[U-Boot-Users] Re: u-boot debugging boot with gdb and bdi2000 in lubbock(PXA255) board

[U-Boot-Users] Re: u-boot debugging boot with gdb and bdi2000 in lubbock(PXA255) board

[Andrea Sorio]

Dear Wolfgang

I followed your advice, but they doesn't work for me:

[u-boot-0.4.6]$ arm-linux-gdb u-boot
GNU gdb 5.1.1
Copyright 2002 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "--host=i386-redhat-linux --target=arm-linux"...
(gdb) target remote bdi:2001
Remote debugging using bdi:2001
0x00000000 in ?? ()
(gdb) symbol-file 
Discard symbol table from `/home/sorio/Devel/UBoot/u-boot-0.4.6/u-boot'? (y or n) y
No symbol file now.
(gdb) add-symbol-file u-boot 0x0
add symbol table from file "u-boot" at
        .text_addr = 0x0
(y or n) y
Reading symbols from u-boot...done.
(gdb) b start_armboot
Breakpoint 1 at 0x80000cb4: file board.c, line 214.
(gdb) x/16i 0x0
0x0:    b       0x54
0x4:    ldr     pc, [pc, #20]   ; 0x20
0x8:    ldr     pc, [pc, #20]   ; 0x24
0xc:    ldr     pc, [pc, #20]   ; 0x28
0x10:   ldr     pc, [pc, #20]   ; 0x2c
0x14:   ldr     pc, [pc, #20]   ; 0x30
0x18:   ldr     pc, [pc, #20]   ; 0x34
0x1c:   ldr     pc, [pc, #20]   ; 0x38
0x20:   andmi   r0, r0, r0, lsl #2
0x24:   andmi   r0, r0, r0, ror #2
0x28:   andmi   r0, r0, r0, asr #3
0x2c:   andmi   r0, r0, r0, lsr #4
0x30:   andmi   r0, r0, r0, lsl #5
0x34:   andmi   r0, r0, r0, ror #5
0x38:   andmi   r0, r0, r0, asr #6
0x3c:   cdple   14, 10, cr11, cr13, cr15, {7}
(gdb) 

The breackpoint is set to "0x80000cb4", so the bdi never stop!
My code is built to run on 0x40000000 address.
Any idea?

Thank you in advance

Best Regards

Andrea Sorio

 In message <3F575E48.1060507@na...> you wrote:
 >
 > Thanks Wolfgang and Kyle for your advice. Now, I can debug the boot 
 > process from the first instruction. Finally I've created two binaries: 
 > u-boot to execute  and u-boot.debug (with TEXTBASE=0x0), to debug the 
 > first part of boot process. When the boot process arrives to the 
 > start_armboot function, I discard the current symbol file(u-boot.debug)  
 > and load the new one(u-boot).
 
 You  don't  need  to  recompile  or  relink  with  TEXTBASE=0x0;  the
 "add-symbol-file"  GDB  command  allows  you  to specify an arbitrary
 target address.
 



-- 
Ing. Andrea Sorio
Hardware & Software Engineer
Free Consultant
http://www.cocay.it
email: a.sorio at cocay.it
Phone: 348 8703449

[U-Boot-Users] Re: u-boot debugging boot with gdb and bdi2000 in lubbock(PXA255) board

[Wolfgang Denk]

Dear Andrea,

in message <3F5D81C3.4050604@cocay.it> you wrote:
> 
> I followed your advice, but they doesn't work for me:

Well, I can only speculate...

> [u-boot-0.4.6]$ arm-linux-gdb u-boot
> GNU gdb 5.1.1

Where did you get this version of a cross debugger from?

> (gdb) symbol-file 
> Discard symbol table from `/home/sorio/Devel/UBoot/u-boot-0.4.6/u-boot'? (y or n) y
> No symbol file now.
> (gdb) add-symbol-file u-boot 0x0
> add symbol table from file "u-boot" at
>         .text_addr = 0x0
> (y or n) y
> Reading symbols from u-boot...done.
> (gdb) b start_armboot
> Breakpoint 1 at 0x80000cb4: file board.c, line 214.

It seems to me that your GDB is broken.

Some versions don't know how to deal with a 0x0 address because  they
confuse this with a NULL pointer.

> The breackpoint is set to "0x80000cb4", so the bdi never stop!

That's a bug in GDB.


Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd at denx.de
The nice thing about  standards  is that there are  so many to choose
from.                                           - Andrew S. Tanenbaum

[U-Boot-Users] Re: u-boot debugging boot with gdb and bdi2000 in lubbock(PXA255) board

[Andrea Sorio]

Dear Wolfgang

>  
>
>>[u-boot-0.4.6]$ arm-linux-gdb u-boot
>>GNU gdb 5.1.1
>>    
>>
>
>Where did you get this version of a cross debugger from?
>
I installed your ELDK! :-)

Thank you in advance

Best regards

Andrea

-- 
Ing. Andrea Sorio
Hardware & Software Engineer
Free Consultant
http://www.cocay.it
email: a.sorio at cocay.it
Phone: 348 8703449

[U-Boot-Users] Re: u-boot debugging boot with gdb and bdi2000 in lubbock(PXA255) board

[Wolfgang Denk]

Dear Andrea,

in message <3F5D8E13.6060609@cocay.it> you wrote:
> 
> >Where did you get this version of a cross debugger from?
> >
> I installed your ELDK! :-)

Yes, this version is still broken.

I think it will be fixed in the next release of the ELDK.

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd at denx.de
You're too beautiful to ignore.  Too much woman.
	-- Kirk to Yeoman Rand, "The Enemy Within", stardate unknown

[U-Boot-Users] Re: u-boot debugging boot with gdb and bdi2000 in lubbock(PXA255) board

[Detlev Zundel]

Hello Andrea,

> [u-boot-0.4.6]$ arm-linux-gdb u-boot
> GNU gdb 5.1.1
> Copyright 2002 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you are
> welcome to change it and/or distribute copies of it under certain conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for details.
> This GDB was configured as "--host=i386-redhat-linux --target=arm-linux"...
> (gdb) target remote bdi:2001
> Remote debugging using bdi:2001
> 0x00000000 in ?? ()
> (gdb) symbol-file
> Discard symbol table from `/home/sorio/Devel/UBoot/u-boot-0.4.6/u-boot'? (y or n) y
> No symbol file now.
> (gdb) add-symbol-file u-boot 0x0
> add symbol table from file "u-boot" at
>         .text_addr = 0x0
> (y or n) y
> Reading symbols from u-boot...done.
> (gdb) b start_armboot
> Breakpoint 1 at 0x80000cb4: file board.c, line 214.
> (gdb) x/16i 0x0
> 0x0:    b       0x54
> 0x4:    ldr     pc, [pc, #20]   ; 0x20
> 0x8:    ldr     pc, [pc, #20]   ; 0x24
> 0xc:    ldr     pc, [pc, #20]   ; 0x28
> 0x10:   ldr     pc, [pc, #20]   ; 0x2c
> 0x14:   ldr     pc, [pc, #20]   ; 0x30
> 0x18:   ldr     pc, [pc, #20]   ; 0x34
> 0x1c:   ldr     pc, [pc, #20]   ; 0x38
> 0x20:   andmi   r0, r0, r0, lsl #2
> 0x24:   andmi   r0, r0, r0, ror #2
> 0x28:   andmi   r0, r0, r0, asr #3
> 0x2c:   andmi   r0, r0, r0, lsr #4
> 0x30:   andmi   r0, r0, r0, lsl #5
> 0x34:   andmi   r0, r0, r0, ror #5
> 0x38:   andmi   r0, r0, r0, asr #6
> 0x3c:   cdple   14, 10, cr11, cr13, cr15, {7}
> (gdb)
> 
> The breackpoint is set to "0x80000cb4", so the bdi never stop!
> My code is built to run on 0x40000000 address.
> Any idea?

It is just something I seem to remember, but can you try to do the
symbol-file manipulations _before_ you give the "target remote"
command?  I know it sounds unlikely, but I think I stumbled across
something like this before.

Cheers
  Detlev

-- 
Greenspun's Tenth Rule of Programming: "Any sufficiently complicated C
or Fortran program contains an ad-hoc, informally-specified bug-ridden
slow implementation of half of Common Lisp."