Re: [patch 1/3] Create spin lock/spin unlock with distinct memorybarrier

[Nick Piggin]

On Sun, Jan 31, 2010 at 03:52:55PM -0500, Mathieu Desnoyers wrote:
> Create the primitive family:
> 
> spin_lock__no_acquire
> spin_unlock__no_release
> spin_lock_irq__no_acquire
> spin_unlock_irq__no_release
> 
> raw_spin_lock__no_acquire
> raw_spin_unlock__no_release
> raw_spin_lock_irq__no_acquire
> raw_spin_unlock_irq__no_release
> 
> smp_acquire__after_spin_lock()
> smp_release__before_spin_unlock()
> smp_mb__after_spin_lock()
> smp_mb__before_spin_unlock()

Wow, someone who likes micro optimising things as much as I do.
However, these have the wrong names.

smp_mb__after_x() means that calling that function after calling x()
will give a smp_mb(), right?

With your functions, this is giving a smp_mb() after calling
x__no_acquire().

I would suggest maybe just don't bother with the __no_acquire
__no_release variants of spin locks, and stick with the expected
semantics for the new smb_mb__xxx functions. x86 still gets the
full benefit.

But, I don't know if this is even worthwhile, given where you are
using it.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/