONFIG_SERIAL_8250_FSL) += 8250_fsl.o
obj-$(CONFIG_SERIAL_8250_DW) += 8250_dw.o
obj-$(CONFIG_SERIAL_8250_EM) += 8250_em.o
include/linux/i2o-dev.h
View file @ d5b4bb4d
......@@ -124,7 +124,7 @@ typedef struct i2o_sg_io_hdr {
#define I2O_BUS_LOCAL 0
#define I2O_BUS_ISA 1
#define I2O_BUS_EISA 2
#define I2O_BUS_MCA 3
/* was I2O_BUS_MCA 3 */
#define I2O_BUS_PCI 4
#define I2O_BUS_PCMCIA 5
#define I2O_BUS_NUBUS 6
......
include/linux/mca-legacy.h deleted 100644 → 0
View file @ c80ddb52
/* -*- mode: c; c-basic-offset: 8 -*- */
/* This is the function prototypes for the old legacy MCA interface
*
* Please move your driver to the new sysfs based one instead */
#ifndef _LINUX_MCA_LEGACY_H
#define _LINUX_MCA_LEGACY_H
#include <linux/mca.h>
#warning "MCA legacy - please move your driver to the new sysfs api"
/* MCA_NOTFOUND is an error condition. The other two indicate
* motherboard POS registers contain the adapter. They might be
* returned by the mca_find_adapter() function, and can be used as
* arguments to mca_read_stored_pos(). I'm not going to allow direct
* access to the motherboard registers until we run across an adapter
* that requires it. We don't know enough about them to know if it's
* safe.
*
* See Documentation/mca.txt or one of the existing drivers for
* more information.
*/
#define MCA_NOTFOUND (-1)
/* Returns the slot of the first enabled adapter matching id. User can
* specify a starting slot beyond zero, to deal with detecting multiple
* devices. Returns MCA_NOTFOUND if id not found. Also checks the
* integrated adapters.
*/
extern int mca_find_adapter(int id, int start);
extern int mca_find_unused_adapter(int id, int start);
extern int mca_mark_as_used(int slot);
extern void mca_mark_as_unused(int slot);
/* gets a byte out of POS register (stored in memory) */
extern unsigned char mca_read_stored_pos(int slot, int reg);
/* This can be expanded later. Right now, it gives us a way of
* getting meaningful information into the MCA_info structure,
* so we can have a more interesting /proc/mca.
*/
extern void mca_set_adapter_name(int slot, char* name);
/* These routines actually mess with the hardware POS registers. They
* temporarily disable the device (and interrupts), so make sure you know
* what you're doing if you use them. Furthermore, writing to a POS may
* result in two devices trying to share a resource, which in turn can
* result in multiple devices sharing memory spaces, IRQs, or even trashing
* hardware. YOU HAVE BEEN WARNED.
*
* You can only access slots with this. Motherboard registers are off
* limits.
*/
/* read a byte from the specified POS register. */
extern unsigned char mca_read_pos(int slot, int reg);
/* write a byte to the specified POS register. */
extern void mca_write_pos(int slot, int reg, unsigned char byte);
#endif
include/linux/mca.h deleted 100644 → 0
View file @ c80ddb52
/*
* Header for Microchannel Architecture Bus
* Written by Martin Kolinek, February 1996
*/
#ifndef _LINUX_MCA_H
#define _LINUX_MCA_H
#include <linux/device.h>
#ifdef CONFIG_MCA
#include <asm/mca.h>
extern int MCA_bus;
#else
#define MCA_bus 0
#endif
/* This sets up an information callback for /proc/mca/slot?. The
* function is called with the buffer, slot, and device pointer (or
* some equally informative context information, or nothing, if you
* prefer), and is expected to put useful information into the
* buffer. The adapter name, id, and POS registers get printed
* before this is called though, so don't do it again.
*
* This should be called with a NULL procfn when a module
* unregisters, thus preventing kernel crashes and other such
* nastiness.
*/
typedef int (*MCA_ProcFn)(char* buf, int slot, void* dev);
/* Should only be called by the NMI interrupt handler, this will do some
* fancy stuff to figure out what might have generated a NMI.
*/
extern void mca_handle_nmi(void);
enum MCA_AdapterStatus {
MCA_ADAPTER_NORMAL = 0,
MCA_ADAPTER_NONE = 1,
MCA_ADAPTER_DISABLED = 2,
MCA_ADAPTER_ERROR = 3
};
struct mca_device {
u64 dma_mask;
int pos_id;
int slot;
/* index into id_table, set by the bus match routine */
int index;
/* is there a driver installed? 0 - No, 1 - Yes */
int driver_loaded;
/* POS registers */
unsigned char pos[8];
/* if a pseudo adapter of the motherboard, this is the motherboard
* register value to use for setup cycles */
short pos_register;
enum MCA_AdapterStatus status;
#ifdef CONFIG_MCA_PROC_FS
/* name of the proc/mca file */
char procname[8];
/* /proc info callback */
MCA_ProcFn procfn;
/* device/context info for proc callback */
void *proc_dev;
#endif
struct device dev;
char name[32];
};
#define to_mca_device(mdev) container_of(mdev, struct mca_device, dev)
struct mca_bus_accessor_functions {
unsigned char (*mca_read_pos)(struct mca_device *, int reg);
void (*mca_write_pos)(struct mca_device *, int reg,
unsigned char byte);
int (*mca_transform_irq)(struct mca_device *, int irq);
int (*mca_transform_ioport)(struct mca_device *,
int region);
void * (*mca_transform_memory)(struct mca_device *,
void *memory);
};
struct mca_bus {
u64 default_dma_mask;
int number;
struct mca_bus_accessor_functions f;
struct device dev;
char name[32];
};
#define to_mca_bus(mdev) container_of(mdev, struct mca_bus, dev)
struct mca_driver {
const short *id_table;
void *driver_data;
int integrated_id;
struct device_driver driver;
};
#define to_mca_driver(mdriver) container_of(mdriver, struct mca_driver, driver)
/* Ongoing supported API functions */
extern struct mca_device *mca_find_device_by_slot(int slot);
extern int mca_system_init(void);
extern struct mca_bus *mca_attach_bus(int);
extern unsigned char mca_device_read_stored_pos(struct mca_device *mca_dev,
int reg);
extern unsigned char mca_device_read_pos(struct mca_device *mca_dev, int reg);
extern void mca_device_write_pos(struct mca_device *mca_dev, int reg,
unsigned char byte);
extern int mca_device_transform_irq(struct mca_device *mca_dev, int irq);
extern int mca_device_transform_ioport(struct mca_device *mca_dev, int port);
extern void *mca_device_transform_memory(struct mca_device *mca_dev,
void *mem);
extern int mca_device_claimed(struct mca_device *mca_dev);
extern void mca_device_set_claim(struct mca_device *mca_dev, int val);
extern void mca_device_set_name(struct mca_device *mca_dev, const char *name);
static inline char *mca_device_get_name(struct mca_device *mca_dev)
{
return mca_dev ? mca_dev->name : NULL;
}
extern enum MCA_AdapterStatus mca_device_status(struct mca_device *mca_dev);
extern struct bus_type mca_bus_type;
extern int mca_register_driver(struct mca_driver *drv);
extern int mca_register_driver_integrated(struct mca_driver *, int);
extern void mca_unregister_driver(struct mca_driver *drv);
/* WARNING: only called by the boot time device setup */
extern int mca_register_device(int bus, struct mca_device *mca_dev);
#ifdef CONFIG_MCA_PROC_FS
extern void mca_do_proc_init(void);
extern void mca_set_adapter_procfn(int slot, MCA_ProcFn, void* dev);
#else
static inline void mca_do_proc_init(void)
{
}
static inline void mca_set_adapter_procfn(int slot, MCA_ProcFn fn, void* dev)
{
}
#endif
#endif /* _LINUX_MCA_H */
include/linux/serial.h
View file @ d5b4bb4d
......@@ -130,7 +130,6 @@ struct serial_uart_config {
#define ASYNCB_CHECK_CD 25 /* i.e., CLOCAL */
#define ASYNCB_SHARE_IRQ 24 /* for multifunction cards, no longer used */
#define ASYNCB_CONS_FLOW 23 /* flow control for console */
#define ASYNCB_BOOT_ONLYMCA 22 /* Probe only if MCA bus */
#define ASYNCB_FIRST_KERNEL 22
#define ASYNC_HUP_NOTIFY (1U << ASYNCB_HUP_NOTIFY)
......@@ -166,7 +165,6 @@ struct serial_uart_config {
#define ASYNC_CHECK_CD (1U << ASYNCB_CHECK_CD)
#define ASYNC_SHARE_IRQ (1U << ASYNCB_SHARE_IRQ)
#define ASYNC_CONS_FLOW (1U << ASYNCB_CONS_FLOW)
#define ASYNC_BOOT_ONLYMCA (1U << ASYNCB_BOOT_ONLYMCA)
#define ASYNC_INTERNAL_FLAGS (~((1U << ASYNCB_FIRST_KERNEL) - 1))
/*
......
include/linux/serial_8250.h
View file @ d5b4bb4d
......@@ -55,7 +55,6 @@ enum {
PLAT8250_DEV_BOCA,
PLAT8250_DEV_EXAR_ST16C554,
PLAT8250_DEV_HUB6,
PLAT8250_DEV_MCA,
PLAT8250_DEV_AU1X00,
PLAT8250_DEV_SM501,
};
......
scripts/kconfig/mconf.c
View file @ d5b4bb4d
......@@ -240,7 +240,7 @@ search_help[] = N_(
"Defined at drivers/pci/Kconfig:47\n"
"Depends on: X86_LOCAL_APIC && X86_IO_APIC || IA64\n"
"Location:\n"
" -> Bus options (PCI, PCMCIA, EISA, MCA, ISA)\n"
" -> Bus options (PCI, PCMCIA, EISA, ISA)\n"
" -> PCI support (PCI [=y])\n"
" -> PCI access mode (<choice> [=y])\n"
"Selects: LIBCRC32\n"
......
scripts/kconfig/nconf.c
View file @ d5b4bb4d
......@@ -223,7 +223,7 @@ search_help[] = N_(
"Defined at drivers/pci/Kconfig:47\n"
"Depends on: X86_LOCAL_APIC && X86_IO_APIC || IA64\n"
"Location:\n"
" -> Bus options (PCI, PCMCIA, EISA, MCA, ISA)\n"
" -> Bus options (PCI, PCMCIA, EISA, ISA)\n"
" -> PCI support (PCI [ = y])\n"
" -> PCI access mode (<choice> [ = y])\n"
"Selects: LIBCRC32\n"
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7