GDB error for angel on ARM oki board

GDB error for angel on ARM oki board

[Charles-Henri Balet]

Hello,
I have continue to test my problem with the insight 5.1.1 with an OKI board
with ARM 1.20 angel debugger.
I have compiled the GDB 5.1.1 and place the command : "target rdi com1", no
effect, when I press the
reset switch on the oki board an error message appear : "Unexpected read on
channel 4, length 192"
this message appear on each press on the reset switch, any has an idea ?
Another question, how can I find the "DEBUG" define in the GDB source code
for see more message on
the run of this arm-elf-gdb, I have see on the source code of the rdi
process that if DEBUG is defined, few
message appear for follow the problem, can you give me the source file where
this DEBUG is defined !

many thanks

balet charles-henri

Re: GDB error for angel on ARM oki board

[Keith Seitz]

See my previous response on this thread. We need to ascertain whether this
is a backend problem or a gui problem.

Once we know that, we should be able to find out what is wrong.

On Tue, 23 Apr 2002, Charles-Henri Balet wrote:

> I have compiled the GDB 5.1.1 and place the command : "target rdi com1", no
> effect, when I press the
> reset switch on the oki board an error message appear : "Unexpected read on
> channel 4, length 192"

Are you setting the baudrate? (set remotebaud XXXX) If you are using
either the command line or the console window to attach, you MUST set this
before attempting to attach.

> this message appear on each press on the reset switch, any has an idea ?

Doesn't ring any bells.

> Another question, how can I find the "DEBUG" define in the GDB source code
> for see more message on
> the run of this arm-elf-gdb, I have see on the source code of the rdi
> process that if DEBUG is defined, few
> message appear for follow the problem, can you give me the source file where
> this DEBUG is defined !

Probably add "-DDEBUG" to CFLAGS in rdi-share directory?

Keith

Insight for X11 under Windoze

[Kai Ruottu]

Hi,
 I remember Cygwin some time including X11-libs, so producing X11-clients running
under Windoze should be possible to produce... Insight can be built as an X11-
application. Linux or any other non-Windoze host provides the needed X11-server to
show the GUI... The claims about Windoze+Cygwin being 'one kind of Unix' could then
be more strong, although a X11-server for it may still be missing...

 So, has anyone tried to build Insight as a X11-app running on Win32, in order to
'remotely debug' Windoze-apps running on Windoze from a Linux etc. host ?

 Perhaps people stopped to buy 'X11-terminals' in early 90's, but the idea about
seeing and controlling X11-apps running on another system should still be there...
But perhaps I'm not aware of something disabling this 'bright' (or 'trash') idea
from working... Anyone knows something about this ?

Cheers, Kai

PS. Someone asked about the possibility to 'remotely debug' Windoze-apps on the
'gnu.gcc.help' NG, so my 'bright idea' comes from that...

Re: GDB error for angel on ARM oki board

[Keith Seitz]

On Wed, 24 Apr 2002, Charles-Henri Balet wrote:


> another infos for the gdb error, this is the message receive when start gdb
> and connect

> $ arm-elf-gdb
> 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=i686-pc-cygwin --target=arm-elf".
> (gdb) set remotebaud 9600
> (gdb) target rdi com1
> RDI_open: undefined error message, should reset target
> RDI_open failed
>

Excellent. This is exactly what I wanted to know. You are having problems
with the command line, so you need to send this to gdb@sources.redhat.com,
where they deal with backend problems with the debugger. They should be
able to help you there.

I think that you can add the debug stuff into rdi-share by adding
"-DDEBUG" to your CFLAGS, i.e., make CFLAGS="-g -DDEBUG" in gdb directory.

> (gdb) target rdi com1
> Unexpected read on channel 4, length 192                                //
> for each reset, I obtain this message
> Unexpected read on channel 4, length 192

This isn't so unexpected. The RDI code that ARM "donated" to us really
sucks. I don't believe that it can handle re-opening the target like this.
You may need to do "detach" first.

> "I have tested after this with the ARM SDT 2.50 and the board work fine with
> no error to 9600 and 19200 baud"

Ok, this is good to know: this removes the possibility that your cable or
something is at fault.

> the message returned at the connection of the ARM SDT with the board is :
>
> Angel Debug Monitor V1.20 (ARM Ltd unreleased) for ML671000 CPU Board(OKI)
> Built for ARM7TDMI Serial(x1), IRQ
> Unreleased build Mar 14 2000
> Serial Rate:   9600
>
> You think, ARM has make an proprietary Angel for OKI ???? and not fully
> compatible ? and has the little or big endian
> mode any importance ?

It is possible, that ARM has a special Angel protocol for OKI, but I don't
know; it has been a while since I last played with ARM and RDI.

You're best bet is to ask over on the gdb list. There are bound to be more
people over there that can help. Don't forget to send them the info above
(gdb -nw session and that it works using ARM's tools). This is very
important for them.

Good luck.
Keith