This is the mail archive of the ecos-discuss@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

FW: redboot booting linux on i386 pc ..



-----Original Message-----
From: Henry Zhu [mailto:zhu_henry@immenstar.com] 
Sent: Wednesday, January 19, 2005 7:18 PM
To: 'Srinivasa R Kanduru'
Subject: RE: [ECOS] redboot booting linux on i386 pc ..

Hi,
I think I got a similar problem when I tried redboot on my Pentium pc.
Redboot boots up. Application images were downloaded through xymodem and
tftp server. But execution of the application image got a strang response
like this:
 +$O5...................#60
Then it hanged.
It looks like the serial console output was confused with GBD stub output.
So, instead of starting the app by redboot, I connected to the target
through gdb via serial port or Ethernet, started the application by gdb
command "set $pc=<app_entry_point>", the application was executed
successfully.
The reason was probably depending on how the applications are built, what
eCos configuration options are selected such as USE_ROM_MONITOR, etc. 
Right now, I am using the lastest eCos sources. I could build RAM
applications using ROM gdb stubs without problem. I am wondering how vmlinux
can be compiled with similar features to use ROM gdb stubs so we can debug
linux kernel conveniently.

Regards,


-----Original Message-----
From: ecos-discuss-owner@ecos.sourceware.org
[mailto:ecos-discuss-owner@ecos.sourceware.org] On Behalf Of Srinivasa R
Kanduru
Sent: Wednesday, January 19, 2005 6:03 PM
To: ecos-discuss@sources.redhat.com
Cc: skanduru@gmail.com
Subject: [ECOS] redboot booting linux on i386 pc ..

Hi,

I am new to eCos and have been playing around with ecos/redboot to
boot vmlinux off the net. Redboot boots up fine and the networking is
configured. I can load vmlinux using tftp protocol. The output on the
monitor shows up like this:

RAM: 0x100000-0x040ffc00, [0x0018efd0-0x01000000] available

I have changed the start address of the kernel to be at 2 MB so as not
to crash redboot. After loading the kernel it prints ..

Entry point 0xc0205000, address range : 0xc0200000-0xc03bdec0

Now if I do 'go 0xc0205000', it gives an error message

$T0408:005020C0;04:e0ffff00;#b5

How do I interpret this ? Does this mean that it got a 'ILL' signal
and the value at 0xc0205000 is 0xffffe0 which is not valid ?

I am unable to get access to serial console. I set the
CYGNUM_HAL_VIRTUAL_VECTOR_COMM_CHANNELS to include both the serial
consoles. The monitor takes the channel id of 2.

When I switch to any of the other channels (channel 1 for ex) eCos
dies I think as I don't see any output on the serial console. (The
baudrate was set correctly using kermit)

Any idea ??

How do I connect to the serial console ?

I was able to connect to the target using gdb thru' ethernet, but I
cannot debug the problem as the redboot freezes when 'gdb' is active.

Is there anyway I can have both 'gdb' and 'redboot' on video console
active simultaneously ?

Since this is a regular pc and not a target board, the debugging
options might be limited. Any help is greatly appreciated.

-- 
     Regards,
     Srinivasa Kanduru

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



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


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]