Re: Gdb with MPC85xx

Re: Gdb with MPC85xx

[Kumar Gala]

On Fri, 3 Feb 2006, Fahd Abidi wrote:

> Hello,
> 
> I am wondering if someone here is using GDB to debug code on an 85xx
> platform with the BDI. I have been running into several issues with GDB
> 5.x and 6.x where the E500 core does not seem to either step properly to
> the next line of code in the kernel or uboot or gets lost altogether.
> Each version of GDB seems to have its own set of bugs. If someone has
> been using a version of GDB that works pretty well then let me know what
> version it is and I can try and download it and build it and see if it
> behaves better.

Are you able to get the registers in GDB properly?  The e500 introduces a 
different GDB register protocol than standard PPC.  This either requires 
the latest GDB and possibly BDI firmware.  I forget how BDI handles the 
different protocol layouts.  I'm pretty sure they added support for "full 
e500 GDB" recently.

You also, have to build u-boot and your kernel to handle the quicky debug 
support on 85xx.

- kumar

Gdb with MPC85xx

[Fahd Abidi]

Hello,

I am wondering if someone here is using GDB to debug code on an 85xx
platform with the BDI. I have been running into several issues with GDB
5.x and 6.x where the E500 core does not seem to either step properly to
the next line of code in the kernel or uboot or gets lost altogether.
Each version of GDB seems to have its own set of bugs. If someone has
been using a version of GDB that works pretty well then let me know what
version it is and I can try and download it and build it and see if it
behaves better.

This is probably more of a tools question then a Linux question, I'm
hoping I can get an answer anyway.

Thanks,

Fahd
=20

RE: Gdb with MPC85xx

[atul.sabharwal]

To the best of my knowledge, I had could single step but not run to a
breakpoint in the initial u-boot code on PPC8540( an E500 core based)
system
Esp. in the code in start.S.  I had a problem where the flash was busted
and
The start.S was running from a DDR2 memory at 600+MHz rate.  So, I had
to lock the TLB and define LAW for the DDR memory in BDI so that we
could run entire u-boot, kernel, ramdisk from RAM. =20

It seemed like a BDI bug to me unless the processor is designed to
always boot from a slow deviceflash or eeprom) till the initialization
sequence is=20
Completed.  Did not try extensive debug with it, so not sure. If it
helps.

--
Atul
-----Original Message-----
From: linuxppc-embedded-bounces@ozlabs.org
[mailto:linuxppc-embedded-bounces@ozlabs.org] On Behalf Of Fahd Abidi
Sent: Friday, February 03, 2006 10:41 AM
To: linuxppc-embedded@ozlabs.org
Subject: Gdb with MPC85xx

Hello,

I am wondering if someone here is using GDB to debug code on an 85xx
platform with the BDI. I have been running into several issues with GDB
5.x and 6.x where the E500 core does not seem to either step properly to
the next line of code in the kernel or uboot or gets lost altogether.
Each version of GDB seems to have its own set of bugs. If someone has
been using a version of GDB that works pretty well then let me know what
version it is and I can try and download it and build it and see if it
behaves better.

This is probably more of a tools question then a Linux question, I'm
hoping I can get an answer anyway.

Thanks,

Fahd
=20
_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded

Re: Gdb with MPC85xx

[Dan Malek]

On Feb 3, 2006, at 3:46 PM, <atul.sabharwal@exgate.tek.com> wrote:

> To the best of my knowledge, I had could single step but not run to a
> breakpoint in the initial u-boot code on PPC8540( an E500 core based)

This is a problem with many processors and has been discussed
at length too often.  The quick response is that various initialization
code in both boot roms and Linux will disable/reset or otherwise
mess up debug registers the BDI has configured for it's use.  So,
it isn't the fault of the BDI2000, you may have to temporarily disable
or modify the code under test in such a way to not do this.  There are
several tricky places to debug code, that are completely inaccessible
without the use of the BDI2000, but that also require some common
sense and minor environment changes to properly debug.  The e500
core (and Book-E in general) significantly complicated the ability
to debug in some modes of development, so you are most
likely to see problems with these processors than others.

Thanks.

	-- Dan

RE: Gdb with MPC85xx

[atul.sabharwal]

Thanks Dan. It helps clarify some debug problems.  I was having similar
problem with BDI2000 to program the AM29F flash part on a 885 reference
board.  Finally gave up.  It's probably not supported or the config file
was
wrong.  My actual board uses different flash, so BDI & the board work
well.

There is a bug in ddd (at least debugging u-boot) when code relocates
from flash to RAM (in_ram routine).  The command add-symbol-file u-boot
<text address> to reload symbols did not work due to some ddd bug (it
could not find bounds of current function).  Switched to manual decode
using BDI
peek/poke and objdump.

--
Atul