On Fri, May 15, 2009 1:51 am, raz ben yehuda wrote: > > On Thu, 2009-05-14 at 22:22 +1000, Neil Brown wrote: >> On Thursday May 14, neilb@xxxxxxx wrote: >> > >> > Then the loop could be >> > for (i = 0; i < conf->nr_strip_zones; i++) >> > if (sector < conf->strip_zone[i+1]) >> > return conf->strip_zone[i] >> > >> > Save our selves an 'add' that way :-) >> >> That should have been >> >> > >> > Then the loop could be >> > for (i = 0; i < conf->nr_strip_zones; i++) >> > if (sector < conf->strip_zone[i+1]->zone_start) >> > return conf->strip_zone[i] >> > >> > Save ourselves an 'add' that way :-) >> >> and as we aren't using ->sectors in this loop any more, I think it >> can be discarded completely, just making struct strip_zone 20% smaller >> and improving cache performance. >> >> What do you think? > hamm.. please do not do it now. I think i need it for the reshape > because I have to traverse each zone up to its end. You don't lose that ability. Then end of one zone matches the beginning of the next zone. So run from zone[0].zone_start to zone[1].zone_start I was thinking further about this and realised that in the first zone, zone_start is always 0. And we added a sentinal zone for which only the zone_start value is set. This seems silly. Instead we can just store 'zone_end' in each zone (or possibly start_of_next_zone) and then we don't need the sentinal and don't waste the zone_start at the beginning. NeilBrown >> NeilBrown > > > -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
