Re: [ECOS] interrupt_end

[Nick Garnett <nickg at redhat dot com>]

Motoya Kurotsu <kurotsu@allied-telesis.co.jp> writes:

> > So I think that the hal_cpu_except_enable is not correct. 
> 
> Or hal_cpu_int_enable instead of hal_cpu_except_enable in vectors.S.
> 

That is certainly worth trying. However, the code has been like this
for about 3 years, so there may be some unknown dependency on having
interrupts disabled somewhere. I wouldn't want to make such a change
to the main source tree without extensive testing on all the current
supported boards.

interrupt_end() can be called either with interrupts enabled or
disabled, it shouldn't care. The only advantage to turning interrupts
back on during the ISR is that nested interrupts can be taken. But
that also depends to some extent on the interrupt controller.

-- 
Nick Garnett, eCos Kernel Architect
Red Hat, Cambridge, UK


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