On Mon, Mar 28, 2016 at 1:55 AM, Robert Haas <robertmhaas@gmail.com> wrote: > > On Sun, Mar 27, 2016 at 8:00 AM, Dilip Kumar <dilipbalaut@gmail.com> wrote: > > > Results: > > ---------- > > 1. With this performance is little less than v14 but the problem of extra > > relation size is solved. > > 2. With this we can conclude that extra size of relation in v14 is because > > some while extending the pages, its not immediately available and at end > > some of the pages are left unused. > > I agree with that conclusion. I'm not quite sure where that leaves > us, though. We can go back to v13, but why isn't that producing extra > pages? It seems like it should: whenever a bulk extend rolls over to > a new FSM page, concurrent backends will search either the old or the > new one but not both. >
I have not debugged the flow, but by looking at v13 code, it looks like it will search both old and new. In function GetPageWithFreeSpaceExtended()->fsm_search_from_addr()->fsm_search_avail(), the basic idea of search is: Start the search from the target slot. At every step, move one
node to the right, then climb up to the parent. Stop when we reach a node with enough free space (as we must, since the root has enough space).
So shouldn't it be able to find the new FSM page where the bulk extend rolls over?