On Mon, 2013-12-02 at 15:23 +0100, Paolo Bonzini wrote:
> This avoids useless masking and shifting when a single call to the
> MemoryRegion ops will do. It cuts 30 cycles off the common case
> of memory dispatch (out of ~150).
>
> Paolo Bonzini (4):
> memory: cache min/max_access_size
> memor
This avoids useless masking and shifting when a single call to the
MemoryRegion ops will do. It cuts 30 cycles off the common case
of memory dispatch (out of ~150).
Paolo Bonzini (4):
memory: cache min/max_access_size
memory: streamline common case for memory dispatch
memory: hoist coalesce
