summaryrefslogtreecommitdiff
path: root/arch/tile/mm
diff options
context:
space:
mode:
authorPeter Zijlstra <a.p.zijlstra@chello.nl>2010-10-27 15:32:58 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2010-10-27 18:03:05 -0700
commit20273941f2129aa5a432796d98a276ed73d60782 (patch)
tree85da0d951ac10f239f81ad7f69559fdeb692095b /arch/tile/mm
parenta8e23a291852cd7c4fb5ca696dbb93912185ad10 (diff)
mm: fix race in kunmap_atomic()
Christoph reported a nice splat which illustrated a race in the new stack based kmap_atomic implementation. The problem is that we pop our stack slot before we're completely done resetting its state -- in particular clearing the PTE (sometimes that's CONFIG_DEBUG_HIGHMEM). If an interrupt happens before we actually clear the PTE used for the last slot, that interrupt can reuse the slot in a dirty state, which triggers a BUG in kmap_atomic(). Fix this by introducing kmap_atomic_idx() which reports the current slot index without actually releasing it and use that to find the PTE and delay the _pop() until after we're completely done. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reported-by: Christoph Hellwig <hch@infradead.org> Acked-by: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'arch/tile/mm')
-rw-r--r--arch/tile/mm/highmem.c3
1 files changed, 2 insertions, 1 deletions
diff --git a/arch/tile/mm/highmem.c b/arch/tile/mm/highmem.c
index 8ef6595e162c..abb57331cf6e 100644
--- a/arch/tile/mm/highmem.c
+++ b/arch/tile/mm/highmem.c
@@ -241,7 +241,7 @@ void __kunmap_atomic(void *kvaddr)
pte_t pteval = *pte;
int idx, type;
- type = kmap_atomic_idx_pop();
+ type = kmap_atomic_idx();
idx = type + KM_TYPE_NR*smp_processor_id();
/*
@@ -252,6 +252,7 @@ void __kunmap_atomic(void *kvaddr)
BUG_ON(!pte_present(pteval) && !pte_migrating(pteval));
kmap_atomic_unregister(pte_page(pteval), vaddr);
kpte_clear_flush(pte, vaddr);
+ kmap_atomic_idx_pop();
} else {
/* Must be a lowmem page */
BUG_ON(vaddr < PAGE_OFFSET);