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]

Re: RedBoot on V850 problem


savin@torina.fe.uni-lj.si wrote:
> 
> Hi all!
> 
> I have adopted the eCos CEB port to work with
> NEC StartWARE-V850/SA1 (20MHz) board. GDB stub is
> in flash and downloading / debugging works fine.
> 
> Now I'am trying to get RedBoot running...

No one has ported redboot (yet).

> I download it with gdb and then use minicom, I can
> see the prompt and type in commands. The problem
> arises when downloading a program - if I just
> type the command 'load -m xm' and watch it gives
> me the 'C' char every 10-15 seconds (isn't this to slow?).

Yes.
 
> Then I implemented the HAL_DELAY_US function using the
> code from hal/common/current/src/hal_if.c and adding the
> interval timer initialization.

Using that code? Better to use something similar to that found in other
target HALs. The one in hal_if.c is a fallback used if nothing else can be
found, which is why it gets it wrong.
 
> NOTE: there seems to be a problem with the current implementation
>       of delay_us in hal/common/current/src/hal_if.c -
>       if we have a long delay (usecs=250000) and the
>       CYGNUM_KERNEL_COUNTERS_RTC_PERIOD is a big number (in
>       my case it is 50000), then the following line gives us
>       an integer overflow (usecs and usec_ticks being of type cyg_int32)
> 
>       usec_ticks = usecs*CYGNUM_KERNEL_COUNTERS_RTC_PERIOD;

Firstly that's only in a kernel config (not relevant for RedBoot).
Secondly, delay_us is only intended for short delays.

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


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