On Mon, Oct 11, 2010 at 3:25 PM, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote: > Replacement patches are a bit cruel to people who've already reviewed > the previous version. I always turn them into deltas so I can see what > was changed. It is below. Thanks Andrew. Sorry for the trouble, I'll know to avoid this next time. > How well was the new swapin path tested? Not as well as the file backed path - it's not gotten real production use yet. However the plan is that this change will be in google's next kernel update, so it will get a lot more more testing soon (starting in ~1 week). I did basic testing by dirtying an anon VMA larger that memory, then accessing it in random order while another thread runs a mmap/munmap loop, and checking that things behave as expected there (i.e. the patch allows the mmap/munmap thread to progress without waiting for the other thread swap-ins). -- Michel "Walken" Lespinasse A program is never fully debugged until the last user dies. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href