On 05/13/2014 11:45 AM, Mel Gorman wrote: > The test_bit operations in get/set pageblock flags are expensive. This patch > reads the bitmap on a word basis and use shifts and masks to isolate the bits > of interest. Similarly masks are used to set a local copy of the bitmap and then > use cmpxchg to update the bitmap if there have been no other changes made in > parallel. > > In a test running dd onto tmpfs the overhead of the pageblock-related > functions went from 1.27% in profiles to 0.5%. > > Signed-off-by: Mel Gorman <mgorman@xxxxxxx> > Acked-by: Vlastimil Babka <vbabka@xxxxxxx> Hi, I've tested if this closes the race I've been previously trying to fix with the series in http://marc.info/?l=linux-mm&m=139359694028925&w=2 And indeed with this patch I wasn't able to reproduce it in my stress test (which adds lots of memory isolation calls) anymore. So thanks to Mel I can dump my series in the trashcan :P Therefore I believe something like below should be added to the changelog, and put to stable as well. Thanks, Vlastimil -----8<----- In addition to the performance benefits, this patch closes races that are possible between: a) get_ and set_pageblock_migratetype(), where get_pageblock_migratetype() reads part of the bits before and other part of the bits after set_pageblock_migratetype() has updated them. b) set_pageblock_migratetype() and set_pageblock_skip(), where the non-atomic read-modify-update set bit operation in set_pageblock_skip() will cause lost updates to some bits changed in the set_pageblock_migratetype(). Joonsoo Kim first reported the case a) via code inspection. Vlastimil Babka's testing with a debug patch showed that either a) or b) occurs roughly once per mmtests' stress-highalloc benchmark (although not necessarily in the same pageblock). Furthermore during development of unrelated compaction patches, it was observed that frequent calls to {start,undo}_isolate_page_range() the race occurs several thousands of times and has resulted in NULL pointer dereferences in move_freepages() and free_one_page() in places where free_list[migratetype] is manipulated by e.g. list_move(). Further debugging confirmed that migratetype had invalid value of 6, causing out of bounds access to the free_list array. That confirmed that the race exist, although it may be extremely rare, and currently only fatal where page isolation is performed due to memory hot remove. Races on pageblocks being updated by set_pageblock_migratetype(), where both old and new migratetype are lower MIGRATE_RESERVE, currently cannot result in an invalid value being observed, although theoretically they may still lead to unexpected creation or destruction of MIGRATE_RESERVE pageblocks. Furthermore, things could get suddenly worse when memory isolation is used more, or when new migratetypes are added. After this patch, the race has no longer been observed in testing. Reported-by: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx> Reported-and-tested-by: Vlastimil Babka <vbabka@xxxxxxx> Cc: <stable@xxxxxxxxxxxxxxx> -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>