On Mon, Dec 17, 2018 at 11:55 PM Christoph Hellwig <[email protected]> wrote: > > On Fri, Nov 30, 2018 at 01:32:02PM +0530, Anup Patel wrote: > > This patch does following optimizations: > > 1. Pre-compute hart base for each context handler > > 2. Pre-compute enable base for each context handler > > 3. Have enable lock for each context handler instead > > of global plic_toggle_lock > > All of which is pretty obvious from reading the patch. The big question > that needs to be answered in the changelog is why you do that.
To compute enable_base and hart_base is two integer additions and one integer multiplication. This micro-optimization simply avoids this repeated operations. Regards, Anup
