Messages in this thread | | | Date | Tue, 10 Apr 2012 11:22:30 +0200 | From | Per Olofsson <> | Subject | Re: [PATCH v9]: Hibernation: fix the number of pages used for hibernate/thaw buffering |
| |
On 2012-04-10 11:19, Bojan Smojver wrote: > On Tue, 2012-04-10 at 10:40 +0200, Per Olofsson wrote: >> Nitpick: it should be faster to calculate "nr_free_pages() - >> nr_free_highpages()", at least on systems without high pages. But >> perhaps it does not matter. > > Actually, nr_free_highpages() does a loop very similar to the one in the > patch. So, you would then have that and nr_free_pages() invocation. > That's why I went with just the loop. > > Make sense? >
If you don't have any high pages (e.g. x86_64), then nr_free_highpages() is defined to be 0 and is optimized away completely.
Still, it seems you are not calculating free pages that often anymore so it should not matter much.
-- Pelle
| |