Re: [Xen-devel] [PATCH] x86: hold mm->page_table_lock while doing vmalloc_sync

[Jeremy Fitzhardinge <jeremy@goop.org>]

 On 10/14/2010 01:56 PM, Jeremy Fitzhardinge wrote:
> Take mm->page_table_lock while syncing the vmalloc region.  This prevents
> a race with the Xen pagetable pin/unpin code, which expects that the
> page_table_lock is already held.  If this race occurs, then Xen can see
> an inconsistent page type (a page can either be read/write or a pagetable
> page, and pin/unpin converts it between them), which will cause either
> the pin or the set_p[gm]d to fail; either will crash the kernel.

I've merged this into tip/x86/mm, which had some conflicting changes
(and fixed up some whitespace issues in one of those changes):

The following changes since commit a416e9e1dde0fbcf20cda59df284cc0dcf2aadc4:

  x86-32: Fix sparse warning for the __PHYSICAL_MASK calculation (2010-10-07 16:36:17 -0700)

are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen.git x86-mm

Jeremy Fitzhardinge (2):
      x86/mm: fix bogus whitespace in sync_global_pgds()
      x86: hold mm->page_table_lock while doing vmalloc_sync

 arch/x86/include/asm/pgtable.h |    2 +
 arch/x86/mm/fault.c            |   11 ++++++++-
 arch/x86/mm/init_64.c          |   51 ++++++++++++++++++++++-----------------
 arch/x86/mm/pgtable.c          |   20 +++++++++++++--
 4 files changed, 58 insertions(+), 26 deletions(-)

	J