Hi to all,
I have create a linuxbios file for my embeded mother board call Adavantech PCM-5823, and I have some problem with the fonctionality of the superIO winbond W83977AF/TF, when I put LinuxBIOS in my chip, I have nothing on screen and no information on the COM port.
In linuxBIOS I used the superIO winbond W83977TF. According to the documentation of the winbond W83977AF and W83977TF the chip is virtually identical, but the are some difference on the number of I/O programable port.
For now I'm blocked. If somebody has a solution for my problem, I thank it by advance.
Regards
Ccases
On Mon, Apr 02, 2007 at 06:03:07PM +0200, ccases@navista.fr wrote:
I have create a linuxbios file for my embeded mother board call Adavantech PCM-5823, and I have some problem with the fonctionality of the superIO winbond W83977AF/TF, when I put LinuxBIOS in my chip, I have nothing on screen and no information on the COM port.
VGA is initialized a lot later, getting serial output is the first step, usually.
In linuxBIOS I used the superIO winbond W83977TF. According to the documentation of the winbond W83977AF and W83977TF the chip is virtually identical, but the are some difference on the number of I/O programable port.
For now I'm blocked. If somebody has a solution for my problem, I thank it by advance.
Without knowing which code you're using we cannot say very much. Can you post the full patch of the code you're trying to use?
Also, are you sure your null-modem cable is ok? Is it attached to the correct COM port? Can you connect to another machine when Linux runs (using minicom for example)? Are your BAUD rate settings correct?
Uwe.
On Mon, Apr 02, 2007 at 06:03:07PM +0200, ccases@navista.fr wrote:
I have create a linuxbios file for my embeded mother board call Adavantech PCM-5823, and I have some problem with the fonctionality of the superIO winbond W83977AF/TF, when I put LinuxBIOS in my chip, I have nothing on screen and no information on the COM port.
VGA is initialized a lot later, getting serial output is the first step, usually.
Yes I know another person already said it to me.
In linuxBIOS I used the superIO winbond W83977TF. According to the documentation of the winbond W83977AF and W83977TF the chip is virtually identical, but the are some difference on the number of I/O programable port.
For now I'm blocked. If somebody has a solution for my problem, I thank it by advance.
Without knowing which code you're using we cannot say very much. Can you post the full patch of the code you're trying to use?
The code is attach on mail.
Also, are you sure your null-modem cable is ok? Is it attached to the correct COM port? Can you connect to another machine when Linux runs (using minicom for example)? Are your BAUD rate settings correct?
My cable is ok I have test it. It is attached on com port one towards another machine on com port one using Hyperterminal mode with this parameters 115200 baud, 8bit, no parity, 1 bit stop on com one. The baud rate is the same in ll of my config file.
Thanks by advance for your help.
Regards.
Ccases
On Tue, Apr 03, 2007 at 09:17:43AM +0200, ccases@navista.fr wrote:
Without knowing which code you're using we cannot say very much. Can you post the full patch of the code you're trying to use?
The code is attach on mail.
Try to use the w83627hf in v2, that might work out of the box. If it doesn't try to use 0x370, 0x3f0, or 0x2e in this line in auto.c:
#define SERIAL_DEV PNP_DEV(0x3f0, W83977AF_SP1)
Page 128 (section 9.2) of the datasheet might help if you still have problems afterwards...
http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83977F_AF-A....
HTH, Uwe.
Uwe Hermann wrote:
On Tue, Apr 03, 2007 at 09:17:43AM +0200, ccases@navista.fr wrote:
Without knowing which code you're using we cannot say very much. Can you post the full patch of the code you're trying to use?
The code is attach on mail.
Try to use the w83627hf in v2, that might work out of the box. If it doesn't try to use 0x370, 0x3f0, or 0x2e in this line in auto.c:
#define SERIAL_DEV PNP_DEV(0x3f0, W83977AF_SP1)
Page 128 (section 9.2) of the datasheet might help if you still have problems afterwards...
http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83977F_AF-A....
HTH, Uwe.
You may also need to do early_mtrr_init() (if it's even valid on geode) before trying to bring up the serial port, I had to do this on my pesky C3-based PCChips board.
Also, grab the datasheet for the w83977tf (google should find this easily) if you go datasheet diving. Comparing the init sequence of the two chips might give you some hint as to where they differ.
-Corey
Uwe Hermann wrote:
On Tue, Apr 03, 2007 at 09:17:43AM +0200, ccases@navista.fr wrote:
Without knowing which code you're using we cannot say very much. Can you post the full patch of the code you're trying to use?
The code is attach on mail.
Try to use the w83627hf in v2, that might work out of the box. If it doesn't try to use 0x370, 0x3f0, or 0x2e in this line in auto.c:
#define SERIAL_DEV PNP_DEV(0x3f0, W83977AF_SP1)
Page 128 (section 9.2) of the datasheet might help if you still have problems afterwards...
http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83977F_AF-A....
HTH, Uwe.
You may also need to do early_mtrr_init() (if it's even valid on geode) before trying to bring up the serial port, I had to do this on my pesky C3-based PCChips board.
I will try to set up the early_mtrr_init(), if it's possible.
Also, grab the datasheet for the w83977tf (google should find this easily) if you go datasheet diving. Comparing the init sequence of the two chips might give you some hint as to where they differ.
if I look the both documentation, and I see some differ on the number of I/O port programmable (23 for the TF against 14 for the AF), after that I haven't see other difference.
Thanks for your help
Regards
Ccases
Uwe Hermann wrote:
On Tue, Apr 03, 2007 at 09:17:43AM +0200, ccases@navista.fr wrote:
Without knowing which code you're using we cannot say very much. Can you post the full patch of the code you're trying to use?
The code is attach on mail.
Try to use the w83627hf in v2, that might work out of the box. If it doesn't try to use 0x370, 0x3f0, or 0x2e in this line in auto.c:
#define SERIAL_DEV PNP_DEV(0x3f0, W83977AF_SP1)
Page 128 (section 9.2) of the datasheet might help if you still have problems afterwards...
http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83977F_AF-A....
HTH, Uwe.
You may also need to do early_mtrr_init() (if it's even valid on geode) before trying to bring up the serial port, I had to do this on my pesky C3-based PCChips board.
Also, grab the datasheet for the w83977tf (google should find this easily) if you go datasheet diving. Comparing the init sequence of the two chips might give you some hint as to where they differ.
-Corey
-- linuxbios mailing list linuxbios@linuxbios.org http://www.linuxbios.org/mailman/listinfo/linuxbios
I have test to use the early_mtrr_init(), no positive result. No com port for now.
If somebody has an idea
Thanks by advance
Regards
CCases
ccases@navista.fr wrote:
I have test to use the early_mtrr_init(), no positive result. No com port for now.
If somebody has an idea
Check in your factory BIOS, make sure it isn't set up to swap COM 1 and 2? Or even better, check for output from the second serial port? I'm pretty much out of ideas.
Do you have a POST card? If so, set up a post code at the start of w83977af_early_serial() so you can make sure LinuxBIOS is in fact making it that far. Do a post code with outb(0x[code], 0x80);
-Corey
On Tue, Apr 03, 2007 at 09:17:43AM +0200, ccases@navista.fr wrote:
Without knowing which code you're using we cannot say very much. Can
you
post the full patch of the code you're trying to use?
The code is attach on mail.
Try to use the w83627hf in v2, that might work out of the box. If it doesn't try to use 0x370, 0x3f0, or 0x2e in this line in auto.c:
#define SERIAL_DEV PNP_DEV(0x3f0, W83977AF_SP1)
Page 128 (section 9.2) of the datasheet might help if you still have problems afterwards...
http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83977F_AF-A....
HTH, Uwe.
http://www.hermann-uwe.de | http://www.holsham-traders.de http://www.crazy-hacks.org | http://www.unmaintained-free-software.org -- linuxbios mailing list linuxbios@linuxbios.org http://www.linuxbios.org/mailman/listinfo/linuxbios
Thanks for your help, I will test all that as soon as I would have five minutes.
Regards
Ccases
On Tue, Apr 03, 2007 at 09:17:43AM +0200, ccases@navista.fr wrote:
Without knowing which code you're using we cannot say very much. Can
you
post the full patch of the code you're trying to use?
The code is attach on mail.
Try to use the w83627hf in v2, that might work out of the box. If it doesn't try to use 0x370, 0x3f0, or 0x2e in this line in auto.c:
#define SERIAL_DEV PNP_DEV(0x3f0, W83977AF_SP1)
Page 128 (section 9.2) of the datasheet might help if you still have problems afterwards...
http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83977F_AF-A....
HTH, Uwe.
http://www.hermann-uwe.de | http://www.holsham-traders.de http://www.crazy-hacks.org | http://www.unmaintained-free-software.org -- linuxbios mailing list linuxbios@linuxbios.org http://www.linuxbios.org/mailman/listinfo/linuxbios
I have test all of this information, and the results is the same like before. No data on the com port ...
If somebody has an Idea
Thanks by advance for your help
regards
Ccases
-
ccases＠navista.fr -
Corey Osgood -
Uwe Hermann