linux-kernel - [PATCH v1 2/5] mm/memory_hotplug: enforce section alignment when onlining/offlining

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 16 Aug 2018 12:06:25 +0200
From:   David Hildenbrand <david@...hat.com>
To:     linux-kernel@...r.kernel.org
Cc:     linux-mm@...ck.org, Andrew Morton <akpm@...ux-foundation.org>,
        Michal Hocko <mhocko@...e.com>,
        Vlastimil Babka <vbabka@...e.cz>,
        Stephen Rothwell <sfr@...b.auug.org.au>,
        Pavel Tatashin <pasha.tatashin@...cle.com>,
        Kemi Wang <kemi.wang@...el.com>,
        David Rientjes <rientjes@...gle.com>,
        Jia He <jia.he@...-semitech.com>,
        Oscar Salvador <osalvador@...e.de>,
        Petr Tesarik <ptesarik@...e.com>,
        Andrey Ryabinin <aryabinin@...tuozzo.com>,
        Dan Williams <dan.j.williams@...el.com>,
        David Hildenbrand <david@...hat.com>,
        Mathieu Malaterre <malat@...ian.org>,
        Baoquan He <bhe@...hat.com>,
        Wei Yang <richard.weiyang@...il.com>,
        Ross Zwisler <zwisler@...nel.org>,
        "Kirill A . Shutemov" <kirill.shutemov@...ux.intel.com>
Subject: [PATCH v1 2/5] mm/memory_hotplug: enforce section alignment when onlining/offlining

onlining/offlining code works on whole sections, so let's enforce that.
Existing code only allows to add memory in memory block size. And only
whole memory blocks can be onlined/offlined. Memory blocks are always
aligned to sections, so this should not break anything.

online_pages/offline_pages will implicitly mark whole sections
online/offline, so the code really can only handle such granularities.

(especially offlining code cannot deal with pageblock_nr_pages but
 theoretically only MAX_ORDER-1)

Signed-off-by: David Hildenbrand <david@...hat.com>
---
 mm/memory_hotplug.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 090cf474de87..30d2fa42b0bb 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -897,6 +897,11 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages, int online_typ
 	struct memory_notify arg;
 	struct memory_block *mem;
 
+	if (!IS_ALIGNED(pfn, PAGES_PER_SECTION))
+		return -EINVAL;
+	if (!IS_ALIGNED(nr_pages, PAGES_PER_SECTION))
+		return -EINVAL;
+
 	/*
 	 * We can't use pfn_to_nid() because nid might be stored in struct page
 	 * which is not yet initialized. Instead, we find nid from memory block.
@@ -1600,10 +1605,9 @@ int offline_pages(unsigned long start_pfn, unsigned long nr_pages)
 	struct zone *zone;
 	struct memory_notify arg;
 
-	/* at least, alignment against pageblock is necessary */
-	if (!IS_ALIGNED(start_pfn, pageblock_nr_pages))
+	if (!IS_ALIGNED(start_pfn, PAGES_PER_SECTION))
 		return -EINVAL;
-	if (!IS_ALIGNED(end_pfn, pageblock_nr_pages))
+	if (!IS_ALIGNED(nr_pages, PAGES_PER_SECTION))
 		return -EINVAL;
 	/* This makes hotplug much easier...and readable.
 	   we assume this for now. .*/
-- 
2.17.1