From: Aaro Koskinen <aaro.koskinen@xxxxxx> Date: Mon, 14 Apr 2014 21:43:53 +0300 > Just for the archives, I got one of these again with 3.14: > > [68674.536190] ------------[ cut here ]------------ > [68674.590467] WARNING: CPU: 0 PID: 14600 at /home/aaro/los/work/shared/linux-v3.14/mm/mmap.c:2738 exit_mmap+0x138/0x160() > [68674.719635] Modules linked in: > [68674.756022] CPU: 0 PID: 14600 Comm: rm Not tainted 3.14.0-ultra-los_0a2b #1 > [68674.839349] Call Trace: > [68674.868507] [00000000004b9c78] exit_mmap+0x138/0x160 > [68674.928931] [00000000004503cc] mmput+0x2c/0xc0 > [68674.983103] [0000000000452e98] do_exit+0x1b8/0x800 > [68675.041409] [000000000045406c] do_group_exit+0x2c/0xa0 > [68675.103897] [00000000004540f4] SyS_exit_group+0x14/0x20 > [68675.167439] [0000000000406074] linux_sparc_syscall32+0x34/0x60 > [68675.238258] ---[ end trace 8a52741fbdb89d8e ]--- > [68675.293470] BUG: Bad rss-counter state mm:ffffff001df3d900 idx:1 val:1 Yes, I have reports of this going back several releases and I started trying to figure out what causes this. I suspect there is something that runs during exit_mmap() that indirectly faults in new pages, and that's how the rss-counter ends up being non-zero at the end of exit_mmap(). I'll let you know if I figure out exactly what the problem is. -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html