![]() |
|
Message-ID: <cef01110-dc23-4442-f277-88d1d3662e00@huawei.com> Date: Mon, 12 Feb 2018 18:24:20 +0200 From: Igor Stoppa <igor.stoppa@...wei.com> To: Matthew Wilcox <willy@...radead.org> CC: <rdunlap@...radead.org>, <corbet@....net>, <keescook@...omium.org>, <mhocko@...nel.org>, <labbott@...hat.com>, <jglisse@...hat.com>, <hch@...radead.org>, <cl@...ux.com>, <linux-security-module@...r.kernel.org>, <linux-mm@...ck.org>, <linux-kernel@...r.kernel.org>, <kernel-hardening@...ts.openwall.com> Subject: Re: [PATCH 3/6] struct page: add field for vm_struct On 11/02/18 23:16, Matthew Wilcox wrote: > On Sun, Feb 11, 2018 at 05:19:17AM +0200, Igor Stoppa wrote: >> The struct page has a "mapping" field, which can be re-used, to store a >> pointer to the parent area. This will avoid more expensive searches. >> >> As example, the function find_vm_area is reimplemented, to take advantage >> of the newly introduced field. > > Umm. Is it more efficient? You're replacing an rb-tree search with a > page-table walk. You eliminate a spinlock, which is great, but is the > page-table walk more efficient? I suppose it'll depend on the depth of > the rb-tree, and (at least on x86), the page tables should already be > in cache. I thought the tradeoff favorable. How to verify it? > Unrelated to this patch, I'm working on a patch to give us page_type, > and I think I'll allocate a bit to mark pages which are vmalloced. pmalloced too? -- igor
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.