Re: [ECOS] RedBoot, GDB_STUB and serial interrupts

[Jonathan Larmour <jlarmour at redhat dot com>]

Chris Zankel wrote:
> 
> I am currently in the process of porting RedBoot to a new
> architecture

Xtensa support would be neat! 

> 
> In one of the mails mentione above Gary Thomas wrote:
> 
>    You should _not_ have the options required to include this
>    code set. The options CYGDBG_HAL_DEBUG_GDB_BREAK_SUPPORT
>    and CYGDBG_HAL_DEBUG_GDB_CTRLC_SUPPORT should only be enabled
>    for "normal" eCos applications.

Well, I think what Gary was thinking was that
CYGDBG_HAL_DEBUG_GDB_BREAK_SUPPORT is enabled but has little effect. See
below...
 
> But because BREAK_SUPPORT is now enabled hal_ctrlc_isr_init()
> is also compiled and enables the interrupt for the serial port:
> 
> hal_ctrlc_isr_init(void)
> ...
> 
>      // We can only enable interrupts on a valid debug channel.
>      if (__chan)
>          CYGACC_COMM_IF_CONTROL(*__chan, __COMMCTL_IRQ_ENABLE);
> 
> 
> This effecively disables debugging in RedBoot because the
> interrupt routine only checks for a ctrl-c character and
> dismisses all other characters.

For a start that stuff in hal_ctrlc_isr_init is #ifndef
CYGSEM_HAL_ROM_MONITOR, which a ROM redboot defines.

But in any case, RedBoot runs with interrupts disabled always. Everything
is done by polling.

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