From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Nicola Bergamin [BW]" <nicola.bergamin@bluewind.it>
To: ecos-discuss@sources.redhat.com
Subject: [ECOS] Insight + eCos +Jeeni
Date: Thu, 23 Aug 2001 07:53:00 -0000
Message-id: <3B8534CD.10282.1932FCC@localhost>
X-SW-Source: 2001-08/msg00762.html

I am using Jeeni  with Cirrus Logic 
EP7209(ARM720), plus Gnu GDB toolchain.


I don't know how to force GDB to load the code 
starting at phisical address 0 !!!
Example:
I try to use Insight to debug the hello.exe eCos 
example.
We have phisical RAM at zero(ncs0) (this 
replaces program ROM, for debugging pourposes).

I cannot understand where phisically GDB tries 
do load the code.
Ok, the debug informations say that the LMA 
start address is 0xe000,0000, but it is obvious 
that the real, phisical load address should be  
the neat 0x0 (Reset vector).

The disassembled code is here:
_____________________________________

hello.exe:     file format elf32-littlearm

Disassembly of section .rom_vectors:

e0000000 <__exception_handlers>:
e0000000:	ea00000e 	b	e0000040 <_stext>
e0000004:	e59ff018 	ldr	pc, [pc, #18]	; 
e0000024 <.undefined_instruction>
e0000008:	e59ff018 	ldr	pc, [pc, #18]	; 
e0000028 <.software_interrupt>
e000000c:	e59ff018 	ldr	pc, [pc, #18]	; 
e000002c <.abort_prefetch>
e0000010:	e59ff018 	ldr	pc, [pc, #18]	; 
e0000030 <.abort_data>
e0000014:	00000000 	andeq	r0, r0, r0
e0000018:	e59ff018 	ldr	pc, [pc, #18]	; 
e0000038 <.IRQ>
e000001c:	e59ff018 	ldr	pc, [pc, #18]	; 
e000003c <.FIQ>

e0000020 <vectors>:
e0000020:	00000060 	andeq	r0, r0, r0, rrx

....
....
....
________________________________________



The software SHOULD be loaded at 0x0 (we have 
phisically mapped RAM at zero, for debug), and, 
after few instructions, should be remapped to 
0xe000,0000 (MMU remapping, as usual in eCos 
environment)

But...
after load, GDB does not show me the 0x00 
address (RESET VECTOR), instead, it shows me 
already the PC= 0xe00000040 (that is 
LMA Logical Memory address, "as if" remapping 
had already occurred ).

ADW, in comparison, shows correctly the 
addresses at 0x0, and, after some STEPs , shows 
addresses at 0xe0000040. This should 
be the correct  behaviour. 


Thanks for patience


------------------------------------------
 Nicola Bergamin
 BlueWind Embedded Systems Design
 Via Steffani, 7/B
 I-31033 Castelfranco Veneto (TV)

 Office: +39 0 423 723431
 Fax   : +39 0 423 744738
 GSM   : +39 335 7556736
 mailto:nicola.bergamin@bluewind.it
 http://www.bluewind.it
------------------------------------------