From: Laura Abbott <labbott@redhat.com> To: Juerg Haefliger <juerg.haefliger@hpe.com>, linux-kernel@vger.kernel.org, linux-mm@kvack.org Cc: vpk@cs.brown.edu, Kees Cook <keescook@chromium.org> Subject: Re: [RFC PATCH] Add support for eXclusive Page Frame Ownership (XPFO) Date: Mon, 28 Mar 2016 12:29:59 -0700 [thread overview] Message-ID: <56F98637.4070705@redhat.com> (raw) In-Reply-To: <56EFB2DB.3090602@hpe.com> On 03/21/2016 01:37 AM, Juerg Haefliger wrote: ... >>> +void xpfo_free_page(struct page *page, int order) >>> +{ >>> + int i; >>> + unsigned long kaddr; >>> + >>> + for (i = 0; i < (1 << order); i++) { >>> + >>> + /* The page frame was previously allocated to user space */ >>> + if (TEST_AND_CLEAR_XPFO_FLAG(user, page + i)) { >>> + kaddr = (unsigned long)page_address(page + i); >>> + >>> + /* Clear the page and mark it accordingly */ >>> + clear_page((void *)kaddr); >> >> Clearing the page isn't related to XPFO. There's other work ongoing to >> do clearing of the page on free. > > It's not strictly related to XPFO but adds another layer of security. Do you > happen to have a pointer to the ongoing work that you mentioned? > > The work was merged for the 4.6 merge window https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=8823b1dbc05fab1a8bec275eeae4709257c2661d This is a separate option to clear the page. ... >>> @@ -2072,10 +2076,11 @@ void free_hot_cold_page(struct page *page, bool cold) >>> } >>> >>> pcp = &this_cpu_ptr(zone->pageset)->pcp; >>> - if (!cold) >>> + if (!cold && !xpfo_test_kernel(page)) >>> list_add(&page->lru, &pcp->lists[migratetype]); >>> else >>> list_add_tail(&page->lru, &pcp->lists[migratetype]); >>> + >> >> What's the advantage of this? > > Allocating a page to userspace that was previously allocated to kernel space > requires an expensive TLB shootdown. The above will put previously > kernel-allocated pages in the cold page cache to postpone their allocation as > long as possible to minimize TLB shootdowns. > > That makes sense. You probably want to make this a separate commmit with this explanation as the commit text. >>> pcp->count++; >>> if (pcp->count >= pcp->high) { >>> unsigned long batch = READ_ONCE(pcp->batch); >>> > > Thanks for the review and comments! It's highly appreciated. > > ...Juerg > > >> Thanks, >> Laura
WARNING: multiple messages have this Message-ID (diff)
From: Laura Abbott <labbott@redhat.com> To: Juerg Haefliger <juerg.haefliger@hpe.com>, linux-kernel@vger.kernel.org, linux-mm@kvack.org Cc: vpk@cs.brown.edu, Kees Cook <keescook@chromium.org> Subject: Re: [RFC PATCH] Add support for eXclusive Page Frame Ownership (XPFO) Date: Mon, 28 Mar 2016 12:29:59 -0700 [thread overview] Message-ID: <56F98637.4070705@redhat.com> (raw) In-Reply-To: <56EFB2DB.3090602@hpe.com> On 03/21/2016 01:37 AM, Juerg Haefliger wrote: ... >>> +void xpfo_free_page(struct page *page, int order) >>> +{ >>> + int i; >>> + unsigned long kaddr; >>> + >>> + for (i = 0; i < (1 << order); i++) { >>> + >>> + /* The page frame was previously allocated to user space */ >>> + if (TEST_AND_CLEAR_XPFO_FLAG(user, page + i)) { >>> + kaddr = (unsigned long)page_address(page + i); >>> + >>> + /* Clear the page and mark it accordingly */ >>> + clear_page((void *)kaddr); >> >> Clearing the page isn't related to XPFO. There's other work ongoing to >> do clearing of the page on free. > > It's not strictly related to XPFO but adds another layer of security. Do you > happen to have a pointer to the ongoing work that you mentioned? > > The work was merged for the 4.6 merge window https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=8823b1dbc05fab1a8bec275eeae4709257c2661d This is a separate option to clear the page. ... >>> @@ -2072,10 +2076,11 @@ void free_hot_cold_page(struct page *page, bool cold) >>> } >>> >>> pcp = &this_cpu_ptr(zone->pageset)->pcp; >>> - if (!cold) >>> + if (!cold && !xpfo_test_kernel(page)) >>> list_add(&page->lru, &pcp->lists[migratetype]); >>> else >>> list_add_tail(&page->lru, &pcp->lists[migratetype]); >>> + >> >> What's the advantage of this? > > Allocating a page to userspace that was previously allocated to kernel space > requires an expensive TLB shootdown. The above will put previously > kernel-allocated pages in the cold page cache to postpone their allocation as > long as possible to minimize TLB shootdowns. > > That makes sense. You probably want to make this a separate commmit with this explanation as the commit text. >>> pcp->count++; >>> if (pcp->count >= pcp->high) { >>> unsigned long batch = READ_ONCE(pcp->batch); >>> > > Thanks for the review and comments! It's highly appreciated. > > ...Juerg > > >> Thanks, >> Laura -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2016-03-28 19:30 UTC|newest] Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-02-26 14:21 [RFC PATCH] Add support for eXclusive Page Frame Ownership (XPFO) Juerg Haefliger 2016-02-26 14:21 ` Juerg Haefliger 2016-03-01 1:31 ` Laura Abbott 2016-03-01 1:31 ` Laura Abbott 2016-03-21 8:37 ` Juerg Haefliger 2016-03-21 8:37 ` Juerg Haefliger 2016-03-28 19:29 ` Laura Abbott [this message] 2016-03-28 19:29 ` Laura Abbott 2016-03-01 2:10 ` Balbir Singh 2016-03-01 2:10 ` Balbir Singh 2016-03-21 8:44 ` Juerg Haefliger 2016-03-21 8:44 ` Juerg Haefliger 2016-04-01 0:21 ` Balbir Singh 2016-04-01 0:21 ` Balbir Singh 2016-09-02 11:39 ` [RFC PATCH v2 0/3] " Juerg Haefliger 2016-09-02 11:39 ` [kernel-hardening] " Juerg Haefliger 2016-09-02 11:39 ` Juerg Haefliger 2016-09-02 11:39 ` [RFC PATCH v2 1/3] " Juerg Haefliger 2016-09-02 11:39 ` [kernel-hardening] " Juerg Haefliger 2016-09-02 11:39 ` Juerg Haefliger 2016-09-02 11:39 ` [RFC PATCH v2 2/3] xpfo: Only put previous userspace pages into the hot cache Juerg Haefliger 2016-09-02 11:39 ` [kernel-hardening] " Juerg Haefliger 2016-09-02 11:39 ` Juerg Haefliger 2016-09-02 20:39 ` Dave Hansen 2016-09-02 20:39 ` [kernel-hardening] " Dave Hansen 2016-09-02 20:39 ` Dave Hansen 2016-09-05 11:54 ` Juerg Haefliger 2016-09-05 11:54 ` [kernel-hardening] " Juerg Haefliger 2016-09-02 11:39 ` [RFC PATCH v2 3/3] block: Always use a bounce buffer when XPFO is enabled Juerg Haefliger 2016-09-02 11:39 ` [kernel-hardening] " Juerg Haefliger 2016-09-02 11:39 ` Juerg Haefliger 2016-09-14 7:18 ` [RFC PATCH v2 0/3] Add support for eXclusive Page Frame Ownership (XPFO) Juerg Haefliger 2016-09-14 7:18 ` [kernel-hardening] " Juerg Haefliger 2016-09-14 7:18 ` Juerg Haefliger 2016-09-14 7:18 ` [RFC PATCH v2 1/3] " Juerg Haefliger 2016-09-14 7:18 ` [kernel-hardening] " Juerg Haefliger 2016-09-14 7:18 ` Juerg Haefliger 2016-09-14 7:19 ` [RFC PATCH v2 2/3] xpfo: Only put previous userspace pages into the hot cache Juerg Haefliger 2016-09-14 7:19 ` [kernel-hardening] " Juerg Haefliger 2016-09-14 7:19 ` Juerg Haefliger 2016-09-14 14:33 ` [kernel-hardening] " Dave Hansen 2016-09-14 14:33 ` Dave Hansen 2016-09-14 14:40 ` Juerg Haefliger 2016-09-14 14:48 ` Dave Hansen 2016-09-14 14:48 ` Dave Hansen 2016-09-21 5:32 ` Juerg Haefliger 2016-09-14 7:19 ` [RFC PATCH v2 3/3] block: Always use a bounce buffer when XPFO is enabled Juerg Haefliger 2016-09-14 7:19 ` [kernel-hardening] " Juerg Haefliger 2016-09-14 7:19 ` Juerg Haefliger 2016-09-14 7:33 ` Christoph Hellwig 2016-09-14 7:33 ` [kernel-hardening] " Christoph Hellwig 2016-09-14 7:33 ` Christoph Hellwig 2016-09-14 7:23 ` [RFC PATCH v2 0/3] Add support for eXclusive Page Frame Ownership (XPFO) Juerg Haefliger 2016-09-14 7:23 ` [kernel-hardening] " Juerg Haefliger 2016-09-14 9:36 ` [kernel-hardening] " Mark Rutland 2016-09-14 9:36 ` Mark Rutland 2016-09-14 9:49 ` Mark Rutland 2016-09-14 9:49 ` Mark Rutland 2016-11-04 14:45 ` [RFC PATCH v3 0/2] " Juerg Haefliger 2016-11-04 14:45 ` [kernel-hardening] " Juerg Haefliger 2016-11-04 14:45 ` Juerg Haefliger 2016-11-04 14:45 ` [RFC PATCH v3 1/2] " Juerg Haefliger 2016-11-04 14:45 ` [kernel-hardening] " Juerg Haefliger 2016-11-04 14:45 ` Juerg Haefliger 2016-11-04 14:50 ` Christoph Hellwig 2016-11-04 14:50 ` [kernel-hardening] " Christoph Hellwig 2016-11-04 14:50 ` Christoph Hellwig 2016-11-10 5:53 ` [kernel-hardening] " ZhaoJunmin Zhao(Junmin) 2016-11-10 5:53 ` ZhaoJunmin Zhao(Junmin) 2016-11-10 5:53 ` ZhaoJunmin Zhao(Junmin) 2016-11-10 19:11 ` Kees Cook 2016-11-10 19:11 ` [kernel-hardening] " Kees Cook 2016-11-10 19:11 ` Kees Cook 2016-11-15 11:15 ` Juerg Haefliger 2016-11-15 11:15 ` [kernel-hardening] " Juerg Haefliger 2016-11-15 11:15 ` Juerg Haefliger 2016-11-10 19:24 ` Kees Cook 2016-11-10 19:24 ` [kernel-hardening] " Kees Cook 2016-11-10 19:24 ` Kees Cook 2016-11-15 11:18 ` Juerg Haefliger 2016-11-15 11:18 ` [kernel-hardening] " Juerg Haefliger 2016-11-15 11:18 ` Juerg Haefliger 2016-11-24 10:56 ` AKASHI Takahiro 2016-11-24 10:56 ` [kernel-hardening] " AKASHI Takahiro 2016-11-24 10:56 ` AKASHI Takahiro 2016-11-28 11:15 ` Juerg Haefliger 2016-11-28 11:15 ` [kernel-hardening] " Juerg Haefliger 2016-12-09 9:02 ` AKASHI Takahiro 2016-12-09 9:02 ` [kernel-hardening] " AKASHI Takahiro 2016-12-09 9:02 ` AKASHI Takahiro 2016-11-04 14:45 ` [RFC PATCH v3 2/2] xpfo: Only put previous userspace pages into the hot cache Juerg Haefliger 2016-11-04 14:45 ` [kernel-hardening] " Juerg Haefliger 2016-11-04 14:45 ` Juerg Haefliger
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=56F98637.4070705@redhat.com \ --to=labbott@redhat.com \ --cc=juerg.haefliger@hpe.com \ --cc=keescook@chromium.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=vpk@cs.brown.edu \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.