[ 01/99] mm: fix the TLB range flushed when __tlb_remove_page() runs out of slots

From: Greg Kroah-Hartman
Date: Fri Aug 02 2013 - 06:57:54 EST

3.10-stable review patch. If anyone has any objections, please let me know.


From: Vineet Gupta <Vineet.Gupta1@xxxxxxxxxxxx>

commit e6c495a96ce02574e765d5140039a64c8d4e8c9e upstream.

zap_pte_range loops from @addr to @end. In the middle, if it runs out of
batching slots, TLB entries needs to be flushed for @start to @interim,
NOT @interim to @end.

Since ARC port doesn't use page free batching I can't test it myself but
this seems like the right thing to do.

Observed this when working on a fix for the issue at thread:

Signed-off-by: Vineet Gupta <vgupta@xxxxxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Acked-by: Catalin Marinas <catalin.marinas@xxxxxxx>
Cc: Max Filippov <jcmvbkbc@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

mm/memory.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)

--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1101,6 +1101,7 @@ static unsigned long zap_pte_range(struc
spinlock_t *ptl;
pte_t *start_pte;
pte_t *pte;
+ unsigned long range_start = addr;

@@ -1206,12 +1207,14 @@ again:
force_flush = 0;

- tlb->start = addr;
- tlb->end = end;
+ tlb->start = range_start;
+ tlb->end = addr;
- if (addr != end)
+ if (addr != end) {
+ range_start = addr;
goto again;
+ }

return addr;

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/