Messages in this thread | | | Date | Thu, 11 May 2017 10:35:38 -0400 (EDT) | Subject | Re: [v3 0/9] parallelized "struct page" zeroing | From | David Miller <> |
| |
From: Michal Hocko <mhocko@kernel.org> Date: Thu, 11 May 2017 10:05:38 +0200
> Anyway, do you agree that doing the struct page initialization along > with other writes to it shouldn't add a measurable overhead comparing > to pre-zeroing of larger block of struct pages? We already have an > exclusive cache line and doing one 64B write along with few other stores > should be basically the same.
Yes, it should be reasonably cheap.
| |