linux-kernel - Re: [PATCH v2 1/3] livepatch: Fix crash with !CONFIG_DEBUG_SET_MODULE

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

Message-ID: <alpine.LNX.2.00.1511061112410.23146@pobox.suse.cz>
Date:	Fri, 6 Nov 2015 11:14:06 +0100 (CET)
From:	Jiri Kosina <jikos@...nel.org>
To:	Josh Poimboeuf <jpoimboe@...hat.com>
cc:	Seth Jennings <sjenning@...hat.com>,
	Vojtech Pavlik <vojtech@...e.com>,
	linux-kernel@...r.kernel.org, live-patching@...r.kernel.org,
	"Cyril B." <cbay@...aysdata.com>,
	Rusty Russell <rusty@...tcorp.com.au>,
	Miroslav Benes <mbenes@...e.cz>
Subject: Re: [PATCH v2 1/3] livepatch: Fix crash with
 !CONFIG_DEBUG_SET_MODULE_RONX

On Thu, 5 Nov 2015, Josh Poimboeuf wrote:

> When loading a patch module on a kernel with
> !CONFIG_DEBUG_SET_MODULE_RONX, the following crash occurs:
> 
>   [  205.988776] livepatch: enabling patch 'kpatch_meminfo_string'
>   [  205.989829] BUG: unable to handle kernel paging request at ffffffffa08d2fc0
>   [  205.989863] IP: [<ffffffff8154fecb>] do_init_module+0x8c/0x1ba
>   [  205.989888] PGD 1a10067 PUD 1a11063 PMD 7bcde067 PTE 3740e161
>   [  205.989915] Oops: 0003 [#1] SMP
>   [  205.990187] CPU: 2 PID: 14570 Comm: insmod Tainted: G           O  K 4.1.12
>   [  205.990214] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.8.1-20150318_183358- 04/01/2014
>   [  205.990249] task: ffff8800374aaa90 ti: ffff8800794b8000 task.ti: ffff8800794b8000
>   [  205.990276] RIP: 0010:[<ffffffff8154fecb>]  [<ffffffff8154fecb>] do_init_module+0x8c/0x1ba
>   [  205.990307] RSP: 0018:ffff8800794bbd58  EFLAGS: 00010246
>   [  205.990327] RAX: 0000000000000000 RBX: ffffffffa08d2fc0 RCX: 0000000000000000
>   [  205.990356] RDX: 01ffff8000000080 RSI: 0000000000000000 RDI: ffffffff81a54b40
>   [  205.990382] RBP: ffff88007b4c4d80 R08: 0000000000000007 R09: 0000000000000000
>   [  205.990408] R10: 0000000000000008 R11: ffffea0001f18840 R12: 0000000000000000
>   [  205.990433] R13: 0000000000000001 R14: ffffffffa08d2fc0 R15: ffff88007bd0bc40
>   [  205.990459] FS:  00007f1128fbc700(0000) GS:ffff88007fc80000(0000) knlGS:0000000000000000
>   [  205.990488] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>   [  205.990509] CR2: ffffffffa08d2fc0 CR3: 000000002606e000 CR4: 00000000001406e0
>   [  205.990536] Stack:
>   [  205.990545]  ffff8800794bbec8 0000000000000001 ffffffffa08d3010 ffffffff810ecea9
>   [  205.990576]  ffffffff810e8e40 000000000005f360 ffff88007bd0bc50 ffffffffa08d3240
>   [  205.990608]  ffffffffa08d52c0 ffffffffa08d3210 ffff8800794bbed8 ffff8800794bbf1c
>   [  205.990639] Call Trace:
>   [  205.990651]  [<ffffffff810ecea9>] ? load_module+0x1e59/0x23a0
>   [  205.990672]  [<ffffffff810e8e40>] ? store_uevent+0x40/0x40
>   [  205.990693]  [<ffffffff810e99b5>] ? copy_module_from_fd.isra.49+0xb5/0x140
>   [  205.990718]  [<ffffffff810ed5bd>] ? SyS_finit_module+0x7d/0xa0
>   [  205.990741]  [<ffffffff81556832>] ? system_call_fastpath+0x16/0x75
>   [  205.990763] Code: f9 00 00 00 74 23 49 c7 c0 92 e1 60 81 48 8d 53 18 89 c1 4c 89 c6 48 c7 c7 f0 85 7d 81 31 c0 e8 71 fa ff ff e8 58 0e 00 00 31 f6 <c7> 03 00 00 00 00 48 89 da 48 c7 c7 20 c7 a5 81 e8 d0 ec b3 ff
>   [  205.990916] RIP  [<ffffffff8154fecb>] do_init_module+0x8c/0x1ba
>   [  205.990940]  RSP <ffff8800794bbd58>
>   [  205.990953] CR2: ffffffffa08d2fc0
> 
> With !CONFIG_DEBUG_SET_MODULE_RONX, module text and rodata pages are
> writable, and the debug_align() macro allows the module struct to share
> a page with executable text.  When klp_write_module_reloc() calls
> set_memory_ro() on the page, it effectively turns the module struct into
> a read-only structure, resulting in a page fault when load_module() does
> "mod->state = MODULE_STATE_LIVE".
> 
> Reported-by: Cyril B. <cbay@...aysdata.com>
> Tested-by: Cyril B. <cbay@...aysdata.com>
> Signed-off-by: Josh Poimboeuf <jpoimboe@...hat.com>

I've cherry-picked this one into livepatching.git#for-4.4/upstream, 
leaving 2/3 and 3/3 for further discussion with Rusty first.

Thanks,

-- 
Jiri Kosina
SUSE Labs

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/