On 03/18/2013 02:19 PM, Yinghai Lu wrote: > On Mon, Mar 18, 2013 at 12:14 PM, H. Peter Anvin <hpa@xxxxxxxxx> wrote: > >> Instead, try to explain why 5 is the correct value in the current code >> and how it is (or should be!) derived. > > initial mapped size is PMD_SIZE, aka 2M. > if we use step_size to be PUD_SIZE aka 1G, as most worse case > that 1G is cross the 1G boundary, and PG_LEVEL_2M is not set, > we will need 1+1+512 pages (aka 2M + 8k) to map 1G range with PTE. > So i picked (30-21)/2 to get 5. > > Please check attached patch. > > Thanks > > Yinghai > This still seems very opaque. I need to look at it and see if it makes more sense in context. -hpa -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>