This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
Re: Redboot-GDB
- From: Mark Salter <msalter at redhat dot com>
- To: embeddedeng at hotmail dot com
- Cc: ecos-discuss at sources dot redhat dot com
- Date: Sat, 28 Jun 2003 13:19:45 -0400 (EDT)
- Subject: Re: [ECOS] Redboot-GDB
- References: <Law15-F49SvX5w0Yg8300004472@hotmail.com>
>>>>> Michael Anburaj writes:
> Hi,
>>
>> an ARM based platform. Ultimately, the bug was in my initialization of the
>> interrupt controller on my target. IIRC, I didn't mask all of the
>> interrupts that should have been masked.
> In my case, I am masking off all the interrupts at initialization. But,
> later on Cyg_RealTimeClock::Cyg_RealTimeClock() of
> kernel/current/src/clock.cxx gets called & this inturn calls
> HAL_CLOCK_INITIALIZE(), which unmasks the RTC <or the ecos system timer>
> interrupt. Is this control path ok for Redboot? If yes, or no; please
> comment on the 4 questions I earlier asked.
Cyg_RealTimeClock::Cyg_RealTimeClock() should not be included in
RedBoot. HAL_CLOCK_INITIALIZE should not unmask RTC interrrupt.
The unmasking is done in Cyg_RealTimeClock::Cyg_RealTimeClock()
after HAL_CLOCK_INITIALIZE is called.
This (RTC interrupt) is almost certainly your problem. RedBoot will
not have a handler installed for it.
How are you configuring/building RedBoot?
You should be doing something like:
% ecosconfig new <target_name> redboot
% ecosconfig import ${ECOS_REPOSITORY}/hal/arm/<target_dir>/current/misc/redboot_ROM.ecm
% ecosconfig tree
% make
--Mark
--
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss