From: Gerald Schaefer <gerald.schaefer@de.ibm.com> To: Andrew Morton <akpm@linux-foundation.org>, Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Michal Hocko <mhocko@suse.cz>, "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>, Vlastimil Babka <vbabka@suse.cz>, Mike Kravetz <mike.kravetz@oracle.com>, "Aneesh Kumar K . V" <aneesh.kumar@linux.vnet.ibm.com>, Martin Schwidefsky <schwidefsky@de.ibm.com>, Heiko Carstens <heiko.carstens@de.ibm.com>, Gerald Schaefer <gerald.schaefer@de.ibm.com> Subject: [PATCH 1/1] mm/hugetlb: fix memory offline with hugepage size > memory block size Date: Tue, 20 Sep 2016 17:53:54 +0200 [thread overview] Message-ID: <20160920155354.54403-2-gerald.schaefer@de.ibm.com> (raw) In-Reply-To: <20160920155354.54403-1-gerald.schaefer@de.ibm.com> dissolve_free_huge_pages() will either run into the VM_BUG_ON() or a list corruption and addressing exception when trying to set a memory block offline that is part (but not the first part) of a gigantic hugetlb page with a size > memory block size. When no other smaller hugepage sizes are present, the VM_BUG_ON() will trigger directly. In the other case we will run into an addressing exception later, because dissolve_free_huge_page() will not use the head page of the compound hugetlb page which will result in a NULL hstate from page_hstate(). list_del() would also not work well on a tail page. To fix this, first remove the VM_BUG_ON() because it is wrong, and then use the compound head page in dissolve_free_huge_page(). Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> --- mm/hugetlb.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 87e11d8..65e723c 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1441,15 +1441,17 @@ static int free_pool_huge_page(struct hstate *h, nodemask_t *nodes_allowed, */ static void dissolve_free_huge_page(struct page *page) { + struct page *head = compound_head(page); + spin_lock(&hugetlb_lock); - if (PageHuge(page) && !page_count(page)) { - struct hstate *h = page_hstate(page); - int nid = page_to_nid(page); - list_del(&page->lru); + if (!page_count(head)) { + struct hstate *h = page_hstate(head); + int nid = page_to_nid(head); + list_del(&head->lru); h->free_huge_pages--; h->free_huge_pages_node[nid]--; h->max_huge_pages--; - update_and_free_page(h, page); + update_and_free_page(h, head); } spin_unlock(&hugetlb_lock); } @@ -1466,9 +1468,9 @@ void dissolve_free_huge_pages(unsigned long start_pfn, unsigned long end_pfn) if (!hugepages_supported()) return; - VM_BUG_ON(!IS_ALIGNED(start_pfn, 1 << minimum_order)); for (pfn = start_pfn; pfn < end_pfn; pfn += 1 << minimum_order) - dissolve_free_huge_page(pfn_to_page(pfn)); + if (PageHuge(pfn_to_page(pfn))) + dissolve_free_huge_page(pfn_to_page(pfn)); } /* -- 2.8.4
WARNING: multiple messages have this Message-ID (diff)
From: Gerald Schaefer <gerald.schaefer@de.ibm.com> To: Andrew Morton <akpm@linux-foundation.org>, Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Michal Hocko <mhocko@suse.cz>, "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>, Vlastimil Babka <vbabka@suse.cz>, Mike Kravetz <mike.kravetz@oracle.com>, "Aneesh Kumar K . V" <aneesh.kumar@linux.vnet.ibm.com>, Martin Schwidefsky <schwidefsky@de.ibm.com>, Heiko Carstens <heiko.carstens@de.ibm.com>, Gerald Schaefer <gerald.schaefer@de.ibm.com> Subject: [PATCH 1/1] mm/hugetlb: fix memory offline with hugepage size > memory block size Date: Tue, 20 Sep 2016 17:53:54 +0200 [thread overview] Message-ID: <20160920155354.54403-2-gerald.schaefer@de.ibm.com> (raw) In-Reply-To: <20160920155354.54403-1-gerald.schaefer@de.ibm.com> dissolve_free_huge_pages() will either run into the VM_BUG_ON() or a list corruption and addressing exception when trying to set a memory block offline that is part (but not the first part) of a gigantic hugetlb page with a size > memory block size. When no other smaller hugepage sizes are present, the VM_BUG_ON() will trigger directly. In the other case we will run into an addressing exception later, because dissolve_free_huge_page() will not use the head page of the compound hugetlb page which will result in a NULL hstate from page_hstate(). list_del() would also not work well on a tail page. To fix this, first remove the VM_BUG_ON() because it is wrong, and then use the compound head page in dissolve_free_huge_page(). Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> --- mm/hugetlb.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 87e11d8..65e723c 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1441,15 +1441,17 @@ static int free_pool_huge_page(struct hstate *h, nodemask_t *nodes_allowed, */ static void dissolve_free_huge_page(struct page *page) { + struct page *head = compound_head(page); + spin_lock(&hugetlb_lock); - if (PageHuge(page) && !page_count(page)) { - struct hstate *h = page_hstate(page); - int nid = page_to_nid(page); - list_del(&page->lru); + if (!page_count(head)) { + struct hstate *h = page_hstate(head); + int nid = page_to_nid(head); + list_del(&head->lru); h->free_huge_pages--; h->free_huge_pages_node[nid]--; h->max_huge_pages--; - update_and_free_page(h, page); + update_and_free_page(h, head); } spin_unlock(&hugetlb_lock); } @@ -1466,9 +1468,9 @@ void dissolve_free_huge_pages(unsigned long start_pfn, unsigned long end_pfn) if (!hugepages_supported()) return; - VM_BUG_ON(!IS_ALIGNED(start_pfn, 1 << minimum_order)); for (pfn = start_pfn; pfn < end_pfn; pfn += 1 << minimum_order) - dissolve_free_huge_page(pfn_to_page(pfn)); + if (PageHuge(pfn_to_page(pfn))) + dissolve_free_huge_page(pfn_to_page(pfn)); } /* -- 2.8.4 -- 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:[~2016-09-20 15:54 UTC|newest] Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-09-20 15:53 [PATCH 0/1] memory offline issues with hugepage size > memory block size Gerald Schaefer 2016-09-20 15:53 ` Gerald Schaefer 2016-09-20 15:53 ` Gerald Schaefer [this message] 2016-09-20 15:53 ` [PATCH 1/1] mm/hugetlb: fix memory offline " Gerald Schaefer 2016-09-21 6:29 ` Hillf Danton 2016-09-21 6:29 ` Hillf Danton 2016-09-21 12:35 ` [PATCH v2 " Gerald Schaefer 2016-09-21 12:35 ` Gerald Schaefer 2016-09-21 13:17 ` Rui Teng 2016-09-21 13:17 ` Rui Teng 2016-09-21 15:13 ` Gerald Schaefer 2016-09-21 15:13 ` Gerald Schaefer 2016-09-22 7:58 ` Hillf Danton 2016-09-22 7:58 ` Hillf Danton 2016-09-22 9:51 ` Michal Hocko 2016-09-22 9:51 ` Michal Hocko 2016-09-22 13:45 ` Gerald Schaefer 2016-09-22 13:45 ` Gerald Schaefer 2016-09-22 16:29 ` [PATCH v3] " Gerald Schaefer 2016-09-22 16:29 ` Gerald Schaefer 2016-09-22 18:12 ` Dave Hansen 2016-09-22 18:12 ` Dave Hansen 2016-09-22 19:13 ` Mike Kravetz 2016-09-22 19:13 ` Mike Kravetz 2016-09-23 10:36 ` Gerald Schaefer 2016-09-23 10:36 ` Gerald Schaefer 2016-09-23 6:40 ` [PATCH v2 1/1] " Rui Teng 2016-09-23 6:40 ` Rui Teng 2016-09-23 11:03 ` Gerald Schaefer 2016-09-23 11:03 ` Gerald Schaefer 2016-09-26 2:49 ` Rui Teng 2016-09-26 2:49 ` Rui Teng 2016-09-20 17:37 ` [PATCH 0/1] memory offline issues " Mike Kravetz 2016-09-20 17:37 ` Mike Kravetz 2016-09-20 17:45 ` Dave Hansen 2016-09-20 17:45 ` Dave Hansen 2016-09-21 9:49 ` Vlastimil Babka 2016-09-21 9:49 ` Vlastimil Babka 2016-09-21 10:34 ` Gerald Schaefer 2016-09-21 10:34 ` Gerald Schaefer 2016-09-21 10:30 ` Gerald Schaefer 2016-09-21 10:30 ` Gerald Schaefer 2016-09-21 18:20 ` Michal Hocko 2016-09-21 18:20 ` Michal Hocko 2016-09-21 18:27 ` Dave Hansen 2016-09-21 18:27 ` Dave Hansen 2016-09-21 19:22 ` Michal Hocko 2016-09-21 19:22 ` Michal Hocko
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=20160920155354.54403-2-gerald.schaefer@de.ibm.com \ --to=gerald.schaefer@de.ibm.com \ --cc=akpm@linux-foundation.org \ --cc=aneesh.kumar@linux.vnet.ibm.com \ --cc=heiko.carstens@de.ibm.com \ --cc=kirill.shutemov@linux.intel.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mhocko@suse.cz \ --cc=mike.kravetz@oracle.com \ --cc=n-horiguchi@ah.jp.nec.com \ --cc=schwidefsky@de.ibm.com \ --cc=vbabka@suse.cz \ /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.