headers
Adam Lackorzynski | 1 May 14:20
Picon
Favicon

Re: Unable to mount /dev/root

Hi,

On Wed Apr 30, 2008 at 20:43:17 +0400, Igor V. Ruzanov wrote:
> >> # CONFIG_HOTPLUG is not set
> >
> > You need CONFIG_HOTPLUG for udev to work correctly, without this option
> > you may not get the device-nodes even if you compiled the right drivers
> > in your kernel.
> >
> Thanks for your hint! I set CONFIG_HOTPLUG to 'y'. Also i added some 
> derivative options:
> 
> CONFIG_HOTPLUG_PCI=y
> CONFIG_HOTPLUG_PCI_FAKE=m
> CONFIG_HOTPLUG_PCI_CPCI=y
> # CONFIG_HOTPLUG_PCI_CPCI_ZT5550 is not set
> # CONFIG_HOTPLUG_PCI_CPCI_GENERIC is not set
> CONFIG_HOTPLUG_PCI_SHPC=m
> 
> With them some error messages were disapperred from boot log but the 
> system persists to be dropped into panic after rest of messages:
> 
> mount: could not find filesystem /dev/root
> [...]
> switchroot: mount failed: No such file or directory
> Kernel panic - not syncing: No such file or directory
> 
> 
> For little debug i unpacked my initrd image and modified 'init' script 
> with several changes:
(Continue reading)

Permalink | Reply | 14 comments |
headers
Igor V. Ruzanov | 3 May 16:45
Picon

Re: Unable to mount /dev/root

On Thu, 1 May 2008, Adam Lackorzynski wrote:

> Hi,
>
> On Wed Apr 30, 2008 at 20:43:17 +0400, Igor V. Ruzanov wrote:
>>>> # CONFIG_HOTPLUG is not set
>>>
>>> You need CONFIG_HOTPLUG for udev to work correctly, without this option
>>> you may not get the device-nodes even if you compiled the right drivers
>>> in your kernel.
>>>
>> Thanks for your hint! I set CONFIG_HOTPLUG to 'y'. Also i added some
>> derivative options:
>>
>> CONFIG_HOTPLUG_PCI=y
>> CONFIG_HOTPLUG_PCI_FAKE=m
>> CONFIG_HOTPLUG_PCI_CPCI=y
>> # CONFIG_HOTPLUG_PCI_CPCI_ZT5550 is not set
>> # CONFIG_HOTPLUG_PCI_CPCI_GENERIC is not set
>> CONFIG_HOTPLUG_PCI_SHPC=m
>>
>> With them some error messages were disapperred from boot log but the
>> system persists to be dropped into panic after rest of messages:
>>
>> mount: could not find filesystem /dev/root
>> [...]
>> switchroot: mount failed: No such file or directory
>> Kernel panic - not syncing: No such file or directory
>>
>>
(Continue reading)

Permalink | Reply | 13 comments |
headers
盛淑蕙 | 5 May 10:16
Picon

unhandled page fault

Hi,
I tried to compile l4linux (kernel 2.6.24) under ubuntu 8.04 (The
version of gcc compiler is 4.2.3) and it failed to boot under l4env
with Fiasco microkernel. Here is my step to compile l4linux

1. make O=/usr/l4/build/linux x86-native_defconfig
2. enable Support for frame buffer devices and Framebuffer Console support
3. make O=/usr/l4/build/linux

and then edited the menu.lst
title L4Linux-2.6
root (hd0,0)
kernel /boot/grub/common/bootstrap -modaddr 0x02000000
module /boot/grub/common/fiasco -nokdb -serial_esc
module /boot/grub/common/sigma0
module /boot/grub/common/roottask task modname "bmodfs" attached 4 modules
module /boot/grub/common/names
module /boot/grub/common/log
module /boot/grub/common/dm_phys --isa=0x00800000
module /boot/grub/common/simple_ts -t 300
module /boot/grub/common/rtc
module /boot/grub/common/l4io
module /boot/grub/common/bmodfs
   module /boot/grub/common/linux.cfg
   module /boot/grub/common/libloader.s.so
   module /boot/grub/common/libld-l4.s.so
   module /boot/grub/common/vmlinuz
module /boot/grub/common/loader --fprov=BMODFS linux.cfg

and linux.cfg
(Continue reading)

Permalink | Reply | 1 comment |
headers
Adam Lackorzynski | 5 May 21:05
Picon
Favicon

Re: Unable to mount /dev/root


On Sat May 03, 2008 at 18:45:47 +0400, Igor V. Ruzanov wrote:
> No, i don't see such logs. Exactly output of rest of the boot looks like:
> 
> [...some messages about loading of usb modules, hotplug configuring...]
> Loading scsi_mod.ko module
> SCSI subsystem initialized
> Loading sd_mod.ko module
> Driver 'sd' needs updating - please use bus_type methods
> Loading libata.ko module
> Loading ahci.ko module
> Waiting for driver initialization
> Creating root device
> Mounting root filesystem
> mount: could not find filesystem '/dev/root'
> [...and still several messages about impossibility of mounting /sys and 
> /proc filesystems before terminating of init and panic state...]
> 
> I'm trying to research problem towards of PCI probing through mmconfig.. 
> In my kernel config is CONFIG_PCI_GOMMCONFIG=y used. GODIRECT resulted 
> __unknown_pf(). It might be due to some mmconfig failures during l4linux 
> boot and incorrect work of ahci driver as result.
> 
> Unfortunately i cannot provide you full log of l4linux boot stages because 
> i haven't any methods (physical COM-port with attached machine as a 
> console with X11 and some console application with text buffering). Does 
> theoretically (at least) mmconfig can make problems within l4env?

It's not unlikely but I would need to look myself. Can you tell me what
the problem with GODIRECT is? That's the mode that is usually supposed
(Continue reading)

Permalink | Reply | 12 comments |
headers
Adam Lackorzynski | 5 May 21:28
Picon
Favicon

Re: unhandled page fault


On Mon May 05, 2008 at 16:16:42 +0800, ²±²Q¿· wrote:
> I tried to compile l4linux (kernel 2.6.24) under ubuntu 8.04 (The
> version of gcc compiler is 4.2.3) and it failed to boot under l4env
> with Fiasco microkernel. Here is my step to compile l4linux
> 
> 1. make O=/usr/l4/build/linux x86-native_defconfig
> 2. enable Support for frame buffer devices and Framebuffer Console support
> 3. make O=/usr/l4/build/linux
> 
> and then edited the menu.lst
> title L4Linux-2.6
> root (hd0,0)
> kernel /boot/grub/common/bootstrap -modaddr 0x02000000
> module /boot/grub/common/fiasco -nokdb -serial_esc
> module /boot/grub/common/sigma0
> module /boot/grub/common/roottask task modname "bmodfs" attached 4 modules
> module /boot/grub/common/names
> module /boot/grub/common/log
> module /boot/grub/common/dm_phys --isa=0x00800000
> module /boot/grub/common/simple_ts -t 300
> module /boot/grub/common/rtc
> module /boot/grub/common/l4io
> module /boot/grub/common/bmodfs
>    module /boot/grub/common/linux.cfg
>    module /boot/grub/common/libloader.s.so
>    module /boot/grub/common/libld-l4.s.so
>    module /boot/grub/common/vmlinuz
> module /boot/grub/common/loader --fprov=BMODFS linux.cfg
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Igor V. Ruzanov | 6 May 15:08
Picon

Re: Unable to mount /dev/root

Hi, Adam!

On Mon, 5 May 2008, Adam Lackorzynski wrote:

>
> It's not unlikely but I would need to look myself. Can you tell me what
> the problem with GODIRECT is? That's the mode that is usually supposed
> to work.

1)First of all i would like to show you my boot log starting from Fiasco 
loading stage (i all the same attached another machine with xterm via 
serial cable made by my hands):

Welcome to Fiasco(ia32)!
DD-L4(v2)/ia32 microkernel (C) 1998-2008 TU Dresden
Rev: r345 compiled with gcc 4.1.1 for Intel Pentium 4    [] 
Performance-critical config option(s) detected:
   CONFIG_NDEBUG is off
   CONFIG_NO_FRAME_PTR is off

Enabling special fully nested mode for PIC
Found Local APIC version 0x14 id 0x00
APIC ESR value before/after enabling: 00000000/00000000
Using the Local APIC timer on vector 3d (Periodic Mode) for scheduling
SERIAL ESC: allocated IRQ 4 for serial uart
SERIAL ESC: allocated IRQ 4 for serial uart
Not using serial hack in slow timer handler.
Absolute KIP Syscalls using: Sysenter
CPU: GenuineIntel (6:F:D:0) Model: Unknown CPU at 1994 MHz
(Continue reading)

Permalink | Reply | 11 comments |
headers
Adam Lackorzynski | 6 May 22:38
Picon
Favicon

Re: Unable to mount /dev/root


On Tue May 06, 2008 at 17:08:26 +0400, Igor V. Ruzanov wrote:
> On Mon, 5 May 2008, Adam Lackorzynski wrote:
> 
> >
> > It's not unlikely but I would need to look myself. Can you tell me what
> > the problem with GODIRECT is? That's the mode that is usually supposed
> > to work.
> 
> 1)First of all i would like to show you my boot log starting from Fiasco 
> loading stage (i all the same attached another machine with xterm via 
> serial cable made by my hands):

> l4lx    | L4RM: [PF] read at 0xf0cfb000, ip 00697406, src F.02
> l4lx    | [F.0] l4rm/lib/src/pagefault.c:81:__unknown_pf():
> l4lx    |  unhandled page fault
> 
>      ---------------------------------------------------------IP: a00b53a6          --PANIC, 'g' for exit
> [l4lx.rm] (f.00) jdb:
> 
> It happens every time when CONFIG_PCI_GODIRECT opthion is set. I tried 
> (quiet useless but for experiment clearence) to point pci=conf2 boot 
> option (as in example above) but it didn't led to any good changes. As i 
> can analyze the boot log there are no any error messages showing the need 
> to fix someting in L4 configs.

Ah, now I got it I think. You need to allow L4Linux to access those
ports, i.e. add 'ioport [ 0x0cf8, 0x0cff ]' to the loader config script
to allow access to the PCI ports. That should fix it. For other devices
this works equally.
(Continue reading)

Permalink | Reply | 10 comments |
headers
Igor V. Ruzanov | 7 May 11:23
Picon

Re: Unable to mount /dev/root

>
> Ah, now I got it I think. You need to allow L4Linux to access those
> ports, i.e. add 'ioport [ 0x0cf8, 0x0cff ]' to the loader config script
> to allow access to the PCI ports. That should fix it. For other devices
> this works equally.
>

I've added command ioport to l4linux26.cfg:
sleep 1
task "vmlinuz-2.6.24-l4" "earlyprintk=yes mem=128M root=LABEL=/ l4env_rd=initrd-2.6.24-l4.img"
ioport [ 0x0cf8, 0x0cff ]
all_sects_writable
allow_vga
allow_cli

but it still comes to old behaviour, with the same failure:
l4lx    | L4RM: [PF] read at 0xf0cfb000, ip 0069730c, src F.02
l4lx    | [F.0] l4rm/lib/src/pagefault.c:81:__unknown_pf():
l4lx    |  unhandled page fault

I've built L4Env and L4Linux in Debian system (runned for a long 
time) under VMware. In this case everything was going allmost fine. 
Comparing the logs:

Physical machine:
=================

l4lx    | Using tamed mode.
ROOT: Task #0d is not allowed to execute cli/sti
loader  | vmlinuz-2.6.24-l4,#f: WARNING: Can't map I/O space, ROOT denies page
(Continue reading)

Permalink | Reply | 0 comments |
headers
Igor V. Ruzanov | 7 May 14:36
Picon

Re: Unable to mount /dev/root

Adam,

> i.e. add 'ioport [ 0x0cf8, 0x0cff ]' to the loader config script
> to allow access to the PCI ports. That should fix it. For other devices
> this works equally.
>

I've allmost won at the problem after 'learning' of 'weird' strings
loader  | vmlinuz-2.6.24-l4,#f: WARNING: Can't map I/O space, ROOT denies page
loader  : (result=00004000)
loader  | vmlinuz-2.6.24-l4,#f: Not allowed to perform any I/O

for some time and remember about Fiasco configuration. There is the 
interesting option:
[ ] Enable I/O port protection

that i activated it before. And when i unset this option (as it looks like 
above) and rebuild Fiasco, sd_mod had determined partinions on sda 
drive during L4Linux boot. Now everything is going fine untill boot 
process gets stuck when the system perfoms some I/O on ata-channels. But i 
think this problem lays in plane of more simple tracings of right 
solutions.. at least i hope;)

Thank you for your HELP!!!
Permalink | Reply | 8 comments |
headers
Adam Lackorzynski | 7 May 14:45
Picon
Favicon

Re: Unable to mount /dev/root


On Wed May 07, 2008 at 16:36:16 +0400, Igor V. Ruzanov wrote:
> Adam,
> 
> > i.e. add 'ioport [ 0x0cf8, 0x0cff ]' to the loader config script
> > to allow access to the PCI ports. That should fix it. For other devices
> > this works equally.
> >
> 
> I've allmost won at the problem after 'learning' of 'weird' strings
> loader  | vmlinuz-2.6.24-l4,#f: WARNING: Can't map I/O space, ROOT denies page
> loader  : (result=00004000)
> loader  | vmlinuz-2.6.24-l4,#f: Not allowed to perform any I/O
> 
> for some time and remember about Fiasco configuration. There is the 
> interesting option:
> [ ] Enable I/O port protection
> 
> that i activated it before. And when i unset this option (as it looks like 
> above) and rebuild Fiasco, sd_mod had determined partinions on sda 
> drive during L4Linux boot. Now everything is going fine untill boot 

Heh, I/O port protection is a feature, not the other way around. :)
What I forgot to mention previously is that you also need to add
'task modname "loader" allow_cli' to the roottask command line so that
roottask is giving out those rights to the loader (sorry).

Adam
--

-- 
Adam                 adam <at> os.inf.tu-dresden.de
(Continue reading)

Permalink | Reply | 7 comments |

Gmane