Ar Gwe, 2006-08-18 am 12:32 -0700, ysgrifennodd Dave Hansen:I only doubt what gain we will have in this situation.
It ought to be cheap. Given each set of page structs is an array its a
simple subtract and divide (or with care and people try to pack them
nicely for cache lines - shift) to get to the parallel accounting array.
I wish page structs were just a simple array. ;)
Note I very carefully said "each set of"
It will just be a bit more code, but we'll need this for the two other
memory models: sparsemem and discontigmem. For discontig, we'll just
need pointers in the pg_data_ts and, for sparsemem, we'll likely need
another pointer in the 'struct mem_section'.
Actually I don't believe this is true in either case. Change the code
which allocates the page arrays to allocate (+ sizeof(void *) *
pages_in_array on the end of each array when using UBC. The rest then
seems to come out naturally.