Files
msm-5.15/include/linux
Kirill A. Shutemov dc6c9a35b6 mm: account pmd page tables to the process
Dave noticed that unprivileged process can allocate significant amount of
memory -- >500 MiB on x86_64 -- and stay unnoticed by oom-killer and
memory cgroup.  The trick is to allocate a lot of PMD page tables.  Linux
kernel doesn't account PMD tables to the process, only PTE.

The use-cases below use few tricks to allocate a lot of PMD page tables
while keeping VmRSS and VmPTE low.  oom_score for the process will be 0.

	#include <errno.h>
	#include <stdio.h>
	#include <stdlib.h>
	#include <unistd.h>
	#include <sys/mman.h>
	#include <sys/prctl.h>

	#define PUD_SIZE (1UL << 30)
	#define PMD_SIZE (1UL << 21)

	#define NR_PUD 130000

	int main(void)
	{
		char *addr = NULL;
		unsigned long i;

		prctl(PR_SET_THP_DISABLE);
		for (i = 0; i < NR_PUD ; i++) {
			addr = mmap(addr + PUD_SIZE, PUD_SIZE, PROT_WRITE|PROT_READ,
					MAP_ANONYMOUS|MAP_PRIVATE, -1, 0);
			if (addr == MAP_FAILED) {
				perror("mmap");
				break;
			}
			*addr = 'x';
			munmap(addr, PMD_SIZE);
			mmap(addr, PMD_SIZE, PROT_WRITE|PROT_READ,
					MAP_ANONYMOUS|MAP_PRIVATE|MAP_FIXED, -1, 0);
			if (addr == MAP_FAILED)
				perror("re-mmap"), exit(1);
		}
		printf("PID %d consumed %lu KiB in PMD page tables\n",
				getpid(), i * 4096 >> 10);
		return pause();
	}

The patch addresses the issue by account PMD tables to the process the
same way we account PTE.

The main place where PMD tables is accounted is __pmd_alloc() and
free_pmd_range(). But there're few corner cases:

 - HugeTLB can share PMD page tables. The patch handles by accounting
   the table to all processes who share it.

 - x86 PAE pre-allocates few PMD tables on fork.

 - Architectures with FIRST_USER_ADDRESS > 0. We need to adjust sanity
   check on exit(2).

Accounting only happens on configuration where PMD page table's level is
present (PMD is not folded).  As with nr_ptes we use per-mm counter.  The
counter value is used to calculate baseline for badness score by
oom-killer.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reported-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Reviewed-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: David Rientjes <rientjes@google.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-02-11 17:06:04 -08:00
..
2014-09-22 16:48:47 +09:00
2014-12-02 16:20:59 -08:00
2014-11-24 17:24:08 -05:00
2014-12-15 18:19:03 -08:00
2014-11-25 20:11:12 -08:00
2015-01-29 10:49:28 +02:00
2014-12-18 19:08:10 -08:00
2014-12-19 22:55:06 +01:00
2015-01-21 19:21:30 +01:00
2014-11-10 09:27:30 -07:00
2014-11-04 17:34:15 -08:00
2014-12-31 13:06:50 -05:00
2014-10-09 11:35:48 +03:00
2014-10-08 16:01:41 -04:00
2015-02-10 16:45:56 -08:00
2015-01-27 11:09:13 +01:00
2014-09-29 15:37:01 -04:00
2015-01-15 10:34:54 +01:00
2015-01-15 10:34:54 +01:00
2014-12-18 09:39:51 +01:00
2015-01-04 23:11:43 -05:00
2014-12-17 08:26:51 -05:00
2014-10-09 22:25:58 -04:00
2015-02-11 17:06:02 -08:00
2014-10-24 00:14:36 +02:00
2014-11-25 16:38:32 -05:00
2014-11-25 16:38:32 -05:00
2014-11-04 13:18:52 -07:00
2014-11-04 13:29:38 +00:00
2015-01-06 11:01:13 -08:00
2014-08-08 15:57:26 -07:00
2014-08-08 15:57:31 -07:00
2015-01-22 15:10:56 +01:00
2015-01-03 14:32:57 -05:00
2014-08-08 15:57:24 -07:00
2014-11-06 14:57:27 -08:00
2014-09-23 21:40:48 -07:00
2014-11-28 16:08:16 +01:00
2014-09-16 15:02:55 -06:00
2014-12-09 12:05:24 +02:00
2014-12-15 23:49:28 +02:00