linux-kernel - [PATCH 0/5] hugepage migration fixes (v4)

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-next>] [day] [month] [year] [list]

Message-Id: <1410820799-27278-1-git-send-email-n-horiguchi@ah.jp.nec.com>
Date:	Mon, 15 Sep 2014 18:39:54 -0400
From:	Naoya Horiguchi <n-horiguchi@...jp.nec.com>
To:	Andrew Morton <akpm@...ux-foundation.org>,
	Hugh Dickins <hughd@...gle.com>
Cc:	David Rientjes <rientjes@...gle.com>, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org,
	Naoya Horiguchi <nao.horiguchi@...il.com>
Subject: [PATCH 0/5] hugepage migration fixes (v4)

This is the ver.4 of hugepage migration fix patchset.

Major changes from ver.3 are:
- to drop locking from follow_huge_pud() and follow_huge_addr() in patch 2,
  which was buggy and is not necessary now because they don't support FOLL_GET,
- follow_huge_pmd(FOLL_GET) can pin and return tail pages,
- and I fixed bugs accidentally introduced in patch 3 and 5.
Others are code improvements and comment/description fixes.

Two related topics (not included in this series but to be discussed/done)
- follow_huge_pmd() explicitly uses pmd_lockptr() instead of huge_pte_lock().
  This point shed light on the subtlety in huge_page_size == PMD_HUGE check
  in huge_pte_lockptr(). This seems to make no runtime problem now, but might
  look fragile for example when pmd is folded (where PUD_SIZE == PMD_SIZE)
  or when hugepage in your architecture is not bound by page table (where
  hugepage size happens to equal with PMD_SIZE.)
- code around is_hugetlb_entry_migration() and is_hugetlb_entry_hwpoisoned()
  is not beautiful or optimized. Cleanup is necessary.

This patchset is based on mmotm-2014-09-09-14-42 and shows no regression
in libhugetlbfs test.

Tree: git@...hub.com:Naoya-Horiguchi/linux.git
Branch: mmotm-2014-09-09-14-42/fix_follow_huge_pmd.v4

v2: http://thread.gmane.org/gmane.linux.kernel/1761065
v3: http://thread.gmane.org/gmane.linux.kernel/1776585
---
Summary:

Naoya Horiguchi (5):
      mm/hugetlb: reduce arch dependent code around follow_huge_*
      mm/hugetlb: take page table lock in follow_huge_pmd()
      mm/hugetlb: fix getting refcount 0 page in hugetlb_fault()
      mm/hugetlb: add migration/hwpoisoned entry check in hugetlb_change_protection
      mm/hugetlb: add migration entry check in __unmap_hugepage_range

 arch/arm/mm/hugetlbpage.c     |   6 --
 arch/arm64/mm/hugetlbpage.c   |   6 --
 arch/ia64/mm/hugetlbpage.c    |   6 --
 arch/metag/mm/hugetlbpage.c   |   6 --
 arch/mips/mm/hugetlbpage.c    |  18 ------
 arch/powerpc/mm/hugetlbpage.c |   8 +++
 arch/s390/mm/hugetlbpage.c    |  20 -------
 arch/sh/mm/hugetlbpage.c      |  12 ----
 arch/sparc/mm/hugetlbpage.c   |  12 ----
 arch/tile/mm/hugetlbpage.c    |  28 ----------
 arch/x86/mm/hugetlbpage.c     |  12 ----
 include/linux/hugetlb.h       |   8 +--
 mm/gup.c                      |  25 ++-------
 mm/hugetlb.c                  | 124 ++++++++++++++++++++++++++++--------------
 mm/migrate.c                  |   3 +-
 15 files changed, 101 insertions(+), 193 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/