[LinuxBIOS] IDE become readonly, why?

[Beneo]

Hi all,

I was using LinuxBIOS to boot Linux from IDE from a Broadcom HT,1000 south bridge based  platform , I encounter an issue which is hard to understand. I hope somebody can give me some insight.

During boot, Linux stoped at the point that it displayed a message:
"Configuring kernel parameters:  [  OK  ]"

then I pressed "CTRL-C",  Linux continue to boot, but display a message:
"Starting automount:Cannot create temp file /tmp/autofs.meQYNM, could not make temp file"

then Linux boot Skipped rest of initialization, like probe the network and etc. but Linux did give me a login prompt, I can login using root, but I cannot write any file at root directory, It appears to me Entire Linux file system becomes readonly.

Phoenix BIOS came with that board doesn't have this issue, so I know it should be LinuxBIOS related. I just don't know why.

Does anybody know anything about this issue?

Thanks

Beneo

---- dump from Linux boot.

INIT: version 2.84 booting
                Welcome to Red Hat Linux
                Press 'I' to enter interactive startup.
Mounting proc filesystem:  [  OK  ]
Unmounting initrd:  [  OK  ]
Configuring kernel parameters:  [  OK  ]
INIT: Entering runlevel: 3
Entering non-interactive startup
Starting pcmcia:  [  OK  ]
Starting automount:Cannot create temp file /tmp/autofs.meQYNM
could not make temp file
INIT: Id "1" respawning too fast: disabled for 5 minutes
INIT: Id "2" respawning too fast: disabled for 5 minutes
INIT: Id "3" respawning too fast: disabled for 5 minutes
INIT: Id "4" respawning too fast: disabled for 5 minutes
INIT: Id "5" respawning too fast: disabled for 5 minutes
INIT: Id "6" respawning too fast: disabled for 5 minutes

Red Hat Linux release 9 (Shrike)
Kernel 2.4.20-8smp on an i686

(none) login:

[Anton]

On Thu, 5 Apr 2007 11:27:11 -0700
"Beneo"  wrote:
> Hi all,
>
> I was using LinuxBIOS to boot Linux from IDE from a Broadcom HT,1000 south bridge based  platform , I encounter an issue which is hard to understand. I hope somebody can give me some insight.
>
> During boot, Linux stoped at the point that it displayed a message:
> "Configuring kernel parameters:  [  OK  ]"
>
> then I pressed "CTRL-C",  Linux continue to boot, but display a message:
> "Starting automount:Cannot create temp file /tmp/autofs.meQYNM, could not make temp file"
>
> then Linux boot Skipped rest of initialization, like probe the network and etc. but Linux did give me a login prompt, I can login using root, but I cannot write any file at root directory, It appears to me Entire Linux file system becomes readonly.
>
> Phoenix BIOS came with that board doesn't have this issue, so I know it should be LinuxBIOS related. I just don't know why.
>
> Does anybody know anything about this issue?

Can you show your /etc/inittab?

[Beneo]

Yes, see below:

Beneo

---
[root@(none) etc]# cat inittab
#
# inittab       This file describes how the INIT process should set up
#               the system in a certain run-level.
#
# Author:       Miquel van Smoorenburg, 
#               Modified for RHS Linux by Marc Ewing and Donnie Barnes
#

# Default runlevel. The runlevels used by RHS are:
#   0 - halt (Do NOT set initdefault to this)
#   1 - Single user mode
#   2 - Multiuser, without NFS (The same as 3, if you do not have 
networking)
#   3 - Full multiuser mode
#   4 - unused
#   5 - X11
#   6 - reboot (Do NOT set initdefault to this)
#
#id:5:initdefault:
id:3:initdefault:

# System initialization.
si::sysinit:/etc/rc.d/rc.sysinit

l0:0:wait:/etc/rc.d/rc 0
l1:1:wait:/etc/rc.d/rc 1
l2:2:wait:/etc/rc.d/rc 2
l3:3:wait:/etc/rc.d/rc 3
l4:4:wait:/etc/rc.d/rc 4
l5:5:wait:/etc/rc.d/rc 5
l6:6:wait:/etc/rc.d/rc 6

# Trap CTRL-ALT-DELETE
ca::ctrlaltdel:/sbin/shutdown -t3 -r now

# When our UPS tells us power has failed, assume we have a few minutes
# of power left.  Schedule a shutdown for 2 minutes from now.
# This does, of course, assume you have powerd installed and your
# UPS connected and working correctly.
pf::powerfail:/sbin/shutdown -f -h +2 "Power Failure; System Shutting Down"

# If power was restored before the shutdown kicked in, cancel it.
pr:12345:powerokwait:/sbin/shutdown -c "Power Restored; Shutdown Cancelled"

# Run gettys in standard runlevels
co:2345:respawn:/sbin/agetty ttyS0 115200 vt100
1:2345:respawn:/sbin/mingetty tty1
2:2345:respawn:/sbin/mingetty tty2
3:2345:respawn:/sbin/mingetty tty3
4:2345:respawn:/sbin/mingetty tty4
5:2345:respawn:/sbin/mingetty tty5
6:2345:respawn:/sbin/mingetty tty6

----- Original Message ----- 
From: "Anton" 
To: 
Sent: Thursday, April 05, 2007 11:43 AM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> On Thu, 5 Apr 2007 11:27:11 -0700
> "Beneo"  wrote:
> > Hi all,
> >
> > I was using LinuxBIOS to boot Linux from IDE from a Broadcom HT,1000 
> > south bridge based  platform , I encounter an issue which is hard to 
> > understand. I hope somebody can give me some insight.
> >
> > During boot, Linux stoped at the point that it displayed a message:
> > "Configuring kernel parameters:  [  OK  ]"
> >
> > then I pressed "CTRL-C",  Linux continue to boot, but display a message:
> > "Starting automount:Cannot create temp file /tmp/autofs.meQYNM, could not 
> > make temp file"
> >
> > then Linux boot Skipped rest of initialization, like probe the network 
> > and etc. but Linux did give me a login prompt, I can login using root, 
> > but I cannot write any file at root directory, It appears to me Entire 
> > Linux file system becomes readonly.
> >
> > Phoenix BIOS came with that board doesn't have this issue, so I know it 
> > should be LinuxBIOS related. I just don't know why.
> >
> > Does anybody know anything about this issue?
>
> Can you show your /etc/inittab?

[Russ Whitaker]

On Thu, 5 Apr 2007, Beneo wrote:
> Yes, see below:
>
> Beneo
>
> [root@(none) etc]# cat inittab
> #
> # inittab       This file describes how the INIT process should set up
> #               the system in a certain run-level.
> #
> # Author:       Miquel van Smoorenburg, 
> #               Modified for RHS Linux by Marc Ewing and Donnie Barnes
> #
>
> # Default runlevel. The runlevels used by RHS are:
> #   0 - halt (Do NOT set initdefault to this)
> #   1 - Single user mode
> #   2 - Multiuser, without NFS (The same as 3, if you do not have
> networking)
> #   3 - Full multiuser mode
> #   4 - unused
> #   5 - X11
> #   6 - reboot (Do NOT set initdefault to this)
> #
> #id:5:initdefault:
> id:3:initdefault:
>
> # System initialization.
> si::sysinit:/etc/rc.d/rc.sysinit
>
> l0:0:wait:/etc/rc.d/rc 0
> l1:1:wait:/etc/rc.d/rc 1
> l2:2:wait:/etc/rc.d/rc 2
> l3:3:wait:/etc/rc.d/rc 3
> l4:4:wait:/etc/rc.d/rc 4
> l5:5:wait:/etc/rc.d/rc 5
> l6:6:wait:/etc/rc.d/rc 6
>
> # Trap CTRL-ALT-DELETE
> ca::ctrlaltdel:/sbin/shutdown -t3 -r now
>
> # When our UPS tells us power has failed, assume we have a few minutes
> # of power left.  Schedule a shutdown for 2 minutes from now.
> # This does, of course, assume you have powerd installed and your
> # UPS connected and working correctly.
> pf::powerfail:/sbin/shutdown -f -h +2 "Power Failure; System Shutting Down"
>
> # If power was restored before the shutdown kicked in, cancel it.
> pr:12345:powerokwait:/sbin/shutdown -c "Power Restored; Shutdown Cancelled"
>
> # Run gettys in standard runlevels
> co:2345:respawn:/sbin/agetty ttyS0 115200 vt100
> 1:2345:respawn:/sbin/mingetty tty1
> 2:2345:respawn:/sbin/mingetty tty2
> 3:2345:respawn:/sbin/mingetty tty3
> 4:2345:respawn:/sbin/mingetty tty4
> 5:2345:respawn:/sbin/mingetty tty5
> 6:2345:respawn:/sbin/mingetty tty6
>
> ----- Original Message -----
> From: "Anton" 
> To: 
> Sent: Thursday, April 05, 2007 11:43 AM
> Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > On Thu, 5 Apr 2007 11:27:11 -0700
> > "Beneo"  wrote:
> > > Hi all,
> > >
> > > I was using LinuxBIOS to boot Linux from IDE from a Broadcom HT,1000
> > > south bridge based  platform , I encounter an issue which is hard to
> > > understand. I hope somebody can give me some insight.
> > >
> > > During boot, Linux stoped at the point that it displayed a message:
> > > "Configuring kernel parameters:  [  OK  ]"
> > >
> > > then I pressed "CTRL-C",  Linux continue to boot, but display a message:
> > > "Starting automount:Cannot create temp file /tmp/autofs.meQYNM, could not
> > > make temp file"
> > >
> > > then Linux boot Skipped rest of initialization, like probe the network
> > > and etc. but Linux did give me a login prompt, I can login using root,
> > > but I cannot write any file at root directory, It appears to me Entire
> > > Linux file system becomes readonly.

It may, or may not, be the fault of linuxbios:
Things to check for hints:
    dmesg
    /var/log/sys

Usually the root file system is first mounted ro and the startup routine
checks the file system. If ok, root is then remounted rw. Slackware does
this in rc.S.

Did you edit any rc files? if so, you may have forgotten to chmod +x.

russ

[ST]

Hi Beneo
> then Linux boot Skipped rest of initialization, like probe the network and
> etc. but Linux did give me a login prompt, I can login using root, but I
> cannot write any file at root directory, It appears to me Entire Linux file
> system becomes readonly.

can you change boot parameters to: init=/bin/sh and get an shell?
If you get an shell can you issue an: "mount -o remount,rw /"?

This helps with normal systems but no experience with linuxbios systems yet.

ST

[Beneo]

Hi ST

I used a parameter at filo like this:

hda1:/vmlinuz-2.4.20-8smp initrd=hda1:/initrd-2.4.20-8smp.img ro 
root=/dev/hda3 debug console=tty0 console=ttyS0,115200

Would you tell me how to modify the boot parameter for boot from shell?

Thanks

Beneo

----- Original Message ----- 
From: "ST" 
To: 
Sent: Thursday, April 05, 2007 12:18 PM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> Hi Beneo
> > then Linux boot Skipped rest of initialization, like probe the network 
> > and
> > etc. but Linux did give me a login prompt, I can login using root, but I
> > cannot write any file at root directory, It appears to me Entire Linux 
> > file
> > system becomes readonly.
>
> can you change boot parameters to: init=/bin/sh and get an shell?
> If you get an shell can you issue an: "mount -o remount,rw /"?
>
> This helps with normal systems but no experience with linuxbios systems 
> yet.
>
> ST

[Carl-Daniel Hailfinger]

On 05.04.2007 23:41, Beneo wrote:
> I used a parameter at filo like this:
>
> hda1:/vmlinuz-2.4.20-8smp initrd=hda1:/initrd-2.4.20-8smp.img ro 
> root=/dev/hda3 debug console=tty0 console=ttyS0,115200

Remove "ro". "ro" means readonly.

Regards,
Carl-Daniel

[Beneo]

when boot from factory BIOS, cat /proc/cmdline output this:
ro root=LABEL=/ debug console=ttyS0,115200n8

I removed ro from the line, it still have the same issue. I made up the filo 
linux boot line by look at the grub.conf file in this Linux install.
I don't know, I'm not a expert on this. but the "ro" maybe means mount the 
randisk image readonly, not the root file system?

Come back to my question,  does LinuxBIOS actually setup IDE interface. I 
know it enables IDE from Chipset level, but does it actully program the IDE 
interface, for example using port 1F0, 170 and etc.

Thanks

Beneo

----- Original Message ----- 
From: "Carl-Daniel Hailfinger" 
To: "Beneo" 
Cc: "ST" ; 
Sent: Thursday, April 05, 2007 2:46 PM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> On 05.04.2007 23:41, Beneo wrote:
> > I used a parameter at filo like this:
> >
> > hda1:/vmlinuz-2.4.20-8smp initrd=hda1:/initrd-2.4.20-8smp.img ro
> > root=/dev/hda3 debug console=tty0 console=ttyS0,115200
>
> Remove "ro". "ro" means readonly.
>
> Regards,
> Carl-Daniel
> -- 
> http://www.hailfinger.org/

[Peter Stuge]

On Thu, Apr 05, 2007 at 03:25:38PM -0700, Beneo wrote:
> when boot from factory BIOS, cat /proc/cmdline output this:
> ro root=LABEL=/ debug console=ttyS0,115200n8

Ok. Looks similar to what you're using in FILO.
> I removed ro from the line, it still have the same issue.

Yes, I would expect this.
> I made up the filo linux boot line by look at the grub.conf file
> in this Linux install.

Ok, good.
> I don't know, I'm not a expert on this. but the "ro" maybe means
> mount the randisk image readonly, not the root file system?

It applies to both initrd and root.
> Come back to my question, does LinuxBIOS actually setup IDE
> interface.

Yes, it does do setup.
> I know it enables IDE from Chipset level, but does it actully
> program the IDE interface, for example using port 1F0, 170 and etc.

LinuxBIOS does not access any IDE devices. FILO does however.

But I don't think IDE access is the real problem.

Try looking at the startup scripts and see what they are doing around
the point where they hang. We can only guess why the script is
hanging at this point.

The relevant script may be in the initrd you are using, or in the
root fs.

How long did you let the script "hang" before giving up on it?

Perhaps it's in fact busy working with something that takes too long
because of another problem in LinuxBIOS. If we know what the script
is doing we can make better guesses about why it is hanging.

You could also try to skip using the initrd, but I don't know/like
Red Hat so I don't know if it is in fact required to boot the system.

//Peter

[Beneo]

I kept the Linux hang at that point for 2 or 3 hours, so I think it probably 
would not go anywhere if I don't do something.

thanks for the suggestion, I will search and check what Linux is doing when 
it hang, I will let you know.

thnks

Beneo.

----- Original Message ----- 
From: "Peter Stuge" 
To: 
Sent: Thursday, April 05, 2007 4:16 PM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> On Thu, Apr 05, 2007 at 03:25:38PM -0700, Beneo wrote:
> > when boot from factory BIOS, cat /proc/cmdline output this:
> > ro root=LABEL=/ debug console=ttyS0,115200n8
>
> Ok. Looks similar to what you're using in FILO.
> > I removed ro from the line, it still have the same issue.
>
> Yes, I would expect this.
> > I made up the filo linux boot line by look at the grub.conf file
> > in this Linux install.
>
> Ok, good.
> > I don't know, I'm not a expert on this. but the "ro" maybe means
> > mount the randisk image readonly, not the root file system?
>
> It applies to both initrd and root.
> > Come back to my question, does LinuxBIOS actually setup IDE
> > interface.
>
> Yes, it does do setup.
> > I know it enables IDE from Chipset level, but does it actully
> > program the IDE interface, for example using port 1F0, 170 and etc.
>
> LinuxBIOS does not access any IDE devices. FILO does however.
>
> But I don't think IDE access is the real problem.
>
> Try looking at the startup scripts and see what they are doing around
> the point where they hang. We can only guess why the script is
> hanging at this point.
>
> The relevant script may be in the initrd you are using, or in the
> root fs.
>
> How long did you let the script "hang" before giving up on it?
>
> Perhaps it's in fact busy working with something that takes too long
> because of another problem in LinuxBIOS. If we know what the script
> is doing we can make better guesses about why it is hanging.
>
> You could also try to skip using the initrd, but I don't know/like
> Red Hat so I don't know if it is in fact required to boot the system.
>
> //Peter

[Peter Stuge]

On Thu, Apr 05, 2007 at 04:53:39PM -0700, Beneo wrote:
> I kept the Linux hang at that point for 2 or 3 hours, so I think
> it probably would not go anywhere if I don't do something.

Wow, ok, I think you are right. :)
> thanks for the suggestion, I will search and check what Linux is
> doing when it hang, I will let you know.

Cool. Since you didn't post what happens up to the hanging point I
can't say for sure if it's in the initrd, but I think it probably is.

//Peter

[Beneo]

Peter,

Thanks for the tips, I get it working.
I searched the Linux startup script, it is a RTC related, I rememer it is 
something related to Daylight saving, so I go check the RTC, found the IRQ 
for that RTC is not programmed, once I get that programmed, it can boot all 
the way now.

I still have a network issue, my network card can not be get a IP address 
from a DHCP server, I'm relatively sure the IRQ in MP table is assigned 
correctly, not sure why it hang yet. Other than that, it looks OK.

Thanks for the help

Beneo

----- Original Message ----- 
From: "Peter Stuge" 
To: 
Sent: Thursday, April 05, 2007 5:37 PM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> On Thu, Apr 05, 2007 at 04:53:39PM -0700, Beneo wrote:
> > I kept the Linux hang at that point for 2 or 3 hours, so I think
> > it probably would not go anywhere if I don't do something.
>
> Wow, ok, I think you are right. :)
> > thanks for the suggestion, I will search and check what Linux is
> > doing when it hang, I will let you know.
>
> Cool. Since you didn't post what happens up to the hanging point I
> can't say for sure if it's in the initrd, but I think it probably is.
>
> //Peter

[Peter Stuge]

Hi Beneo,

On Fri, Apr 06, 2007 at 06:58:22PM -0700, Beneo wrote:
> Thanks for the tips, I get it working.

Glad to hear it!
> I searched the Linux startup script, it is a RTC related, I rememer
> it is something related to Daylight saving, so I go check the RTC,
> found the IRQ for that RTC is not programmed, once I get that
> programmed, it can boot all the way now.

Aha! Could you provide a patch for the RTC IRQ issue?
> I still have a network issue, my network card can not be get a IP
> address from a DHCP server, I'm relatively sure the IRQ in MP table
> is assigned correctly, not sure why it hang yet. Other than that,
> it looks OK.

Hm. You can check if interrupts are coming through by configuring an
IP address manually and sending some pings, and then:

cat /proc/interrupts
> Thanks for the help

I'm glad it works. :)

//Peter

[Beneo]

Peter,

For the RTC patch, it is porting for a SMSC SIO, I didn't see LinuxBIOS has 
this SIO porting. The part number is SCH--4703, I don't know how to 
contribute it to LinuxBIOS tree.

I haven't get the chance to take look at my NIC issue yet, I will certainly 
check the interrupt when I get the chance.

Thanks

Beneo

----- Original Message ----- 
From: "Peter Stuge" 
To: 
Sent: Friday, April 06, 2007 9:00 PM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> Hi Beneo,
>
> On Fri, Apr 06, 2007 at 06:58:22PM -0700, Beneo wrote:
> > Thanks for the tips, I get it working.
>
> Glad to hear it!
> > I searched the Linux startup script, it is a RTC related, I rememer
> > it is something related to Daylight saving, so I go check the RTC,
> > found the IRQ for that RTC is not programmed, once I get that
> > programmed, it can boot all the way now.
>
> Aha! Could you provide a patch for the RTC IRQ issue?
> > I still have a network issue, my network card can not be get a IP
> > address from a DHCP server, I'm relatively sure the IRQ in MP table
> > is assigned correctly, not sure why it hang yet. Other than that,
> > it looks OK.
>
> Hm. You can check if interrupts are coming through by configuring an
> IP address manually and sending some pings, and then:
>
> cat /proc/interrupts
> > Thanks for the help
>
> I'm glad it works. :)
>
> //Peter

[Peter Stuge]

On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> For the RTC patch, it is porting for a SMSC SIO, I didn't see
> LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> don't know how to contribute it to LinuxBIOS tree.

Ahh! It's new code, sorry, forgot that.

Please have a look at http://linuxbios.org/Development_Guidelines to
learn most if not all you need to know to submit a patch.

Also, when submitting patches, please make sure they apply cleanly to
the very latest revision of the tree.

The simplest way to do this is to always keep your own tree updated
by running svn up now and then, or when you see a new revision
announced on the list. After an update, there may be conflicts that
need to be resolved manually by you if you have been working on code
that was also changed in the new revision.
> I haven't get the chance to take look at my NIC issue yet, I will
> certainly check the interrupt when I get the chance.

Feel free to ask the list if you run into trouble!

//Peter

[LinuxBIOS] PIC mode or APIC mode

[Beneo]

I was working this Broadcom bcm 5785 based reference board with LinuxBIOS, I 
have a NIC interrupt related issue, the eth0 DHCP detection always fail, so 
I can not get an IP address for that NIC. I did what Peter Stuge suggested 
to dump the /proc/interrupts. It turns out the Linux is using PIC mode, not 
APIC mode. (Linux report multiple CPUs, so I assume Linux took MP Table 
somewhat correctly)

I checked chipset setting, the APIC indeed is enabled. I don't see Linux is 
doing anything for PIC or APIC, I assume Linux kernel will correctly 
initialize PIC or APIC depends on MP Table reporting. right?

If anybody has any idea on why Linux is running at PIC mode, Please let me 
know. I will be very appreciated.

Thanks

Beneo

--- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
[root@localhost root]# cat /proc/interrupts
           CPU0       CPU1
  0:      13538       1521          XT-PIC  timer
  1:          0          0          XT-PIC  keyboard
  2:          0          0          XT-PIC  cascade
  4:        302        153          XT-PIC  serial
  8:          1          0          XT-PIC  rtc
 14:       4729       1803          XT-PIC  ide0
NMI:          0          0
LOC:      14778      14897
ERR:      21478
MIS:          0

--- When using factory BIOS, the cat /proc/interrupts looks like this, it is 
in APIC mode.
[root@localhost root]# cat /proc/interrupts
           CPU0       CPU1
  0:        719       8393    IO-APIC-edge  timer
  1:          0          4    IO-APIC-edge  keyboard
  2:          0          0          XT-PIC  cascade
  4:          5        505    IO-APIC-edge  serial
  8:          0          1    IO-APIC-edge  rtc
 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci, usb-ohci
 12:          1          6    IO-APIC-edge  PS/2 Mouse
 14:       1433       5111    IO-APIC-edge  ide0
 38:          7        112   IO-APIC-level  eth0
NMI:          0          0
LOC:       9020       8869
ERR:          0
MIS:          0

----- Original Message -----
From: "Peter Stuge" 
To: 
Sent: Monday, April 09, 2007 10:23 AM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > don't know how to contribute it to LinuxBIOS tree.
>
> Ahh! It's new code, sorry, forgot that.
>
> Please have a look at http://linuxbios.org/Development_Guidelines to
> learn most if not all you need to know to submit a patch.
>
> Also, when submitting patches, please make sure they apply cleanly to
> the very latest revision of the tree.
>
> The simplest way to do this is to always keep your own tree updated
> by running svn up now and then, or when you see a new revision
> announced on the list. After an update, there may be conflicts that
> need to be resolved manually by you if you have been working on code
> that was also changed in the new revision.
> > I haven't get the chance to take look at my NIC issue yet, I will
> > certainly check the interrupt when I get the chance.
>
> Feel free to ask the list if you run into trouble!
>
> //Peter

[LinuxBIOS] PIC mode or APIC mode

[Lu, Yinghai]

BLAST?

That code in the tree should work.

YH

-----Original Message-----
From: [e-mail]
[mailto:[e-mail]] On Behalf Of Beneo
Sent: Friday, April 13, 2007 10:48 AM
To: Peter Stuge; [e-mail]
Subject: [LinuxBIOS] PIC mode or APIC mode

I was working this Broadcom bcm 5785 based reference board with
LinuxBIOS, I 
have a NIC interrupt related issue, the eth0 DHCP detection always fail,
so 
I can not get an IP address for that NIC. I did what Peter Stuge
suggested 
to dump the /proc/interrupts. It turns out the Linux is using PIC mode,
not 
APIC mode. (Linux report multiple CPUs, so I assume Linux took MP Table 
somewhat correctly)

I checked chipset setting, the APIC indeed is enabled. I don't see Linux
is 
doing anything for PIC or APIC, I assume Linux kernel will correctly 
initialize PIC or APIC depends on MP Table reporting. right?

If anybody has any idea on why Linux is running at PIC mode, Please let
me 
know. I will be very appreciated.

Thanks

Beneo

--- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
[root@localhost root]# cat /proc/interrupts
           CPU0       CPU1
  0:      13538       1521          XT-PIC  timer
  1:          0          0          XT-PIC  keyboard
  2:          0          0          XT-PIC  cascade
  4:        302        153          XT-PIC  serial
  8:          1          0          XT-PIC  rtc
 14:       4729       1803          XT-PIC  ide0
NMI:          0          0
LOC:      14778      14897
ERR:      21478
MIS:          0

--- When using factory BIOS, the cat /proc/interrupts looks like this,
it is 
in APIC mode.
[root@localhost root]# cat /proc/interrupts
           CPU0       CPU1
  0:        719       8393    IO-APIC-edge  timer
  1:          0          4    IO-APIC-edge  keyboard
  2:          0          0          XT-PIC  cascade
  4:          5        505    IO-APIC-edge  serial
  8:          0          1    IO-APIC-edge  rtc
 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci, usb-ohci
 12:          1          6    IO-APIC-edge  PS/2 Mouse
 14:       1433       5111    IO-APIC-edge  ide0
 38:          7        112   IO-APIC-level  eth0
NMI:          0          0
LOC:       9020       8869
ERR:          0
MIS:          0

----- Original Message -----
From: "Peter Stuge" 
To: 
Sent: Monday, April 09, 2007 10:23 AM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > don't know how to contribute it to LinuxBIOS tree.
>
> Ahh! It's new code, sorry, forgot that.
>
> Please have a look at http://linuxbios.org/Development_Guidelines to
> learn most if not all you need to know to submit a patch.
>
> Also, when submitting patches, please make sure they apply cleanly to
> the very latest revision of the tree.
>
> The simplest way to do this is to always keep your own tree updated
> by running svn up now and then, or when you see a new revision
> announced on the list. After an update, there may be conflicts that
> need to be resolved manually by you if you have been working on code
> that was also changed in the new revision.
> > I haven't get the chance to take look at my NIC issue yet, I will
> > certainly check the interrupt when I get the chance.
>
> Feel free to ask the list if you run into trouble!
>
> //Peter

[LinuxBIOS] PIC mode or APIC mode

[Beneo]

It is not Blast, it is newer version of Broadcom board, it is called 
Explosion board.

Beneo

----- Original Message ----- 
From: "Lu, Yinghai" 
To: "Beneo" ; "Peter Stuge" ; 

Sent: Friday, April 13, 2007 10:54 AM
Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> BLAST?
>
> That code in the tree should work.
>
> YH
>
> -----Original Message-----
> From: [e-mail]
> [mailto:[e-mail]] On Behalf Of Beneo
> Sent: Friday, April 13, 2007 10:48 AM
> To: Peter Stuge; [e-mail]
> Subject: [LinuxBIOS] PIC mode or APIC mode
>
> I was working this Broadcom bcm 5785 based reference board with
> LinuxBIOS, I
> have a NIC interrupt related issue, the eth0 DHCP detection always fail,
> so
> I can not get an IP address for that NIC. I did what Peter Stuge
> suggested
> to dump the /proc/interrupts. It turns out the Linux is using PIC mode,
> not
> APIC mode. (Linux report multiple CPUs, so I assume Linux took MP Table
> somewhat correctly)
>
> I checked chipset setting, the APIC indeed is enabled. I don't see Linux
> is
> doing anything for PIC or APIC, I assume Linux kernel will correctly
> initialize PIC or APIC depends on MP Table reporting. right?
>
> If anybody has any idea on why Linux is running at PIC mode, Please let
> me
> know. I will be very appreciated.
>
> Thanks
>
> Beneo
>
> --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:      13538       1521          XT-PIC  timer
>  1:          0          0          XT-PIC  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:        302        153          XT-PIC  serial
>  8:          1          0          XT-PIC  rtc
> 14:       4729       1803          XT-PIC  ide0
> NMI:          0          0
> LOC:      14778      14897
> ERR:      21478
> MIS:          0
>
> --- When using factory BIOS, the cat /proc/interrupts looks like this,
> it is
> in APIC mode.
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:        719       8393    IO-APIC-edge  timer
>  1:          0          4    IO-APIC-edge  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:          5        505    IO-APIC-edge  serial
>  8:          0          1    IO-APIC-edge  rtc
> 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci, usb-ohci
> 12:          1          6    IO-APIC-edge  PS/2 Mouse
> 14:       1433       5111    IO-APIC-edge  ide0
> 38:          7        112   IO-APIC-level  eth0
> NMI:          0          0
> LOC:       9020       8869
> ERR:          0
> MIS:          0
>
> ----- Original Message -----
> From: "Peter Stuge" 
> To: 
> Sent: Monday, April 09, 2007 10:23 AM
> Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > don't know how to contribute it to LinuxBIOS tree.
> >
> > Ahh! It's new code, sorry, forgot that.
> >
> > Please have a look at http://linuxbios.org/Development_Guidelines to
> > learn most if not all you need to know to submit a patch.
> >
> > Also, when submitting patches, please make sure they apply cleanly to
> > the very latest revision of the tree.
> >
> > The simplest way to do this is to always keep your own tree updated
> > by running svn up now and then, or when you see a new revision
> > announced on the list. After an update, there may be conflicts that
> > need to be resolved manually by you if you have been working on code
> > that was also changed in the new revision.
> > > I haven't get the chance to take look at my NIC issue yet, I will
> > > certainly check the interrupt when I get the chance.
> >
> > Feel free to ask the list if you run into trouble!
> >
> > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Lu, Yinghai]

Good.

I can not get serial output from that MB, what is your superio on the
MB?

YH

-----Original Message-----
From: Beneo [mailto:[e-mail]] 
Sent: Friday, April 13, 2007 10:59 AM
To: Lu, Yinghai; Peter Stuge; [e-mail]
Subject: Re: [LinuxBIOS] PIC mode or APIC mode

It is not Blast, it is newer version of Broadcom board, it is called 
Explosion board.

Beneo

----- Original Message ----- 
From: "Lu, Yinghai" 
To: "Beneo" ; "Peter Stuge"
; 

Sent: Friday, April 13, 2007 10:54 AM
Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> BLAST?
>
> That code in the tree should work.
>
> YH
>
> -----Original Message-----
> From: [e-mail]
> [mailto:[e-mail]] On Behalf Of Beneo
> Sent: Friday, April 13, 2007 10:48 AM
> To: Peter Stuge; [e-mail]
> Subject: [LinuxBIOS] PIC mode or APIC mode
>
> I was working this Broadcom bcm 5785 based reference board with
> LinuxBIOS, I
> have a NIC interrupt related issue, the eth0 DHCP detection always

fail,
> so
> I can not get an IP address for that NIC. I did what Peter Stuge
> suggested
> to dump the /proc/interrupts. It turns out the Linux is using PIC

mode,
> not
> APIC mode. (Linux report multiple CPUs, so I assume Linux took MP

Table
> somewhat correctly)
>
> I checked chipset setting, the APIC indeed is enabled. I don't see

Linux
> is
> doing anything for PIC or APIC, I assume Linux kernel will correctly
> initialize PIC or APIC depends on MP Table reporting. right?
>
> If anybody has any idea on why Linux is running at PIC mode, Please

let
> me
> know. I will be very appreciated.
>
> Thanks
>
> Beneo
>
> --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:      13538       1521          XT-PIC  timer
>  1:          0          0          XT-PIC  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:        302        153          XT-PIC  serial
>  8:          1          0          XT-PIC  rtc
> 14:       4729       1803          XT-PIC  ide0
> NMI:          0          0
> LOC:      14778      14897
> ERR:      21478
> MIS:          0
>
> --- When using factory BIOS, the cat /proc/interrupts looks like this,
> it is
> in APIC mode.
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:        719       8393    IO-APIC-edge  timer
>  1:          0          4    IO-APIC-edge  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:          5        505    IO-APIC-edge  serial
>  8:          0          1    IO-APIC-edge  rtc
> 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,

usb-ohci
> 12:          1          6    IO-APIC-edge  PS/2 Mouse
> 14:       1433       5111    IO-APIC-edge  ide0
> 38:          7        112   IO-APIC-level  eth0
> NMI:          0          0
> LOC:       9020       8869
> ERR:          0
> MIS:          0
>
> ----- Original Message -----
> From: "Peter Stuge" 
> To: 
> Sent: Monday, April 09, 2007 10:23 AM
> Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > don't know how to contribute it to LinuxBIOS tree.
> >
> > Ahh! It's new code, sorry, forgot that.
> >
> > Please have a look at http://linuxbios.org/Development_Guidelines to
> > learn most if not all you need to know to submit a patch.
> >
> > Also, when submitting patches, please make sure they apply cleanly to
> > the very latest revision of the tree.
> >
> > The simplest way to do this is to always keep your own tree updated
> > by running svn up now and then, or when you see a new revision
> > announced on the list. After an update, there may be conflicts that
> > need to be resolved manually by you if you have been working on code
> > that was also changed in the new revision.
> > > I haven't get the chance to take look at my NIC issue yet, I will
> > > certainly check the interrupt when I get the chance.
> >
> > Feel free to ask the list if you run into trouble!
> >
> > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Beneo]

it is SMSC SCH 4307, I ported some code for that serail, it works.

Beneo

----- Original Message ----- 
From: "Lu, Yinghai" 
To: "Beneo" ; "Peter Stuge" ; 

Sent: Friday, April 13, 2007 11:01 AM
Subject: RE: [LinuxBIOS] PIC mode or APIC mode

Good.

I can not get serial output from that MB, what is your superio on the
MB?

YH

-----Original Message-----
From: Beneo [mailto:[e-mail]]
Sent: Friday, April 13, 2007 10:59 AM
To: Lu, Yinghai; Peter Stuge; [e-mail]
Subject: Re: [LinuxBIOS] PIC mode or APIC mode

It is not Blast, it is newer version of Broadcom board, it is called
Explosion board.

Beneo

----- Original Message ----- 
From: "Lu, Yinghai" 
To: "Beneo" ; "Peter Stuge"
;

Sent: Friday, April 13, 2007 10:54 AM
Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> BLAST?
>
> That code in the tree should work.
>
> YH
>
> -----Original Message-----
> From: [e-mail]
> [mailto:[e-mail]] On Behalf Of Beneo
> Sent: Friday, April 13, 2007 10:48 AM
> To: Peter Stuge; [e-mail]
> Subject: [LinuxBIOS] PIC mode or APIC mode
>
> I was working this Broadcom bcm 5785 based reference board with
> LinuxBIOS, I
> have a NIC interrupt related issue, the eth0 DHCP detection always

fail,
> so
> I can not get an IP address for that NIC. I did what Peter Stuge
> suggested
> to dump the /proc/interrupts. It turns out the Linux is using PIC

mode,
> not
> APIC mode. (Linux report multiple CPUs, so I assume Linux took MP

Table
> somewhat correctly)
>
> I checked chipset setting, the APIC indeed is enabled. I don't see

Linux
> is
> doing anything for PIC or APIC, I assume Linux kernel will correctly
> initialize PIC or APIC depends on MP Table reporting. right?
>
> If anybody has any idea on why Linux is running at PIC mode, Please

let
> me
> know. I will be very appreciated.
>
> Thanks
>
> Beneo
>
> --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:      13538       1521          XT-PIC  timer
>  1:          0          0          XT-PIC  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:        302        153          XT-PIC  serial
>  8:          1          0          XT-PIC  rtc
> 14:       4729       1803          XT-PIC  ide0
> NMI:          0          0
> LOC:      14778      14897
> ERR:      21478
> MIS:          0
>
> --- When using factory BIOS, the cat /proc/interrupts looks like this,
> it is
> in APIC mode.
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:        719       8393    IO-APIC-edge  timer
>  1:          0          4    IO-APIC-edge  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:          5        505    IO-APIC-edge  serial
>  8:          0          1    IO-APIC-edge  rtc
> 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,

usb-ohci
> 12:          1          6    IO-APIC-edge  PS/2 Mouse
> 14:       1433       5111    IO-APIC-edge  ide0
> 38:          7        112   IO-APIC-level  eth0
> NMI:          0          0
> LOC:       9020       8869
> ERR:          0
> MIS:          0
>
> ----- Original Message -----
> From: "Peter Stuge" 
> To: 
> Sent: Monday, April 09, 2007 10:23 AM
> Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > don't know how to contribute it to LinuxBIOS tree.
> >
> > Ahh! It's new code, sorry, forgot that.
> >
> > Please have a look at http://linuxbios.org/Development_Guidelines to
> > learn most if not all you need to know to submit a patch.
> >
> > Also, when submitting patches, please make sure they apply cleanly to
> > the very latest revision of the tree.
> >
> > The simplest way to do this is to always keep your own tree updated
> > by running svn up now and then, or when you see a new revision
> > announced on the list. After an update, there may be conflicts that
> > need to be resolved manually by you if you have been working on code
> > that was also changed in the new revision.
> > > I haven't get the chance to take look at my NIC issue yet, I will
> > > certainly check the interrupt when I get the chance.
> >
> > Feel free to ask the list if you run into trouble!
> >
> > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Lu, Yinghai]

The MB I got with SCH 4304, and schematic said PC87427...

YH

-----Original Message-----
From: Beneo [mailto:[e-mail]] 
Sent: Friday, April 13, 2007 11:12 AM
To: Lu, Yinghai; Peter Stuge; [e-mail]
Subject: Re: [LinuxBIOS] PIC mode or APIC mode

it is SMSC SCH 4307, I ported some code for that serail, it works.

Beneo

----- Original Message ----- 
From: "Lu, Yinghai" 
To: "Beneo" ; "Peter Stuge"
; 

Sent: Friday, April 13, 2007 11:01 AM
Subject: RE: [LinuxBIOS] PIC mode or APIC mode

Good.

I can not get serial output from that MB, what is your superio on the
MB?

YH

-----Original Message-----
From: Beneo [mailto:[e-mail]]
Sent: Friday, April 13, 2007 10:59 AM
To: Lu, Yinghai; Peter Stuge; [e-mail]
Subject: Re: [LinuxBIOS] PIC mode or APIC mode

It is not Blast, it is newer version of Broadcom board, it is called
Explosion board.

Beneo

----- Original Message ----- 
From: "Lu, Yinghai" 
To: "Beneo" ; "Peter Stuge"
;

Sent: Friday, April 13, 2007 10:54 AM
Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> BLAST?
>
> That code in the tree should work.
>
> YH
>
> -----Original Message-----
> From: [e-mail]
> [mailto:[e-mail]] On Behalf Of Beneo
> Sent: Friday, April 13, 2007 10:48 AM
> To: Peter Stuge; [e-mail]
> Subject: [LinuxBIOS] PIC mode or APIC mode
>
> I was working this Broadcom bcm 5785 based reference board with
> LinuxBIOS, I
> have a NIC interrupt related issue, the eth0 DHCP detection always

fail,
> so
> I can not get an IP address for that NIC. I did what Peter Stuge
> suggested
> to dump the /proc/interrupts. It turns out the Linux is using PIC

mode,
> not
> APIC mode. (Linux report multiple CPUs, so I assume Linux took MP

Table
> somewhat correctly)
>
> I checked chipset setting, the APIC indeed is enabled. I don't see

Linux
> is
> doing anything for PIC or APIC, I assume Linux kernel will correctly
> initialize PIC or APIC depends on MP Table reporting. right?
>
> If anybody has any idea on why Linux is running at PIC mode, Please

let
> me
> know. I will be very appreciated.
>
> Thanks
>
> Beneo
>
> --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:      13538       1521          XT-PIC  timer
>  1:          0          0          XT-PIC  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:        302        153          XT-PIC  serial
>  8:          1          0          XT-PIC  rtc
> 14:       4729       1803          XT-PIC  ide0
> NMI:          0          0
> LOC:      14778      14897
> ERR:      21478
> MIS:          0
>
> --- When using factory BIOS, the cat /proc/interrupts looks like this,
> it is
> in APIC mode.
> [root@localhost root]# cat /proc/interrupts
>           CPU0       CPU1
>  0:        719       8393    IO-APIC-edge  timer
>  1:          0          4    IO-APIC-edge  keyboard
>  2:          0          0          XT-PIC  cascade
>  4:          5        505    IO-APIC-edge  serial
>  8:          0          1    IO-APIC-edge  rtc
> 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,

usb-ohci
> 12:          1          6    IO-APIC-edge  PS/2 Mouse
> 14:       1433       5111    IO-APIC-edge  ide0
> 38:          7        112   IO-APIC-level  eth0
> NMI:          0          0
> LOC:       9020       8869
> ERR:          0
> MIS:          0
>
> ----- Original Message -----
> From: "Peter Stuge" 
> To: 
> Sent: Monday, April 09, 2007 10:23 AM
> Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > don't know how to contribute it to LinuxBIOS tree.
> >
> > Ahh! It's new code, sorry, forgot that.
> >
> > Please have a look at http://linuxbios.org/Development_Guidelines to
> > learn most if not all you need to know to submit a patch.
> >
> > Also, when submitting patches, please make sure they apply cleanly to
> > the very latest revision of the tree.
> >
> > The simplest way to do this is to always keep your own tree updated
> > by running svn up now and then, or when you see a new revision
> > announced on the list. After an update, there may be conflicts that
> > need to be resolved manually by you if you have been working on code
> > that was also changed in the new revision.
> > > I haven't get the chance to take look at my NIC issue yet, I will
> > > certainly check the interrupt when I get the chance.
> >
> > Feel free to ask the list if you run into trouble!
> >
> > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Tom Sylla]

Explosion Rev A shipped with SCH430{4,7}. Explosion Rev B has a PC87427 on it.

On 4/13/07, Lu, Yinghai  wrote:
> The MB I got with SCH 4304, and schematic said PC87427...
>
> YH
>
> -----Original Message-----
> From: Beneo [mailto:[e-mail]]
> Sent: Friday, April 13, 2007 11:12 AM
> To: Lu, Yinghai; Peter Stuge; [e-mail]
> Subject: Re: [LinuxBIOS] PIC mode or APIC mode
>
> it is SMSC SCH 4307, I ported some code for that serail, it works.
>
> Beneo
>
> ----- Original Message -----
> From: "Lu, Yinghai" 
> To: "Beneo" ; "Peter Stuge"
> ;
>
> Sent: Friday, April 13, 2007 11:01 AM
> Subject: RE: [LinuxBIOS] PIC mode or APIC mode
>
> Good.
>
> I can not get serial output from that MB, what is your superio on the
> MB?
>
> YH
>
> -----Original Message-----
> From: Beneo [mailto:[e-mail]]
> Sent: Friday, April 13, 2007 10:59 AM
> To: Lu, Yinghai; Peter Stuge; [e-mail]
> Subject: Re: [LinuxBIOS] PIC mode or APIC mode
>
> It is not Blast, it is newer version of Broadcom board, it is called
> Explosion board.
>
> Beneo
>
> ----- Original Message -----
> From: "Lu, Yinghai" 
> To: "Beneo" ; "Peter Stuge"
> ;
>
> Sent: Friday, April 13, 2007 10:54 AM
> Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > BLAST?
> >
> > That code in the tree should work.
> >
> > YH
> >
> > -----Original Message-----
> > From: [e-mail]
> > [mailto:[e-mail]] On Behalf Of Beneo
> > Sent: Friday, April 13, 2007 10:48 AM
> > To: Peter Stuge; [e-mail]
> > Subject: [LinuxBIOS] PIC mode or APIC mode
> >
> > I was working this Broadcom bcm 5785 based reference board with
> > LinuxBIOS, I
> > have a NIC interrupt related issue, the eth0 DHCP detection always
>
> fail,
> > so
> > I can not get an IP address for that NIC. I did what Peter Stuge
> > suggested
> > to dump the /proc/interrupts. It turns out the Linux is using PIC
>
> mode,
> > not
> > APIC mode. (Linux report multiple CPUs, so I assume Linux took MP
>
> Table
> > somewhat correctly)
> >
> > I checked chipset setting, the APIC indeed is enabled. I don't see
>
> Linux
> > is
> > doing anything for PIC or APIC, I assume Linux kernel will correctly
> > initialize PIC or APIC depends on MP Table reporting. right?
> >
> > If anybody has any idea on why Linux is running at PIC mode, Please
>
> let
> > me
> > know. I will be very appreciated.
> >
> > Thanks
> >
> > Beneo
> >
> > --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
> > [root@localhost root]# cat /proc/interrupts
> >           CPU0       CPU1
> >  0:      13538       1521          XT-PIC  timer
> >  1:          0          0          XT-PIC  keyboard
> >  2:          0          0          XT-PIC  cascade
> >  4:        302        153          XT-PIC  serial
> >  8:          1          0          XT-PIC  rtc
> > 14:       4729       1803          XT-PIC  ide0
> > NMI:          0          0
> > LOC:      14778      14897
> > ERR:      21478
> > MIS:          0
> >
> > --- When using factory BIOS, the cat /proc/interrupts looks like this,
> > it is
> > in APIC mode.
> > [root@localhost root]# cat /proc/interrupts
> >           CPU0       CPU1
> >  0:        719       8393    IO-APIC-edge  timer
> >  1:          0          4    IO-APIC-edge  keyboard
> >  2:          0          0          XT-PIC  cascade
> >  4:          5        505    IO-APIC-edge  serial
> >  8:          0          1    IO-APIC-edge  rtc
> > 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,
>
> usb-ohci
> > 12:          1          6    IO-APIC-edge  PS/2 Mouse
> > 14:       1433       5111    IO-APIC-edge  ide0
> > 38:          7        112   IO-APIC-level  eth0
> > NMI:          0          0
> > LOC:       9020       8869
> > ERR:          0
> > MIS:          0
> >
> > ----- Original Message -----
> > From: "Peter Stuge" 
> > To: 
> > Sent: Monday, April 09, 2007 10:23 AM
> > Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > > don't know how to contribute it to LinuxBIOS tree.
> > >
> > > Ahh! It's new code, sorry, forgot that.
> > >
> > > Please have a look at http://linuxbios.org/Development_Guidelines to
> > > learn most if not all you need to know to submit a patch.
> > >
> > > Also, when submitting patches, please make sure they apply cleanly to
> > > the very latest revision of the tree.
> > >
> > > The simplest way to do this is to always keep your own tree updated
> > > by running svn up now and then, or when you see a new revision
> > > announced on the list. After an update, there may be conflicts that
> > > need to be resolved manually by you if you have been working on code
> > > that was also changed in the new revision.
> > > > I haven't get the chance to take look at my NIC issue yet, I will
> > > > certainly check the interrupt when I get the chance.
> > >
> > > Feel free to ask the list if you run into trouble!
> > >
> > > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Tom Sylla]

woops, had that swapped. Rev A has 87427, B has 4307.

On 4/13/07, Tom Sylla  wrote:
> Explosion Rev A shipped with SCH430{4,7}. Explosion Rev B has a PC87427 on it.
>
> On 4/13/07, Lu, Yinghai  wrote:
> > The MB I got with SCH 4304, and schematic said PC87427...
> >
> > YH
> >
> > -----Original Message-----
> > From: Beneo [mailto:[e-mail]]
> > Sent: Friday, April 13, 2007 11:12 AM
> > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> >
> > it is SMSC SCH 4307, I ported some code for that serail, it works.
> >
> > Beneo
> >
> > ----- Original Message -----
> > From: "Lu, Yinghai" 
> > To: "Beneo" ; "Peter Stuge"
> > ;
> >
> > Sent: Friday, April 13, 2007 11:01 AM
> > Subject: RE: [LinuxBIOS] PIC mode or APIC mode
> >
> > Good.
> >
> > I can not get serial output from that MB, what is your superio on the
> > MB?
> >
> > YH
> >
> > -----Original Message-----
> > From: Beneo [mailto:[e-mail]]
> > Sent: Friday, April 13, 2007 10:59 AM
> > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> >
> > It is not Blast, it is newer version of Broadcom board, it is called
> > Explosion board.
> >
> > Beneo
> >
> > ----- Original Message -----
> > From: "Lu, Yinghai" 
> > To: "Beneo" ; "Peter Stuge"
> > ;
> >
> > Sent: Friday, April 13, 2007 10:54 AM
> > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > > BLAST?
> > >
> > > That code in the tree should work.
> > >
> > > YH
> > >
> > > -----Original Message-----
> > > From: [e-mail]
> > > [mailto:[e-mail]] On Behalf Of Beneo
> > > Sent: Friday, April 13, 2007 10:48 AM
> > > To: Peter Stuge; [e-mail]
> > > Subject: [LinuxBIOS] PIC mode or APIC mode
> > >
> > > I was working this Broadcom bcm 5785 based reference board with
> > > LinuxBIOS, I
> > > have a NIC interrupt related issue, the eth0 DHCP detection always
> >
> > fail,
> > > so
> > > I can not get an IP address for that NIC. I did what Peter Stuge
> > > suggested
> > > to dump the /proc/interrupts. It turns out the Linux is using PIC
> >
> > mode,
> > > not
> > > APIC mode. (Linux report multiple CPUs, so I assume Linux took MP
> >
> > Table
> > > somewhat correctly)
> > >
> > > I checked chipset setting, the APIC indeed is enabled. I don't see
> >
> > Linux
> > > is
> > > doing anything for PIC or APIC, I assume Linux kernel will correctly
> > > initialize PIC or APIC depends on MP Table reporting. right?
> > >
> > > If anybody has any idea on why Linux is running at PIC mode, Please
> >
> > let
> > > me
> > > know. I will be very appreciated.
> > >
> > > Thanks
> > >
> > > Beneo
> > >
> > > --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
> > > [root@localhost root]# cat /proc/interrupts
> > >           CPU0       CPU1
> > >  0:      13538       1521          XT-PIC  timer
> > >  1:          0          0          XT-PIC  keyboard
> > >  2:          0          0          XT-PIC  cascade
> > >  4:        302        153          XT-PIC  serial
> > >  8:          1          0          XT-PIC  rtc
> > > 14:       4729       1803          XT-PIC  ide0
> > > NMI:          0          0
> > > LOC:      14778      14897
> > > ERR:      21478
> > > MIS:          0
> > >
> > > --- When using factory BIOS, the cat /proc/interrupts looks like this,
> > > it is
> > > in APIC mode.
> > > [root@localhost root]# cat /proc/interrupts
> > >           CPU0       CPU1
> > >  0:        719       8393    IO-APIC-edge  timer
> > >  1:          0          4    IO-APIC-edge  keyboard
> > >  2:          0          0          XT-PIC  cascade
> > >  4:          5        505    IO-APIC-edge  serial
> > >  8:          0          1    IO-APIC-edge  rtc
> > > 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,
> >
> > usb-ohci
> > > 12:          1          6    IO-APIC-edge  PS/2 Mouse
> > > 14:       1433       5111    IO-APIC-edge  ide0
> > > 38:          7        112   IO-APIC-level  eth0
> > > NMI:          0          0
> > > LOC:       9020       8869
> > > ERR:          0
> > > MIS:          0
> > >
> > > ----- Original Message -----
> > > From: "Peter Stuge" 
> > > To: 
> > > Sent: Monday, April 09, 2007 10:23 AM
> > > Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > > > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > > > don't know how to contribute it to LinuxBIOS tree.
> > > >
> > > > Ahh! It's new code, sorry, forgot that.
> > > >
> > > > Please have a look at http://linuxbios.org/Development_Guidelines to
> > > > learn most if not all you need to know to submit a patch.
> > > >
> > > > Also, when submitting patches, please make sure they apply cleanly to
> > > > the very latest revision of the tree.
> > > >
> > > > The simplest way to do this is to always keep your own tree updated
> > > > by running svn up now and then, or when you see a new revision
> > > > announced on the list. After an update, there may be conflicts that
> > > > need to be resolved manually by you if you have been working on code
> > > > that was also changed in the new revision.
> > > > > I haven't get the chance to take look at my NIC issue yet, I will
> > > > > certainly check the interrupt when I get the chance.
> > > >
> > > > Feel free to ask the list if you run into trouble!
> > > >
> > > > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Beneo]

I've got probably is a rev B, it is dead now,
But I captured some dump before it went dead, I still try to understand why 
it runs in PIC mode... Does LinuxBIOS need to program PIC or APIC, I don't 
LinuxBIOS was doing that.. Would it cause problem if leave PIC/APIC 
unprogrammed?

Beneo

----- Original Message ----- 
From: "Tom Sylla" 
To: "Lu, Yinghai" 
Cc: ; "Peter Stuge" ; 
"Beneo" 
Sent: Friday, April 13, 2007 11:55 AM
Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> woops, had that swapped. Rev A has 87427, B has 4307.
>
> On 4/13/07, Tom Sylla  wrote:
> > Explosion Rev A shipped with SCH430{4,7}. Explosion Rev B has a PC87427 
> > on it.
> >
> > On 4/13/07, Lu, Yinghai  wrote:
> > > The MB I got with SCH 4304, and schematic said PC87427...
> > >
> > > YH
> > >
> > > -----Original Message-----
> > > From: Beneo [mailto:[e-mail]]
> > > Sent: Friday, April 13, 2007 11:12 AM
> > > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > >
> > > it is SMSC SCH 4307, I ported some code for that serail, it works.
> > >
> > > Beneo
> > >
> > > ----- Original Message -----
> > > From: "Lu, Yinghai" 
> > > To: "Beneo" ; "Peter Stuge"
> > > ;
> > >
> > > Sent: Friday, April 13, 2007 11:01 AM
> > > Subject: RE: [LinuxBIOS] PIC mode or APIC mode
> > >
> > > Good.
> > >
> > > I can not get serial output from that MB, what is your superio on the
> > > MB?
> > >
> > > YH
> > >
> > > -----Original Message-----
> > > From: Beneo [mailto:[e-mail]]
> > > Sent: Friday, April 13, 2007 10:59 AM
> > > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > >
> > > It is not Blast, it is newer version of Broadcom board, it is called
> > > Explosion board.
> > >
> > > Beneo
> > >
> > > ----- Original Message -----
> > > From: "Lu, Yinghai" 
> > > To: "Beneo" ; "Peter Stuge"
> > > ;
> > >
> > > Sent: Friday, April 13, 2007 10:54 AM
> > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > > > BLAST?
> > > >
> > > > That code in the tree should work.
> > > >
> > > > YH
> > > >
> > > > -----Original Message-----
> > > > From: [e-mail]
> > > > [mailto:[e-mail]] On Behalf Of Beneo
> > > > Sent: Friday, April 13, 2007 10:48 AM
> > > > To: Peter Stuge; [e-mail]
> > > > Subject: [LinuxBIOS] PIC mode or APIC mode
> > > >
> > > > I was working this Broadcom bcm 5785 based reference board with
> > > > LinuxBIOS, I
> > > > have a NIC interrupt related issue, the eth0 DHCP detection always
> > >
> > > fail,
> > > > so
> > > > I can not get an IP address for that NIC. I did what Peter Stuge
> > > > suggested
> > > > to dump the /proc/interrupts. It turns out the Linux is using PIC
> > >
> > > mode,
> > > > not
> > > > APIC mode. (Linux report multiple CPUs, so I assume Linux took MP
> > >
> > > Table
> > > > somewhat correctly)
> > > >
> > > > I checked chipset setting, the APIC indeed is enabled. I don't see
> > >
> > > Linux
> > > > is
> > > > doing anything for PIC or APIC, I assume Linux kernel will correctly
> > > > initialize PIC or APIC depends on MP Table reporting. right?
> > > >
> > > > If anybody has any idea on why Linux is running at PIC mode, Please
> > >
> > > let
> > > > me
> > > > know. I will be very appreciated.
> > > >
> > > > Thanks
> > > >
> > > > Beneo
> > > >
> > > > --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode --
> > > > [root@localhost root]# cat /proc/interrupts
> > > >           CPU0       CPU1
> > > >  0:      13538       1521          XT-PIC  timer
> > > >  1:          0          0          XT-PIC  keyboard
> > > >  2:          0          0          XT-PIC  cascade
> > > >  4:        302        153          XT-PIC  serial
> > > >  8:          1          0          XT-PIC  rtc
> > > > 14:       4729       1803          XT-PIC  ide0
> > > > NMI:          0          0
> > > > LOC:      14778      14897
> > > > ERR:      21478
> > > > MIS:          0
> > > >
> > > > --- When using factory BIOS, the cat /proc/interrupts looks like 
> > > > this,
> > > > it is
> > > > in APIC mode.
> > > > [root@localhost root]# cat /proc/interrupts
> > > >           CPU0       CPU1
> > > >  0:        719       8393    IO-APIC-edge  timer
> > > >  1:          0          4    IO-APIC-edge  keyboard
> > > >  2:          0          0          XT-PIC  cascade
> > > >  4:          5        505    IO-APIC-edge  serial
> > > >  8:          0          1    IO-APIC-edge  rtc
> > > > 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,
> > >
> > > usb-ohci
> > > > 12:          1          6    IO-APIC-edge  PS/2 Mouse
> > > > 14:       1433       5111    IO-APIC-edge  ide0
> > > > 38:          7        112   IO-APIC-level  eth0
> > > > NMI:          0          0
> > > > LOC:       9020       8869
> > > > ERR:          0
> > > > MIS:          0
> > > >
> > > > ----- Original Message -----
> > > > From: "Peter Stuge" 
> > > > To: 
> > > > Sent: Monday, April 09, 2007 10:23 AM
> > > > Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > > > > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > > > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > > > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > > > > don't know how to contribute it to LinuxBIOS tree.
> > > > >
> > > > > Ahh! It's new code, sorry, forgot that.
> > > > >
> > > > > Please have a look at http://linuxbios.org/Development_Guidelines to
> > > > > learn most if not all you need to know to submit a patch.
> > > > >
> > > > > Also, when submitting patches, please make sure they apply cleanly 
> > > > > to
> > > > > the very latest revision of the tree.
> > > > >
> > > > > The simplest way to do this is to always keep your own tree updated
> > > > > by running svn up now and then, or when you see a new revision
> > > > > announced on the list. After an update, there may be conflicts that
> > > > > need to be resolved manually by you if you have been working on code
> > > > > that was also changed in the new revision.
> > > > > > I haven't get the chance to take look at my NIC issue yet, I will
> > > > > > certainly check the interrupt when I get the chance.
> > > > >
> > > > > Feel free to ask the list if you run into trouble!
> > > > >
> > > > > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Lu, Yinghai]

Did you change anything for BLAST except for rev F support?

I enabled APIC mode for BLAST.

YH

-----Original Message-----
From: Beneo [mailto:[e-mail]] 
Sent: Friday, April 13, 2007 12:53 PM
To: Tom Sylla; Lu, Yinghai
Cc: [e-mail]; Peter Stuge
Subject: Re: [LinuxBIOS] PIC mode or APIC mode

I've got probably is a rev B, it is dead now,
But I captured some dump before it went dead, I still try to understand
why 
it runs in PIC mode... Does LinuxBIOS need to program PIC or APIC, I
don't 
LinuxBIOS was doing that.. Would it cause problem if leave PIC/APIC 
unprogrammed?

Beneo

----- Original Message ----- 
From: "Tom Sylla" 
To: "Lu, Yinghai" 
Cc: ; "Peter Stuge" ; 
"Beneo" 
Sent: Friday, April 13, 2007 11:55 AM
Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> woops, had that swapped. Rev A has 87427, B has 4307.
>
> On 4/13/07, Tom Sylla  wrote:
> > Explosion Rev A shipped with SCH430{4,7}. Explosion Rev B has a

PC87427
> > on it.
> >
> > On 4/13/07, Lu, Yinghai  wrote:
> > > The MB I got with SCH 4304, and schematic said PC87427...
> > >
> > > YH
> > >
> > > -----Original Message-----
> > > From: Beneo [mailto:[e-mail]]
> > > Sent: Friday, April 13, 2007 11:12 AM
> > > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > >
> > > it is SMSC SCH 4307, I ported some code for that serail, it works.
> > >
> > > Beneo
> > >
> > > ----- Original Message -----
> > > From: "Lu, Yinghai" 
> > > To: "Beneo" ; "Peter Stuge"
> > > ;
> > >
> > > Sent: Friday, April 13, 2007 11:01 AM
> > > Subject: RE: [LinuxBIOS] PIC mode or APIC mode
> > >
> > > Good.
> > >
> > > I can not get serial output from that MB, what is your superio on

the
> > > MB?
> > >
> > > YH
> > >
> > > -----Original Message-----
> > > From: Beneo [mailto:[e-mail]]
> > > Sent: Friday, April 13, 2007 10:59 AM
> > > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > >
> > > It is not Blast, it is newer version of Broadcom board, it is

called
> > > Explosion board.
> > >
> > > Beneo
> > >
> > > ----- Original Message -----
> > > From: "Lu, Yinghai" 
> > > To: "Beneo" ; "Peter Stuge"
> > > ;
> > >
> > > Sent: Friday, April 13, 2007 10:54 AM
> > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > > > BLAST?
> > > >
> > > > That code in the tree should work.
> > > >
> > > > YH
> > > >
> > > > -----Original Message-----
> > > > From: [e-mail]
> > > > [mailto:[e-mail]] On Behalf Of Beneo
> > > > Sent: Friday, April 13, 2007 10:48 AM
> > > > To: Peter Stuge; [e-mail]
> > > > Subject: [LinuxBIOS] PIC mode or APIC mode
> > > >
> > > > I was working this Broadcom bcm 5785 based reference board with
> > > > LinuxBIOS, I
> > > > have a NIC interrupt related issue, the eth0 DHCP detection

always
> > > fail,
> > > > so
> > > > I can not get an IP address for that NIC. I did what Peter Stuge
> > > > suggested
> > > > to dump the /proc/interrupts. It turns out the Linux is using PIC
> > >
> > > mode,
> > > > not
> > > > APIC mode. (Linux report multiple CPUs, so I assume Linux took MP
> > >
> > > Table
> > > > somewhat correctly)
> > > >
> > > > I checked chipset setting, the APIC indeed is enabled. I don't

see
> > > Linux
> > > > is
> > > > doing anything for PIC or APIC, I assume Linux kernel will

correctly
> > > > initialize PIC or APIC depends on MP Table reporting. right?
> > > >
> > > > If anybody has any idea on why Linux is running at PIC mode,

Please
> > > let
> > > > me
> > > > know. I will be very appreciated.
> > > >
> > > > Thanks
> > > >
> > > > Beneo
> > > >
> > > > --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode
> > > > [root@localhost root]# cat /proc/interrupts
> > > >           CPU0       CPU1
> > > >  0:      13538       1521          XT-PIC  timer
> > > >  1:          0          0          XT-PIC  keyboard
> > > >  2:          0          0          XT-PIC  cascade
> > > >  4:        302        153          XT-PIC  serial
> > > >  8:          1          0          XT-PIC  rtc
> > > > 14:       4729       1803          XT-PIC  ide0
> > > > NMI:          0          0
> > > > LOC:      14778      14897
> > > > ERR:      21478
> > > > MIS:          0
> > > >
> > > > --- When using factory BIOS, the cat /proc/interrupts looks like 
> > > > this,
> > > > it is
> > > > in APIC mode.
> > > > [root@localhost root]# cat /proc/interrupts
> > > >           CPU0       CPU1
> > > >  0:        719       8393    IO-APIC-edge  timer
> > > >  1:          0          4    IO-APIC-edge  keyboard
> > > >  2:          0          0          XT-PIC  cascade
> > > >  4:          5        505    IO-APIC-edge  serial
> > > >  8:          0          1    IO-APIC-edge  rtc
> > > > 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,
> > >
> > > usb-ohci
> > > > 12:          1          6    IO-APIC-edge  PS/2 Mouse
> > > > 14:       1433       5111    IO-APIC-edge  ide0
> > > > 38:          7        112   IO-APIC-level  eth0
> > > > NMI:          0          0
> > > > LOC:       9020       8869
> > > > ERR:          0
> > > > MIS:          0
> > > >
> > > > ----- Original Message -----
> > > > From: "Peter Stuge" 
> > > > To: 
> > > > Sent: Monday, April 09, 2007 10:23 AM
> > > > Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > > > > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > > > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > > > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > > > > don't know how to contribute it to LinuxBIOS tree.
> > > > >
> > > > > Ahh! It's new code, sorry, forgot that.
> > > > >
> > > > > Please have a look at

http://linuxbios.org/Development_Guidelines to
> > > > > learn most if not all you need to know to submit a patch.
> > > > >
> > > > > Also, when submitting patches, please make sure they apply

cleanly
> > > > > to
> > > > > the very latest revision of the tree.
> > > > >
> > > > > The simplest way to do this is to always keep your own tree

updated
> > > > > by running svn up now and then, or when you see a new revision
> > > > > announced on the list. After an update, there may be conflicts

that
> > > > > need to be resolved manually by you if you have been working on

code
> > > > > that was also changed in the new revision.
> > > > > > I haven't get the chance to take look at my NIC issue yet, I

will
> > > > > > certainly check the interrupt when I get the chance.
> > > > >
> > > > > Feel free to ask the list if you run into trouble!
> > > > >
> > > > > //Peter

[LinuxBIOS] PIC mode or APIC mode

[Beneo]

When you say you enabled APIC mode, it is in the Broadcom southbridge, not 
directly program the APIC or PIC themselves, right?

if so, I didn't change anything in term of Broadcom southbridge setting, I 
changed Config.lb because the HT Topologies were different. I assume that 
should have nothing to do with the issue. I can access APIC using base 
address 0xFEC00000 and etc after LinuxBIOS finishes. So I assume APIC itself 
is enabled.

Except this, do I have to something else?

Thanks

Beneo

----- Original Message ----- 
From: "Lu, Yinghai" 
To: "Beneo" ; "Tom Sylla" 
Cc: "Peter Stuge" ; 
Sent: Friday, April 13, 2007 1:52 PM
Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> Did you change anything for BLAST except for rev F support?
>
> I enabled APIC mode for BLAST.
>
> YH
>
> -----Original Message-----
> From: Beneo [mailto:[e-mail]]
> Sent: Friday, April 13, 2007 12:53 PM
> To: Tom Sylla; Lu, Yinghai
> Cc: [e-mail]; Peter Stuge
> Subject: Re: [LinuxBIOS] PIC mode or APIC mode
>
> I've got probably is a rev B, it is dead now,
> But I captured some dump before it went dead, I still try to understand
> why
> it runs in PIC mode... Does LinuxBIOS need to program PIC or APIC, I
> don't
> LinuxBIOS was doing that.. Would it cause problem if leave PIC/APIC
> unprogrammed?
>
> Beneo
>
> ----- Original Message ----- 
> From: "Tom Sylla" 
> To: "Lu, Yinghai" 
> Cc: ; "Peter Stuge" ;
> "Beneo" 
> Sent: Friday, April 13, 2007 11:55 AM
> Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > woops, had that swapped. Rev A has 87427, B has 4307.
> >
> > On 4/13/07, Tom Sylla  wrote:
> > > Explosion Rev A shipped with SCH430{4,7}. Explosion Rev B has a
>
> PC87427
> > > on it.
> > >
> > > On 4/13/07, Lu, Yinghai  wrote:
> > > > The MB I got with SCH 4304, and schematic said PC87427...
> > > >
> > > > YH
> > > >
> > > > -----Original Message-----
> > > > From: Beneo [mailto:[e-mail]]
> > > > Sent: Friday, April 13, 2007 11:12 AM
> > > > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > > >
> > > > it is SMSC SCH 4307, I ported some code for that serail, it works.
> > > >
> > > > Beneo
> > > >
> > > > ----- Original Message -----
> > > > From: "Lu, Yinghai" 
> > > > To: "Beneo" ; "Peter Stuge"
> > > > ;
> > > >
> > > > Sent: Friday, April 13, 2007 11:01 AM
> > > > Subject: RE: [LinuxBIOS] PIC mode or APIC mode
> > > >
> > > > Good.
> > > >
> > > > I can not get serial output from that MB, what is your superio on
>
> the
> > > > MB?
> > > >
> > > > YH
> > > >
> > > > -----Original Message-----
> > > > From: Beneo [mailto:[e-mail]]
> > > > Sent: Friday, April 13, 2007 10:59 AM
> > > > To: Lu, Yinghai; Peter Stuge; [e-mail]
> > > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > > >
> > > > It is not Blast, it is newer version of Broadcom board, it is
>
> called
> > > > Explosion board.
> > > >
> > > > Beneo
> > > >
> > > > ----- Original Message -----
> > > > From: "Lu, Yinghai" 
> > > > To: "Beneo" ; "Peter Stuge"
> > > > ;
> > > >
> > > > Sent: Friday, April 13, 2007 10:54 AM
> > > > Subject: Re: [LinuxBIOS] PIC mode or APIC mode
> > > > > BLAST?
> > > > >
> > > > > That code in the tree should work.
> > > > >
> > > > > YH
> > > > >
> > > > > -----Original Message-----
> > > > > From: [e-mail]
> > > > > [mailto:[e-mail]] On Behalf Of Beneo
> > > > > Sent: Friday, April 13, 2007 10:48 AM
> > > > > To: Peter Stuge; [e-mail]
> > > > > Subject: [LinuxBIOS] PIC mode or APIC mode
> > > > >
> > > > > I was working this Broadcom bcm 5785 based reference board with
> > > > > LinuxBIOS, I
> > > > > have a NIC interrupt related issue, the eth0 DHCP detection
>
> always
> > > > fail,
> > > > > so
> > > > > I can not get an IP address for that NIC. I did what Peter Stuge
> > > > > suggested
> > > > > to dump the /proc/interrupts. It turns out the Linux is using PIC
> > > >
> > > > mode,
> > > > > not
> > > > > APIC mode. (Linux report multiple CPUs, so I assume Linux took MP
> > > >
> > > > Table
> > > > > somewhat correctly)
> > > > >
> > > > > I checked chipset setting, the APIC indeed is enabled. I don't
>
> see
> > > > Linux
> > > > > is
> > > > > doing anything for PIC or APIC, I assume Linux kernel will
>
> correctly
> > > > > initialize PIC or APIC depends on MP Table reporting. right?
> > > > >
> > > > > If anybody has any idea on why Linux is running at PIC mode,
>
> Please
> > > > let
> > > > > me
> > > > > know. I will be very appreciated.
> > > > >
> > > > > Thanks
> > > > >
> > > > > Beneo
> > > > >
> > > > > --- LinuxBIOS /proc/interrupts look like this, it is in PIC mode
> > > > > [root@localhost root]# cat /proc/interrupts
> > > > >           CPU0       CPU1
> > > > >  0:      13538       1521          XT-PIC  timer
> > > > >  1:          0          0          XT-PIC  keyboard
> > > > >  2:          0          0          XT-PIC  cascade
> > > > >  4:        302        153          XT-PIC  serial
> > > > >  8:          1          0          XT-PIC  rtc
> > > > > 14:       4729       1803          XT-PIC  ide0
> > > > > NMI:          0          0
> > > > > LOC:      14778      14897
> > > > > ERR:      21478
> > > > > MIS:          0
> > > > >
> > > > > --- When using factory BIOS, the cat /proc/interrupts looks like
> > > > > this,
> > > > > it is
> > > > > in APIC mode.
> > > > > [root@localhost root]# cat /proc/interrupts
> > > > >           CPU0       CPU1
> > > > >  0:        719       8393    IO-APIC-edge  timer
> > > > >  1:          0          4    IO-APIC-edge  keyboard
> > > > >  2:          0          0          XT-PIC  cascade
> > > > >  4:          5        505    IO-APIC-edge  serial
> > > > >  8:          0          1    IO-APIC-edge  rtc
> > > > > 10:          0          2   IO-APIC-level  ehci-hcd, usb-ohci,
> > > >
> > > > usb-ohci
> > > > > 12:          1          6    IO-APIC-edge  PS/2 Mouse
> > > > > 14:       1433       5111    IO-APIC-edge  ide0
> > > > > 38:          7        112   IO-APIC-level  eth0
> > > > > NMI:          0          0
> > > > > LOC:       9020       8869
> > > > > ERR:          0
> > > > > MIS:          0
> > > > >
> > > > > ----- Original Message -----
> > > > > From: "Peter Stuge" 
> > > > > To: 
> > > > > Sent: Monday, April 09, 2007 10:23 AM
> > > > > Subject: Re: [LinuxBIOS] IDE become readonly, why?
> > > > > > On Mon, Apr 09, 2007 at 10:02:15AM -0700, Beneo wrote:
> > > > > > > For the RTC patch, it is porting for a SMSC SIO, I didn't see
> > > > > > > LinuxBIOS has this SIO porting. The part number is SCH--4703, I
> > > > > > > don't know how to contribute it to LinuxBIOS tree.
> > > > > >
> > > > > > Ahh! It's new code, sorry, forgot that.
> > > > > >
> > > > > > Please have a look at
>
> http://linuxbios.org/Development_Guidelines to
> > > > > > learn most if not all you need to know to submit a patch.
> > > > > >
> > > > > > Also, when submitting patches, please make sure they apply
>
> cleanly
> > > > > > to
> > > > > > the very latest revision of the tree.
> > > > > >
> > > > > > The simplest way to do this is to always keep your own tree
>
> updated
> > > > > > by running svn up now and then, or when you see a new revision
> > > > > > announced on the list. After an update, there may be conflicts
>
> that
> > > > > > need to be resolved manually by you if you have been working on
>
> code
> > > > > > that was also changed in the new revision.
> > > > > >
> > > > > > > I haven't get the chance to take look at my NIC issue yet, I
>
> will
> > > > > > > certainly check the interrupt when I get the chance.
> > > > > >
> > > > > > Feel free to ask the list if you run into trouble!
> > > > > >
> > > > > > //Peter

[Peter Stuge]

Hi Beneo,

On Thu, Apr 05, 2007 at 02:41:34PM -0700, Beneo wrote:
> I used a parameter at filo like this:
>
> hda1:/vmlinuz-2.4.20-8smp initrd=hda1:/initrd-2.4.20-8smp.img ro
> root=/dev/hda3 debug console=tty0 console=ttyS0,115200

Are these the same parameters as you use with the boot loader when
booting with the factory BIOS? If not, why?
> Would you tell me how to modify the boot parameter for boot from
> shell?

You have to recompile FILO and hence LinuxBIOS if FILO is your
payload.

There's also a setting in FILO to pause and wait for user input
during boot. This is good to not have to recompile/reflash LB over
and over.

//Peter

[Peter Stuge]

Hi,

On Thu, Apr 05, 2007 at 11:27:11AM -0700, Beneo wrote:
> I was using LinuxBIOS to boot Linux from IDE from a Broadcom
> HT,1000 south bridge based  platform , I encounter an issue which
> is hard to understand. I hope somebody can give me some insight.
>
> During boot, Linux stoped at the point that it displayed a message:
> "Configuring kernel parameters:  [  OK  ]"

This message comes from your installed Linux distribution.

By this time in the boot process, LinuxBIOS is out of the picture.
> then I pressed "CTRL-C",  Linux continue to boot, but display a
> message:
> "Starting automount:Cannot create temp file /tmp/autofs.meQYNM,
> could not make temp file"
>
> then Linux boot Skipped rest of initialization, like probe the
> network and etc.

Linux systems are booted with the root filesystem mounted read-only,
then one of the (usually many) startup scripts will re-mount the root
filesystem read-write. Usually after having performed a filesystem
check to make sure the fs is not broken.

It looks like you interrupted the boot process before the startup
scripts had re-mounted the filesystem read-write.
> Phoenix BIOS came with that board doesn't have this issue, so I
> know it should be LinuxBIOS related. I just don't know why.

Again, the BIOS is long gone when any messages from the start scripts
are shown.

I imagine that it's important to hit Ctrl-C during boot at precisely
the right (actually wrong) time in order to trigger this symptom, and
that if you make some more attempts to reproduce it would be equally
difficult to do with LinuxBIOS and the factory BIOS.

Or have you already been able to reproduce the behaviour reliably?

//Peter

[Beneo]

Peter,

I hit CRTL-C becomes Linux Boot stopped after display the "Configuring 
kernel parameters:  [  OK  ]" message. I have to CRTL-C to get the boot to 
continue.

I understand LinuxBIOS will not be called by Linux, (LinuxBIOS is long 
gone.), but rather it is some setup LinuxBIOS did caused this issue as 
Pheonix BIOS didn't have this issue when booting the same Linux from the 
same IDE HD.

So my question is what possible things LinuxBIOS did to cause Linux boot to 
stop at this location?

File system becomes readonly probably is caused Linux boot skipped some 
procedures, like remount. but it is the consequence of the setup issue 
LinuxBIOS.

Does anybody know what possible areas to check in the LinuxBIOS?

Thanks

Beneo

----- Original Message ----- 
From: "Peter Stuge" 
To: "LinuxBIOS mailinglist" 
Sent: Thursday, April 05, 2007 1:28 PM
Subject: Re: [LinuxBIOS] IDE become readonly, why?
> Hi,
>
> On Thu, Apr 05, 2007 at 11:27:11AM -0700, Beneo wrote:
> > I was using LinuxBIOS to boot Linux from IDE from a Broadcom
> > HT,1000 south bridge based  platform , I encounter an issue which
> > is hard to understand. I hope somebody can give me some insight.
> >
> > During boot, Linux stoped at the point that it displayed a message:
> > "Configuring kernel parameters:  [  OK  ]"
>
> This message comes from your installed Linux distribution.
>
> By this time in the boot process, LinuxBIOS is out of the picture.
> > then I pressed "CTRL-C",  Linux continue to boot, but display a
> > message:
> > "Starting automount:Cannot create temp file /tmp/autofs.meQYNM,
> > could not make temp file"
> >
> > then Linux boot Skipped rest of initialization, like probe the
> > network and etc.
>
> Linux systems are booted with the root filesystem mounted read-only,
> then one of the (usually many) startup scripts will re-mount the root
> filesystem read-write. Usually after having performed a filesystem
> check to make sure the fs is not broken.
>
> It looks like you interrupted the boot process before the startup
> scripts had re-mounted the filesystem read-write.
> > Phoenix BIOS came with that board doesn't have this issue, so I
> > know it should be LinuxBIOS related. I just don't know why.
>
> Again, the BIOS is long gone when any messages from the start scripts
> are shown.
>
> I imagine that it's important to hit Ctrl-C during boot at precisely
> the right (actually wrong) time in order to trigger this symptom, and
> that if you make some more attempts to reproduce it would be equally
> difficult to do with LinuxBIOS and the factory BIOS.
>
> Or have you already been able to reproduce the behaviour reliably?
>
> //Peter

[Carl-Daniel Hailfinger]

On 05.04.2007 23:37, Beneo wrote:
> So my question is what possible things LinuxBIOS did to cause Linux boot to 
> stop at this location?

Different boot parameters.
> File system becomes readonly probably is caused Linux boot skipped some 
> procedures, like remount. but it is the consequence of the setup issue 
> LinuxBIOS.

Compare /proc/cmdline of the running system for LinuxBIOS and vendor BIOS.

Regards,
Carl-Daniel

[Stefan Reinauer]

* Beneo  [070405 20:27]:
> Hi all,
>
> I was using LinuxBIOS to boot Linux from IDE from a Broadcom HT,1000 south
> bridge based  platform , I encounter an issue which is hard to understand. I
> hope somebody can give me some insight.

What bootloader are you using?

What kernel command line are you using in linuxbios?
What kernel command line are you using in phoenixbios?

Do you have "ro" in the linuxbios kernel command line?