| 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
| ||
|
Date: Mon, 12 Oct 2015 14:02:49 +0200 From: Michal Kubecek <mkubecek@...e.cz> To: Dmitry Vyukov <dvyukov@...gle.com> Cc: Al Viro <viro@...iv.linux.org.uk>, "linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>, LKML <linux-kernel@...r.kernel.org>, David Miller <davem@...emloft.net>, Hannes Frederic Sowa <hannes@...essinduktion.org>, dhowells@...hat.com, paul@...l-moore.com, salyzyn@...roid.com, sds@...ho.nsa.gov, ying.xue@...driver.com, netdev@...r.kernel.org, syzkaller@...glegroups.com, Kostya Serebryany <kcc@...gle.com>, Alexander Potapenko <glider@...gle.com>, Andrey Konovalov <andreyknvl@...gle.com>, Sasha Levin <sasha.levin@...cle.com>, Julien Tinnes <jln@...gle.com>, Kees Cook <keescook@...gle.com>, Mathias Krause <minipli@...glemail.com>, Jason Baron <jbaron@...mai.com>, Rainer Weikusat <rweikusat@...ileactivedefense.com> Subject: Re: Use-after-free in ep_remove_wait_queue On Mon, Oct 12, 2015 at 01:07:55PM +0200, Dmitry Vyukov wrote: > Hello, > > The following program causes use-after-in kernel: > ... > long r0 = syscall(SYS_mmap, 0x20001000ul, 0x1000ul, 0x3ul, > 0x32ul, 0xfffffffffffffffful, 0x0ul); > long r1 = syscall(SYS_mmap, 0x20000000ul, 0x1000ul, 0x3ul, > 0x32ul, 0xfffffffffffffffful, 0x0ul); > long r2 = syscall(SYS_socketpair, 0x1ul, 0x3ul, 0x1ul, 0x20000ffcul); > long r3 = -1; > if (r2 != -1) > r3 = *(uint32_t*)0x20000ffc; > long r4 = -1; > if (r2 != -1) > r4 = *(uint32_t*)0x20001000; > long r5 = syscall(SYS_epoll_create, 0x1ul); > long r6 = syscall(SYS_mmap, 0x20003000ul, 0x1000ul, 0x3ul, > 0x32ul, 0xfffffffffffffffful, 0x0ul); > long r7 = syscall(SYS_dup3, r4, r3, 0x80000ul); > *(uint32_t*)0x20003000 = 0x6; > *(uint32_t*)0x20003004 = 0x2; > *(uint64_t*)0x20003008 = 0x6; > long r11 = syscall(SYS_epoll_ctl, r5, 0x1ul, r3, 0x20003000ul); > long r12 = syscall(SYS_mmap, 0x20002000ul, 0x1000ul, 0x3ul, > 0x32ul, 0xfffffffffffffffful, 0x0ul); > memcpy((void*)0x20002000, "\x00", 1); > long r14 = syscall(SYS_write, r7, 0x20002000ul, 0x1ul); > *(uint64_t*)0x20001a4d = 0x5; > long r16 = syscall(SYS_epoll_pwait, r5, 0x20004000ul, 0x1ul, > 0x1ul, 0x20001a4dul, 0x8ul); > return 0; ... > Call Trace: > [< inline >] __list_del include/linux/list.h:89 > [< inline >] list_del include/linux/list.h:107 > [< inline >] __remove_wait_queue include/linux/wait.h:145 > [<ffffffff811bfbab>] remove_wait_queue+0xfb/0x120 kernel/sched/wait.c:50 > [< inline >] ep_remove_wait_queue fs/eventpoll.c:524 > [<ffffffff8154094b>] ep_unregister_pollwait.isra.7+0x10b/0x1c0 > fs/eventpoll.c:542 > [<ffffffff81541417>] ep_free+0x97/0x190 fs/eventpoll.c:759 (discriminator 3) > [<ffffffff81541554>] ep_eventpoll_release+0x44/0x60 fs/eventpoll.c:791 > [<ffffffff814772ad>] __fput+0x21d/0x6e0 fs/file_table.c:208 > [<ffffffff814777f5>] ____fput+0x15/0x20 fs/file_table.c:244 > [<ffffffff81150094>] task_work_run+0x164/0x1f0 kernel/task_work.c:115 > (discriminator 1) > [< inline >] exit_task_work include/linux/task_work.h:21 > [<ffffffff810fe7ae>] do_exit+0xa4e/0x2d40 kernel/exit.c:746 > [<ffffffff81104426>] do_group_exit+0xf6/0x340 kernel/exit.c:874 > [< inline >] SYSC_exit_group kernel/exit.c:885 > [<ffffffff8110468d>] SyS_exit_group+0x1d/0x20 kernel/exit.c:883 > [<ffffffff82e424d1>] entry_SYSCALL_64_fastpath+0x31/0x95 > arch/x86/entry/entry_64.S:187 > > INFO: Allocated in sk_prot_alloc+0x69/0x340 age=6 cpu=1 pid=10568 > [< none >] __slab_alloc+0x426/0x470 mm/slub.c:2402 > [< inline >] slab_alloc_node mm/slub.c:2470 > [< inline >] slab_alloc mm/slub.c:2512 > [< none >] kmem_cache_alloc+0x10d/0x140 mm/slub.c:2517 > [< none >] sk_prot_alloc+0x69/0x340 net/core/sock.c:1329 > [< none >] sk_alloc+0x33/0x280 net/core/sock.c:1404 > [< none >] unix_create1+0x5e/0x3d0 net/unix/af_unix.c:648 > [< none >] unix_create+0x14d/0x1f0 net/unix/af_unix.c:707 > [< none >] __sock_create+0x1f1/0x4c0 net/socket.c:1169 > [< inline >] sock_create net/socket.c:1209 > [< inline >] SYSC_socketpair net/socket.c:1281 > [< none >] SyS_socketpair+0x10f/0x4d0 net/socket.c:1260 > [< none >] entry_SYSCALL_64_fastpath+0x31/0x95 > arch/x86/entry/entry_64.S:187 > > INFO: Freed in sk_destruct+0x2e9/0x400 age=9 cpu=1 pid=10568 > [< none >] __slab_free+0x12d/0x280 mm/slub.c:2587 (discriminator 1) > [< inline >] slab_free mm/slub.c:2736 > [< none >] kmem_cache_free+0x161/0x180 mm/slub.c:2745 > [< inline >] sk_prot_free net/core/sock.c:1374 > [< none >] sk_destruct+0x2e9/0x400 net/core/sock.c:1452 > [< none >] __sk_free+0x57/0x200 net/core/sock.c:1460 > [< none >] sk_free+0x30/0x40 net/core/sock.c:1471 > [< inline >] sock_put include/net/sock.h:1593 > [< none >] unix_dgram_sendmsg+0xeaf/0x1100 net/unix/af_unix.c:1571 > [< inline >] sock_sendmsg_nosec net/socket.c:610 > [< none >] sock_sendmsg+0xca/0x110 net/socket.c:620 > [< none >] sock_write_iter+0x216/0x3a0 net/socket.c:819 > [< inline >] new_sync_write fs/read_write.c:478 > [< none >] __vfs_write+0x2ed/0x3d0 fs/read_write.c:491 > [< none >] vfs_write+0x173/0x4a0 fs/read_write.c:538 > [< inline >] SYSC_write fs/read_write.c:585 > [< none >] SyS_write+0x108/0x220 fs/read_write.c:577 > [< none >] entry_SYSCALL_64_fastpath+0x31/0x95 > arch/x86/entry/entry_64.S:187 Probably the issue discussed in http://thread.gmane.org/gmane.linux.kernel/2057497/ and previous related threads. Michal Kubecek -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@...r.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists