[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJ+HfNhnmvTHsE7cgi8pROjprsYvCZuzZDmYmpiUx4jqtHboUw@mail.gmail.com>
Date: Sun, 10 Jun 2018 11:31:55 +0200
From: Björn Töpel <bjorn.topel@...il.com>
To: penguin-kernel@...ove.sakura.ne.jp
Cc: syzbot+4abadc5d69117b346506@...kaller.appspotmail.com,
Björn Töpel <bjorn.topel@...el.com>,
"Karlsson, Magnus" <magnus.karlsson@...el.com>,
David Miller <davem@...emloft.net>,
LKML <linux-kernel@...r.kernel.org>,
Netdev <netdev@...r.kernel.org>, syzkaller-bugs@...glegroups.com
Subject: Re: WARNING: kmalloc bug in xdp_umem_create
Den sön 10 juni 2018 kl 04:53 skrev Tetsuo Handa
<penguin-kernel@...ove.sakura.ne.jp>:
>
> On 2018/06/10 7:47, syzbot wrote:
> > Hello,
> >
> > syzbot found the following crash on:
> >
> > HEAD commit: 7d3bf613e99a Merge tag 'libnvdimm-for-4.18' of git://git.k..
> > git tree: upstream
> > console output: https://syzkaller.appspot.com/x/log.txt?x=1073f68f800000
> > kernel config: https://syzkaller.appspot.com/x/.config?x=f04d8d0a2afb789a
> > dashboard link: https://syzkaller.appspot.com/bug?extid=4abadc5d69117b346506
> > compiler: gcc (GCC) 8.0.1 20180413 (experimental)
> > syzkaller repro:https://syzkaller.appspot.com/x/repro.syz?x=13c9756f800000
> > C reproducer: https://syzkaller.appspot.com/x/repro.c?x=16366f9f800000
> >
> > IMPORTANT: if you fix the bug, please add the following tag to the commit:
> > Reported-by: syzbot+4abadc5d69117b346506@...kaller.appspotmail.com
> >
> > random: sshd: uninitialized urandom read (32 bytes read)
> > random: sshd: uninitialized urandom read (32 bytes read)
> > random: sshd: uninitialized urandom read (32 bytes read)
> > random: sshd: uninitialized urandom read (32 bytes read)
> > random: sshd: uninitialized urandom read (32 bytes read)
> > WARNING: CPU: 1 PID: 4537 at mm/slab_common.c:996 kmalloc_slab+0x56/0x70 mm/slab_common.c:996
> > Kernel panic - not syncing: panic_on_warn set ...
>
> syzbot gave up upon kmalloc(), but actually error handling path has
> NULL pointer dereference bug.
>
Thanks Tetsuo! This crash has been fixed by Daniel Borkmann in commit
c09290c56376 ("bpf, xdp: fix crash in xdp_umem_unaccount_pages").
Björn
> ----------
> #include <sys/socket.h>
> #include <unistd.h>
> #define PF_XDP 44
> #define SOL_XDP 283
> #define XDP_UMEM_REG 4
>
> int main(int argc, char *argv[])
> {
> int fd = socket(PF_XDP, SOCK_RAW, 0);
> struct xdp_umem_reg {
> unsigned long long addr;
> unsigned long long len;
> unsigned int chunk_size;
> unsigned int headroom;
> } arg = {
> 0x20000000,
> 0x200002000,
> 0x800,
> 2
> };
> setsockopt(fd, SOL_XDP, XDP_UMEM_REG, &arg, sizeof(arg));
> return 0;
> }
> ----------
>
> [ 95.172962] WARNING: CPU: 3 PID: 2891 at mm/page_alloc.c:4065 __alloc_pages_nodemask+0x283/0xdf0
> [ 95.175179] Modules linked in: pcspkr sg vmw_vmci i2c_piix4 sd_mod ata_generic pata_acpi ahci libahci vmwgfx drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm drm ata_piix mptspi scsi_transport_spi i2c_core mptscsih e1000 mptbase libata serio_raw
> [ 95.180614] CPU: 3 PID: 2891 Comm: a.out Kdump: loaded Not tainted 4.17.0+ #421
> [ 95.182351] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 05/19/2017
> [ 95.184909] RIP: 0010:__alloc_pages_nodemask+0x283/0xdf0
> [ 95.186319] Code: 00 00 04 00 41 0f 44 c6 48 3b 5c 24 78 c6 84 24 90 00 00 00 00 0f 85 50 0b 00 00 41 83 fd 0a 76 1d f6 c4 02 0f 85 3b ff ff ff <0f> 0b e9 34 ff ff ff 0f 0b 0f 1f 40 00 e9 10 fe ff ff 0f 0b 89 c2
> [ 95.190997] RSP: 0018:ffffc900008efd20 EFLAGS: 00010246
> [ 95.192257] RAX: 000000000060c0c0 RBX: 0000000000000000 RCX: ffff88013f7fe920
> [ 95.194005] RDX: 0000000000000000 RSI: 0000000000000002 RDI: 0000000000000000
> [ 95.195697] RBP: 000000000060c0c0 R08: 0000000000000001 R09: ffffffffffffef81
> [ 95.197393] R10: 000000000000000d R11: 0000000000000e8c R12: 0000000000000001
> [ 95.199084] R13: 000000000000000d R14: 000000000060c0c0 R15: 0000000000000000
> [ 95.200735] FS: 00007f8387e61740(0000) GS:ffff88013f4c0000(0000) knlGS:0000000000000000
> [ 95.203441] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 95.205726] CR2: 0000000020000040 CR3: 0000000133e2c006 CR4: 00000000001606e0
> [ 95.207743] Call Trace:
> [ 95.208427] ? __lock_acquire+0x22a/0x1830
> [ 95.209391] ? kmalloc_order+0x15/0x60
> [ 95.210266] ? __kmalloc+0x20a/0x210
> [ 95.211104] ? xdp_umem_create+0x16e/0x3c0
> [ 95.212095] ? xsk_setsockopt+0x153/0x1a0
> [ 95.213143] ? __sys_setsockopt+0x67/0xb0
> [ 95.214058] ? __x64_sys_setsockopt+0x1b/0x20
> [ 95.215040] ? do_syscall_64+0x4f/0x1f0
> [ 95.215890] ? entry_SYSCALL_64_after_hwframe+0x49/0xbe
> [ 95.217079] irq event stamp: 5296
> [ 95.217785] hardirqs last enabled at (5295): [<ffffffff810b2a77>] __raw_spin_lock_init+0x17/0x50
> [ 95.220381] hardirqs last disabled at (5296): [<ffffffff81800f33>] error_entry+0x73/0xc0
> [ 95.222447] softirqs last enabled at (5284): [<ffffffff81a00183>] __do_softirq+0x183/0x204
> [ 95.224328] softirqs last disabled at (5277): [<ffffffff81061bcd>] irq_exit+0xcd/0xf0
> [ 95.226065] ---[ end trace 75b6f67917663997 ]---
> [ 95.227250] BUG: unable to handle kernel NULL pointer dereference at 0000000000000060
> [ 95.229101] PGD 1342eb067 P4D 1342eb067 PUD 1314a2067 PMD 0
> [ 95.230398] Oops: 0002 [#1] SMP DEBUG_PAGEALLOC
> [ 95.231418] CPU: 3 PID: 2891 Comm: a.out Kdump: loaded Tainted: G W 4.17.0+ #421
> [ 95.233474] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 05/19/2017
> [ 95.236636] RIP: 0010:xdp_umem_create+0x228/0x3c0
> [ 95.237867] Code: f4 ff ff ff e8 b9 f9 ff ff 48 8b bb 90 00 00 00 e8 3d d9 a7 ff 48 c7 83 90 00 00 00 00 00 00 00 48 8b 43 30 8b 93 98 00 00 00 <f0> 48 29 50 60 48 8b 7b 30 49 63 ec e8 57 10 92 ff 48 8b 7b 38 e8
> [ 95.241945] RSP: 0018:ffffc900008efe88 EFLAGS: 00010246
> [ 95.243236] RAX: 0000000000000000 RBX: ffff880133401288 RCX: 000000000060c0c0
> [ 95.244789] RDX: 0000000000200002 RSI: 0000000001000010 RDI: 0000000000000000
> [ 95.247382] RBP: 0000000000200002 R08: 0000000000000001 R09: ffffffffffffef81
> [ 95.249735] R10: 000000000000000d R11: 0000000000000e8c R12: 00000000fffffff4
> [ 95.252391] R13: 0000000000000040 R14: 0000000020000000 R15: 00000000000007c0
> [ 95.255280] FS: 00007f8387e61740(0000) GS:ffff88013f4c0000(0000) knlGS:0000000000000000
> [ 95.257918] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 95.260068] CR2: 0000000000000060 CR3: 0000000133e2c006 CR4: 00000000001606e0
> [ 95.262535] Call Trace:
> [ 95.263900] ? xsk_setsockopt+0x153/0x1a0
> [ 95.265495] ? __sys_setsockopt+0x67/0xb0
> [ 95.267108] ? __x64_sys_setsockopt+0x1b/0x20
> [ 95.269532] ? do_syscall_64+0x4f/0x1f0
> [ 95.271474] ? entry_SYSCALL_64_after_hwframe+0x49/0xbe
> [ 95.273292] Modules linked in: pcspkr sg vmw_vmci i2c_piix4 sd_mod ata_generic pata_acpi ahci libahci vmwgfx drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm drm ata_piix mptspi scsi_transport_spi i2c_core mptscsih e1000 mptbase libata serio_raw
> [ 95.279548] CR2: 0000000000000060
> [ 95.281044] ---[ end trace 75b6f67917663998 ]---
> [ 95.283132] RIP: 0010:xdp_umem_create+0x228/0x3c0
> [ 95.285257] Code: f4 ff ff ff e8 b9 f9 ff ff 48 8b bb 90 00 00 00 e8 3d d9 a7 ff 48 c7 83 90 00 00 00 00 00 00 00 48 8b 43 30 8b 93 98 00 00 00 <f0> 48 29 50 60 48 8b 7b 30 49 63 ec e8 57 10 92 ff 48 8b 7b 38 e8
> [ 95.291487] RSP: 0018:ffffc900008efe88 EFLAGS: 00010246
> [ 95.293429] RAX: 0000000000000000 RBX: ffff880133401288 RCX: 000000000060c0c0
> [ 95.295761] RDX: 0000000000200002 RSI: 0000000001000010 RDI: 0000000000000000
> [ 95.298072] RBP: 0000000000200002 R08: 0000000000000001 R09: ffffffffffffef81
> [ 95.300403] R10: 000000000000000d R11: 0000000000000e8c R12: 00000000fffffff4
> [ 95.303699] R13: 0000000000000040 R14: 0000000020000000 R15: 00000000000007c0
> [ 95.306178] FS: 00007f8387e61740(0000) GS:ffff88013f4c0000(0000) knlGS:0000000000000000
> [ 95.308645] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 95.310782] CR2: 0000000000000060 CR3: 0000000133e2c006 CR4: 00000000001606e0
>
> xdp_umem_create+0x228/0x3c0:
> arch_atomic64_sub at arch/x86/include/asm/atomic64_64.h:60
> (inlined by) atomic64_sub at include/asm-generic/atomic-instrumented.h:145
> (inlined by) atomic_long_sub at include/asm-generic/atomic-long.h:199
> (inlined by) xdp_umem_unaccount_pages at net/xdp/xdp_umem.c:135
> (inlined by) xdp_umem_reg at net/xdp/xdp_umem.c:334
> (inlined by) xdp_umem_create at net/xdp/xdp_umem.c:349
Powered by blists - more mailing lists