linux-kernel - kasan: paging percpu + kasan causes a double fault

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-next>] [day] [month] [year] [list]

Date:   Mon, 8 Jul 2019 11:05:32 -0400
From:   Dennis Zhou <dennis@...nel.org>
To:     Andrey Ryabinin <aryabinin@...tuozzo.com>,
        Alexander Potapenko <glider@...gle.com>,
        Dmitry Vyukov <dvyukov@...gle.com>
Cc:     Tejun Heo <tj@...nel.org>,
        Kefeng Wang <wangkefeng.wang@...wei.com>,
        kasan-dev@...glegroups.com, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: kasan: paging percpu + kasan causes a double fault

Hi Andrey, Alexander, and Dmitry,

It was reported to me that when percpu is ran with param
percpu_alloc=page or the embed allocation scheme fails and falls back to
page that a double fault occurs.

I don't know much about how kasan works, but a difference between the
two is that we manually reserve vm area via vm_area_register_early().
I guessed it had something to do with the stack canary or the irq_stack,
and manually mapped the shadow vm area with kasan_add_zero_shadow(), but
that didn't seem to do the trick.

RIP resolves to the fixed_percpu_data declaration.

Double fault below:
[    0.000000] PANIC: double fault, error_code: 0x0
[    0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.2.0-rc7-00007-ge0afe6d4d12c-dirty #299
[    0.000000] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014
[    0.000000] RIP: 0010:no_context+0x38/0x4b0
[    0.000000] Code: df 41 57 41 56 4c 8d bf 88 00 00 00 41 55 49 89 d5 41 54 49 89 f4 55 48 89 fd 4c8
[    0.000000] RSP: 0000:ffffc8ffffffff28 EFLAGS: 00010096
[    0.000000] RAX: dffffc0000000000 RBX: ffffc8ffffffff50 RCX: 000000000000000b
[    0.000000] RDX: fffff52000000030 RSI: 0000000000000003 RDI: ffffc90000000130
[    0.000000] RBP: ffffc900000000a8 R08: 0000000000000001 R09: 0000000000000000
[    0.000000] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000003
[    0.000000] R13: fffff52000000030 R14: 0000000000000000 R15: ffffc90000000130
[    0.000000] FS:  0000000000000000(0000) GS:ffffc90000000000(0000) knlGS:0000000000000000
[    0.000000] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    0.000000] CR2: ffffc8ffffffff18 CR3: 0000000002e0d001 CR4: 00000000000606b0
[    0.000000] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[    0.000000] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[    0.000000] Call Trace:
[    0.000000] Kernel panic - not syncing: Machine halted.
[    0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.2.0-rc7-00007-ge0afe6d4d12c-dirty #299
[    0.000000] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014
[    0.000000] Call Trace:
[    0.000000]  <#DF>
[    0.000000]  dump_stack+0x5b/0x90
[    0.000000]  panic+0x17e/0x36e
[    0.000000]  ? __warn_printk+0xdb/0xdb
[    0.000000]  ? spurious_kernel_fault_check+0x1a/0x60
[    0.000000]  df_debug+0x2e/0x39
[    0.000000]  do_double_fault+0x89/0xb0
[    0.000000]  double_fault+0x1e/0x30
[    0.000000] RIP: 0010:no_context+0x38/0x4b0
[    0.000000] Code: df 41 57 41 56 4c 8d bf 88 00 00 00 41 55 49 89 d5 41 54 49 89 f4 55 48 89 fd 4c8
[    0.000000] RSP: 0000:ffffc8ffffffff28 EFLAGS: 00010096
[    0.000000] RAX: dffffc0000000000 RBX: ffffc8ffffffff50 RCX: 000000000000000b
[    0.000000] RDX: fffff52000000030 RSI: 0000000000000003 RDI: ffffc90000000130
[    0.000000] RBP: ffffc900000000a8 R08: 0000000000000001 R09: 0000000000000000
[    0.000000] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000003
[ 0.000000] R13: fffff52000000030 R14: 0000000000000000 R15: ffffc90000000130

Thanks,
Dennis