[ECOS] GDBstub porting for vrc4375

[ECOS] GDBstub porting for vrc4375

[Ling Su]

Dear All,

As you may know, I am porting GDB stub for NEC DDB-VRC4375 based on VRC4373
source code. I got following questions, anybody can help me?

<1>. After try to burn code into EPROM and test it on board for lots of
times, I began to use PMON to debug my GDB stub in RAM startup. I disabled
MMU, Cache. But I found there are some conflicts when the GDB stub and PMON
try to access serial port together, so the program usually dead in trying to
write char to serial port. I am not sure if there is any method I can slove
this? I hope I can use another serial port to connect with GDB in this
debugging session, is that possible? there are two ports on the board. But I
am not sure to avoid the PMON and the monitor to access the serial port
simutenously.

<2>. There are a 7-segment display on board, according to the document the
address is 0x7c000000, I am trying to write to the address after system
initialization, but it doesn't work. Anything wrong?

Thanks!

-Ling

Re: [ECOS] GDBstub porting for vrc4375

[Jonathan Larmour]

Ling Su wrote:
> 
> 
> <1>. After try to burn code into EPROM and test it on board for lots of
> times, I began to use PMON to debug my GDB stub in RAM startup. I disabled
> MMU, Cache. But I found there are some conflicts when the GDB stub and PMON
> try to access serial port together, so the program usually dead in trying to
> write char to serial port. I am not sure if there is any method I can slove
> this? I hope I can use another serial port to connect with GDB in this
> debugging session, is that possible? there are two ports on the board. But I
> am not sure to avoid the PMON and the monitor to access the serial port
> simutenously.

You could try changing CYGHWR_HAL_MIPS_VR4300_VRC4373_GDB_PORT in your
configuration.

> <2>. There are a 7-segment display on board, according to the document the
> address is 0x7c000000, I am trying to write to the address after system
> initialization, but it doesn't work. Anything wrong?

Maybe PMON doesn't set up the MMU mappings for that?

Jifl
-- 
Red Hat, Rustat House, Clifton Road, Cambridge, UK. Tel: +44 (1223) 271062
Maybe this world is another planet's Hell -Aldous Huxley || Opinions==mine

Re: [ECOS] GDBstub porting for vrc4375

[Jesper Skov]

>>>>> "Jonathan" == Jonathan Larmour <jlarmour@redhat.com> writes:

>> <2>. There are a 7-segment display on board, according to the
>> document the address is 0x7c000000, I am trying to write to the
>> address after system initialization, but it doesn't work. Anything
>> wrong?

Jonathan> Maybe PMON doesn't set up the MMU mappings for that?

Or the CPU runs in big-endian mode and the LED is 8-bit addressed so
you need to poke at 0x7c000000^3

Jesper

Re: [ECOS] GDBstub porting for vrc4375

[Ling Su]

> >> <2>. There are a 7-segment display on board, according to the
> >> document the address is 0x7c000000, I am trying to write to the
> >> address after system initialization, but it doesn't work. Anything
> >> wrong?
>
> Jonathan> Maybe PMON doesn't set up the MMU mappings for that?
>
> Or the CPU runs in big-endian mode and the LED is 8-bit addressed so
> you need to poke at 0x7c000000^3
>

In the document of this board, it mentioned the mapping and virtual address
is 0xc7000000 (Sorry, I made a typo last time).  Actually I tried to write
to (*(cyg_uint8 *)(0x7c000000 + i)), in which i ranges from 0 to 0xFFFF, I
guess I should hit the right address. :-/

Regards,
-Ling