[LinuxBIOS] Add support for Via vt82c686a

[Corey Osgood]

Okay, I'm hoping that I'm doing this right. This is basically just a 
hacked up version of the vt8231 to support the vt82c686a. I can't 
confirm that it works entirely, but it does work far enough to provide 
serial output. I've added what I hope are the correct GPL headers.

Signed off by: Corey Osgood

[Uwe Hermann]

Hi,

On Sun, Jan 14, 2007 at 03:57:28PM -0500, Corey Osgood wrote:
> Okay, I'm hoping that I'm doing this right. This is basically just a 
> hacked up version of the vt8231 to support the vt82c686a. I can't 
> confirm that it works entirely, but it does work far enough to provide 
> serial output. I've added what I hope are the correct GPL headers.
>
> Signed off by: Corey Osgood

Sign-off-by, see
http://linuxbios.org/Development_Guidelines#Sign-off_Procedure

Most files are missing the GPL header, but every file needs one.
You need to check the svn history and add the proper copyright lines, as
the original files you based your code on don't have the header either.

Let me know if you need help, I can provide patches for this if you want...
> Property changes on: src/southbridge/via/vt82c686a/vt82c686a_early_smbus.c
> ___________________________________________________________________
> Name: svn:executable

Not necessary, these files won't be executed.

HTH, Uwe.

[Corey Osgood]

In trac, who are ebiederm and mwrwilkinson? I can't seem to find names 
to go with handles...

-Corey

[David H. Barr]

On 1/14/07, Corey Osgood  wrote:
> In trac, who are ebiederm and mwrwilkinson? I can't seem to find names
> to go with handles...

ebiederm has got to be Eric W. Beiderman.  Last I remember he was
ebiederman at lnxi dot com.

-dhbarr.

[Peter Stuge]

On Sun, Jan 14, 2007 at 10:24:45PM -0500, Corey Osgood wrote:
> Thanks! And mwrwilkinson appears to be Mark Wilkinson. I found an old
>    sourceforge email for him, so I've used that for now, if anyone has an
>    updated email for him I'll fix it.

My archive shows [e-mail] on 07 Sep 2006, but I
can't say which address he would prefer.
> References
>
> Visible links
> mailto:[e-mail]
> mailto:[e-mail]

Please avoid sending HTML emails to the list. Thanks! :)

//Peter

[Mark Wilkinson]

Hi Peter,
The address I'd prefer is '[e-mail]' - com will work 
but takes a longer route inside my company's e-mail system :-)

I'm not even sure the sourceforge one is still working for me.

Regards
Mark.

Peter Stuge wrote:
> On Sun, Jan 14, 2007 at 10:24:45PM -0500, Corey Osgood wrote:
> > Thanks! And mwrwilkinson appears to be Mark Wilkinson. I found an old
> >   sourceforge email for him, so I've used that for now, if anyone has an
> >   updated email for him I'll fix it.
>
> My archive shows [e-mail] on 07 Sep 2006, but I
> can't say which address he would prefer.
> > References
> >
> > Visible links
> > mailto:[e-mail]
> > mailto:[e-mail]
>
> Please avoid sending HTML emails to the list. Thanks! :)
>
> //Peter

[Segher Boessenkool]

> > Sign-off-by, see
> > http://linuxbios.org/Development_Guidelines#Sign-off_Procedure

Signed-off-by: even.
> As far as the GPL headers go, should everyone who's ever made a change 
> to that file be included? I'm assuming yes, and working on 
> straightening it out.

Only people who made substantive, non-mechanical changes need
to be listed, other changes aren't copyrightable anyway.  You
can of course list everyone anyway, heh.

Segher

[Uwe Hermann]

Hi,

On Mon, Jan 15, 2007 at 04:57:59PM +0100, Segher Boessenkool wrote:
> > > Sign-off-by, see
> > > http://linuxbios.org/Development_Guidelines#Sign-off_Procedure
>
> Signed-off-by: even.

Yep, typo on my side, sorry.
> > As far as the GPL headers go, should everyone who's ever made a change 
> > to that file be included? I'm assuming yes, and working on 
> > straightening it out.
>
> Only people who made substantive, non-mechanical changes need
> to be listed, other changes aren't copyrightable anyway.  You
> can of course list everyone anyway, heh.

Yeah, everyone who contributed a non-trivial amount of code, something
like 4 or more (non-trivial) lines of code maybe...
Merely fixing typos etc. doesn't make you a copyright holder
on (part of) the code.

Cheers, Uwe.

[Corey Osgood]

Okay, looking further into the archives, and into the very /origins/ of 
these files, in the original freebios cvs (on sourceforge), it appears 
that there were a lot more people involved than I've included. This 
patch has everyone whos made any change to the linuxbios svn whatsoever, 
I don't think any of them are too trivial, but I haven't got the time to 
go back and check right now. Most of that content has, over time, been 
rewritten, but some of it's still intact (mainly in the vtX_ide.c). If 
someone wants to trace all the changes, be my guest, but I haven't got 
the time right now. I've updated Mark's email, thanks for getting ahold 
of him.  Also, by Ron's definition, I probably don't even qualify as a 
copyright holder...if anyone has a problem with including myself, let me 
know and I am completely willing to remove it (or it can be removed by 
someone else), as I don't think I've written any actual code, just 
changed a few variables here and there.

Signed-off-by: Corey Osgood

[Adam Talbot]

Do we need to update the website to reflect the state of the the via 
vt82c686a, under the "Supported Chipsets and Devices"
-Adam

Corey Osgood wrote:
> Okay, looking further into the archives, and into the very /origins/ 
> of these files, in the original freebios cvs (on sourceforge), it 
> appears that there were a lot more people involved than I've included. 
> This patch has everyone whos made any change to the linuxbios svn 
> whatsoever, I don't think any of them are too trivial, but I haven't 
> got the time to go back and check right now. Most of that content has, 
> over time, been rewritten, but some of it's still intact (mainly in 
> the vtX_ide.c). If someone wants to trace all the changes, be my 
> guest, but I haven't got the time right now. I've updated Mark's 
> email, thanks for getting ahold of him.  Also, by Ron's definition, I 
> probably don't even qualify as a copyright holder...if anyone has a 
> problem with including myself, let me know and I am completely willing 
> to remove it (or it can be removed by someone else), as I don't think 
> I've written any actual code, just changed a few variables here and 
> there.
> Signed-off-by: Corey Osgood
> Index: src/southbridge/via/vt82c686a/vt82c686a_early_smbus.c
>
> --- src/southbridge/via/vt82c686a/vt82c686a_early_smbus.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a_early_smbus.c	(revision 0)
> @@ -0,0 +1,317 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2003 Ron Minnich
> Copyright (C) 2003 Eric W. Beiderman
> Copyright (C) 2004 Mark Wilkinson
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +#define SMBUS_IO_BASE 0x5000
>
> +#define SMBHSTSTAT 0x0
> +#define SMBSLVSTAT 0x1
> +#define SMBHSTCTL  0x2
> +#define SMBHSTCMD  0x3
> +#define SMBXMITADD 0x4
> +#define SMBHSTDAT0 0x5
> +#define SMBHSTDAT1 0x6
> +#define SMBBLKDAT  0x7
> +#define SMBSLVCTL  0x8
> +#define SMBTRNSADD 0x9
> +#define SMBSLVDATA 0xa
> +#define SMLINK_PIN_CTL 0xe
> +#define SMBUS_PIN_CTL  0xf
>
> +/* Define register settings */
> +#define HOST_RESET 0xff
> +#define DIMM_BASE 0xa0		// 1010000 is base for DIMM in SMBus
> +#define READ_CMD  0x01		// 1 in the 0 bit of SMBHSTADD states to READ
>
> +#define SMBUS_TIMEOUT (100*1000*10)
>
> +static void enable_smbus(void)
> +{
> device_t dev;
> unsigned char c;
> /* Power management controller */
> dev = pci_locate_device(PCI_ID(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4), 0);
> if (dev == PCI_DEV_INVALID) {
> die("SMBUS controller not found\r\n");
> }
> // set IO base address to SMBUS_IO_BASE
> pci_write_config32(dev, 0x90, SMBUS_IO_BASE | 1);
> // Enable SMBus
> c = pci_read_config8(dev, 0xd2);
> c |= 5;
> pci_write_config8(dev, 0xd2, c);
> /* make it work for I/O ...
> */
> dev = pci_locate_device(PCI_ID(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4), 0);
> c = pci_read_config8(dev, 4);
> c |= 1;
> pci_write_config8(dev, 4, c);
> print_debug_hex8(c);
> print_debug(" is the comm register\r\n");
> print_debug("SMBus controller enabled\r\n");
>
> +}
>
> +static inline void smbus_delay(void)
> +{
> outb(0x80, 0x80);
>
> +}
>
> +static int smbus_wait_until_active(void)
> +{
> unsigned long loops;
> loops = SMBUS_TIMEOUT;
> do {
> unsigned char val;
> smbus_delay();
> val = inb(SMBUS_IO_BASE + SMBHSTSTAT);
> if ((val & 1)) {
> 	break;
> }
> } while (--loops);
> return loops ? 0 : -4;
>
> +}
>
> +static int smbus_wait_until_ready(void)
> +{
> unsigned long loops;
> loops = SMBUS_TIMEOUT;
> do {
> unsigned char val;
> smbus_delay();
> val = inb(SMBUS_IO_BASE + SMBHSTSTAT);
> if ((val & 1) == 0) {
> 	break;
> }
> if (loops == (SMBUS_TIMEOUT / 2)) {
> 	outb(inb(SMBUS_IO_BASE + SMBHSTSTAT), SMBUS_IO_BASE + SMBHSTSTAT);
> }
> } while (--loops);
> return loops ? 0 : -2;
>
> +}
>
> +static int smbus_wait_until_done(void)
> +{
> unsigned long loops;
> loops = SMBUS_TIMEOUT;
> do {
> unsigned char val;
> smbus_delay();
> val = inb(SMBUS_IO_BASE + SMBHSTSTAT);
> if ((val & 1) == 0) {
> 	break;
> }
> } while (--loops);
> return loops ? 0 : -3;
>
> +}
>
> +void smbus_reset(void)
> +{
> outb(HOST_RESET, SMBUS_IO_BASE + SMBHSTSTAT);
> outb(HOST_RESET, SMBUS_IO_BASE + SMBHSTSTAT);
> outb(HOST_RESET, SMBUS_IO_BASE + SMBHSTSTAT);
> outb(HOST_RESET, SMBUS_IO_BASE + SMBHSTSTAT);
> smbus_wait_until_ready();
> print_debug("After reset status ");
> print_debug_hex8(inb(SMBUS_IO_BASE + SMBHSTSTAT));
> print_debug("\r\n");
>
> +}
>
> +static void smbus_print_error(unsigned char host_status_register)
> +{
> print_err("smbus_error: ");
> print_err_hex8(host_status_register);
> print_err("\r\n");
> if (host_status_register & (1 << 4)) {
> print_err("Interrup/SMI# was Failed Bus Transaction\r\n");
> }
> if (host_status_register & (1 << 3)) {
> print_err("Bus Error\r\n");
> }
> if (host_status_register & (1 << 2)) {
> print_err("Device Error\r\n");
> }
> if (host_status_register & (1 << 1)) {
> print_err("Interrupt/SMI# was Successful Completion\r\n");
> }
> if (host_status_register & (1 << 0)) {
> print_err("Host Busy\r\n");
> }
>
> +}
>
> +/*
> Copied from intel/i82801dbm early smbus code - suggested by rgm.
> Modifications/check against i2c-viapro driver code from linux-2.4.22
> and VT8231 Reference Docs - mw.
> */
>
> +static int smbus_read_byte(unsigned device, unsigned address)
> +{
> unsigned char global_control_register;
> unsigned char global_status_register;
> unsigned char byte;
> if (smbus_wait_until_ready() < 0) {
> outb(inb(SMBUS_IO_BASE + SMBHSTSTAT), SMBUS_IO_BASE + SMBHSTSTAT);
> if (smbus_wait_until_ready() < 0) {
> 	return -2;
> }
> }
> /* setup transaction */
> /* disable interrupts */
> outb(inb(SMBUS_IO_BASE + SMBHSTCTL) & 0xfe, SMBUS_IO_BASE + SMBHSTCTL);
> /* set the device I'm talking too */
> outb(((device & 0x7f) << 1) | 1, SMBUS_IO_BASE + SMBXMITADD);
> /* set the command/address... */
> outb(address & 0xFF, SMBUS_IO_BASE + SMBHSTCMD);
> /* set up for a byte data read */
> outb((inb(SMBUS_IO_BASE + SMBHSTCTL) & 0xe3) | (0x2 << 2), SMBUS_IO_BASE + SMBHSTCTL);
> /* clear any lingering errors, so the transaction will run */
> outb(inb(SMBUS_IO_BASE + SMBHSTSTAT), SMBUS_IO_BASE + SMBHSTSTAT);
> /* clear the data byte... */
> outb(0, SMBUS_IO_BASE + SMBHSTDAT0);
> /* start a byte read, with interrupts disabled */
> outb((inb(SMBUS_IO_BASE + SMBHSTCTL) | 0x40), SMBUS_IO_BASE + SMBHSTCTL);
> /* poll for it to start */
> if (smbus_wait_until_active() < 0) {
> return -4;
> }
> /* poll for transaction completion */
> if (smbus_wait_until_done() < 0) {
> return -3;
> }
> /* Ignore the Host Busy & Command Complete ? */
> global_status_register = inb(SMBUS_IO_BASE + SMBHSTSTAT) & ~((1 << 1) | (1 << 0));
> /* read results of transaction */
> byte = inb(SMBUS_IO_BASE + SMBHSTDAT0);
> if (global_status_register != 0) {
> return -1;
> }
> return byte;
>
> +}
>
> +#if 0
> +/* SMBus routines borrowed from VIA's Trident Driver */
> +/* this works, so I am not going to touch it for now -- rgm */
> +static unsigned char smbus_read_byte(unsigned char devAdr, unsigned char bIndex)
> +{
> unsigned int i;
> unsigned char bData;
> unsigned char sts = 0;
> /* clear host status */
> outb(0xff, SMBUS_IO_BASE);
> /* check SMBUS ready */
> for (i = 0; i < SMBUS_TIMEOUT; i++)
> if ((inb(SMBUS_IO_BASE) & 0x01) == 0)
> 	break;
> /* set host command */
> outb(bIndex, SMBUS_IO_BASE + 3);
> /* set slave address */
> outb(devAdr | 0x01, SMBUS_IO_BASE + 4);
> /* start */
> outb(0x48, SMBUS_IO_BASE + 2);
> /* SMBUS Wait Ready */
> for (i = 0; i < SMBUS_TIMEOUT; i++)
> if (((sts = inb(SMBUS_IO_BASE)) & 0x01) == 0)
> 	break;
> if ((sts & ~3) != 0) {
> smbus_print_error(sts);
> return 0;
> }
> bData = inb(SMBUS_IO_BASE + 5);
> return bData;
>
> +}
> +#endif
> +/* for reference, here is the fancier version which we will use at some
> point
> */
>
> +# if 0
> +int smbus_read_byte(unsigned device, unsigned address, unsigned char *result)
> +{
> unsigned char host_status_register;
> unsigned char byte;
> reset();
> smbus_wait_until_ready();
> /* setup transaction */
> /* disable interrupts */
> outb(inb(SMBUS_IO_BASE + SMBHSTCTL) & (~1), SMBUS_IO_BASE + SMBHSTCTL);
> /* set the device I'm talking too */
> outb(((device & 0x7f) << 1) | 1, SMBUS_IO_BASE + SMBXMITADD);
> /* set the command/address... */
> outb(address & 0xFF, SMBUS_IO_BASE + SMBHSTCMD);
> /* set up for a byte data read */
> outb((inb(SMBUS_IO_BASE + SMBHSTCTL) & 0xE3) | (0x2 << 2), SMBUS_IO_BASE + SMBHSTCTL);
> /* clear any lingering errors, so the transaction will run */
> outb(inb(SMBUS_IO_BASE + SMBHSTSTAT), SMBUS_IO_BASE + SMBHSTSTAT);
> /* clear the data byte... */
> outb(0, SMBUS_IO_BASE + SMBHSTDAT0);
> /* start the command */
> outb((inb(SMBUS_IO_BASE + SMBHSTCTL) | 0x40), SMBUS_IO_BASE + SMBHSTCTL);
> /* poll for transaction completion */
> smbus_wait_until_done();
> host_status_register = inb(SMBUS_IO_BASE + SMBHSTSTAT);
> /* Ignore the In Use Status... */
> host_status_register &= ~(1 << 6);
> /* read results of transaction */
> byte = inb(SMBUS_IO_BASE + SMBHSTDAT0);
> smbus_print_error(byte);
> *result = byte;
> return host_status_register != 0x02;
>
> +}
>
> +#endif
> Index: src/southbridge/via/vt82c686a/Config.lb
> ===================================================================
> --- src/southbridge/via/vt82c686a/Config.lb	(revision 0)
> +++ src/southbridge/via/vt82c686a/Config.lb	(revision 0)
> @@ -0,0 +1,8 @@
> +config chip.h
> +driver vt82c686a.o
> +driver vt82c686a_lpc.o
> +driver vt82c686a_acpi.o
> +driver vt82c686a_ide.o
> +#driver vt82c686a_nic.o
> +#driver vt82c686a_usb.o
> Index: src/southbridge/via/vt82c686a/vt82c686a_nic.c
>
> --- src/southbridge/via/vt82c686a/vt82c686a_nic.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a_nic.c	(revision 0)
> @@ -0,0 +1,62 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +#include <console/console.h>
> +#include <device/device.h>
> +#include <device/pci.h>
> +#include <device/pci_ops.h>
> +#include <device/pci_ids.h>
>
> +/*
> Enable the ethernet device and turn off stepping (because it is integrated
> inside the southbridge)
> Since my vt82c686 doesn't have ethernet, this will fail. but it will be fun
> to watch */
>
> +static void nic_init(struct device *dev)
> +{
> uint8_t		byte;
> printk_debug("Configuring VIA LAN\n");
> /* We don't need stepping - though the device supports it */
> byte = pci_read_config8(dev, PCI_COMMAND);
> byte &= ~PCI_COMMAND_WAIT;
> pci_write_config8(dev, PCI_COMMAND, byte);
>
> +}
>
> +static struct device_operations nic_ops = {
> .read_resources   = pci_dev_read_resources,
> .set_resources    = pci_dev_set_resources,
> .enable_resources = pci_dev_enable_resources,
> .init		  = nic_init,
> .enable           = 0,
> .ops_pci          = 0,
>
> +};
>
> +/* If anyone ever has a vt82c686a with integrated nic, they'll need to modify
> this deviceid to make it work */
>
> +static struct pci_driver northbridge_driver __pci_driver = {
> .ops	= &nic_ops,
> .vendor = PCI_VENDOR_ID_VIA,
> .device = PCI_DEVICE_ID_VIA_8233_7,
>
> +};
> Index: src/southbridge/via/vt82c686a/vt82c686a_usb.c
> ===================================================================
> --- src/southbridge/via/vt82c686a/vt82c686a_usb.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a_usb.c	(revision 0)
> @@ -0,0 +1,72 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +static void usb_on(int enable)
> +{
> unsigned char regval;
> /* Base 82c686 controller */
> device_t dev0 = dev_find_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686, 0);
> /* USB controller 1 */
> device_t dev2 = dev_find_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_2, 0);
> /* USB controller 2 */
> device_t dev3 = dev_find_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_2, dev2);
> /* enable USB1 */
> if(dev2) {
> if (enable) {
> 	pci_write_config8(dev2, 0x3c, 0x05);
> 	pci_write_config8(dev2, 0x04, 0x07);
> } else {
> 	pci_write_config8(dev2, 0x3c, 0x00);
> 	pci_write_config8(dev2, 0x04, 0x00);
> }
> }
> if(dev0) {
> regval = pci_read_config8(dev0, 0x50);
> if (enable) 
> 	regval &= ~(0x10);    
> else
> 	regval |= 0x10;    	      
> pci_write_config8(dev0, 0x50, regval);
> }
> /* enable USB2 */
> if(dev3) {
> if (enable) {
> 	pci_write_config8(dev3, 0x3c, 0x05);
> 	pci_write_config8(dev3, 0x04, 0x07);
> } else {
> 	pci_write_config8(dev3, 0x3c, 0x00);
> 	pci_write_config8(dev3, 0x04, 0x00);
> }
> }
> if(dev0) {
> regval = pci_read_config8(dev0, 0x50);
> if (enable) 
> 	regval &= ~(0x20);    
> else
> 	regval |= 0x20;    
> pci_write_config8(dev0, 0x50, regval);
> }
>
> +}
> Index: src/southbridge/via/vt82c686a/vt82c686a_early_serial.c
> ===================================================================
> --- src/southbridge/via/vt82c686a/vt82c686a_early_serial.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a_early_serial.c	(revision 0)
> @@ -0,0 +1,97 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2003 Ron Minnich
> Copyright (C) 2003 Eric W. Beiderman
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +/*
> Enable the serial evices on the VIA
> */
>
> +/* The base address is 0x15c, 0x2e, depending on config bytes */
> +#include <device/pci_ids.h>
> +#define SIO_BASE 0x3f0
> +#define SIO_DATA  SIO_BASE+1
>
> +static void vt82c686a_writesuper(uint8_t reg, uint8_t val) 
> +{
> outb(reg, SIO_BASE);
> outb(val, SIO_DATA);
>
> +}
>
> +static void vt82c686a_writesiobyte(uint16_t reg, uint8_t val) 
> +{
> outb(val, reg);
>
> +}
>
> +static void vt82c686a_writesioword(uint16_t reg, uint16_t val) 
> +{
> outw(val, reg);
>
> +}
>
> +/* regs we use: 85, and the southbridge devfn is defined by the
> mainboard
> */
>
> +static void enable_vt82c686a_serial(void) 
> +{
> unsigned long x;
> uint8_t c;
> device_t dev;
> outb(6, 0x80);
> dev = pci_locate_device(PCI_ID(PCI_VENDOR_ID_VIA,PCI_DEVICE_ID_VIA_82C686), 0);
> if (dev == PCI_DEV_INVALID) {
> outb(7, 0x80);
> die("Serial controller not found\r\n"); //umm, pointless???
> }
> /* first, you have to enable the superio and superio config.
>   put a 6 reg 80
> */
> c = pci_read_config8(dev, 0x50);
> c |= 6;
> pci_write_config8(dev, 0x50, c);
> outb(2, 0x80);
> // now go ahead and set up com1.
> // set address
> vt82c686a_writesuper(0xf4, 0xfe);
> // enable serial out
> vt82c686a_writesuper(0xf2, 7);
> // That's it for the sio stuff.
> //	movl	$SUPERIOCONFIG, %eax
> //	movb	$9, %dl
> //	PCI_WRITE_CONFIG_BYTE
> // set up reg to set baud rate.
> vt82c686a_writesiobyte(0x3fb, 0x80);
> // Set 115 kb
> vt82c686a_writesioword(0x3f8, 1);
> // Set 9.6 kb
> //	WRITESIOWORD(0x3f8, 12)
> // now set no parity, one stop, 8 bits
> vt82c686a_writesiobyte(0x3fb, 3);
> // now turn on RTS, DRT
> vt82c686a_writesiobyte(0x3fc, 3);
> // Enable interrupts
> vt82c686a_writesiobyte(0x3f9, 0xf);
> // should be done. Dump a char for fun.
> vt82c686a_writesiobyte(0x3f8, 48);
>
> +}
> Index: src/southbridge/via/vt82c686a/vt82c686a_acpi.c
> ===================================================================
> --- src/southbridge/via/vt82c686a/vt82c686a_acpi.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a_acpi.c	(revision 0)
> @@ -0,0 +1,64 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +#include <console/console.h>
> +#include <device/device.h>
> +#include <device/pci.h>
> +#include <device/pci_ops.h>
> +#include <device/pci_ids.h>
>
> +static void acpi_init(struct device *dev)
> +{
> printk_debug("Configuring VIA ACPI\n");
> // Set ACPI base address to IO 0x4000
> pci_write_config32(dev, 0x48, 0x4001);
> // Enable ACPI access (and setup like award)
> pci_write_config8(dev, 0x41, 0x84);
> // Set hardware monitor base address to IO 0x6000
> pci_write_config32(dev, 0x70, 0x6001);
> // Enable hardware monitor (and setup like award)
> pci_write_config8(dev, 0x74, 0x01);
> // set IO base address to 0x5000
> pci_write_config32(dev, 0x90, 0x5001);
> // Enable SMBus
> pci_write_config8(dev, 0xd2, 0x01);
>
> +}
>
> +static struct device_operations acpi_ops = {
> .read_resources   = pci_dev_read_resources,
> .set_resources    = pci_dev_set_resources,
> .enable_resources = pci_dev_enable_resources,
> .init		  = acpi_init,
> .enable           = 0,
> .ops_pci          = 0,
>
> +};
>
> +static struct pci_driver northbridge_driver __pci_driver = {
> .ops	= &acpi_ops,
> .vendor = PCI_VENDOR_ID_VIA,
> .device = PCI_DEVICE_ID_VIA_82C686_4, /* For 686b, this is different (3057)*/
>
> +};
> Index: src/southbridge/via/vt82c686a/chip.h
> ===================================================================
> --- src/southbridge/via/vt82c686a/chip.h	(revision 0)
> +++ src/southbridge/via/vt82c686a/chip.h	(revision 0)
> @@ -0,0 +1,41 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2003 Ron Minnich
> Copyright (C) 2003 Eric W. Beiderman
> Copyright (C) 2004 Mark Wilkinson
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2005 Stefan Reinauer
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +#ifndef _SOUTHBRIDGE_VIA_VT82C686A
> +#define _SOUTHBRIDGE_VIA_VT82C686A
>
> +extern struct chip_operations southbridge_via_vt82c686a_ops;
>
> +struct southbridge_via_vt82c686a_config {
> /* enables of Non-PCI devices */
> int enable_native_ide;
> int enable_com_ports;
> int enable_keyboard;
> /* currently not parsed but needed by densitron dpx114 */
> int enable_usb;
> int enable_nvram;
>
> +};
>
> +#endif /* _SOUTHBRIDGE_VIA_VT82C686A */
> Index: src/southbridge/via/vt82c686a/vt82c686a_lpc.c
> ===================================================================
> --- src/southbridge/via/vt82c686a/vt82c686a_lpc.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a_lpc.c	(revision 0)
> @@ -0,0 +1,174 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +#include <console/console.h>
> +#include <device/device.h>
> +#include <device/pci.h>
> +#include <device/pci_ops.h>
> +#include <device/pci_ids.h>
>
> +#include <pc80/mc146818rtc.h>
>
> +#include "chip.h"
>
> +/* PIRQ init
> */
>
> +void pci_assign_irqs(unsigned bus, unsigned slot, const unsigned char pIntAtoD[4]);
> +static const unsigned char southbridgeIrqs[4] = { 11, 5, 10, 12 };
> +static const unsigned char enetIrqs[4] = { 11, 5, 10, 12 };
> +static const unsigned char slotIrqs[4] = { 5, 10, 12, 11 };
>
> +/*
> Our IDSEL mappings are as follows
> PCI slot is AD31          (device 15) (00:14.0)
> Southbridge is AD28       (device 12) (00:11.0)
>
> +*/
> +static void pci_routing_fixup(struct device *dev)
> +{
> printk_info("%s: dev is %p\n", __FUNCTION__, dev);
> if (dev) {
> /* initialize PCI interupts - these assignments depend
>    on the PCB routing of PINTA-D 
>    PINTA = IRQ11
>    PINTB = IRQ5
>    PINTC = IRQ10
>    PINTD = IRQ12
> */
> pci_write_config8(dev, 0x55, 0xb0);
> pci_write_config8(dev, 0x56, 0xa5);
> pci_write_config8(dev, 0x57, 0xc0);
> }
> // Standard southbridge components
> printk_info("setting southbridge\n");
> pci_assign_irqs(0, 0x11, southbridgeIrqs);
> // Ethernet built into southbridge
> printk_info("setting ethernet\n");
> pci_assign_irqs(0, 0x12, enetIrqs);
> // PCI slot
> printk_info("setting pci slot\n");
> pci_assign_irqs(0, 0x14, slotIrqs);
> printk_info("%s: DONE\n", __FUNCTION__);
>
> +}
>
> +static void vt82c686_init(struct device *dev)
> +{
> unsigned char enables;
> struct southbridge_via_vt82c686a_config *conf = dev->chip_info;
> printk_debug("vt82c686a init\n");
> // enable the internal I/O decode
> enables = pci_read_config8(dev, 0x6C);
> enables |= 0x80;
> pci_write_config8(dev, 0x6C, enables);
> // Map 4MB of FLASH into the address space
> pci_write_config8(dev, 0x41, 0x7f);
> // Set bit 6 of 0x40, because Award does it (IO recovery time)
> // IMPORTANT FIX - EISA 0x4d0 decoding must be on so that PCI
> // interrupts can be properly marked as level triggered.
> enables = pci_read_config8(dev, 0x40);
> pci_write_config8(dev, 0x40, enables);
> // Set 0x42 to 0xf0 to match Award bios
> enables = pci_read_config8(dev, 0x42);
> enables |= 0xf0;
> pci_write_config8(dev, 0x42, enables);
> // Set bit 3 of 0x4a, to match award (dummy pci request)
> enables = pci_read_config8(dev, 0x4a);
> enables |= 0x08;
> pci_write_config8(dev, 0x4a, enables);
> // Set bit 3 of 0x4f to match award (use INIT# as cpu reset)
> enables = pci_read_config8(dev, 0x4f);
> enables |= 0x08;
> pci_write_config8(dev, 0x4f, enables);
> // Set 0x58 to 0x03 to match Award
> pci_write_config8(dev, 0x58, 0x03);
> // enable the ethernet/RTC
> if (dev) {
> enables = pci_read_config8(dev, 0x51);
> enables |= 0x18; 
> pci_write_config8(dev, 0x51, enables);
> }
> // enable IDE, since Linux won't do it.
> // First do some more things to devfn (17,0)
> // note: this should already be cleared, according to the book.
> enables = pci_read_config8(dev, 0x50);
> printk_debug("IDE enable in reg. 50 is 0x%x\n", enables);
> enables &= ~8; // need manifest constant here!
> printk_debug("set IDE reg. 50 to 0x%x\n", enables);
> pci_write_config8(dev, 0x50, enables);
> // set default interrupt values (IDE)
> enables = pci_read_config8(dev, 0x4c);
> printk_debug("IRQs in reg. 4c are 0x%x\n", enables & 0xf);
> // clear out whatever was there.
> enables &= ~0xf;
> enables |= 4;
> printk_debug("setting reg. 4c to 0x%x\n", enables);
> pci_write_config8(dev, 0x4c, enables);
> // set up the serial port interrupts.
> // com2 to 3, com1 to 4
> pci_write_config8(dev, 0x46, 0x04);
> pci_write_config8(dev, 0x47, 0x03);
> pci_write_config8(dev, 0x6e, 0x98);
> /* set up isa bus -- i/o recovery time, rom write enable, extend-ale */
> pci_write_config8(dev, 0x40, 0x54);
> //ethernet_fixup();
> // Start the rtc
> rtc_init(0);
>
> +}
>
> +static void southbridge_init(struct device *dev)
> +{
> vt82c686a_init(dev);
> pci_routing_fixup(dev);
>
> +}
>
> +static struct device_operations vt82c686_lpc_ops = {
> .read_resources   = pci_dev_read_resources,
> .set_resources    = pci_dev_set_resources,
> .enable_resources = pci_dev_enable_resources,
> .init		  = &southbridge_init,
> .scan_bus	  = scan_static_bus,
> .enable           = 0,
> .ops_pci          = 0,
>
> +};
>
> +static struct pci_driver lpc_driver __pci_driver = {
> .ops	= &vt82c686a_lpc_ops,
> .vendor = PCI_VENDOR_ID_VIA,
> .device = PCI_DEVICE_ID_VIA_82C686,
>
> +};
> Index: src/southbridge/via/vt82c686a/vt82c686a.c
> ===================================================================
> --- src/southbridge/via/vt82c686a/vt82c686a.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a.c	(revision 0)
> @@ -0,0 +1,97 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2003-2004 Ron Minnich
> Copyright (C) 2003-2004 Eric W. Beiderman
> Copyright (C) 2004 Mark Wilkinson
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2006 Uwe Hermann
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +#include <console/console.h>
> +#include <device/device.h>
> +#include <device/pci.h>
> +#include <device/pci_ops.h>
> +#include <device/pci_ids.h>
>
> +#include <pc80/mc146818rtc.h>
> +#include <pc80/keyboard.h>
>
> +#include "chip.h"
>
> +/* Base 8231 controller */
> +static device_t lpc_dev;
>
> +void hard_reset(void)
> +{
>    printk_err("NO HARD RESET ON VT82C686A! FIX ME!\n");
>
> +}
>
> +static void keyboard_on(void)
> +{
> unsigned char regval;
> if (lpc_dev) {
> regval = pci_read_config8(lpc_dev, 0x51);
> regval |= 0x0f; 
> pci_write_config8(lpc_dev, 0x51, regval);
> }
> init_pc_keyboard(0x60, 0x64, 0);
>
> +}
>
> +static void com_port_on(void)
> +{
> +#if 0
> // enable com1 and com2.
> enables = pci_read_config8(dev, 0x6e);
> /* 0x80 is enable com port b, 0x10 is to make it com2, 0x8
> * is enable com port a as com1 kevinh/Ispiri - Old code
> * thought 0x01 would make it com1, that was wrong enables =
> * 0x80 | 0x10 | 0x8 ; pci_write_config8(dev, 0x6e,
> * enables); // note: this is also a redo of some port of
> * assembly, but we want everything up.
> */
> /* set com1 to 115 kbaud not clear how to do this yet.
> * forget it; done in assembly.
> */
>
> +#endif
> +}
>
> +/* FixME: to be removed ? */
> +static void vt82c686a_enable(struct device *dev)
> +{
> struct southbridge_via_vt82c686a_config *conf = dev->chip_info;
> if (!lpc_dev) {
> /* the first time called, enable devices not on PCI bus
>  * FIXME: is that device struct there yet? */
> lpc_dev = dev_find_device(PCI_VENDOR_ID_VIA,
> 			  PCI_DEVICE_ID_VIA_82C686, 0);
> if (conf->enable_keyboard)
> 	keyboard_on();
> if (conf->enable_com_ports)
> 	com_port_on();
> }
>
> +}
>
> +struct chip_operations southbridge_via_vt82c686a_ops = {
> CHIP_NAME("VIA VT82C686A Southbridge")
> .enable_dev	= vt82c686a_enable,
>
> +};
> Index: src/southbridge/via/vt82c686a/vt82c686a_ide.c
> ===================================================================
> --- src/southbridge/via/vt82c686a/vt82c686a_ide.c	(revision 0)
> +++ src/southbridge/via/vt82c686a/vt82c686a_ide.c	(revision 0)
> @@ -0,0 +1,129 @@
> +/*
> This file is part of the LinuxBIOS project.
> Copyright (C) 2005 Li-Ta Lo
> Copyright (C) 2006 Ron Minnich
> Copyright (C) 2007 Corey Osgood
> 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., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
> */
>
> +#include <console/console.h>
> +#include <device/device.h>
> +#include <device/pci.h>
> +#include <device/pci_ops.h>
> +#include <device/pci_ids.h>
> +#include "chip.h"
>
> +static void ide_init(struct device *dev)
> +{
> struct southbridge_via_vt82c686a_config *conf = (struct southbridge_via_vt82c686a_config *)dev->chip_info;
> unsigned char enables;
> if (!conf->enable_native_ide) {
> // Run the IDE controller in 'compatiblity mode - i.e. don't use PCI
> // interrupts.  Using PCI ints confuses linux for some reason.
> printk_info("%s: enabling compatibility IDE addresses\n", __FUNCTION__);
> enables = pci_read_config8(dev, 0x42);
> printk_debug("enables in reg 0x42 0x%x\n", enables);
> enables &= ~0xc0;		// compatability mode
> pci_write_config8(dev, 0x42, enables);
> enables = pci_read_config8(dev, 0x42);
> printk_debug("enables in reg 0x42 read back as 0x%x\n", enables);
> }
> enables = pci_read_config8(dev, 0x40);
> printk_debug("enables in reg 0x40 0x%x\n", enables);
> enables |= 3;
> pci_write_config8(dev, 0x40, enables);
> enables = pci_read_config8(dev, 0x40);
> printk_debug("enables in reg 0x40 read back as 0x%x\n", enables);
> // Enable prefetch buffers
> enables = pci_read_config8(dev, 0x41);
> enables |= 0xf0;
> pci_write_config8(dev, 0x41, enables);
> // Lower thresholds (cause award does it)
> enables = pci_read_config8(dev, 0x43);
> enables &= ~0x0f;
> enables |=  0x05;
> pci_write_config8(dev, 0x43, enables);
> // PIO read prefetch counter (cause award does it)
> pci_write_config8(dev, 0x44, 0x18);
> // Use memory read multiple
> pci_write_config8(dev, 0x45, 0x1c);
> // address decoding.
> // we want "flexible", i.e. 1f0-1f7 etc. or native PCI
> // [e-mail] - the standard linux drivers seem ass slow when
> // used in native mode - I've changed back to classic
> enables = pci_read_config8(dev, 0x9);
> printk_debug("enables in reg 0x9 0x%x\n", enables);
> // by the book, set the low-order nibble to 0xa.
> if (conf->enable_native_ide) {
> enables &= ~0xf;
> // cf/cg silicon needs an 'f' here. 
> enables |= 0xf;
> } else {
> enables &= ~0x5;
> }
> pci_write_config8(dev, 0x9, enables);
> enables = pci_read_config8(dev, 0x9);
> printk_debug("enables in reg 0x9 read back as 0x%x\n", enables);
> // standard bios sets master bit.
> enables = pci_read_config8(dev, 0x4);
> printk_debug("command in reg 0x4 0x%x\n", enables);
> enables |= 7;
> // No need for stepping - [e-mail]
> enables &= ~0x80;
> pci_write_config8(dev, 0x4, enables);
> enables = pci_read_config8(dev, 0x4);
> printk_debug("command in reg 0x4 reads back as 0x%x\n", enables);
> if (!conf->enable_native_ide) {
> // Use compatability mode - per award bios
> pci_write_config32(dev, 0x10, 0x0);
> pci_write_config32(dev, 0x14, 0x0);
> pci_write_config32(dev, 0x18, 0x0);
> pci_write_config32(dev, 0x1c, 0x0);
> // Force interrupts to use compat mode - just like Award bios
> pci_write_config8(dev, 0x3d, 00);
> pci_write_config8(dev, 0x3c, 0xff);
> }
>
> +}
>
> +static struct device_operations ide_ops = {
> .read_resources   = pci_dev_read_resources,
> .set_resources    = pci_dev_set_resources,
> .enable_resources = pci_dev_enable_resources,
> .init		  = ide_init,
> .enable           = 0,
> .ops_pci          = 0,
>
> +};
>
> +static struct pci_driver northbridge_driver __pci_driver = {
> .ops	= &ide_ops,
> .vendor = PCI_VENDOR_ID_VIA,
> .device = PCI_DEVICE_ID_VIA_82C586_1,
>
> +};

[Uwe Hermann]

On Mon, Jan 15, 2007 at 05:21:13PM -0800, Adam Talbot wrote:
> Do we need to update the website to reflect the state of the the via 
> vt82c686a, under the "Supported Chipsets and Devices"

No yet, IMHO. We'll do that when the code gets committed.
Btw, please don't fully quote such huge patches ;-)

Uwe.

[Corey Osgood]

Please scratch the patch I submitted. vt82c686a and b are the same, the 
docs lied (and this is the first time I've looked at it on the machines, 
stupid me). I'm reworking a few things and cleaning up a few more 
things, and I'll submit a working, tested patch this weekend, if all 
goes well.

On a related note, does anyone know if any northbridge init is required 
to bring up the southbridge (and in this case also superio)? And is the 
rcn dc1100s from v1 working or not? I couldn't get it to run (no serial 
or video output at all) on the tyan s2507, which seems to be the same 
hardware.

[Uwe Hermann]

Hi,

On Thu, Feb 01, 2007 at 02:45:22AM -0500, Corey Osgood wrote:
> Please scratch the patch I submitted. vt82c686a and b are the same, the 
> docs lied (and this is the first time I've looked at it on the machines, 
> stupid me). I'm reworking a few things and cleaning up a few more 
> things, and I'll submit a working, tested patch this weekend, if all 
> goes well.

OK, great!
> On a related note, does anyone know if any northbridge init is required 
> to bring up the southbridge (and in this case also superio)? And is the 
> rcn dc1100s from v1 working or not? I couldn't get it to run (no serial 
> or video output at all) on the tyan s2507, which seems to be the same 
> hardware.

I guess you need the northbridge (RAM init) for anything else to work.
I have a K7T Turbo which has a VIA VT82C686B southbridge, but
unfortunately the northbridge is a VIA Apollo KT133A (VT8363A) which
is not supported, and I don't have VIA data sheets...

Uwe.

[Corey Osgood]

Uwe Hermann wrote:
> Hi,
>
> On Thu, Feb 01, 2007 at 02:45:22AM -0500, Corey Osgood wrote:
> > Please scratch the patch I submitted. vt82c686a and b are the same, the 
> > docs lied (and this is the first time I've looked at it on the machines, 
> > stupid me). I'm reworking a few things and cleaning up a few more 
> > things, and I'll submit a working, tested patch this weekend, if all 
> > goes well.
>
> OK, great!

So, this weekend just happened to be both the super bowl and my 21st
birthday, so I didn't get as much as I'd hoped to done. What I did get
means that either my serial setup on the monitoring PC is messed up
(very possible) or else this quick hack from vt8231 isn't going to work,
and I need to go back and work from v1, which I can only assume to be
working. I make no more promises, it will be done when it's done (duke
nukem forever anyone?)
> > On a related note, does anyone know if any northbridge init is required 
> > to bring up the southbridge (and in this case also superio)? And is the 
> > rcn dc1100s from v1 working or not? I couldn't get it to run (no serial 
> > or video output at all) on the tyan s2507, which seems to be the same 
> > hardware.
>
> I guess you need the northbridge (RAM init) for anything else to work.
> I have a K7T Turbo which has a VIA VT82C686B southbridge, but
> unfortunately the northbridge is a VIA Apollo KT133A (VT8363A) which
> is not supported, and I don't have VIA data sheets...

The vt82c686b datasheet is here (I also found an older version on some
obscure page of via's site, but can't find it now):
http://www.datasheets.org.uk/search.php?q=VT82C686B&sType=part
As for the 8363a, rom.by has the datasheet for both it and my 694x (and
ALi chipsets!), but I can't be sure if they're, well, supposed to or
not, so I'm avoiding touching them.

-Corey