Re: Queston about Walnut

Re: Queston about Walnut

[Wolfgang Denk]

In message <000e01c388dc$eb525510$0202a8c0@homevl9biy3v7e> you wrote:
>
> > Who gave you such (obviously wrong) instructions?
>
>         I take the example of u-boot to create the "uImage" and download it
> to the board.

Please show me which U-Boot document uses 0x00400000 as load or entry
point address so I can fix this immediately. It is WRONG  and  CANNOT
work.

> > For PowerPC systems, both  Load  Address  and  Entry  Point  must  be
> > specified as 0.
>
>         I have tried 0 before. But I get an exception in return. Could you
> kindly give me some instructions how to run a kernel on this board?

Please start by giving a precise description of  what  you  did.  and
which error messages you got.

You "get an exception" - when? When running "mkimage"?  When  booting
the kernel? Which sort of exception? What is the exact error message?
Please include a complete log of your actions and all output.

Best regards,

Wolfgang Denk

--
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd@denx.de
Totally illogical, there was no chance.
	-- Spock, "The Galileo Seven", stardate 2822.3

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Jacky Lam]

> > > Who gave you such (obviously wrong) instructions?
> >
> >         I take the example of u-boot to create the "uImage" and download
it
> > to the board.
>
> Please show me which U-Boot document uses 0x00400000 as load or entry
> point address so I can fix this immediately. It is WRONG  and  CANNOT
> work.

    I take the "step" from u-boot's README file only. Because I got an
exception, I try to load it to other address. I use 0x00400000 because
zImage seems to be loaded at there....anyway, just a guess.

>
> Please start by giving a precise description of  what  you  did.  and
> which error messages you got.

    Here is the exact step that I have done so far:

    <Compile toolcahin by crosstools>
    <Patch 2.4.22 kernel with patch on
http://www.kernel.org/pub/linux/kernel/ports/ppc/2.4/linuxppc-2.4.22.patch.bz2>
    <Compile kernel with cross toolchain>
    powerpc-405-linux-gnu-objcopy -O binary -R .note -R .comment -S vmlinux
linux.bin
    gzip -9  linux.bin
    ./mkimage -A ppc -O linux -T kernel -C gzip -a 0x00000000 -e
0x00000000 -n "test" -d linux.bin.gz uImage
    <run ftpd on host>
    <boot up the board and enter command prompt of u-boot>

    Here is the u-boot log:

    U-Boot 0.4.0 (Sep 29 2003 - 11:44:31)

CPU:   IBM PowerPC 405GP Rev. D at 200 MHz (PLB=100, OPB=50, EBC=33 MHz)
           PCI sync clock at 33 MHz, external PCI arbiter enabled
           16 kB I-Cache 8 kB D-Cache
Board: ### No HW ID - assuming WALNUT405
I2C:   ready
DRAM:  64 MB
FLASH: 512 kB
In:    serial
Out:   serial
Err:   serial
KGDB:  kgdb ready
ready
BEDBUG:ready
Hit any key to stop autoboot:  0
=> tftpboot
ENET Speed is 100 Mbps - FULL duplex connection
TFTP from server 10.5.6.76; our IP address is 10.5.3.45
Filename 'uImage'.
Load address: 0x400000
Loading: #################################################################
         #######################################################
done
Bytes transferred = 612748 (9598c hex)
=> bootm
## Booting image at 00400000 ...
   Image Name:   test
   Created:      2003-10-03   1:08:06 UTC
   Image Type:   PowerPC Linux Kernel Image (gzip compressed)
   Data Size:    612684 Bytes = 598.3 kB
   Load Address: 00000000
   Entry Point:  00000000
   Verifying Checksum ... OK
   Uncompressing Kernel Image ... OK
id mach(): done
MMU:enter
MMU:hw init
MMU:mapin

<hang>


>
> You "get an exception" - when? When running "mkimage"?  When  booting
> the kernel? Which sort of exception? What is the exact error message?
> Please include a complete log of your actions and all output.

    Maybe I do something wrong before. Now, it just hangs. Thanks your
suggestion about the load address.



Best Regards,
Jacky


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Jacky Lam]

> the kernel you use - vanilla+ppc-patch - is based on linuxppc-2.4, which
> is not appropriate for your system. [At least it was like that some time
> ago. Tom, could you please comment.] Try using the linuxppc_2_4_devel
> tree from Bitkeeper or from our FTP server (see http://www.denx.de).

    I just tried. The result is the same. Using the original walnut_config,
it hangs in mapin_ram().
    If I use minimal config, it hangs in __sti().

    Does anyone still working on Walnut platform and successfully boot up
2.4 kernel on that? If no, could someone remember what is the last known
working toolchain and kernel on Walnut?

    I use toolchain gcc 3.2.3, glibc 2.2.5 because I have some trouble to
compile to newest one (it requests my host gcc >=3.1). Is this related? If
yes, I would tried to reinstall my RedHat to newer version, but it would be
painful.....

    Thanks.

Best regards,
Jacky


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Matt Porter]

On Fri, Oct 03, 2003 at 05:40:46PM +0800, Jacky Lam wrote:
>
> > the kernel you use - vanilla+ppc-patch - is based on linuxppc-2.4, which
> > is not appropriate for your system. [At least it was like that some time
> > ago. Tom, could you please comment.] Try using the linuxppc_2_4_devel
> > tree from Bitkeeper or from our FTP server (see http://www.denx.de).
>
>     I just tried. The result is the same. Using the original walnut_config,
> it hangs in mapin_ram().
>     If I use minimal config, it hangs in __sti().
>
>     Does anyone still working on Walnut platform and successfully boot up
> 2.4 kernel on that? If no, could someone remember what is the last known
> working toolchain and kernel on Walnut?

I noticed this about a week ago.  It is some kind of merge problem
between linux-2.4 and linuxppc-2.4.  linux-2.4 boot fine on walnut
for me.  I didn't have time to investigate but might take a look at
it again soon.

-Matt

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Jacky Lam]

----- Original Message -----
From: "Matt Porter" <mporter@kernel.crashing.org>
To: "Jacky Lam" <jackylam@astri.org>
Cc: <linuxppc-embedded@lists.linuxppc.org>
Sent: Friday, October 03, 2003 11:07 PM
Subject: Re: Queston about Walnut


>
> On Fri, Oct 03, 2003 at 05:40:46PM +0800, Jacky Lam wrote:
> >
> > > the kernel you use - vanilla+ppc-patch - is based on linuxppc-2.4,
which
> > > is not appropriate for your system. [At least it was like that some
time
> > > ago. Tom, could you please comment.] Try using the linuxppc_2_4_devel
> > > tree from Bitkeeper or from our FTP server (see http://www.denx.de).
> >
> >     I just tried. The result is the same. Using the original
walnut_config,
> > it hangs in mapin_ram().
> >     If I use minimal config, it hangs in __sti().
> >
> >     Does anyone still working on Walnut platform and successfully boot
up
> > 2.4 kernel on that? If no, could someone remember what is the last known
> > working toolchain and kernel on Walnut?
>
> I noticed this about a week ago.  It is some kind of merge problem
> between linux-2.4 and linuxppc-2.4.  linux-2.4 boot fine on walnut
> for me.  I didn't have time to investigate but might take a look at
> it again soon.

    Really? That's so great! Do you have any idea about this? Maybe I can
try to investigate as well. Also, do you remember what is the version just
before this merge problem occurs? Thanks.

Best Regards,
Jacky


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Matt Porter]

On Sat, Oct 04, 2003 at 12:34:37AM +0800, Jacky Lam wrote:
> > On Fri, Oct 03, 2003 at 05:40:46PM +0800, Jacky Lam wrote:
> > >
> > > > the kernel you use - vanilla+ppc-patch - is based on linuxppc-2.4,
> which
> > > > is not appropriate for your system. [At least it was like that some
> time
> > > > ago. Tom, could you please comment.] Try using the linuxppc_2_4_devel
> > > > tree from Bitkeeper or from our FTP server (see http://www.denx.de).
> > >
> > >     I just tried. The result is the same. Using the original
> walnut_config,
> > > it hangs in mapin_ram().
> > >     If I use minimal config, it hangs in __sti().
> > >
> > >     Does anyone still working on Walnut platform and successfully boot
> up
> > > 2.4 kernel on that? If no, could someone remember what is the last known
> > > working toolchain and kernel on Walnut?
> >
> > I noticed this about a week ago.  It is some kind of merge problem
> > between linux-2.4 and linuxppc-2.4.  linux-2.4 boot fine on walnut
> > for me.  I didn't have time to investigate but might take a look at
> > it again soon.
>
>     Really? That's so great! Do you have any idea about this? Maybe I can
> try to investigate as well. Also, do you remember what is the version just
> before this merge problem occurs? Thanks.

Well, maybe I should have cleaned my tree from a 440 build, I had
an old CONFIG_440 hanging around causing the kernel to not be
happy.  At any rate, I verified that current linuxppc-2.4 boots
on walnut from stock openbios.

-Matt

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Dan Kegel]

Jacky Lam wrote:
>     Does anyone still working on Walnut platform and successfully boot up
> 2.4 kernel on that? If no, could someone remember what is the last known
> working toolchain and kernel on Walnut?

We used 2.4.17 from linuxppc_2_4_devel.

 >     I use toolchain gcc 3.2.3, glibc 2.2.5 because I have some trouble to
 > compile to newest one (it requests my host gcc >=3.1). Is this related? If
 > yes, I would tried to reinstall my RedHat to newer version, but it would be
 > painful.....

Why painful?  It's as easy as this:

----------
wget ftp://ftp.gnu.org/gnu/gcc/gcc-3.3.1.tar.gz
tar -xzvf gcc-3.3.1.tar.gz
mkdir build-gcc
cd build-gcc
../gcc-3.3.1/configure --prefix=/usr/local/gcc-3.3.1 --enable-languages=c,c++
make
sudo make install
----------

Then put /usr/local/gcc-3.3.1/bin in your PATH before anything else, and you're done.

BTW here's the (awful) script I used to grab the kernel.  (It reveals my
total lack of knowledge of bitkeeper; I think 'bk export' is preferred
to my clumsy removal of CVS directories.)

#!/bin/sh

if [ -z $1 ]; then
         echo "Usage: `basename $0` TAG"
         echo "Example: `basename $0` v2.4.18"
         echo "Purpose: fetch a kernel source tarball from the linuxppc_2_4_devel tree"
         exit 1
fi
set -x -e

rm -rf linux linuxppc_2_4_devel-$1 || /bin/true

# Download bk from http://www.bitkeeper.com
bk clone -r$1 bk://ppc.bkserver.net/linuxppc_2_4_devel linuxppc_2_4_devel-$1
cd linuxppc_2_4_devel-$1
find . -type d -exec bk get {} \;

# Hey!  nonxargs version fails with 'find: no such file foo/SCCS' after deleting foo/SCCS!
# Bug in gnu find 4.1.7?  Or in kernel 2.4.18?  Who knows...
#find . -name SCCS -exec rm {} \;
find . -name SCCS -print | xargs rm -rf

rm -rf RESYNC/
rm -rf PENDING/
rm -rf BitKeeper/
rm -f ChangeSet
cd ..
mv linuxppc_2_4_devel-$1 linux



--
Dan Kegel
http://www.kegel.com
http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Jacky Lam]

----- Original Message -----
From: "Dan Kegel" <dank@kegel.com>
To: "Jacky Lam" <jackylam@astri.org>
Cc: <linuxppc-embedded@lists.linuxppc.org>
Sent: Saturday, October 04, 2003 12:10 AM
Subject: Re: Queston about Walnut



>  >     I use toolchain gcc 3.2.3, glibc 2.2.5 because I have some trouble
to
>  > compile to newest one (it requests my host gcc >=3.1). Is this related?
If
>  > yes, I would tried to reinstall my RedHat to newer version, but it
would be
>  > painful.....
>
> Why painful?  It's as easy as this:

    No. The story is like this....

    I upgrade my gcc to 3.3.1 on my host and start to compile my cross
toolchain. However, while it is compiling gcc, it has strange error that I
have never seen (seems gcc has made something that ld can't understand). So,
I guess gcc 3.3.1 may not work very well with my old glibc (provided in
RedHat 7.2). I don't know how to upgrade glibc safely and make it compatible
to existing applications. That's why I would like to upgrade the whole
Redhat.....and this is painful...

>
> ----------
> wget ftp://ftp.gnu.org/gnu/gcc/gcc-3.3.1.tar.gz
> tar -xzvf gcc-3.3.1.tar.gz
> mkdir build-gcc
> cd build-gcc
>
../gcc-3.3.1/configure --prefix=/usr/local/gcc-3.3.1 --enable-languages=c,c+
+
> make
> sudo make install
> ----------
>
> Then put /usr/local/gcc-3.3.1/bin in your PATH before anything else, and
you're done.
>
> BTW here's the (awful) script I used to grab the kernel.  (It reveals my
> total lack of knowledge of bitkeeper; I think 'bk export' is preferred
> to my clumsy removal of CVS directories.)
>
> #!/bin/sh
>
> if [ -z $1 ]; then
>          echo "Usage: `basename $0` TAG"
>          echo "Example: `basename $0` v2.4.18"
>          echo "Purpose: fetch a kernel source tarball from the
linuxppc_2_4_devel tree"
>          exit 1
> fi
> set -x -e
>
> rm -rf linux linuxppc_2_4_devel-$1 || /bin/true
>
> # Download bk from http://www.bitkeeper.com
> bk clone -r$1 bk://ppc.bkserver.net/linuxppc_2_4_devel
linuxppc_2_4_devel-$1
> cd linuxppc_2_4_devel-$1
> find . -type d -exec bk get {} \;
>
> # Hey!  nonxargs version fails with 'find: no such file foo/SCCS' after
deleting foo/SCCS!
> # Bug in gnu find 4.1.7?  Or in kernel 2.4.18?  Who knows...
> #find . -name SCCS -exec rm {} \;
> find . -name SCCS -print | xargs rm -rf
>
> rm -rf RESYNC/
> rm -rf PENDING/
> rm -rf BitKeeper/
> rm -f ChangeSet
> cd ..
> mv linuxppc_2_4_devel-$1 linux
>
>
>
> --
> Dan Kegel
> http://www.kegel.com
> http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045
>
>
>
        I use CVS to grab the linuxppc-2.4-devel module. Is it the same as
that? Thanks.

Best regards,
Jacky


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Dan Kegel]

Jacky Lam wrote:
>     I upgrade my gcc to 3.3.1 on my host and start to compile my cross
> toolchain. However, while it is compiling gcc, it has strange error that I
> have never seen (seems gcc has made something that ld can't understand). So,
> I guess gcc 3.3.1 may not work very well with my old glibc (provided in
> RedHat 7.2). I don't know how to upgrade glibc safely and make it compatible
> to existing applications. That's why I would like to upgrade the whole
> Redhat.....and this is painful...

I see.  Well, you can always build and install a newer binutils;
that's not painful, and might get you past the problem.
(If you post the error message, we might be able to help.)

But I agree that doing a fresh install of e.g. Red Hat 9 is
nicer than keeping an old system alive by little patches
(unless you're going for an uptime record).

>># Download bk from http://www.bitkeeper.com
>>bk clone -r$1 bk://ppc.bkserver.net/linuxppc_2_4_devel
>
>         I use CVS to grab the linuxppc-2.4-devel module. Is it the same as
> that? Thanks.

Probably.  I dunno.
- Dan


--
Dan Kegel
http://www.kegel.com
http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Kernel versions

[Gary Thomas]

Tom,

For more than a year now, I've been working from the linuxppc_2_4_devel
tree, sending updates which have been merged in, etc.  I just looked at
the recently discussed 2.4 tree (http://ppc.bkbits.net:8080/linuxppc-2.4)
and I see none of this work present.

So my questions are:
* What is the "proper" tree to be working from (2.4, 2.6)?
* What is the merge path from the work in 2_4_devel to these other trees?

... I wouldn't want all of my efforts to just get lost.

Thanks.

--
Gary Thomas <gary@mlbassoc.com>
MLB Associates


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Kernel versions

[Tom Rini]

On Fri, Oct 03, 2003 at 05:36:11PM -0600, Gary Thomas wrote:

> Tom,
>
> For more than a year now, I've been working from the linuxppc_2_4_devel
> tree, sending updates which have been merged in, etc.  I just looked at
> the recently discussed 2.4 tree (http://ppc.bkbits.net:8080/linuxppc-2.4)
> and I see none of this work present.
>
> So my questions are:
> * What is the "proper" tree to be working from (2.4, 2.6)?
> * What is the merge path from the work in 2_4_devel to these other trees?

For 2.6, a tree based off of Linus' BK (or CVS or SVN or tarball of
that) or current 2.6 release.  For 2.4, it depends on the type of board:
8xx: Should be mostly merged into kernel.org now, but if you need
something, pull it up and out separately.
'classic' mpc10x based: All of the library type bits of code
(mpc10x_common, gen550) are in kernel.org, so it should be trivial to
pull a platform out of _devel
gt64260 (and gt64360): At OLS Paul talked to Mark about what needs to
bee done to make him happy with the code and Mark has expressed a desire
to do some (other?) cleanups as well to the 64260 code, which should
take into account the 64360 when possible.
'other' classic: Most likely needs someone with the hardware to state
that it's still working.
8260: Needs things to be pulled up and out of 2_4_devel.

And by 'pull it up and out' I mean to make another patch / ChangeSet of
just those changes (or many, if there's many difference pieces to it)
that would go in before the patch for a given board.

Now, since 2.4 (2_4_devel, even) and 2.6 are not in sync for some code
(gt64260) and not all platforms work in 2.6 (8xx, 8260 is in almost
working shape again) it might not be an equal amount of work needed to
get something into 2.4 and 2.6, but it would be appreciated if patches
come in at roughly the same time for both (or 2.6 first, even).

The linuxppc-2.4 tree contains what's gone to Marcelo already, along
with some things that are being polished up.  At the moment, I don't
think there's anything special there, but I haven't looked.

> ... I wouldn't want all of my efforts to just get lost.

It won't be.

--
Tom Rini
http://gate.crashing.org/~trini/

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Jacky Lam]

        I trace the code in kernel and I found some strange behaviors.
        In the first case (using walnut_config provided in kernel source),
kernel hangs at mapin_ram();
        However, if I turn off all unnecssary options and leave only serial
console, my kernel can run until init/main.c: kmem_cache_init();

        As both hangs in memory related function, I tried to run a memory
test on my board in u-boot. But it hangs.

        => mtest 0 3ffffff
        Pattern 00000000  Writing...
        <hangs>

    I tried smaller range (ffffff), it still hangs.

    I am sure my board's memory is in good condition because VxWorks can run
very well on it. Have any idea? Thanks.

Best regards,
Jacky





----- Original Message -----
From: "Wolfgang Denk" <wd@denx.de>
To: "Jacky Lam" <jackylam@astri.org>
Cc: <linuxppc-embedded@lists.linuxppc.org>
Sent: Thursday, October 02, 2003 8:17 PM
Subject: Re: Queston about Walnut


>
> In message <000e01c388dc$eb525510$0202a8c0@homevl9biy3v7e> you wrote:
> >
> > > Who gave you such (obviously wrong) instructions?
> >
> >         I take the example of u-boot to create the "uImage" and download
it
> > to the board.
>
> Please show me which U-Boot document uses 0x00400000 as load or entry
> point address so I can fix this immediately. It is WRONG  and  CANNOT
> work.
>
> > > For PowerPC systems, both  Load  Address  and  Entry  Point  must  be
> > > specified as 0.
> >
> >         I have tried 0 before. But I get an exception in return. Could
you
> > kindly give me some instructions how to run a kernel on this board?
>
> Please start by giving a precise description of  what  you  did.  and
> which error messages you got.
>
> You "get an exception" - when? When running "mkimage"?  When  booting
> the kernel? Which sort of exception? What is the exact error message?
> Please include a complete log of your actions and all output.
>
> Best regards,
>
> Wolfgang Denk
>
> --
> Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
> Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd@denx.de
> Totally illogical, there was no chance.
> -- Spock, "The Galileo Seven", stardate 2822.3
>


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Jacky Lam]

----- Original Message -----
From: "Jacky Lam" <jackylam@astri.org>
To: "Wolfgang Denk" <wd@denx.de>
Cc: <linuxppc-embedded@lists.linuxppc.org>
Sent: Friday, October 03, 2003 12:11 PM
Subject: Re: Queston about Walnut


>
>         I trace the code in kernel and I found some strange behaviors.
>         In the first case (using walnut_config provided in kernel source),
> kernel hangs at mapin_ram();
>         However, if I turn off all unnecssary options and leave only
serial
> console, my kernel can run until init/main.c: kmem_cache_init();

        Sorry, it should be sti() just after kmem_cache_init()  instead.
>
>         As both hangs in memory related function, I tried to run a memory
> test on my board in u-boot. But it hangs.
>
>         => mtest 0 3ffffff
>         Pattern 00000000  Writing...
>         <hangs>
>
>     I tried smaller range (ffffff), it still hangs.
>
>     I am sure my board's memory is in good condition because VxWorks can
run
> very well on it. Have any idea? Thanks.
>
> Best regards,
> Jacky
>
>
>
>
>
> ----- Original Message -----
> From: "Wolfgang Denk" <wd@denx.de>
> To: "Jacky Lam" <jackylam@astri.org>
> Cc: <linuxppc-embedded@lists.linuxppc.org>
> Sent: Thursday, October 02, 2003 8:17 PM
> Subject: Re: Queston about Walnut
>
>
> >
> > In message <000e01c388dc$eb525510$0202a8c0@homevl9biy3v7e> you wrote:
> > >
> > > > Who gave you such (obviously wrong) instructions?
> > >
> > >         I take the example of u-boot to create the "uImage" and
download
> it
> > > to the board.
> >
> > Please show me which U-Boot document uses 0x00400000 as load or entry
> > point address so I can fix this immediately. It is WRONG  and  CANNOT
> > work.
> >
> > > > For PowerPC systems, both  Load  Address  and  Entry  Point  must
be
> > > > specified as 0.
> > >
> > >         I have tried 0 before. But I get an exception in return. Could
> you
> > > kindly give me some instructions how to run a kernel on this board?
> >
> > Please start by giving a precise description of  what  you  did.  and
> > which error messages you got.
> >
> > You "get an exception" - when? When running "mkimage"?  When  booting
> > the kernel? Which sort of exception? What is the exact error message?
> > Please include a complete log of your actions and all output.
> >
> > Best regards,
> >
> > Wolfgang Denk
> >
> > --
> > Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
> > Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd@denx.de
> > Totally illogical, there was no chance.
> > -- Spock, "The Galileo Seven", stardate 2822.3
> >
>
>


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

RE: Queston about Walnut

[ARIBAUD Albert]

>
> > I would suggest inspecting the timings for memory access.
> > If this is SDRAM, you must have had to program the UPM for
> > it, I suppose. Might be that VxWorks has it right, and you
> > have set it up differently.
> >
> > Also, writing to address 0 might be a Bad Ideam (tm), if you
> > have set this area to hold interrupt vectors.
> >
> > Did you try single write (say, "mw 100000 0") ?
>
>     Thanks for you suggestion. Not only single write, I can
> now write to anywhere except the first 1MB of SDRAM. I think
> it is occupied by u-boot?

Possibly, but you should check that for yourself. There is
documentation on what u-boot does to RAM upon power-on.

> So, I think memory is ok now.

Seems right.

>     Do you have any idea it hangs at __sti()?

As I said, I know next to nothing about the Walnut. My own
embedded uboot/Linux experience is as follows:

I had a 855T-based custom board, so I took u-boot, duplicated
the TQM855L-specific files into a custom directory *and checked
every single setting* against my own hardware. Once u-boot
started OK on my board, Linux followed almost without trouble.

Only problems were with a weird bootup IMMR address setting
(settled by modifying u-boot's startup code) and FEC MII
interrupt mishandling (settled by careful option setting in
the kernel).

You should ensure that you have access to both software (u-boot
and Linux) *and* hardware (both PPC and your board) resources.

HTH,

Albert.

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

RE: Queston about Walnut

[ARIBAUD Albert]

(disclaimer : I work on 8xx, not 4xx)

I would suggest inspecting the timings for memory access.
If this is SDRAM, you must have had to program the UPM for
it, I suppose. Might be that VxWorks has it right, and you
have set it up differently.

Also, writing to address 0 might be a Bad Ideam (tm), if you
have set this area to hold interrupt vectors.

Did you try single write (say, "mw 100000 0") ?

Albert.

P.S. I think that if you write directly to linuxppc-embedded,
Wolfgang will cease receiving you mails twice (which he probably
does now since you e-mail him *and* CC: to the list).

> -----Message d'origine-----
> De : owner-linuxppc-embedded@lists.linuxppc.org
> [mailto:owner-linuxppc-embedded@lists.linuxppc.org]De la part de Jacky
> Lam
> Envoyé : vendredi 3 octobre 2003 06:11
> À : Wolfgang Denk
> Cc : linuxppc-embedded@lists.linuxppc.org
> Objet : Re: Queston about Walnut

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Jacky Lam]

----- Original Message -----
From: "ARIBAUD Albert" <a.aribaud@giat-industries.fr>
To: <linuxppc-embedded@lists.linuxppc.org>
Sent: Friday, October 03, 2003 3:57 PM
Subject: RE: Queston about Walnut


>
> (disclaimer : I work on 8xx, not 4xx)
>
> I would suggest inspecting the timings for memory access.
> If this is SDRAM, you must have had to program the UPM for
> it, I suppose. Might be that VxWorks has it right, and you
> have set it up differently.
>
> Also, writing to address 0 might be a Bad Ideam (tm), if you
> have set this area to hold interrupt vectors.
>
> Did you try single write (say, "mw 100000 0") ?

    Thanks for you suggestion. Not only single write, I can now write to
anywhere except the first 1MB of SDRAM. I think it is occupied by u-boot?
So, I think memory is ok now.

    Do you have any idea it hangs at __sti()?

        mfmsr   r3              /* Get current state */
        ori     r3,r3,MSR_EE    /* Turn on 'EE' bit */
        SYNC                   /* Some chip revs have problems here... */
        mtmsr   r3        <------Hangs after executing it      /* Update
machine state */
        blr

    Is there any relationship with the previous instruction "SYNC"? What
revision will have problem there?

    Thanks.


>
> Albert.
>
> P.S. I think that if you write directly to linuxppc-embedded,
> Wolfgang will cease receiving you mails twice (which he probably
> does now since you e-mail him *and* CC: to the list).

    Thanks for remind me. I will be careful next time.

Best regards,
Jacky


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Queston about Walnut

[Jacky Lam]

Dear all,

    I get a IBM 405 GP evaluation board recently. After I compile the
toolchain and kernel 2.4.22 using the crosstools. I try to boot the kernel
on the board by u-boot. I follow the instruction of uboot to create a uImage
as follows:

    powerpc-405-linux-gnu-objcopy -O binary -R .note -R .comment -S vmlinux
linux.bin
    gzip -9  linux.bin
    ./mkimage -A ppc -O linux -T kernel -C gzip -a 0x00400000 -e
0x00400000 -n "test" -d linux.bin.gz uImage

    and load the image by tftpboot. But after I type bootm, only the
following is displayed and the whole system hangs:

    ## Booting image at 00400000 ...
   Image Name:   test
   Created:      2003-10-02   8:51:50 UTC
   Image Type:   PowerPC Linux Kernel Image (gzip compressed)
   Data Size:    612743 Bytes = 598.4 kB
   Load Address: 00400000
   Entry Point:  00400000
   Verifying Checksum ... OK
   Uncompressing Kernel Image ... OK
   <Hangs afterwards...>

    I haven't modified any source yet. I don't know even it is needed.
    Could anyone kindly tell me what's wrong about my steps? Thanks so much.

Regards,
Jacky


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Queston about Walnut

[Wolfgang Denk]

In message <012801c388c5$f4378540$2803050a@JackyLam> you wrote:
>
>     I get a IBM 405 GP evaluation board recently. After I compile the
> toolchain and kernel 2.4.22 using the crosstools. I try to boot the kernel
> on the board by u-boot. I follow the instruction of uboot to create a uImage
> as follows:

Who gave you such (obviously wrong) instructions?

>     ## Booting image at 00400000 ...
>    Image Name:   test
>    Created:      2003-10-02   8:51:50 UTC
>    Image Type:   PowerPC Linux Kernel Image (gzip compressed)
>    Data Size:    612743 Bytes = 598.4 kB
>    Load Address: 00400000
>    Entry Point:  00400000

For PowerPC systems, both  Load  Address  and  Entry  Point  must  be
specified as 0.

Best regards,

Wolfgang Denk

--
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd@denx.de
It may be that your whole purpose in life is simply  to  serve  as  a
warning to others.

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/