2.6 & pdc_console pb on c110: "Warning: unable to open an initial console."
Hi all,
I try to see if pdc_console could be operational on my c110 with kernel 2.6.0-pa5.
So I uncheck serial_mux and check pdc_console with make menuconfig (edit also pdc_cons.c to define EARLY_BOOTUP_DEBUG) Then I create the special file "mknod /dev/ttyB0 c 11 0", add "TB0:12345:respawn:/sbin/getty 38400 ttyB0" in /etc/inittab
(I also try to apply Richard patch <http://lists.parisc-linux.org/pipermail/parisc-linux/2002-May/016384.html>)
But no way to make it works?
Linux version 2.6.0-pa5 (root@hpalin) (gcc version 3.3.3 20031206 (prerelease) 3FP[0] enabled: Rev 1 Model 11
The 32-bit Kernel has started...
Initialized PDC Console for debugging.
Determining PDC firmware type: System Map.
model 000058e0 00000481 00000000 00000002 77e47570 100000f1 00000004 0000008a 0avers 0000000d CPUID vers 11 rev 13 (0x0000016d)
model 9000/777/C110
Total Memory: 128 Mb
[snip]
Soft power switch enabled, polling @ 0xf0140000.
Console: switching to colour frame buffer device 128x48
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
Generic RTC Driver v1.07
Serial: 8250/16550 driver $Revision: 1.90 $ 13 ports, IRQ sharing enabled
ttyS0 at MMIO 0xffd05800 (irq = 90) is a 16550A
ttyS1 at MMIO 0xffe02800 (irq = 121) is a 16550A
[snip]
Freeing unused kernel memory: 504k freed
Warning: unable to open an initial console.
Adding 131832k swap on /dev/md1. Priority:-1 extents:1
EXT3 FS on md2, internal journal
[snip]
....eth0: link ok.
Debian GNU/Linux testing/unstable hpalin ttyS0
hpalin login:
In th syslog I also notice:
[snip]
Dec 27 21:45:03 hpalin init: Id "TB0" respawning too fast: disabled for 5 minutes Dec 27 21:52:15 hpalin init: Id "TB0" respawning too fast: disabled for 5 minutes
[snip]
And don't find any such device with:
# cat /proc/devices Character devices: 1 mem 2 pty 3 ttyp 4 /dev/vc/0 4 tty 4 ttyS 5 /dev/tty 5 /dev/console 5 /dev/ptmx 6 lp 7 vcs 9 st 10 misc 13 input 14 sound 21 sg 29 fb 36 netlink 128 ptm 136 pts 180 usb [snip]
Any idea? (what should I do wrongly?)
Thanks in advance for attention,
Joel
Quoted 18 lines · Joel Soete
I try to see if
pdc_consolecould be operational on my c110 with kernel 2.6.0-pa5.So I uncheck
serial_muxand checkpdc_consolewith make menuconfig (edit alsopdc_cons.cto defineEARLY_BOOTUP_DEBUG) Then I create the special file "mknod/dev/ttyB0c 11 0", add "TB0:12345:respawn:/sbin/getty 38400 ttyB0" in/etc/inittab(I also try to apply Richard patch <http://lists.parisc-linux.org/pipermail/parisc-linux/2002-May/016384.html>)
But no way to make it works?
Linux version 2.6.0-pa5 (root@hpalin) (gcc version 3.3.3 20031206 (prerelease) 3FP[0] enabled: Rev 1 Model 11
The 32-bit Kernel has started...
Initialized PDC Console for debugging.
Determining PDC firmware type: System Map.
model000058e000000481 00000000 0000000277e47570100000f1000000040000008a0avers0000000dCPUID vers 11 rev 13 (0x0000016d)
model 9000/777/C110
Total Memory: 128 Mb
[snip]
Quoted 8 lines
Soft power switch enabled, polling @
0xf0140000.
Console: switching to colour frame buffer device 128x48
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
Generic RTC Driver v1.07
Serial: 8250/16550 driver $Revision: 1.90 $ 13 ports, IRQ sharing enabled ttyS0 at MMIO0xffd05800(irq = 90) is a 16550A
ttyS1 at MMIO0xffe02800(irq = 121) is a 16550A
Your machine has standard serial ports. So why do you want to switch on serial_mux stuff when it's not needed (and not possible since the device is not present in your machine) ?
Helge
Quoted 30 lines · Helge Deller
Quoted 18 lines · Joel Soete
I try to see if
pdc_consolecould be operational on my c110 with kernel 2.6.0-pa5.So I uncheck
serial_muxand checkpdc_consolewith make menuconfig (edit alsopdc_cons.cto defineEARLY_BOOTUP_DEBUG) Then I create the special file "mknod/dev/ttyB0c 11 0", add "TB0:12345:respawn:/sbin/getty 38400 ttyB0" in/etc/inittab(I also try to apply Richard patch <http://lists.parisc-linux.org/pipermail/parisc-linux/2002-May/016384.html>)
But no way to make it works?
Linux version 2.6.0-pa5 (root@hpalin) (gcc version 3.3.3 20031206 (prerelease) 3FP[0] enabled: Rev 1 Model 11
The 32-bit Kernel has started...
Initialized PDC Console for debugging.
Determining PDC firmware type: System Map.
model000058e000000481 00000000 0000000277e47570100000f1000000040000008a0avers0000000dCPUID vers 11 rev 13 (0x0000016d)
model 9000/777/C110
Total Memory: 128 Mb
[snip]Quoted 8 lines
Soft power switch enabled, polling @
0xf0140000.
Console: switching to colour frame buffer device 128x48
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
Generic RTC Driver v1.07
Serial: 8250/16550 driver $Revision: 1.90 $ 13 ports, IRQ sharing enabled ttyS0 at MMIO0xffd05800(irq = 90) is a 16550A
ttyS1 at MMIO0xffe02800(irq = 121) is a 16550A
Quoted 6 lines
Your machine has standard serial ports. So why do you want to switch on
serial_muxstuff when it's not needed (and not possible since the device is not present in your machine) ?Helge
Helge,
Sorry I don't understand: where do you read that I switch on serial mux?
My goal is to test pdc_console mainly for EARLY_BOOTUP_DEBUG facility (for later b2k 64bit kernel pb). Also afaik otc I have to switch off serial mux (well i say uncheck):drivers/serial/Kconfig
[snip]
config PDC_CONSOLE
bool "PDC software console support"
depends on PARISC && !SERIAL_MUX && VT
default n
[snip]?
Thanks in advance for more info,
Joel
Quoted 9 lines · Joel Soete
My goal is to test
pdc_consolemainly forEARLY_BOOTUP_DEBUGfacility (for later b2k 64bit kernel pb). Also afaik otc I have to switch off serial mux (well i say uncheck):drivers/serial/Kconfig
[snip]
configPDC_CONSOLE
bool "PDC software console support"
depends on PARISC && !SERIAL_MUX && VT
default n
[snip]?
To debug such problems I often just turn on EARLY_BOOTUP_DEBUG in arch/parisc/kernel/pdc_cons.c, line 15:
/* Define EARLY_BOOTUP_DEBUG to debug kernel related boot problems.
* On production kernels EARLY_BOOTUP_DEBUG should be undefined. */
#define EARLY_BOOTUP_DEBUG
HTH,
Helge
Anyway, the PDC driver should work for all architectures, independent of the serial type (this was my concept of this driver).
Is the driver fully ported to the 2.6 kernel ?
In the first log, the line "Warning: unable to open an initial console." was shown, so the open on this device fails !
Futher there is no PDC driver registered under proc/devices !
I have never tested a 2.6 kernel on a HP.
Christoph
Quoted 41 lines · Helge Deller
Quoted 18 lines · Joel Soete
I try to see if
pdc_consolecould be operational on my c110 with kernel 2.6.0-pa5.So I uncheck
serial_muxand checkpdc_consolewith make menuconfig (edit alsopdc_cons.cto defineEARLY_BOOTUP_DEBUG) Then I create the special file "mknod/dev/ttyB0c 11 0", add "TB0:12345:respawn:/sbin/getty 38400 ttyB0" in/etc/inittab(I also try to apply Richard patch <http://lists.parisc-linux.org/pipermail/parisc-linux/2002-May/016384.html>)
But no way to make it works?
Linux version 2.6.0-pa5 (root@hpalin) (gcc version 3.3.3 20031206 (prerelease) 3FP[0] enabled: Rev 1 Model 11
The 32-bit Kernel has started...
Initialized PDC Console for debugging.
Determining PDC firmware type: System Map.
model000058e000000481 00000000 0000000277e47570100000f1000000040000008a0avers0000000dCPUID vers 11 rev 13 (0x0000016d)
model 9000/777/C110
Total Memory: 128 Mb
[snip]Quoted 8 lines
Soft power switch enabled, polling @
0xf0140000.
Console: switching to colour frame buffer device 128x48
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
Generic RTC Driver v1.07
Serial: 8250/16550 driver $Revision: 1.90 $ 13 ports, IRQ sharing enabled ttyS0 at MMIO0xffd05800(irq = 90) is a 16550A
ttyS1 at MMIO0xffe02800(irq = 121) is a 16550AYour machine has standard serial ports. So why do you want to switch on
serial_muxstuff when it's not needed (and not possible since the device is not present in your machine) ?Helge
_______________________________________________
parisc-linux mailing list
[email protected]
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux
Signature · 3 lines
------------------------------------------------------- private: [email protected] company: [email protected]
Hello Christoph,
Quoted 8 lines
Anyway, the PDC driver should work for all architectures, independent of the serial type (this was my concept of this driver).
Is the driver fully ported to the 2.6 kernel ?
In the first log, the line "Warning: unable to open an initial console." was shown, so the open on this device fails !
Futher there is no PDC driver registered under proc/devices !
That a better explanation ;) (afaik there are some changes from 2.4 to 2.6 driver's implementation and also some paper about the subject, I will see)
Quoted 1 line
I have never tested a 2.6 kernel on a HP.
Another question: I also try it with 2.4 (32 or 64 bit kernel) running on a b2k.
But this b2k is equiped with an ide cdrom and when I try to boot a kernel with builtin pdc driver (2.4 and 2.6), the boot process hung (the green led of the cdrom stay continuously bright, so it seems a pb of initialization in the pdc_console driver?. This pb also occurs even if I unplug the cdrom-drive or remove ide-cdrom driver from kernels. otc kernels boot well with builtin ide-cdrom driver without pdc_console driver)
Any other idea?
Thanks again,
Joel
Christoph
Quoted 4 lines · Helge Deller
Quoted 1 line · Joel Soete
I try to see if
pdc_consolecould be operational on my c110 with kernel
2.6.0-pa5.
Quoted 2 lines
Quoted 2 lines
So I uncheck
serial_muxand checkpdc_consolewith make menuconfig (edit
also pdc_cons.c to define EARLY_BOOTUP_DEBUG)
Quoted 9 lines
Quoted 1 line
Then I create the special file "mknod
/dev/ttyB0c 11 0", add "TB0:12345:respawn:/sbin/getty 38400 ttyB0" in/etc/inittabQuoted 6 lines
(I also try to apply Richard patch <http://lists.parisc-linux.org/pipermail/parisc-linux/2002-May/016384.html>)
But no way to make it works?
Linux
ersion 2.6.0-pa5 (root@hpalin) (gcc version 3.3.3 20031206 (prerelease)
3FP[0] enabled: Rev 1 Model 11
Quoted 5 lines
Quoted 3 lines
The 32-bit Kernel has started...
Initialized PDC Console for debugging.
Determining PDC firmware type: SysteMap.
Quoted 1 line
model
000058e000000481 00000000 0000000277e47570100000f1000000040000008a
0avers 0000000d
Quoted 28 lines
Quoted 5 lines
CPUID vers 11 rev 13 (
0x0000016d)
model 9000/777/C110
Total Memory: 128 Mb
[snip]
Soft power switch enabled, polling @0xf0140000.Console: switching to colour frame buffer device 128x48
Quoted 5 lines
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
Generic RTC Driver v1.07
Serial: 8250/16550 driver $Revision: 1.90 $ 13 ports, IRQ sharing enabled ttyS0 at MMO
0xffd05800(irq = 90) is a 16550AQuoted 1 line
ttyS1 at MMIO
0xffe02800(irq = 121) is a 16550AYour machine has standard serial ports. So why do you want to switch on
serial_muxstuff when it's not needed (and not possible since the device is
not present in your machine) ?Helge
_______________________________________________
parisc-linux mailing list
[email protected]
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux
Signature · 1 line
------
Quoted 1 line
------------------------------------------------
private: [email protected]
company: [email protected]
_______________________________________________
parisc-linux mailing list
[email protected]
http://list
Quoted 1 line
.parisc-linux.org/mailman/listinfo/parisc-linux
-------------------------------------------------------------------------
Tiscali ADSL: 12 mois à 29,50 €/mois! L'Internet rapide, c'est pour tout le monde.
http://reg.tiscali.be/default.asp?lg=fr
Hello,
sorry have no idea for the hang up on boot!
BUT: experiments with the MUX driver has show, that using not ttyB0 (but 1,2...) there system also hangs up.
BUT: This is no explanation for this problem here. The PDC console driver calls the "BIOS-like" routines for printing out (and polling keys !).
All the "printk()" are printed directly (without my PDC console driver), and everything running via "/dev/console" or "/dev/ttyB0" goes through the driver. Is the kernel so far, that an `open( "/dev/console"' is already called (just done in kernel before calling "/sbin/init" or "/bin/sh") ?
Christoph
Quoted 112 lines · Joel Soete
Hello Christoph,
Quoted 7 lines
Anyway, the PDC driver should work for all architectures, independent of the serial type (this was my concept of this driver).
Is the driver fully ported to the 2.6 kernel ?
In the first log, the line "Warning: unable to open an initial console." was shown, so the open on this device fails !
Futher there is no PDC driver registered under proc/devices !That a better explanation ;) (afaik there are some changes from 2.4 to 2.6 driver's implementation and also some paper about the subject, I will see)
Quoted 1 line
I have never tested a 2.6 kernel on a HP.
Another question: I also try it with 2.4 (32 or 64 bit kernel) running on a b2k.
But this b2k is equiped with an ide cdrom and when I try to boot a kernel with builtin pdc driver (2.4 and 2.6), the boot process hung (the green led of the cdrom stay continuously bright, so it seems a pb of initialization in thepdc_consoledriver?. This pb also occurs even if I unplug the cdrom-drive or remove ide-cdrom driver from kernels. otc kernels boot well with builtin ide-cdrom driver withoutpdc_consoledriver)Any other idea?
Thanks again,
JoelChristoph
Quoted 6 lines · Helge Deller
Quoted 1 line · Joel Soete
I try to see if
pdc_consolecould be operational on my c110 with kernel2.6.0-pa5.
Quoted 2 lines
Quoted 1 line
So I uncheck
serial_muxand checkpdc_consolewith make menuconfig (editalso
pdc_cons.cto defineEARLY_BOOTUP_DEBUG)Quoted 11 lines
Quoted 1 line
Then I create the special file "mknod
/dev/ttyB0c 11 0", add "TB0:12345:respawn:/sbin/getty 38400 ttyB0" in/etc/inittabQuoted 5 lines
(I also try to apply Richard patch <http://lists.parisc-linux.org/pipermail/parisc-linux/2002-May/016384.html>)
But no way to make it works?
Linux
ersion 2.6.0-pa5 (root@hpalin) (gcc version 3.3.3 20031206 (prerelease)
3FP[0] enabled: Rev 1 Model 11
Quoted 8 lines
Quoted 3 lines
The 32-bit Kernel has started...
Initialized PDC Console for debugging.
Determining PDC firmware type: SysteMap.
Quoted 1 line
model
000058e000000481 00000000 0000000277e47570100000f1000000040000008a0avers
0000000dQuoted 31 lines
Quoted 5 lines
CPUID vers 11 rev 13 (
0x0000016d)
model 9000/777/C110
Total Memory: 128 Mb
[snip]
Soft power switch enabled, polling @0xf0140000.Console: switching to colour frame buffer device 128x48
Quoted 5 lines
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
Generic RTC Driver v1.07
Serial: 8250/16550 driver $Revision: 1.90 $ 13 ports, IRQ sharing enabled ttyS0 at MMO
0xffd05800(irq = 90) is a 16550AQuoted 1 line
ttyS1 at MMIO
0xffe02800(irq = 121) is a 16550AYour machine has standard serial ports. So why do you want to switch on
serial_muxstuff when it's not needed (and not possible since the device is
not present in your machine) ?Helge
_______________________________________________
parisc-linux mailing list
[email protected]
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux
Signature · 3 lines
------------------------------------------------------- private: [email protected] company: [email protected]
Joel,
Even though you should not need the pdc_console for what you want to do, pdc_console should work in theory. It probably needs to be updated to work with the 2.6 kernel. I will add this to my TODO list.
Thanks,
- Ryan
Quoted 82 lines · Joel Soete
Hi all,
I try to see if
pdc_consolecould be operational on my c110 with kernel 2.6.0-pa5.So I uncheck
serial_muxand checkpdc_consolewith make menuconfig (edit alsopdc_cons.cto defineEARLY_BOOTUP_DEBUG) Then I create the special file "mknod/dev/ttyB0c 11 0", add "TB0:12345:respawn:/sbin/getty 38400 ttyB0" in/etc/inittab(I also try to apply Richard patch <http://lists.parisc-linux.org/pipermail/parisc-linux/2002-May/016384.html>)
But no way to make it works?
Linux version 2.6.0-pa5 (root@hpalin) (gcc version 3.3.3 20031206 (prerelease) 3FP[0] enabled: Rev 1 Model 11
The 32-bit Kernel has started...
Initialized PDC Console for debugging.
Determining PDC firmware type: System Map.
model000058e000000481 00000000 0000000277e47570100000f1000000040000008a0avers0000000dCPUID vers 11 rev 13 (0x0000016d)
model 9000/777/C110
Total Memory: 128 Mb
[snip]
Soft power switch enabled, polling @0xf0140000.
Console: switching to colour frame buffer device 128x48
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
Generic RTC Driver v1.07
Serial: 8250/16550 driver $Revision: 1.90 $ 13 ports, IRQ sharing enabled
ttyS0 at MMIO0xffd05800(irq = 90) is a 16550A
ttyS1 at MMIO0xffe02800(irq = 121) is a 16550A
[snip]
Freeing unused kernel memory: 504k freed
Warning: unable to open an initial console.
Adding 131832k swap on/dev/md1. Priority:-1 extents:1
EXT3 FS on md2, internal journal
[snip]
....eth0: link ok.Debian GNU/Linux testing/unstable hpalin ttyS0
hpalin login:
In th syslog I also notice:
[snip]Dec 27 21:45:03 hpalin init: Id "TB0" respawning too fast: disabled for 5 minutes Dec 27 21:52:15 hpalin init: Id "TB0" respawning too fast: disabled for 5 minutes[snip]
And don't find any such device with:
# cat /proc/devices Character devices: 1 mem 2 pty 3 ttyp 4 /dev/vc/0 4 tty 4 ttyS 5 /dev/tty 5 /dev/console 5 /dev/ptmx 6 lp 7 vcs 9 st 10 misc 13 input 14 sound 21 sg 29 fb 36 netlink 128 ptm 136 pts 180 usb [snip]Any idea? (what should I do wrongly?)
Thanks in advance for attention,
Joel_______________________________________________
parisc-linux mailing list
[email protected]
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux
Hi Ryan,
Quoted 3 lines · Ryan Bradetich
Joel,
Even though you should not need the
pdc_consolefor what you want to do,
Well, I had the opportunity to test pdc_console on my b2k but as well as in 2.4 it make hung the kernel at boot time:
the cdrom green led stay continuously lighted (if I add or remove ide-cdrom support in the kernel)
Unfortunately I never reach to get any relevant toc info to go further in this pb invetigation :(. Now basicaly 2.6 is running on this system and so I will try to add some printk() to locate more accurately this event.
Quoted 3 lines
pdc_consoleshould work in theory. It probably needs to be updated to work with the 2.6 kernel. I will add this to my TODO list.
Ah ok I will waiting some news so.
Thanks also and happy new year,
Joel