Re: start_kernel(): bug: interrupts were enabled early

[Russell King <rmk@arm.linux.org.uk>]

On Wed, Mar 31, 2010 at 02:57:20PM -0700, H. Peter Anvin wrote:
> On 03/31/2010 02:54 PM, Russell King wrote:
> > On Wed, Mar 31, 2010 at 04:42:25PM -0500, Christoph Lameter wrote:
> >> Do not run the checks while we are in a single threaded context?
> >>
> >> I thought we had some dynamic code patching thingamy that could change
> >> those when we go to smp mode?
> > 
> > You have to remember that on embedded architectures, such as ARM,
> > where XIP is supported we can't change the text segment at run time -
> > which means dynamic code patching won't work.
> > 
> > However, the kernel should still work in such situations.
> > 
> 
> The question still remains what the incremental cost is of doing
> irqsave/irqrestore.

Compared to irq disable/enable, it wouldn't be much higher; saving
can be done by a direct register to register move, so that should be
relatively cheap.  The restore may be a little bit more depending on
the CPU arch version, but not significant.

So there shouldn't be a problem from ARM POV to switch to using
irqsave/irqrestore.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of: