linux/arch/x86/xen
Jeremy Fitzhardinge 64141da587 vmalloc: eagerly clear ptes on vunmap
On stock 2.6.37-rc4, running:

  # mount lilith:/export /mnt/lilith
  # find  /mnt/lilith/ -type f -print0 | xargs -0 file

crashes the machine fairly quickly under Xen.  Often it results in oops
messages, but the couple of times I tried just now, it just hung quietly
and made Xen print some rude messages:

    (XEN) mm.c:2389:d80 Bad type (saw 7400000000000001 != exp
    3000000000000000) for mfn 1d7058 (pfn 18fa7)
    (XEN) mm.c:964:d80 Attempt to create linear p.t. with write perms
    (XEN) mm.c:2389:d80 Bad type (saw 7400000000000010 != exp
    1000000000000000) for mfn 1d2e04 (pfn 1d1fb)
    (XEN) mm.c:2965:d80 Error while pinning mfn 1d2e04

Which means the domain tried to map a pagetable page RW, which would
allow it to map arbitrary memory, so Xen stopped it.  This is because
vm_unmap_ram() left some pages mapped in the vmalloc area after NFS had
finished with them, and those pages got recycled as pagetable pages
while still having these RW aliases.

Removing those mappings immediately removes the Xen-visible aliases, and
so it has no problem with those pages being reused as pagetable pages.
Deferring the TLB flush doesn't upset Xen because it can flush the TLB
itself as needed to maintain its invariants.

When unmapping a region in the vmalloc space, clear the ptes
immediately.  There's no point in deferring this because there's no
amortization benefit.

The TLBs are left dirty, and they are flushed lazily to amortize the
cost of the IPIs.

This specific motivation for this patch is an oops-causing regression
since 2.6.36 when using NFS under Xen, triggered by the NFS client's use
of vm_map_ram() introduced in 56e4ebf877 ("NFS: readdir with vmapped
pages") .  XFS also uses vm_map_ram() and could cause similar problems.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Bryan Schumaker <bjschuma@netapp.com>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Cc: Alex Elder <aelder@sgi.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-12-02 14:51:15 -08:00
..
debugfs.c llseek: automatically add .llseek fop 2010-10-15 15:53:27 +02:00
debugfs.h xen: add debugfs support 2008-08-21 13:52:58 +02:00
enlighten.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen 2010-11-25 08:35:53 +09:00
grant-table.c xen: make grant table arch portable 2008-04-24 23:57:32 +02:00
irq.c x86: Move irq_init to x86_init_ops 2009-08-31 09:35:45 +02:00
Kconfig Merge branch 'stable/xen-pcifront-0.8.2' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen 2010-10-28 17:11:17 -07:00
Makefile Merge branch 'stable/xen-swiotlb-0.8.6' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen 2010-08-12 09:09:41 -07:00
mmu.c vmalloc: eagerly clear ptes on vunmap 2010-12-02 14:51:15 -08:00
mmu.h xen: make install_p2mtop_page() static 2010-10-22 12:57:23 -07:00
multicalls.c x86, xen: do multicall callbacks with interrupts disabled 2009-02-16 08:56:41 +01:00
multicalls.h Merge commit 'v2.6.29-rc4' into core/percpu 2009-02-09 14:58:11 +01:00
pci-swiotlb-xen.c Merge branch 'stable/xen-pcifront-0.8.2' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen 2010-10-28 17:11:17 -07:00
platform-pci-unplug.c xen: pvhvm: rename xen_emul_unplug=ignore to =unnnecessary 2010-08-23 11:59:29 +01:00
setup.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen 2010-11-25 08:35:53 +09:00
smp.c Merge branch 'stable/xen-pcifront-0.8.2' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen 2010-10-28 17:11:17 -07:00
spinlock.c Fix IRQ flag handling naming 2010-10-07 14:08:55 +01:00
suspend.c x86: Use xen_vcpuop_clockevent, xen_clocksource and xen wallclock. 2010-07-26 23:13:25 -07:00
time.c xen: do not initialize PV timers on HVM if !xen_have_vector_callback 2010-10-05 13:39:23 +01:00
vdso.h
xen-asm_32.S percpu: remove per_cpu__ prefix. 2009-10-29 22:34:15 +09:00
xen-asm_64.S xen: use iret for return from 64b kernel to 32b usermode 2009-12-03 11:14:54 -08:00
xen-asm.h xen: make direct versions of irq_enable/disable/save/restore to common code 2009-02-04 16:59:04 -08:00
xen-asm.S x86: style cleanups for xen assemblies 2009-02-05 20:25:41 +01:00
xen-head.S x86: use _types.h headers in asm where available 2009-02-13 11:35:01 -08:00
xen-ops.h xen: add support for PAT 2010-10-22 12:57:31 -07:00