Il 14/07/2013 12:23, liu ping fan ha scritto:
if the refcount ops are frequent enough, I strongly suspect cacheline
bouncing has a bigger effect than the memory barriers.
When out of biglock, object_ref/unref to pin the Device will be quite
often, and can it be marked frequent? Or how can we
On Sun, Jul 14, 2013 at 1:57 PM, Paolo Bonzini pbonz...@redhat.com wrote:
Il 14/07/2013 04:53, Liu Ping Fan ha scritto:
Refcnt's atomic inc/dec ops are frequent and its idiom need no seq_cst
order. So to get better performance, it worth to adopt _relaxed
other than _seq_cst memory model on
Refcnt's atomic inc/dec ops are frequent and its idiom need no seq_cst
order. So to get better performance, it worth to adopt _relaxed
other than _seq_cst memory model on them.
We resort to gcc builtins. If gcc supports C11 memory model, __atomic_*
buitlins is used, otherwise __sync_* builtins.
Il 14/07/2013 04:53, Liu Ping Fan ha scritto:
Refcnt's atomic inc/dec ops are frequent and its idiom need no seq_cst
order. So to get better performance, it worth to adopt _relaxed
other than _seq_cst memory model on them.
We resort to gcc builtins. If gcc supports C11 memory model,