|
|
sx.txt -- specialix SX/SI multiport serial driver readme.
|
sx.txt -- specialix SX/SI multiport serial driver readme.
|
|
|
|
|
|
|
Copyright (C) 1997 Roger Wolff (R.E.Wolff@BitWizard.nl)
|
Copyright (C) 1997 Roger Wolff (R.E.Wolff@BitWizard.nl)
|
|
|
Specialix pays for the development and support of this driver.
|
Specialix pays for the development and support of this driver.
|
Please DO contact support@specialix.co.uk if you require
|
Please DO contact support@specialix.co.uk if you require
|
support.
|
support.
|
|
|
This driver was developed in the BitWizard linux device
|
This driver was developed in the BitWizard linux device
|
driver service. If you require a linux device driver for your
|
driver service. If you require a linux device driver for your
|
product, please contact devices@BitWizard.nl for a quote.
|
product, please contact devices@BitWizard.nl for a quote.
|
|
|
(History)
|
(History)
|
There used to be an SI driver by Simon Allan. This is a complete
|
There used to be an SI driver by Simon Allan. This is a complete
|
rewrite from scratch. Just a few lines-of-code have been snatched.
|
rewrite from scratch. Just a few lines-of-code have been snatched.
|
|
|
(Sources)
|
(Sources)
|
Specialix document number 6210028: SX Host Card and Download Code
|
Specialix document number 6210028: SX Host Card and Download Code
|
Software Functional Specification.
|
Software Functional Specification.
|
|
|
(Copying)
|
(Copying)
|
This program is free software; you can redistribute it and/or
|
This program is free software; you can redistribute it and/or
|
modify it under the terms of the GNU General Public License as
|
modify it under the terms of the GNU General Public License as
|
published by the Free Software Foundation; either version 2 of
|
published by the Free Software Foundation; either version 2 of
|
the License, or (at your option) any later version.
|
the License, or (at your option) any later version.
|
|
|
This program is distributed in the hope that it will be
|
This program is distributed in the hope that it will be
|
useful, but WITHOUT ANY WARRANTY; without even the implied
|
useful, but WITHOUT ANY WARRANTY; without even the implied
|
warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
|
warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
|
PURPOSE. See the GNU General Public License for more details.
|
PURPOSE. See the GNU General Public License for more details.
|
|
|
You should have received a copy of the GNU General Public
|
You should have received a copy of the GNU General Public
|
License along with this program; if not, write to the Free
|
License along with this program; if not, write to the Free
|
Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139,
|
Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139,
|
USA.
|
USA.
|
|
|
(Addendum)
|
(Addendum)
|
I'd appreciate it that if you have fixes, that you send them
|
I'd appreciate it that if you have fixes, that you send them
|
to me first.
|
to me first.
|
|
|
|
|
Introduction
|
Introduction
|
============
|
============
|
|
|
This file contains some random information, that I like to have online
|
This file contains some random information, that I like to have online
|
instead of in a manual that can get lost. Ever misplace your Linux
|
instead of in a manual that can get lost. Ever misplace your Linux
|
kernel sources? And the manual of one of the boards in your computer?
|
kernel sources? And the manual of one of the boards in your computer?
|
|
|
|
|
Theory of operation
|
Theory of operation
|
===================
|
===================
|
|
|
An important thing to know is that the driver itself doesn't have the
|
An important thing to know is that the driver itself doesn't have the
|
firmware for the card. This means that you need the separate package
|
firmware for the card. This means that you need the separate package
|
"sx_firmware". For now you can get the source at
|
"sx_firmware". For now you can get the source at
|
|
|
ftp://ftp.bitwizard.nl/specialix/sx_firmware_.tgz
|
ftp://ftp.bitwizard.nl/specialix/sx_firmware_.tgz
|
|
|
The firmware load needs a "misc" device, so you'll need to enable the
|
The firmware load needs a "misc" device, so you'll need to enable the
|
"Support for user misc device modules" in your kernel configuration.
|
"Support for user misc device modules" in your kernel configuration.
|
The misc device needs to be called "/dev/specialix_sxctl". It needs
|
The misc device needs to be called "/dev/specialix_sxctl". It needs
|
misc major 10, and minor number 167 (assigned by HPA). The section
|
misc major 10, and minor number 167 (assigned by HPA). The section
|
on creating device files below also creates this device.
|
on creating device files below also creates this device.
|
|
|
After loading the sx.o module into your kernel, the driver will report
|
After loading the sx.o module into your kernel, the driver will report
|
the number of cards detected, but because it doesn't have any
|
the number of cards detected, but because it doesn't have any
|
firmware, it will not be able to determine the number of ports. Only
|
firmware, it will not be able to determine the number of ports. Only
|
when you then run "sx_firmware" will the firmware be downloaded and
|
when you then run "sx_firmware" will the firmware be downloaded and
|
the rest of the driver initialized. At that time the sx_firmware
|
the rest of the driver initialized. At that time the sx_firmware
|
program will report the number of ports installed.
|
program will report the number of ports installed.
|
|
|
In contrast with many other multi port serial cards, some of the data
|
In contrast with many other multi port serial cards, some of the data
|
structures are only allocated when the card knows the number of ports
|
structures are only allocated when the card knows the number of ports
|
that are connected. This means we won't waste memory for 120 port
|
that are connected. This means we won't waste memory for 120 port
|
descriptor structures when you only have 8 ports. If you experience
|
descriptor structures when you only have 8 ports. If you experience
|
problems due to this, please report them: I haven't seen any.
|
problems due to this, please report them: I haven't seen any.
|
|
|
|
|
Interrupts
|
Interrupts
|
==========
|
==========
|
|
|
A multi port serial card, would generate a horrendous amount of
|
A multi port serial card, would generate a horrendous amount of
|
interrupts if it would interrupt the CPU for every received
|
interrupts if it would interrupt the CPU for every received
|
character. Even more than 10 years ago, the trick not to use
|
character. Even more than 10 years ago, the trick not to use
|
interrupts but to poll the serial cards was invented.
|
interrupts but to poll the serial cards was invented.
|
|
|
The SX card allow us to do this two ways. First the card limits its
|
The SX card allow us to do this two ways. First the card limits its
|
own interrupt rate to a rate that won't overwhelm the CPU. Secondly,
|
own interrupt rate to a rate that won't overwhelm the CPU. Secondly,
|
we could forget about the cards interrupt completely and use the
|
we could forget about the cards interrupt completely and use the
|
internal timer for this purpose.
|
internal timer for this purpose.
|
|
|
Polling the card can take up to a few percent of your CPU. Using the
|
Polling the card can take up to a few percent of your CPU. Using the
|
interrupts would be better if you have most of the ports idle. Using
|
interrupts would be better if you have most of the ports idle. Using
|
timer-based polling is better if your card almost always has work to
|
timer-based polling is better if your card almost always has work to
|
do. You save the separate interrupt in that case.
|
do. You save the separate interrupt in that case.
|
|
|
In any case, it doesn't really matter all that much.
|
In any case, it doesn't really matter all that much.
|
|
|
The most common problem with interrupts is that for ISA cards in a PCI
|
The most common problem with interrupts is that for ISA cards in a PCI
|
system the BIOS has to be told to configure that interrupt as "legacy
|
system the BIOS has to be told to configure that interrupt as "legacy
|
ISA". Otherwise the card can pull on the interrupt line all it wants
|
ISA". Otherwise the card can pull on the interrupt line all it wants
|
but the CPU won't see this.
|
but the CPU won't see this.
|
|
|
If you can't get the interrupt to work, remember that polling mode is
|
If you can't get the interrupt to work, remember that polling mode is
|
more efficient (provided you actually use the card intensively).
|
more efficient (provided you actually use the card intensively).
|
|
|
|
|
Allowed Configurations
|
Allowed Configurations
|
======================
|
======================
|
|
|
Some configurations are disallowed. Even though at a glance they might
|
Some configurations are disallowed. Even though at a glance they might
|
seem to work, they are known to lockup the bus between the host card
|
seem to work, they are known to lockup the bus between the host card
|
and the device concentrators. You should respect the drivers decision
|
and the device concentrators. You should respect the drivers decision
|
not to support certain configurations. It's there for a reason.
|
not to support certain configurations. It's there for a reason.
|
|
|
Warning: Seriously technical stuff ahead. Executive summary: Don't use
|
Warning: Seriously technical stuff ahead. Executive summary: Don't use
|
SX cards except configured at a 64k boundary. Skip the next paragraph.
|
SX cards except configured at a 64k boundary. Skip the next paragraph.
|
|
|
The SX cards can theoretically be placed at a 32k boundary. So for
|
The SX cards can theoretically be placed at a 32k boundary. So for
|
instance you can put an SX card at 0xc8000-0xd7fff. This is not a
|
instance you can put an SX card at 0xc8000-0xd7fff. This is not a
|
"recommended configuration". ISA cards have to tell the bus controller
|
"recommended configuration". ISA cards have to tell the bus controller
|
how they like their timing. Due to timing issues they have to do this
|
how they like their timing. Due to timing issues they have to do this
|
based on which 64k window the address falls into. This means that the
|
based on which 64k window the address falls into. This means that the
|
32k window below and above the SX card have to use exactly the same
|
32k window below and above the SX card have to use exactly the same
|
timing as the SX card. That reportedly works for other SX cards. But
|
timing as the SX card. That reportedly works for other SX cards. But
|
you're still left with two useless 32k windows that should not be used
|
you're still left with two useless 32k windows that should not be used
|
by anybody else.
|
by anybody else.
|
|
|
|
|
Configuring the driver
|
Configuring the driver
|
======================
|
======================
|
|
|
PCI cards are always detected. The driver auto-probes for ISA cards at
|
PCI cards are always detected. The driver auto-probes for ISA cards at
|
some sensible addresses. Please report if the auto-probe causes trouble
|
some sensible addresses. Please report if the auto-probe causes trouble
|
in your system, or when a card isn't detected.
|
in your system, or when a card isn't detected.
|
|
|
I'm afraid I haven't implemented "kernel command line parameters" yet.
|
I'm afraid I haven't implemented "kernel command line parameters" yet.
|
This means that if the default doesn't work for you, you shouldn't use
|
This means that if the default doesn't work for you, you shouldn't use
|
the compiled-into-the-kernel version of the driver. Use a module
|
the compiled-into-the-kernel version of the driver. Use a module
|
instead. If you convince me that you need this, I'll make it for
|
instead. If you convince me that you need this, I'll make it for
|
you. Deal?
|
you. Deal?
|
|
|
I'm afraid that the module parameters are a bit clumsy. If you have a
|
I'm afraid that the module parameters are a bit clumsy. If you have a
|
better idea, please tell me.
|
better idea, please tell me.
|
|
|
You can specify several parameters:
|
You can specify several parameters:
|
|
|
sx_poll: number of jiffies between timer-based polls.
|
sx_poll: number of jiffies between timer-based polls.
|
|
|
Set this to "0" to disable timer based polls.
|
Set this to "0" to disable timer based polls.
|
Initialization of cards without a working interrupt
|
Initialization of cards without a working interrupt
|
will fail.
|
will fail.
|
|
|
Set this to "1" if you want a polling driver.
|
Set this to "1" if you want a polling driver.
|
(on Intel: 100 polls per second). If you don't use
|
(on Intel: 100 polls per second). If you don't use
|
fast baud rates, you might consider a value like "5".
|
fast baud rates, you might consider a value like "5".
|
(If you don't know how to do the math, use 1).
|
(If you don't know how to do the math, use 1).
|
|
|
sx_slowpoll: Number of jiffies between timer-based polls.
|
sx_slowpoll: Number of jiffies between timer-based polls.
|
Set this to "100" to poll once a second.
|
Set this to "100" to poll once a second.
|
This should get the card out of a stall if the driver
|
This should get the card out of a stall if the driver
|
ever misses an interrupt. I've never seen this happen,
|
ever misses an interrupt. I've never seen this happen,
|
and if it does, that's a bug. Tell me.
|
and if it does, that's a bug. Tell me.
|
|
|
sx_maxints: Number of interrupts to request from the card.
|
sx_maxints: Number of interrupts to request from the card.
|
The card normally limits interrupts to about 100 per
|
The card normally limits interrupts to about 100 per
|
second to offload the host CPU. You can increase this
|
second to offload the host CPU. You can increase this
|
number to reduce latency on the card a little.
|
number to reduce latency on the card a little.
|
Note that if you give a very high number you can overload
|
Note that if you give a very high number you can overload
|
your CPU as well as the CPU on the host card. This setting
|
your CPU as well as the CPU on the host card. This setting
|
is inaccurate and not recommended for SI cards (But it
|
is inaccurate and not recommended for SI cards (But it
|
works).
|
works).
|
|
|
sx_irqmask: The mask of allowable IRQs to use. I suggest you set
|
sx_irqmask: The mask of allowable IRQs to use. I suggest you set
|
this to 0 (disable IRQs all together) and use polling if
|
this to 0 (disable IRQs all together) and use polling if
|
the assignment of IRQs becomes problematic. This is defined
|
the assignment of IRQs becomes problematic. This is defined
|
as the sum of (1 << irq) 's that you want to allow. So
|
as the sum of (1 << irq) 's that you want to allow. So
|
sx_irqmask of 8 (1 << 3) specifies that only irq 3 may
|
sx_irqmask of 8 (1 << 3) specifies that only irq 3 may
|
be used by the SX driver. If you want to specify to the
|
be used by the SX driver. If you want to specify to the
|
driver: "Either irq 11 or 12 is ok for you to use", then
|
driver: "Either irq 11 or 12 is ok for you to use", then
|
specify (1 << 11) | (1 << 12) = 0x1800 .
|
specify (1 << 11) | (1 << 12) = 0x1800 .
|
|
|
sx_debug: You can enable different sorts of debug traces with this.
|
sx_debug: You can enable different sorts of debug traces with this.
|
At "-1" all debugging traces are active. You'll get several
|
At "-1" all debugging traces are active. You'll get several
|
times more debugging output than you'll get characters
|
times more debugging output than you'll get characters
|
transmitted.
|
transmitted.
|
|
|
|
|
Baud rates
|
Baud rates
|
==========
|
==========
|
|
|
Theoretically new SXDCs should be capable of more than 460k
|
Theoretically new SXDCs should be capable of more than 460k
|
baud. However the line drivers usually give up before that. Also the
|
baud. However the line drivers usually give up before that. Also the
|
CPU on the card may not be able to handle 8 channels going at full
|
CPU on the card may not be able to handle 8 channels going at full
|
blast at that speed. Moreover, the buffers are not large enough to
|
blast at that speed. Moreover, the buffers are not large enough to
|
allow operation with 100 interrupts per second. You'll have to realize
|
allow operation with 100 interrupts per second. You'll have to realize
|
that the card has a 256 byte buffer, so you'll have to increase the
|
that the card has a 256 byte buffer, so you'll have to increase the
|
number of interrupts per second if you have more than 256*100 bytes
|
number of interrupts per second if you have more than 256*100 bytes
|
per second to transmit. If you do any performance testing in this
|
per second to transmit. If you do any performance testing in this
|
area, I'd be glad to hear from you...
|
area, I'd be glad to hear from you...
|
|
|
(Psst Linux users..... I think the Linux driver is more efficient than
|
(Psst Linux users..... I think the Linux driver is more efficient than
|
the driver for other OSes. If you can and want to benchmark them
|
the driver for other OSes. If you can and want to benchmark them
|
against each other, be my guest, and report your findings...... :-)
|
against each other, be my guest, and report your findings...... :-)
|
|
|
|
|
Ports and devices
|
Ports and devices
|
=================
|
=================
|
|
|
Port 0 is the top connector on the module closest to the host
|
Port 0 is the top connector on the module closest to the host
|
card. Oh, the ports on the SXDCs and TAs are labelled from 1 to 8
|
card. Oh, the ports on the SXDCs and TAs are labelled from 1 to 8
|
instead of from 0 to 7, as they are numbered by linux. I'm stubborn in
|
instead of from 0 to 7, as they are numbered by linux. I'm stubborn in
|
this: I know for sure that I wouldn't be able to calculate which port
|
this: I know for sure that I wouldn't be able to calculate which port
|
is which anymore if I would change that....
|
is which anymore if I would change that....
|
|
|
|
|
Devices:
|
Devices:
|
|
|
You should make the device files as follows:
|
You should make the device files as follows:
|
|
|
#!/bin/sh
|
#!/bin/sh
|
# (I recommend that you cut-and-paste this into a file and run that)
|
# (I recommend that you cut-and-paste this into a file and run that)
|
cd /dev
|
cd /dev
|
t=0
|
t=0
|
mknod specialix_sxctl c 10 167
|
mknod specialix_sxctl c 10 167
|
while [ $t -lt 64 ]
|
while [ $t -lt 64 ]
|
do
|
do
|
echo -n "$t "
|
echo -n "$t "
|
mknod ttyX$t c 32 $t
|
mknod ttyX$t c 32 $t
|
mknod cux$t c 33 $t
|
mknod cux$t c 33 $t
|
t=`expr $t + 1`
|
t=`expr $t + 1`
|
done
|
done
|
echo ""
|
echo ""
|
rm /etc/psdevtab
|
rm /etc/psdevtab
|
ps > /dev/null
|
ps > /dev/null
|
|
|
|
|
This creates 64 devices. If you have more, increase the constant on
|
This creates 64 devices. If you have more, increase the constant on
|
the line with "while". The devices start at 0, as is customary on
|
the line with "while". The devices start at 0, as is customary on
|
Linux. Specialix seems to like starting the numbering at 1.
|
Linux. Specialix seems to like starting the numbering at 1.
|
|
|
If your system doesn't come with these devices pre-installed, bug your
|
If your system doesn't come with these devices pre-installed, bug your
|
linux-vendor about this. They should have these devices
|
linux-vendor about this. They should have these devices
|
"pre-installed" before the new millennium. The "ps" stuff at the end
|
"pre-installed" before the new millennium. The "ps" stuff at the end
|
is to "tell" ps that the new devices exist.
|
is to "tell" ps that the new devices exist.
|
|
|
Officially the maximum number of cards per computer is 4. This driver
|
Officially the maximum number of cards per computer is 4. This driver
|
however supports as many cards in one machine as you want. You'll run
|
however supports as many cards in one machine as you want. You'll run
|
out of interrupts after a few, but you can switch to polled operation
|
out of interrupts after a few, but you can switch to polled operation
|
then. At about 256 ports (More than 8 cards), we run out of minor
|
then. At about 256 ports (More than 8 cards), we run out of minor
|
device numbers. Sorry. I suggest you buy a second computer.... (Or
|
device numbers. Sorry. I suggest you buy a second computer.... (Or
|
switch to RIO).
|
switch to RIO).
|
|
|
------------------------------------------------------------------------
|
------------------------------------------------------------------------
|
|
|
|
|
Fixed bugs and restrictions:
|
Fixed bugs and restrictions:
|
- Hangup processing.
|
- Hangup processing.
|
-- Done.
|
-- Done.
|
|
|
- the write path in generic_serial (lockup / oops).
|
- the write path in generic_serial (lockup / oops).
|
-- Done (Ugly: not the way I want it. Copied from serial.c).
|
-- Done (Ugly: not the way I want it. Copied from serial.c).
|
|
|
- write buffer isn't flushed at close.
|
- write buffer isn't flushed at close.
|
-- Done. I still seem to loose a few chars at close.
|
-- Done. I still seem to loose a few chars at close.
|
Sorry. I think that this is a firmware issue. (-> Specialix)
|
Sorry. I think that this is a firmware issue. (-> Specialix)
|
|
|
- drain hardware before changing termios
|
- drain hardware before changing termios
|
- Change debug on the fly.
|
- Change debug on the fly.
|
- ISA free irq -1. (no firmware loaded).
|
- ISA free irq -1. (no firmware loaded).
|
- adding c8000 as a probe address. Added warning.
|
- adding c8000 as a probe address. Added warning.
|
- Add a RAMtest for the RAM on the card.c
|
- Add a RAMtest for the RAM on the card.c
|
- Crash when opening a port "way" of the number of allowed ports.
|
- Crash when opening a port "way" of the number of allowed ports.
|
(for example opening port 60 when there are only 24 ports attached)
|
(for example opening port 60 when there are only 24 ports attached)
|
- Sometimes the use-count strays a bit. After a few hours of
|
- Sometimes the use-count strays a bit. After a few hours of
|
testing the use count is sometimes "3". If you are not like
|
testing the use count is sometimes "3". If you are not like
|
me and can remember what you did to get it that way, I'd
|
me and can remember what you did to get it that way, I'd
|
appreciate an Email. Possibly fixed. Tell me if anyone still
|
appreciate an Email. Possibly fixed. Tell me if anyone still
|
sees this.
|
sees this.
|
- TAs don't work right if you don't connect all the modem control
|
- TAs don't work right if you don't connect all the modem control
|
signals. SXDCs do. T225 firmware problem -> Specialix.
|
signals. SXDCs do. T225 firmware problem -> Specialix.
|
(Mostly fixed now, I think. Tell me if you encounter this!)
|
(Mostly fixed now, I think. Tell me if you encounter this!)
|
|
|
Bugs & restrictions:
|
Bugs & restrictions:
|
|
|
- Arbitrary baud rates. Requires firmware update. (-> Specialix)
|
- Arbitrary baud rates. Requires firmware update. (-> Specialix)
|
|
|
- Low latency (mostly firmware, -> Specialix)
|
- Low latency (mostly firmware, -> Specialix)
|
|
|
|
|
|
|
|
|
