linux-kernel - Re: kernel BUG at /src/linux-dev/mm/mempolicy.c:1738! on v3.16-rc1

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Fri, 20 Jun 2014 15:46:39 -0400
From:	Naoya Horiguchi <n-horiguchi@...jp.nec.com>
To:	Christoph Lameter <cl@...two.org>
Cc:	linux-mm@...ck.org, linux-kernel@...r.kernel.org,
	Andrew Morton <akpm@...ux-foundation.org>,
	Hugh Dickins <hughd@...gle.com>,
	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
	Naoya Horiguchi <nao.horiguchi@...il.com>
Subject: Re: kernel BUG at /src/linux-dev/mm/mempolicy.c:1738! on v3.16-rc1

On Fri, Jun 20, 2014 at 09:24:36AM -0500, Christoph Lameter wrote:
> On Thu, 19 Jun 2014, Naoya Horiguchi wrote:
>
> > I'm suspecting that mbind_range() do something wrong around vma handling,
> > but I don't have enough luck yet. Anyone has an idea?
>
> Well memory policy data corrupted. This looks like you were trying to do
> page migration via mbind()?

Right.

> Could we get some more details as to what is
> going on here? Specifically the parameters passed to mbind would be
> interesting.

My view about the kernel behavior was in another email a few hours ago.
And as for what userspace did, I attach the reproducer below. It's simply
doing mbind(mode=MPOL_BIND, flags=MPOL_MF_MOVE_ALL) on random address/length/node.

What I did to trigger the bug is like below:

  while true ; do
      dd if=/dev/urandom of=testfile bs=4096 count=1000
      for i in $(seq 10) ; do
          ./mbind_bug_reproducer testfile > /dev/null &
      done
      sleep 3
      pkill -SIGUSR1 -f mbind_bug_reproducer
  done

mbind_bug_reproducer.c
---
#include <stdio.h>
#include <signal.h>
#include <sys/mman.h>
#include <numa.h>
#include <numaif.h>
#include <fcntl.h>

#define ADDR_INPUT 0x700000000000
#define PS 4096
#define err(x) perror(x),exit(EXIT_FAILURE)
#define errmsg(x, ...) fprintf(stderr, x, ##__VA_ARGS__),exit(EXIT_FAILURE)

int flag = 1;

void sig_handle_flag(int signo) { flag = 0; }

void set_new_nodes(struct bitmask *mask, unsigned long node) {
	numa_bitmask_clearall(mask);
	numa_bitmask_setbit(mask, node);
}

int main(int argc, char *argv[]) {
	int nr = 1000;
	int fd = -1;
	char *pfile;
	struct timeval tv;
	struct bitmask *nodes;
	unsigned long nr_nodes;
	unsigned long memsize = nr * PS;

	nr_nodes = numa_max_node() + 1; /* numa_num_possible_nodes(); */
	nodes = numa_bitmask_alloc(nr_nodes);
	if (nr_nodes < 2)
		errmsg("A minimum of 2 nodes is required for this test.\n");

	gettimeofday(&tv, NULL);
	srandom(tv.tv_usec);

	fd = open(argv[1], O_RDWR, S_IRWXU);
	if (fd < 0)
		err("open");
	pfile = mmap((void *)ADDR_INPUT, memsize, PROT_READ|PROT_WRITE,
		      MAP_SHARED, fd, 0);
	if (pfile == (void*)-1L)
		err("mmap");

	signal(SIGUSR1, sig_handle_flag);

	while (flag) {
		int node;
		unsigned long offset;
		unsigned long length;

		memset(pfile, 'a', memsize);

		node = random() % nr_nodes;
		set_new_nodes(nodes, random() & nr_nodes);
		offset = (random() % nr) * PS;
		length = (random() % (nr - offset/PS)) * PS;
		printf("[%d] node:%x, offset:%x, length:%x\n",
		       getpid(), node, offset, length);
		mbind(pfile + offset, length, MPOL_BIND, nodes->maskp,
		      nodes->size + 1, MPOL_MF_MOVE_ALL);
	}

	munmap(pfile, memsize);
	return 0;
}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/