[ECOS] MIPS32 gdb vectors question

[Tim Michals]

Ok, I have narrowed down the problem
1) Using a ROMRAM configuration for MIPS32 and had to change
to
hal_vsr_table = 0x80000300;
hal_virtual_vector_table = 0x80000400;
from:
hal_vsr_table = 0x80000200;
hal_virtual_vector_table = 0x80000300;
A slight collision with the debug vector.
2) had a memory configuration issue pushing the _hal_registers past memory.
(Fixed(
3) GDB issue using 3.2
Ignoring packet error, continuing...
Remote packet too long:
0000000000000000ffffffffb80600000000000000000001ffffffff
8003e2b0ffffffff8000d594ffffffff83fffff0ffffffff800452200000000000000001ffff
ffff
fff0e51200000000000f4240ffffffffb802801800000000000027100000000000000064ffff
ffff
80040000000000000800000000000000000000aa0000000000000002ffffffff800004000000
0000
00000000ffffffff80040000ffffffff800400000000000000000000ffffffff80000400ffff
ffff
fffffffd00000000000000000000000000000001000000002e21088c0000000069a43a40ffff
ffff
80045220ffffffff83ffffd00000000000000002ffffffff80000ee40000000010006c020000
0000
000f4240000000000000000000000000000000000000000010000024ffffffff8000d5980000
0000
0000000000000000000000000000000000000000000000000000000000000000000000000000
0000
0000000000000000000000000000000000000000000000000000000
Remote packet too long:
0000000000000000ffffffffb80600000000000000000001ffffffff
8003e2b0ffffffff8000d594ffffffff83fffff0ffffffff800452200000000000000001ffff
ffff
fff0e51200000000000f4240ffffffffb802801800000000000027100000000000000064ffff
ffff
80040000000000000800000000000000000000aa0000000000000002ffffffff800004000000
0000
00000000ffffffff80040000ffffffff800400000000000000000000ffffffff80000400ffff
ffff
fffffffd00000000000000000000000000000001000000002e21088c0000000069a43a40ffff
ffff
80045220ffffffff83ffffd00000000000000002ffffffff80000ee40000000010006c020000
0000
000f4240000000000000000000000000000000000000000010000024ffffffff8000d5980000
0000
0000000000000000000000000000000000000000000000000000000000000000000000000000
0000
0000000000000000000000000000000000000000000000000000000
Ignoring packet error, continuing...
0x00000000 in ?? ()
(gdb)

The serial baudrate is 38400, it gets past several packets (changeThread,
Query, etc), then GDB requests for registers and then the HAL dumps this
message to gdb and gdb does not like it. Have you seen this? sugguestions?

Thanks,
tmichals


----- Original Message -----
From: "Jonathan Larmour" <jifl@eCosCentric.com>
To: "Tim Michals" <t.michals@attbi.com>
Cc: <ecos-discuss@sources.redhat.com>
Sent: Wednesday, February 26, 2003 12:26 PM
Subject: Re: [ECOS] MIPS32 gdb vectors question


> Tim Michals wrote:
> > I'm looking into why MIPS32/GDB does not respond to GDB commands.  Just
> > wanted to check to make sure the correct vector processing code is
called,
> > hence, debug_vector is not called, but other_vector is.
>
> Correct. That's because that's what the break command says it does :-).
> The debug vector is only used by the debug single step exception, debug
> interrupt exception, debug instruction break exception, debug software
> breakpoint exception and debug data break exception.
>
> Software break is only used with the SDBBP instruction. It's possible GDB
> is putting that instruction in instead of "break 5" (opcode 0x005000d if
> you want to check GDB memory writes with set debug remote 1), but that
> seems highly unlikely as that would break so many things.
>
> Jifl
> --
> eCosCentric    http://www.eCosCentric.com/    The eCos and RedBoot experts
> --[ "You can complain because roses have thorns, or you ]--
> --[  can rejoice because thorns have roses." -Lincoln   ]-- Opinions==mine
>
>



-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss