new topic ??

--- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@...> wrote:
>
> new topic ??
>

A finally conviced this dumb-assed group software to create a new
topic when I click "Start topic".

now where were we ??



--- In ts-7000@yahoogroups.com, "bsmuck2000" <bsmucker@...> wrote:
>
> (Hopefully this is a new thread this time)

Well, it isn't yet ...

>
> Hello,
>
> We are working at transitioning our ts7260-based system to using
the 2.6
> kernel (right now the one supplied by Technologics. One problem that
> I'm running into is that when I boot to 2.6, the yaffs2 file system
on

Was the ts-7260 delivered with YAFFS or YAFFS2? IIRC, YAFFS2 was used
only for the larger flash sizes (2k page sizes).

> the flash is not visible. I can mount it fine, but nothing is there.

Anything in the kernel log after the mount or attempted read?

> If I rewrite the filesystem while in 2.6, it works fine, but when I
boot
> back into 2.4, it again shows a blank filesystem.
>
> Is anyone aware of this incompatibility and how to fix it?

Check the YAFFS2 build options on the 2.4 config, and make sure your
2.6 version is configured the same or at least backwards compatible.

Especially look at YAFFS_AUTO_YAFFS2, YAFFS_YAFFS2,
YAFFS_DISABLE_WIDE_TNODES, and their help text in Kconfig. If you are
mixing yaffs and yaffs2, these config options can help make this
transparent.

You can also turn on some debugging options in yaffs_fs.c (need to
recompile).

>
> For the transition time, it is essential that we be able to access
the
> yaffs filesystem from both 2.4 and 2.6.
>
> Thanks,
>
> Brian
>

Regards, .......... Charlie




	 Thu May 29, 2008 8:52 pm

  Show Message Info
"charliem_1216" <charliem_1216@...>
charliem_1216
  Offline
  Send Email

  Forward  < Prev Message  |  Next Message >

--- In ts-7000@yahoogroups.com, "Razvan Ionut
Stoian" <razvan_ionut_stoian@...> wrote:
>
> --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> >
> >
> >
> >
> > On Thu, 29 May 2008, Razvan Ionut Stoian wrote:
> >
> > > --- In ts-7000@yahoogroups.com, Triffid Hunter
<triffid_hunter@>
> wrote:
> > >>
> > >> On Wed, 28 May 2008, Razvan Ionut Stoian wrote:
> > >>
> > >>> For some reason the system freezes after vmlinux starts
> > >>> to decompress. This is what I got:
> > >>>
> > >>> Uncompressing
> > >>>
Linux.............................................................
> > >>>
> > >
>
  ........................................................................
> > >>>
> > >>> What am I doing wrong?
> > >>
> > >> Mine do that if my toolchain is broken
> > >>
> > >
> > > This is odd. I used the toolchain provided by TS.
> > > Any good alternatives?
> >
> > You can check if you toolchain is ok, by compiling the supplied
2.4
> kernel
> > source, with the default configuration. If the 2.4 kernel fails
to boot
> > then it may be your toolchain. If 2.4 works, then you have a 2.6
issue.
> >
> Hi.
>
> I managed to boot the 2.6 kernel. Apparently, the initial image was
> just too big. Now it boots, but my SBC freezes due to a kernel
panic:
> ----------------------------------------------------
> Kernel panic - not syncing: Attempted to kill init!
> ----------------------------------------------------
>
> Any ideas?
>
> thank you,
> R
>

That's probably the point at which it switches from initrd to the
disk fs. Check you rootfs is compatible with your kernel and
accessible.

Can yo boot to it over nfs using your new kernel?

--- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "Razvan Ionut
> Stoian" <razvan_ionut_stoian@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > >
> > >
> > >
> > >
> > > On Thu, 29 May 2008, Razvan Ionut Stoian wrote:
> > >
> > > > --- In ts-7000@yahoogroups.com, Triffid Hunter
> <triffid_hunter@>
> > wrote:
> > > >>
> > > >> On Wed, 28 May 2008, Razvan Ionut Stoian wrote:
> > > >>
> > > >>> For some reason the system freezes after vmlinux starts
> > > >>> to decompress. This is what I got:
> > > >>>
> > > >>> Uncompressing
> > > >>>
> Linux.............................................................
> > > >>>
> > > >
> >
>
........................................................................
> > > >>>
> > > >>> What am I doing wrong?
> > > >>
> > > >> Mine do that if my toolchain is broken
> > > >>
> > > >
> > > > This is odd. I used the toolchain provided by TS.
> > > > Any good alternatives?
> > >
> > > You can check if you toolchain is ok, by compiling the supplied
> 2.4
> > kernel
> > > source, with the default configuration. If the 2.4 kernel fails
> to boot
> > > then it may be your toolchain. If 2.4 works, then you have a 2.6
> issue.
> > >
> > Hi.
> >
> > I managed to boot the 2.6 kernel. Apparently, the initial image was
> > just too big. Now it boots, but my SBC freezes due to a kernel
> panic:
> > ----------------------------------------------------
> > Kernel panic - not syncing: Attempted to kill init!
> > ----------------------------------------------------
> >
> > Any ideas?
> >
> > thank you,
> > R
> >
>
> That's probably the point at which it switches from initrd to the
> disk fs. Check you rootfs is compatible with your kernel and
> accessible.
>
> Can yo boot to it over nfs using your new kernel?
>
I do not have an NFS server, but it seems that it boots alright.
So the question is how do I make the kernel recognize the root device
hda1?
I parsed the lines in the .config file, but I could not find anything
helpful.

R

On Thu, May 29, 2008 at 12:06 PM, Gonzalo A. de la Vega
<gadelavega@...> wrote:
> On Thu, May 29, 2008 at 10:39 AM, charliem_1216 <charliem_1216@...>
wrote:
>> --- In ts-7000@yahoogroups.com, "Gonzalo A. de la Vega"
>> <gadelavega@...> wrote:
>>>
>>> On Tue, May 27, 2008 at 11:35 PM, charliem_1216 <charliem_1216@...>
>> wrote:
>>> > Hi Gonzalo --
>>> >
>>> > --- In ts-7000@yahoogroups.com, "Gonzalo A. de la Vega"
>>> >
>>> > <gadelavega@> wrote:
>>> >>
>>> >> On Tue, May 27, 2008 at 8:52 PM, Oliver Martin
>>> >> <oliver.martin@> wrote:
>>> >> > Am Tue, 27 May 2008 22:46:36 -0000 schrieb Gonzalo A. de la
>>> > Vega:
>>> >> >
>>> >> >> Allright guys... I give up. How do you get udev to show all the
>>> >> >> devices at startup with TS-Linux? I can boot the 2.6 kernel but
>>> > then
>>> >> >> the /dev/ directory is empty. Isn't udev supposed to manage
>>> > that?
>>> >> >> I can make nodes like /dev/null but the /dev/ttyAM* won't work.
>>> >> >>
>>> >> > If your board comes with 2.4, then it doesn't use udev but
>>> > devfs. You
>>> >> > either have to manually install udev in TS-Linux or build your
>>> > own
>>> >> > "distro" (i.e., root fs). Have a look at ptxdist, it will give
>>> > you
>>> >> > something workable with a bit of experimentation.
>>> >>
>>> >> It's a TS7260 so it comes with 2.4, but I built a 2.6, booted, but
>>> > I
>>> >> get no devices, despite having installed udev. I'll try ptxdist,
>>> > at
>>> >> least to check how they handle this.
>>> >>
>>> >> Thanks!
>>> >>
>>> >> Gonzalo
>>> >>
>>> >
>>> > I always thought that udev was a little overkill for embedded
>>> > systems, where you generally know ahead of time what devices you
>>> > will need, and can make the device nodes the old fashioned way.
>>> > That's what I did for my 2.6 development, but IIRC there were some
>>> > devices required that were not immediately obvious.
>>> >
>>> > Regards, ...... Charlie
>>>
>>> Agree about udev on embedded systems. So you created the nodes
>>> manually and it worked?
>>>
>>
>> Yep, worked great once I got everything that was needed. From memory,
>> check for device nodes: console, ptmx, tty, tty#, ttyS#, ttyAM#, sd?
>>
>> And links: ft, stdin, stdout, stderr
>>
>> And directories: pts (to mount devpts FS), maybe shmem (to mount tmpfs
>> FS).
>>
>> I probably forgot some, but look into Documentation/devices.txt in any
>> recent kernel tree for more details. IIRC I had trouble until I
>> figured out pts and ptmx.
>>
>> Regards, ..... Charlie
>
> Thanks Charlie,
> I tried creating the ttyAM# devices and didn't work, but not I'm sure
> I did something wrong. I'll try again right away.

I was using the wrong mayor,minor pair!!! It works now, so we don't need udev.
BTW: in the previous message it should say "... but now I' m sure..."

--- In ts-7000@yahoogroups.com, "Gonzalo A. de la
Vega" <gadelavega@...> wrote:
>
> On Thu, May 29, 2008 at 12:06 PM, Gonzalo A. de la Vega
> <gadelavega@...> wrote:
> > On Thu, May 29, 2008 at 10:39 AM, charliem_1216
<charliem_1216@...> wrote:
> >> --- In ts-7000@yahoogroups.com, "Gonzalo A. de la Vega"
> >> <gadelavega@> wrote:
> >>>
> >>> On Tue, May 27, 2008 at 11:35 PM, charliem_1216
<charliem_1216@>
> >> wrote:
> >>> > Hi Gonzalo --
> >>> >
> >>> > --- In ts-7000@yahoogroups.com, "Gonzalo A. de la Vega"
> >>> >
> >>> > <gadelavega@> wrote:
> >>> >>
> >>> >> On Tue, May 27, 2008 at 8:52 PM, Oliver Martin
> >>> >> <oliver.martin@> wrote:
> >>> >> > Am Tue, 27 May 2008 22:46:36 -0000 schrieb Gonzalo A. de
la
> >>> > Vega:
> >>> >> >
> >>> >> >> Allright guys... I give up. How do you get udev to show
all the
> >>> >> >> devices at startup with TS-Linux? I can boot the 2.6
kernel but
> >>> > then
> >>> >> >> the /dev/ directory is empty. Isn't udev supposed to
manage
> >>> > that?
> >>> >> >> I can make nodes like /dev/null but the /dev/ttyAM*
won't work.
> >>> >> >>
> >>> >> > If your board comes with 2.4, then it doesn't use udev but
> >>> > devfs. You
> >>> >> > either have to manually install udev in TS-Linux or build
your
> >>> > own
> >>> >> > "distro" (i.e., root fs). Have a look at ptxdist, it will
give
> >>> > you
> >>> >> > something workable with a bit of experimentation.
> >>> >>
> >>> >> It's a TS7260 so it comes with 2.4, but I built a 2.6,
booted, but
> >>> > I
> >>> >> get no devices, despite having installed udev. I'll try
ptxdist,
> >>> > at
> >>> >> least to check how they handle this.
> >>> >>
> >>> >> Thanks!
> >>> >>
> >>> >> Gonzalo
> >>> >>
> >>> >
> >>> > I always thought that udev was a little overkill for embedded
> >>> > systems, where you generally know ahead of time what devices
you
> >>> > will need, and can make the device nodes the old fashioned
way.
> >>> > That's what I did for my 2.6 development, but IIRC there
were some
> >>> > devices required that were not immediately obvious.
> >>> >
> >>> > Regards, ...... Charlie
> >>>
> >>> Agree about udev on embedded systems. So you created the nodes
> >>> manually and it worked?
> >>>
> >>
> >> Yep, worked great once I got everything that was needed. From
memory,
> >> check for device nodes: console, ptmx, tty, tty#, ttyS#,
ttyAM#, sd?
> >>
> >> And links: ft, stdin, stdout, stderr
> >>
> >> And directories: pts (to mount devpts FS), maybe shmem (to
mount tmpfs
> >> FS).
> >>
> >> I probably forgot some, but look into Documentation/devices.txt
in any
> >> recent kernel tree for more details. IIRC I had trouble until I
> >> figured out pts and ptmx.
> >>
> >> Regards, ..... Charlie
> >
> > Thanks Charlie,
> > I tried creating the ttyAM# devices and didn't work, but not I'm
sure
> > I did something wrong. I'll try again right away.
>
> I was using the wrong mayor,minor pair!!! It works now, so we
don't need udev.
> BTW: in the previous message it should say "... but now I' m
sure..."
>

Hi,

great to see you sussed it.

Could you post details of all the devices , that will save me
duplicationg what you've just done.

Thx.

On Thu, May 29, 2008 at 5:34 PM, j.chitte <j.chitte@...> wrote:
> --- In ts-7000@yahoogroups.com, "Gonzalo A. de la
> Vega" <gadelavega@...> wrote:
>>
>> On Thu, May 29, 2008 at 12:06 PM, Gonzalo A. de la Vega
>> <gadelavega@...> wrote:
>> > On Thu, May 29, 2008 at 10:39 AM, charliem_1216
> <charliem_1216@...> wrote:
>> >> --- In ts-7000@yahoogroups.com, "Gonzalo A. de la Vega"
>> >> <gadelavega@> wrote:
>> >>>
>> >>> On Tue, May 27, 2008 at 11:35 PM, charliem_1216
> <charliem_1216@>
>> >> wrote:
>> >>> > Hi Gonzalo --
>> >>> >
>> >>> > --- In ts-7000@yahoogroups.com, "Gonzalo A. de la Vega"
>> >>> >
>> >>> > <gadelavega@> wrote:
>> >>> >>
>> >>> >> On Tue, May 27, 2008 at 8:52 PM, Oliver Martin
>> >>> >> <oliver.martin@> wrote:
>> >>> >> > Am Tue, 27 May 2008 22:46:36 -0000 schrieb Gonzalo A. de
> la
>> >>> > Vega:
>> >>> >> >
>> >>> >> >> Allright guys... I give up. How do you get udev to show
> all the
>> >>> >> >> devices at startup with TS-Linux? I can boot the 2.6
> kernel but
>> >>> > then
>> >>> >> >> the /dev/ directory is empty. Isn't udev supposed to
> manage
>> >>> > that?
>> >>> >> >> I can make nodes like /dev/null but the /dev/ttyAM*
> won't work.
>> >>> >> >>
>> >>> >> > If your board comes with 2.4, then it doesn't use udev but
>> >>> > devfs. You
>> >>> >> > either have to manually install udev in TS-Linux or build
> your
>> >>> > own
>> >>> >> > "distro" (i.e., root fs). Have a look at ptxdist, it will
> give
>> >>> > you
>> >>> >> > something workable with a bit of experimentation.
>> >>> >>
>> >>> >> It's a TS7260 so it comes with 2.4, but I built a 2.6,
> booted, but
>> >>> > I
>> >>> >> get no devices, despite having installed udev. I'll try
> ptxdist,
>> >>> > at
>> >>> >> least to check how they handle this.
>> >>> >>
>> >>> >> Thanks!
>> >>> >>
>> >>> >> Gonzalo
>> >>> >>
>> >>> >
>> >>> > I always thought that udev was a little overkill for embedded
>> >>> > systems, where you generally know ahead of time what devices
> you
>> >>> > will need, and can make the device nodes the old fashioned
> way.
>> >>> > That's what I did for my 2.6 development, but IIRC there
> were some
>> >>> > devices required that were not immediately obvious.
>> >>> >
>> >>> > Regards, ...... Charlie
>> >>>
>> >>> Agree about udev on embedded systems. So you created the nodes
>> >>> manually and it worked?
>> >>>
>> >>
>> >> Yep, worked great once I got everything that was needed. From
> memory,
>> >> check for device nodes: console, ptmx, tty, tty#, ttyS#,
> ttyAM#, sd?
>> >>
>> >> And links: ft, stdin, stdout, stderr
>> >>
>> >> And directories: pts (to mount devpts FS), maybe shmem (to
> mount tmpfs
>> >> FS).
>> >>
>> >> I probably forgot some, but look into Documentation/devices.txt
> in any
>> >> recent kernel tree for more details. IIRC I had trouble until I
>> >> figured out pts and ptmx.
>> >>
>> >> Regards, ..... Charlie
>> >
>> > Thanks Charlie,
>> > I tried creating the ttyAM# devices and didn't work, but not I'm
> sure
>> > I did something wrong. I'll try again right away.
>>
>> I was using the wrong mayor,minor pair!!! It works now, so we
> don't need udev.
>> BTW: in the previous message it should say "... but now I' m
> sure..."
>>
>
> Hi,
>
> great to see you sussed it.
>
> Could you post details of all the devices , that will save me
> duplicationg what you've just done.
>
> Thx.

OK, give ma a day 'cause I'm leaving the lab right now. I'll finish
creating all the devices tomorrow and e-mail the solution, probably as
a script.

G.

--- In ts-7000@yahoogroups.com, "Razvan Ionut Stoian"
<razvan_ionut_stoian@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, "Razvan Ionut
> > Stoian" <razvan_ionut_stoian@> wrote:
> > >
> > > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > > >
> > > >
> > > >
> > > >
> > > > On Thu, 29 May 2008, Razvan Ionut Stoian wrote:
> > > >
> > > > > --- In ts-7000@yahoogroups.com, Triffid Hunter
> > <triffid_hunter@>
> > > wrote:
> > > > >>
> > > > >> On Wed, 28 May 2008, Razvan Ionut Stoian wrote:
> > > > >>
> > > > >>> For some reason the system freezes after vmlinux starts
> > > > >>> to decompress. This is what I got:
> > > > >>>
> > > > >>> Uncompressing
> > > > >>>
> > Linux.............................................................
> > > > >>>
> > > > >
> > >
> >
> ........................................................................
> > > > >>>
> > > > >>> What am I doing wrong?
> > > > >>
> > > > >> Mine do that if my toolchain is broken
> > > > >>
> > > > >
> > > > > This is odd. I used the toolchain provided by TS.
> > > > > Any good alternatives?
> > > >
> > > > You can check if you toolchain is ok, by compiling the supplied
> > 2.4
> > > kernel
> > > > source, with the default configuration. If the 2.4 kernel fails
> > to boot
> > > > then it may be your toolchain. If 2.4 works, then you have a 2.6
> > issue.
> > > >
> > > Hi.
> > >
> > > I managed to boot the 2.6 kernel. Apparently, the initial image was
> > > just too big. Now it boots, but my SBC freezes due to a kernel
> > panic:
> > > ----------------------------------------------------
> > > Kernel panic - not syncing: Attempted to kill init!
> > > ----------------------------------------------------
> > >
> > > Any ideas?
> > >
> > > thank you,
> > > R
> > >
> >
> > That's probably the point at which it switches from initrd to the
> > disk fs. Check you rootfs is compatible with your kernel and
> > accessible.
> >
> > Can yo boot to it over nfs using your new kernel?
> >
> I do not have an NFS server, but it seems that it boots alright.
> So the question is how do I make the kernel recognize the root device
> hda1?
> I parsed the lines in the .config file, but I could not find anything
> helpful.
>
> R
>

I used the debian_sarge_udev filesystem and copied vmlinux to /boot.
Still nothing...

Now I get this message:
================================================================
VFS: Cannot open root device "hda1" or unknown-block(0,0)
Please append a correct "root=" boot option
Kernel panic - not syncing: VFS: UNable to mount root fs on
unknown-block(0,0)
================================================================

I included the ext2/ext3/cramfs modules in the kernel so the
partitions could be recognized.
What am I missing here?

Razvan

--- In ts-7000@yahoogroups.com, "Razvan Ionut
Stoian" <razvan_ionut_stoian@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "Razvan Ionut Stoian"
> <razvan_ionut_stoian@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@> wrote:
> > >
> > > --- In ts-7000@yahoogroups.com, "Razvan Ionut
> > > Stoian" <razvan_ionut_stoian@> wrote:
> > > >
> > > > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > On Thu, 29 May 2008, Razvan Ionut Stoian wrote:
> > > > >
> > > > > > --- In ts-7000@yahoogroups.com, Triffid Hunter
> > > <triffid_hunter@>
> > > > wrote:
> > > > > >>
> > > > > >> On Wed, 28 May 2008, Razvan Ionut Stoian wrote:
> > > > > >>
> > > > > >>> For some reason the system freezes after vmlinux starts
> > > > > >>> to decompress. This is what I got:
> > > > > >>>
> > > > > >>> Uncompressing
> > > > > >>>
> > >
Linux.............................................................
> > > > > >>>
> > > > > >
> > > >
> > >
> >
  ........................................................................
> > > > > >>>
> > > > > >>> What am I doing wrong?
> > > > > >>
> > > > > >> Mine do that if my toolchain is broken
> > > > > >>
> > > > > >
> > > > > > This is odd. I used the toolchain provided by TS.
> > > > > > Any good alternatives?
> > > > >
> > > > > You can check if you toolchain is ok, by compiling the
supplied
> > > 2.4
> > > > kernel
> > > > > source, with the default configuration. If the 2.4 kernel
fails
> > > to boot
> > > > > then it may be your toolchain. If 2.4 works, then you have
a 2.6
> > > issue.
> > > > >
> > > > Hi.
> > > >
> > > > I managed to boot the 2.6 kernel. Apparently, the initial
image was
> > > > just too big. Now it boots, but my SBC freezes due to a
kernel
> > > panic:
> > > > ----------------------------------------------------
> > > > Kernel panic - not syncing: Attempted to kill init!
> > > > ----------------------------------------------------
> > > >
> > > > Any ideas?
> > > >
> > > > thank you,
> > > > R
> > > >
> > >
> > > That's probably the point at which it switches from initrd to
the
> > > disk fs. Check you rootfs is compatible with your kernel and
> > > accessible.
> > >
> > > Can yo boot to it over nfs using your new kernel?
> > >
> > I do not have an NFS server, but it seems that it boots alright.
> > So the question is how do I make the kernel recognize the root
device
> > hda1?
> > I parsed the lines in the .config file, but I could not find
anything
> > helpful.
> >
> > R
> >
>
> I used the debian_sarge_udev filesystem and copied vmlinux to /boot.
> Still nothing...
>
> Now I get this message:
> ================================================================
> VFS: Cannot open root device "hda1" or unknown-block(0,0)
> Please append a correct "root=" boot option
> Kernel panic - not syncing: VFS: UNable to mount root fs on
> unknown-block(0,0)
> ================================================================
>
> I included the ext2/ext3/cramfs modules in the kernel so the
> partitions could be recognized.
> What am I missing here?
>
> Razvan
>


> VFS: Cannot open root device "hda1" or unknown-block(0,0)
hda1 is not a device ;) /dev/hda1 ?

check what you are feeding to redboot.

> Please append a correct "root=" boot option
sounds like a good idea to try too.

Hi List,

On http://bat3.googlecode.com you can find a new version (0.1.3) of my
own bat3 software.  It allows you to load your batteries with 1500mA
which we needed to fast load our 4200 NiMH batteries.  There's one big
fat warning about the code: it doesn't control if the batteries are
loaded or not so I guess you can blow up your batteries if you don't
watch the state.
I normally start the program running `bat3 -s -1` which makes the
program keep on running.  I control the battery using a raw telnet
interface on port 1302.  I watch the current state with the state
command and change the target current with `current 1500`.

Any comments are welcome,

wimpunk.
--
I would love to change the world, but they wont give me the source code.

--- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "Razvan Ionut
> Stoian" <razvan_ionut_stoian@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, "Razvan Ionut Stoian"
> > <razvan_ionut_stoian@> wrote:
> > >
> > > --- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@> wrote:
> > > >
> > > > --- In ts-7000@yahoogroups.com, "Razvan Ionut
> > > > Stoian" <razvan_ionut_stoian@> wrote:
> > > > >
> > > > > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Thu, 29 May 2008, Razvan Ionut Stoian wrote:
> > > > > >
> > > > > > > --- In ts-7000@yahoogroups.com, Triffid Hunter
> > > > <triffid_hunter@>
> > > > > wrote:
> > > > > > >>
> > > > > > >> On Wed, 28 May 2008, Razvan Ionut Stoian wrote:
> > > > > > >>
> > > > > > >>> For some reason the system freezes after vmlinux
starts
> > > > > > >>> to decompress. This is what I got:
> > > > > > >>>
> > > > > > >>> Uncompressing
> > > > > > >>>
> > > >
> Linux.............................................................
> > > > > > >>>
> > > > > > >
> > > > >
> > > >
> > >
>
   ........................................................................
> > > > > > >>>
> > > > > > >>> What am I doing wrong?
> > > > > > >>
> > > > > > >> Mine do that if my toolchain is broken
> > > > > > >>
> > > > > > >
> > > > > > > This is odd. I used the toolchain provided by TS.
> > > > > > > Any good alternatives?
> > > > > >
> > > > > > You can check if you toolchain is ok, by compiling the
> supplied
> > > > 2.4
> > > > > kernel
> > > > > > source, with the default configuration. If the 2.4 kernel
> fails
> > > > to boot
> > > > > > then it may be your toolchain. If 2.4 works, then you
have
> a 2.6
> > > > issue.
> > > > > >
> > > > > Hi.
> > > > >
> > > > > I managed to boot the 2.6 kernel. Apparently, the initial
> image was
> > > > > just too big. Now it boots, but my SBC freezes due to a
> kernel
> > > > panic:
> > > > > ----------------------------------------------------
> > > > > Kernel panic - not syncing: Attempted to kill init!
> > > > > ----------------------------------------------------
> > > > >
> > > > > Any ideas?
> > > > >
> > > > > thank you,
> > > > > R
> > > > >
> > > >
> > > > That's probably the point at which it switches from initrd to
> the
> > > > disk fs. Check you rootfs is compatible with your kernel and
> > > > accessible.
> > > >
> > > > Can yo boot to it over nfs using your new kernel?
> > > >
> > > I do not have an NFS server, but it seems that it boots alright.
> > > So the question is how do I make the kernel recognize the root
> device
> > > hda1?
> > > I parsed the lines in the .config file, but I could not find
> anything
> > > helpful.
> > >
> > > R
> > >
> >
> > I used the debian_sarge_udev filesystem and copied vmlinux to /
boot.
> > Still nothing...
> >
> > Now I get this message:
> > ================================================================
> > VFS: Cannot open root device "hda1" or unknown-block(0,0)
> > Please append a correct "root=" boot option
> > Kernel panic - not syncing: VFS: UNable to mount root fs on
> > unknown-block(0,0)
> > ================================================================
> >
> > I included the ext2/ext3/cramfs modules in the kernel so the
> > partitions could be recognized.
> > What am I missing here?
> >
> > Razvan
> >
>
>
> > VFS: Cannot open root device "hda1" or unknown-block(0,0)
> hda1 is not a device ;) /dev/hda1 ?
>
> check what you are feeding to redboot.
>
> > Please append a correct "root=" boot option
> sounds like a good idea to try too.
>

TS manual states:


The root file system used by the Linux OS can be any of the following:
✔ JFFS/YAFFS file system image in the on-board Flash (if using
RedBoot, it should
include the option root=/dev/mtdblock1 to instruct the kernel to boot
here)
✔ EXT2 file system image in the Compact Flash card (if using
RedBoot, it should include
the option root=/dev/hda)
✔ NFS root, via Ethernet port (if using RedBoot, it should include
the option root=/dev/nfs
nfsroot=<IP>:<DIRECTORY> ip=dhcp)

HTH

Hi,

there seems to be quite a lot of individual work on 2.6.x kernels so
I thought it would be a good idea to focus various segments of code
and knowlege into creating a generic 2.6 kernel for all 72xx boards.

The overall aim being to produce kernel , patchset .config and
modules tarballs to post the list.

If possible it should support all NAND/NOR flash types and SD cards
from the kernel tree , hence gnu-ts kernel (and gnuts to proprietory
closed source modules).

It would seem a good idea to start from a recent kernel unless there
is are any show stoppers.

I'm looking at 2.6.25 where one issue is the disappearance of
read_inode that affects yaffs2 . This appears to be fixed here but I
have not yet verified locally:

http://lists.aleph1.co.uk/lurker/message/
20080423.092213.a8dd7079.en.html

If anyone has any particular reason recommend the use this or other
2.6.x as a base please post comments.

best regards,

jacques.

On Fri, May 30, 2008 at 6:30 AM, j.chitte <j.chitte@...> wrote:
> Hi,
>
> there seems to be quite a lot of individual work on 2.6.x kernels so
> I thought it would be a good idea to focus various segments of code
> and knowlege into creating a generic 2.6 kernel for all 72xx boards.
>
> The overall aim being to produce kernel , patchset .config and
> modules tarballs to post the list.
>
> If possible it should support all NAND/NOR flash types and SD cards
> from the kernel tree , hence gnu-ts kernel (and gnuts to proprietory
> closed source modules).
>
> It would seem a good idea to start from a recent kernel unless there
> is are any show stoppers.
>
> I'm looking at 2.6.25 where one issue is the disappearance of
> read_inode that affects yaffs2 . This appears to be fixed here but I
> have not yet verified locally:
>
> http://lists.aleph1.co.uk/lurker/message/
> 20080423.092213.a8dd7079.en.html
>
> If anyone has any particular reason recommend the use this or other
> 2.6.x as a base please post comments.
>
> best regards,
>
> jacques.

It sounds good to unify our efforts. I'm now working on creating the
needed device nodes so we can actually run 2.6 kernel on a plain SBC
(no external driver nor NFS). I'll probably try udev too, but it
doesn't seem useful. In embedded is kind of an overkill, as Charlie
<charliem_1216@...> said since you're not plugging things all
the time... actually, you never do it.

Gonzalo

Gonzalo.

On Fri, 30 May 2008, Gonzalo A. de la Vega wrote:

> It sounds good to unify our efforts. I'm now working on creating the
> needed device nodes so we can actually run 2.6 kernel on a plain SBC
> (no external driver nor NFS). I'll probably try udev too, but it
> doesn't seem useful. In embedded is kind of an overkill, as Charlie
> <charliem_1216@...> said since you're not plugging things all
> the time... actually, you never do it.

I really like only having /dev/{console,null} on "disk", then mounting a
ramfs over /dev and letting mdev (busybox's udev equiv) populate it for
me. That way, I only use a couple of inodes, and if I do plug in a usb
wifi or usb disk or whatever, it appears without me having to do anything
:)

I often plug/unplug various usb devices into my ts7250, and don't want to
have to look up major/minor numbers every time I do.

I concur on the unification of effort thing - something I'd really like to
have is a binary i686 gcc-4 uclibc-eabi and gnueabi toolchain. All my
efforts to create such a beast have been fruitless so far.

It seems like most of the patches we need have been trickling upstream
which is excellent :)

--- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, "Razvan Ionut
> > Stoian" <razvan_ionut_stoian@> wrote:
> > >
> > > --- In ts-7000@yahoogroups.com, "Razvan Ionut Stoian"
> > > <razvan_ionut_stoian@> wrote:
> > > >
> > > > --- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@> wrote:
> > > > >
> > > > > --- In ts-7000@yahoogroups.com, "Razvan Ionut
> > > > > Stoian" <razvan_ionut_stoian@> wrote:
> > > > > >
> > > > > > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On Thu, 29 May 2008, Razvan Ionut Stoian wrote:
> > > > > > >
> > > > > > > > --- In ts-7000@yahoogroups.com, Triffid Hunter
> > > > > <triffid_hunter@>
> > > > > > wrote:
> > > > > > > >>
> > > > > > > >> On Wed, 28 May 2008, Razvan Ionut Stoian wrote:
> > > > > > > >>
> > > > > > > >>> For some reason the system freezes after vmlinux
> starts
> > > > > > > >>> to decompress. This is what I got:
> > > > > > > >>>
> > > > > > > >>> Uncompressing
> > > > > > > >>>
> > > > >
> > Linux.............................................................
> > > > > > > >>>
> > > > > > > >
> > > > > >
> > > > >
> > > >
> >
>
........................................................................
> > > > > > > >>>
> > > > > > > >>> What am I doing wrong?
> > > > > > > >>
> > > > > > > >> Mine do that if my toolchain is broken
> > > > > > > >>
> > > > > > > >
> > > > > > > > This is odd. I used the toolchain provided by TS.
> > > > > > > > Any good alternatives?
> > > > > > >
> > > > > > > You can check if you toolchain is ok, by compiling the
> > supplied
> > > > > 2.4
> > > > > > kernel
> > > > > > > source, with the default configuration. If the 2.4 kernel
> > fails
> > > > > to boot
> > > > > > > then it may be your toolchain. If 2.4 works, then you
> have
> > a 2.6
> > > > > issue.
> > > > > > >
> > > > > > Hi.
> > > > > >
> > > > > > I managed to boot the 2.6 kernel. Apparently, the initial
> > image was
> > > > > > just too big. Now it boots, but my SBC freezes due to a
> > kernel
> > > > > panic:
> > > > > > ----------------------------------------------------
> > > > > > Kernel panic - not syncing: Attempted to kill init!
> > > > > > ----------------------------------------------------
> > > > > >
> > > > > > Any ideas?
> > > > > >
> > > > > > thank you,
> > > > > > R
> > > > > >
> > > > >
> > > > > That's probably the point at which it switches from initrd to
> > the
> > > > > disk fs. Check you rootfs is compatible with your kernel and
> > > > > accessible.
> > > > >
> > > > > Can yo boot to it over nfs using your new kernel?
> > > > >
> > > > I do not have an NFS server, but it seems that it boots alright.
> > > > So the question is how do I make the kernel recognize the root
> > device
> > > > hda1?
> > > > I parsed the lines in the .config file, but I could not find
> > anything
> > > > helpful.
> > > >
> > > > R
> > > >
> > >
> > > I used the debian_sarge_udev filesystem and copied vmlinux to /
> boot.
> > > Still nothing...
> > >
> > > Now I get this message:
> > > ================================================================
> > > VFS: Cannot open root device "hda1" or unknown-block(0,0)
> > > Please append a correct "root=" boot option
> > > Kernel panic - not syncing: VFS: UNable to mount root fs on
> > > unknown-block(0,0)
> > > ================================================================
> > >
> > > I included the ext2/ext3/cramfs modules in the kernel so the
> > > partitions could be recognized.
> > > What am I missing here?
> > >
> > > Razvan
> > >
> >
> >
> > > VFS: Cannot open root device "hda1" or unknown-block(0,0)
> > hda1 is not a device ;) /dev/hda1 ?
> >
> > check what you are feeding to redboot.
> >
> > > Please append a correct "root=" boot option
> > sounds like a good idea to try too.
> >
>
> TS manual states:
>
>
> The root file system used by the Linux OS can be any of the following:
> ✔ JFFS/YAFFS file system image in the on-board Flash (if using
> RedBoot, it should
> include the option root=/dev/mtdblock1 to instruct the kernel to boot
> here)
> ✔ EXT2 file system image in the Compact Flash card (if using
> RedBoot, it should include
> the option root=/dev/hda)
> ✔ NFS root, via Ethernet port (if using RedBoot, it should
include
> the option root=/dev/nfs
> nfsroot=<IP>:<DIRECTORY> ip=dhcp)
>
> HTH
>

This is the problem. These errors show up even if I give the correct
command line in RedBoot.
RedBoot> load -r -b 0x00218000 -m disk hda1:/boot/vmlinux
RedBoot> exec -c "console=ttyAM0,115200 ip=dhcp root=/dev/hda1"

with my CF card formated as ext2.

R

> TS manual states:
>
>
> The root file system used by the Linux OS can be any of the following:
> ✔ JFFS/YAFFS file system image in the on-board Flash (if using
> RedBoot, it should
> include the option root=/dev/mtdblock1 to instruct the kernel to boot
> here)
> ✔ EXT2 file system image in the Compact Flash card (if using
> RedBoot, it should include
> the option root=/dev/hda)
                      ------^

There would normally be a number there, e.g. /dev/hda1

> ✔ NFS root, via Ethernet port (if using RedBoot, it should include
> the option root=/dev/nfs
> nfsroot=<IP>:<DIRECTORY> ip=dhcp)
>
> HTH
>
>
>

--- In ts-7000@yahoogroups.com, Jim Jackson <jj@...> wrote:
>
>
>
> > TS manual states:
> >
> >
> > The root file system used by the Linux OS can be any of the following:
> > ✔ JFFS/YAFFS file system image in the on-board Flash (if using
> > RedBoot, it should
> > include the option root=/dev/mtdblock1 to instruct the kernel to boot
> > here)
> > ✔ EXT2 file system image in the Compact Flash card (if using
> > RedBoot, it should include
> > the option root=/dev/hda)
>                      ------^
>
> There would normally be a number there, e.g. /dev/hda1
>
> > ✔ NFS root, via Ethernet port (if using RedBoot, it should
include
> > the option root=/dev/nfs
> > nfsroot=<IP>:<DIRECTORY> ip=dhcp)
> >
> > HTH
> >
> >
> >
>

I tried both options root=/dev/hda and root=/dev/hda1 and I still get
kernel panic.

R

--- In ts-7000@yahoogroups.com, "Razvan Ionut
Stoian" <razvan_ionut_stoian@...> wrote:
>
> --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> >
> >
> >
> > > TS manual states:
> > >
> > >
> > > The root file system used by the Linux OS can be any of the
following:
> > > ✔ JFFS/YAFFS file system image in the on-board Flash
(if using
> > > RedBoot, it should
> > > include the option root=/dev/mtdblock1 to instruct the kernel
to boot
> > > here)
> > > ✔ EXT2 file system image in the Compact Flash card (if
using
> > > RedBoot, it should include
> > > the option root=/dev/hda)
> >                      ------^
> >
> > There would normally be a number there, e.g. /dev/hda1
> >
> > > ✔ NFS root, via Ethernet port (if using RedBoot, it
should
> include
> > > the option root=/dev/nfs
> > > nfsroot=<IP>:<DIRECTORY> ip=dhcp)
> > >
> > > HTH
> > >
> > >
> > >
> >
>
> I tried both options root=/dev/hda and root=/dev/hda1 and I still
get
> kernel panic.
>
> R
>

Well you said you did not have nfs server but that really would help
you narrow things down. If you can boot to the root fs via nfs you
know it's consistant and that your problem lies with you CF driver or
fs.

If you dont have nfs I guess that means you are working from windows.
Probably not the best dev platform for a linux target.

I suggest you get a liveCD of some sort , linux from scratch or
partedmagic would do. I'm fairly sure both of those will let you boot
directly to a system having nfs. Mount your rootfs where ever it sits
and export it as nfs. Just a couple of commands really.

It seems you currently have too many unknowns.

--- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "Razvan Ionut
> Stoian" <razvan_ionut_stoian@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > >
> > >
> > >
> > > > TS manual states:
> > > >
> > > >
> > > > The root file system used by the Linux OS can be any of the
> following:
> > > > ✔ JFFS/YAFFS file system image in the on-board Flash
> (if using
> > > > RedBoot, it should
> > > > include the option root=/dev/mtdblock1 to instruct the kernel
> to boot
> > > > here)
> > > > ✔ EXT2 file system image in the Compact Flash card (if
> using
> > > > RedBoot, it should include
> > > > the option root=/dev/hda)
> > >                      ------^
> > >
> > > There would normally be a number there, e.g. /dev/hda1
> > >
> > > > ✔ NFS root, via Ethernet port (if using RedBoot, it
> should
> > include
> > > > the option root=/dev/nfs
> > > > nfsroot=<IP>:<DIRECTORY> ip=dhcp)
> > > >
> > > > HTH
> > > >
> > > >
> > > >
> > >
> >
> > I tried both options root=/dev/hda and root=/dev/hda1 and I still
> get
> > kernel panic.
> >
> > R
> >
>
> Well you said you did not have nfs server but that really would help
> you narrow things down. If you can boot to the root fs via nfs you
> know it's consistant and that your problem lies with you CF driver or
> fs.
>
> If you dont have nfs I guess that means you are working from windows.
> Probably not the best dev platform for a linux target.
>
> I suggest you get a liveCD of some sort , linux from scratch or
> partedmagic would do. I'm fairly sure both of those will let you boot
> directly to a system having nfs. Mount your rootfs where ever it sits
> and export it as nfs. Just a couple of commands really.
>
> It seems you currently have too many unknowns.
>

No, I am not working from Windoze ... Actually I haven't seen a
winbloze machine in a looooong time.
... and if there is a problem with the CF card driver how could it be
solved? I used the ts2.6 kernel sources for compiling the kernel.

R

Howdy All!

I am trying to recompile wu-ftpd-2.6.2 (with option --disable-dns) on
my TS-7800.

I am not able to run: ./configure --disable-dns

I get the following output:

$ ./configure --disable-dns
checking build system type... Invalid configuration
`armv5tejl-pc-linux-gnu': machine `armv5tejl-pc' not recognized
configure: error: /bin/sh ./config.sub armv5tejl-pc-linux-gnu failed.


I think I need to set: --build=BUILD --host=HOST --target=TARGET
...but I am not sure what to set them to.

Has anybody done this before?

Thanks!
Robert

Hi Robert --

--- In ts-7000@yahoogroups.com, "Robert" <rulbrich@...> wrote:
>
> Howdy All!
>
> I am trying to recompile wu-ftpd-2.6.2 (with option --disable-dns) on
> my TS-7800.
>
> I am not able to run: ./configure --disable-dns
>
> I get the following output:
>
> $ ./configure --disable-dns
> checking build system type... Invalid configuration
> `armv5tejl-pc-linux-gnu': machine `armv5tejl-pc' not recognized
> configure: error: /bin/sh ./config.sub armv5tejl-pc-linux-gnu failed.
>
>
> I think I need to set: --build=BUILD --host=HOST --target=TARGET
> ...but I am not sure what to set them to.

No, I think these are only needed if you are cross-compiling, which
you are not doing.

>
> Has anybody done this before?
>
> Thanks!
> Robert
>

Debian carries almost 140 kB of (compressed!) diffs to the source of
this package fixing bugs, vulnerabilities and documentation[1].  Have
you applied these diffs?  I noticed massive updates to configure
files, for example.

If that doesn't work, you'll have to do it the old fashioned way, by
editing headers & makefiles.  But you should still patch the original
source, as lots of vulnerabilities have been fixed.

Regards, ........ Charlie

[1] http://packages.debian.org/source/sid/wu-ftpd

On Fri, May 30, 2008 at 9:35 AM, Triffid Hunter
<triffid_hunter@...> wrote:
> On Fri, 30 May 2008, Gonzalo A. de la Vega wrote:
>
>> It sounds good to unify our efforts. I'm now working on creating the
>> needed device nodes so we can actually run 2.6 kernel on a plain SBC
>> (no external driver nor NFS). I'll probably try udev too, but it
>> doesn't seem useful. In embedded is kind of an overkill, as Charlie
>> <charliem_1216@...> said since you're not plugging things all
>> the time... actually, you never do it.
>
> I really like only having /dev/{console,null} on "disk", then mounting a
> ramfs over /dev and letting mdev (busybox's udev equiv) populate it for
> me.

Oooohhh... so that's the trick, mounting a ramfs... so that's why udev
didn't work.

> That way, I only use a couple of inodes, and if I do plug in a usb
> wifi or usb disk or whatever, it appears without me having to do anything
> :)

Mine are 400km from here, in the middle of the desert, so to get there
I need to travel 5 hours by bus, pick a truck and drive to each
station each 1.5km (about a mile) away from the others. That's why I
never plug anything to them!!! :-D

> I often plug/unplug various usb devices into my ts7250, and don't want to
> have to look up major/minor numbers every time I do.

mdev is certainly your choice.

> I concur on the unification of effort thing - something I'd really like to
> have is a binary i686 gcc-4 uclibc-eabi and gnueabi toolchain. All my
> efforts to create such a beast have been fruitless so far.

I'll give it a try, but if you didn't succeed why would I? ;-)

> It seems like most of the patches we need have been trickling upstream
> which is excellent :)

Yeah, saves a lot of work.

Folks,
I sat down last night to blink my 7400's LEDs using a 2.4.26-ts11
module. I did what I've always done before with arm processors
ioremap_nocache with physical address then use readw and writew.
Nothing happened. Further investigation revealed inw/outw, IO_ADDRESS,
__ioremap approaches. When I wrote my book 6 years there was only
ioremap_*. Can someone explain why these other approaches have cropped
up over the years with 2.4?

Thanks,
Craig


Dr. Craig Hollabaugh, craig@..., 970 240 0509
Author of Embedded Linux: Hardware, Software and Interfacing
www.embeddedlinuxinterfacing.com

Actually, unless you really *need* wu-ftpd for some reason, you'd be
better off avoiding it.
It has a long history of security holes and is basically junk. The
more recent pureftpd and proftpd, for example, are generally more
sound. Their build scripts probably work too!

The answer to your specific question is that if you're building from
debian source on a debian box, the debian way is to fetch all three
source files, then go:
     dpkg-source -x wu*.dsc
     cd wu*
     dpkg-buildpackage -rfakeroot
     # this may fail immediately telling you to install some packages it needs
     cd ..
     dpkg -i wu*.deb

and if you need to fiddle configure options, you probably want to edit
debian/rules before lauching dpkg-buildpackage

If you're not building on a debian filesystem, I'd get the original
source from the original authors, wherever that is.

Good luck - if the problem persists, please ask again.

     M

It seems that the kernel sources on the TS website do not contain the
IDE drivers. I am not a kernel developer, so is there anybody who can
tell me what needs to be done in order to include the CF drivers and
IDE support in TS kernel sources?

Thank you,
R

--- In ts-7000@yahoogroups.com, "Gonzalo A. de la
Vega" <gadelavega@...> wrote:

> > I really like only having /dev/{console,null} on "disk", then
mounting a
> > ramfs over /dev and letting mdev (busybox's udev equiv) populate
it for
> > me.
>
> Oooohhh... so that's the trick, mounting a ramfs... so that's why
udev
> didn't work.
>

No, that's not why udev did not work. It just means it will run in
RAM and not on flash.

--- In ts-7000@yahoogroups.com, "Razvan Ionut
Stoian" <razvan_ionut_stoian@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, "Razvan Ionut
> > Stoian" <razvan_ionut_stoian@> wrote:
> > >
> > > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > > >
> > > >
> > > >
> > > > > TS manual states:
> > > > >
> > > > >
> > > > > The root file system used by the Linux OS can be any of the
> > following:
> > > > > ✔ JFFS/YAFFS file system image in the on-board Flash
> > (if using
> > > > > RedBoot, it should
> > > > > include the option root=/dev/mtdblock1 to instruct the
kernel
> > to boot
> > > > > here)
> > > > > ✔ EXT2 file system image in the Compact Flash card
(if
> > using
> > > > > RedBoot, it should include
> > > > > the option root=/dev/hda)
> > > >                      ------^
> > > >
> > > > There would normally be a number there, e.g. /dev/hda1
> > > >
> > > > > ✔ NFS root, via Ethernet port (if using RedBoot, it
> > should
> > > include
> > > > > the option root=/dev/nfs
> > > > > nfsroot=<IP>:<DIRECTORY> ip=dhcp)
> > > > >
> > > > > HTH
> > > > >
> > > > >
> > > > >
> > > >
> > >
> > > I tried both options root=/dev/hda and root=/dev/hda1 and I
still
> > get
> > > kernel panic.
> > >
> > > R
> > >
> >
> > Well you said you did not have nfs server but that really would
help
> > you narrow things down. If you can boot to the root fs via nfs
you
> > know it's consistant and that your problem lies with you CF
driver or
> > fs.
> >
> > If you dont have nfs I guess that means you are working from
windows.
> > Probably not the best dev platform for a linux target.
> >
> > I suggest you get a liveCD of some sort , linux from scratch or
> > partedmagic would do. I'm fairly sure both of those will let you
boot
> > directly to a system having nfs. Mount your rootfs where ever it
sits
> > and export it as nfs. Just a couple of commands really.
> >
> > It seems you currently have too many unknowns.
> >
>
> No, I am not working from Windoze ... Actually I haven't seen a
> winbloze machine in a looooong time.
> ... and if there is a problem with the CF card driver how could it
be
> solved? I used the ts2.6 kernel sources for compiling the kernel.
>
> R
>

So if you are on linux why don't you have a nfs server? Set one up.
Then you can test your kernel and root-fs. If it works you know to
look to the CF driver having eliminated other problems.

At the moment it seems you have an untested kernel, root-fs and have
not verified the flash image. All you know is that it does not boot.

You need to break the problem down. Just posting "what am I missing"
is not likely to get it fixed.

--- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@...> wrote:
>
> --- In ts-7000@yahoogroups.com, "Razvan Ionut
> Stoian" <razvan_ionut_stoian@> wrote:
> >
> > --- In ts-7000@yahoogroups.com, "j.chitte" <j.chitte@> wrote:
> > >
> > > --- In ts-7000@yahoogroups.com, "Razvan Ionut
> > > Stoian" <razvan_ionut_stoian@> wrote:
> > > >
> > > > --- In ts-7000@yahoogroups.com, Jim Jackson <jj@> wrote:
> > > > >
> > > > >
> > > > >
> > > > > > TS manual states:
> > > > > >
> > > > > >
> > > > > > The root file system used by the Linux OS can be any of the
> > > following:
> > > > > > ✔ JFFS/YAFFS file system image in the on-board Flash
> > > (if using
> > > > > > RedBoot, it should
> > > > > > include the option root=/dev/mtdblock1 to instruct the
> kernel
> > > to boot
> > > > > > here)
> > > > > > ✔ EXT2 file system image in the Compact Flash card
> (if
> > > using
> > > > > > RedBoot, it should include
> > > > > > the option root=/dev/hda)
> > > > >                      ------^
> > > > >
> > > > > There would normally be a number there, e.g. /dev/hda1
> > > > >
> > > > > > ✔ NFS root, via Ethernet port (if using RedBoot, it
> > > should
> > > > include
> > > > > > the option root=/dev/nfs
> > > > > > nfsroot=<IP>:<DIRECTORY> ip=dhcp)
> > > > > >
> > > > > > HTH
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > > >
> > > > I tried both options root=/dev/hda and root=/dev/hda1 and I
> still
> > > get
> > > > kernel panic.
> > > >
> > > > R
> > > >
> > >
> > > Well you said you did not have nfs server but that really would
> help
> > > you narrow things down. If you can boot to the root fs via nfs
> you
> > > know it's consistant and that your problem lies with you CF
> driver or
> > > fs.
> > >
> > > If you dont have nfs I guess that means you are working from
> windows.
> > > Probably not the best dev platform for a linux target.
> > >
> > > I suggest you get a liveCD of some sort , linux from scratch or
> > > partedmagic would do. I'm fairly sure both of those will let you
> boot
> > > directly to a system having nfs. Mount your rootfs where ever it
> sits
> > > and export it as nfs. Just a couple of commands really.
> > >
> > > It seems you currently have too many unknowns.
> > >
> >
> > No, I am not working from Windoze ... Actually I haven't seen a
> > winbloze machine in a looooong time.
> > ... and if there is a problem with the CF card driver how could it
> be
> > solved? I used the ts2.6 kernel sources for compiling the kernel.
> >
> > R
> >
>
> So if you are on linux why don't you have a nfs server? Set one up.
> Then you can test your kernel and root-fs. If it works you know to
> look to the CF driver having eliminated other problems.
>
> At the moment it seems you have an untested kernel, root-fs and have
> not verified the flash image. All you know is that it does not boot.
>
> You need to break the problem down. Just posting "what am I missing"
> is not likely to get it fixed.
>

I got it fixed. The problem was the lack of IDE support in ts'
sources. I compiled a new kernel (http://jandmworks.com/) and
everything runs just fine now.

On Sat, May 31, 2008 at 3:58 PM, j.chitte <j.chitte@...> wrote:
> --- In ts-7000@yahoogroups.com, "Gonzalo A. de la
>
> Vega" <gadelavega@...> wrote:
>
>> > I really like only having /dev/{console,null} on "disk", then
> mounting a
>> > ramfs over /dev and letting mdev (busybox's udev equiv) populate
> it for
>> > me.
>>
>> Oooohhh... so that's the trick, mounting a ramfs... so that's why
> udev
>> didn't work.
>>
>
> No, that's not why udev did not work. It just means it will run in
> RAM and not on flash.

I mean, I couldn't make it work... because I didn't mounted a ramfs.
I'll rephrase then:
...so that's why I couldn't get udev to work.

On Sat, 31 May 2008, Gonzalo A. de la Vega wrote:

> On Sat, May 31, 2008 at 3:58 PM, j.chitte <j.chitte@...> wrote:
>> --- In ts-7000@yahoogroups.com, "Gonzalo A. de la
>>
>> Vega" <gadelavega@...> wrote:
>>
>>>> I really like only having /dev/{console,null} on "disk", then
>> mounting a
>>>> ramfs over /dev and letting mdev (busybox's udev equiv) populate
>> it for
>>>> me.
>>>
>>> Oooohhh... so that's the trick, mounting a ramfs... so that's why
>> udev
>>> didn't work.
>>>
>>
>> No, that's not why udev did not work. It just means it will run in
>> RAM and not on flash.
>
> I mean, I couldn't make it work... because I didn't mounted a ramfs.
> I'll rephrase then:
> ...so that's why I couldn't get udev to work.

Not mounting a ramfs just means that udev will create and remove its nodes
from your flash filesystem instead of ram.. It should make absolutely no
difference to whether or not it works, as j.chitte said...