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

------------------

From: Andrey Ryabinin <[email protected]>

commit d59b1087a98e402ed9a7cc577f4da435f9a555f5 upstream.

Calculation of dirty_ratelimit sometimes is not correct.  E.g.  initial
values of dirty_ratelimit == INIT_BW and step == 0, lead to the
following result:

   UBSAN: Undefined behaviour in ../mm/page-writeback.c:1286:7
   shift exponent 25600 is too large for 64-bit type 'long unsigned int'

The fix is straightforward - make step 0 if the shift exponent is too
big.

Signed-off-by: Andrey Ryabinin <[email protected]>
Cc: Wu Fengguang <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Andy Shevchenko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Signed-off-by: Mel Gorman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

---
 mm/page-writeback.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -1162,6 +1162,7 @@ static void wb_update_dirty_ratelimit(st
        unsigned long balanced_dirty_ratelimit;
        unsigned long step;
        unsigned long x;
+       unsigned long shift;
 
        /*
         * The dirty rate will match the writeout rate in long term, except
@@ -1286,11 +1287,11 @@ static void wb_update_dirty_ratelimit(st
         * rate itself is constantly fluctuating. So decrease the track speed
         * when it gets close to the target. Helps eliminate pointless tremors.
         */
-       step >>= dirty_ratelimit / (2 * step + 1);
-       /*
-        * Limit the tracking speed to avoid overshooting.
-        */
-       step = (step + 7) / 8;
+       shift = dirty_ratelimit / (2 * step + 1);
+       if (shift < BITS_PER_LONG)
+               step = DIV_ROUND_UP(step >> shift, 8);
+       else
+               step = 0;
 
        if (dirty_ratelimit < balanced_dirty_ratelimit)
                dirty_ratelimit += step;


Reply via email to