linux-kernel - Re: [RFC PATCH v2 00/20] arm64: livepatch: Use ORC for dynamic frame pointer validation

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: <0f673713-ce37-1328-59da-2d623a6627e9@linux.microsoft.com>
Date:   Wed, 15 Jun 2022 15:50:42 -0500
From:   "Madhavan T. Venkataraman" <madvenka@...ux.microsoft.com>
To:     Mark Rutland <mark.rutland@....com>,
        "Ivan T. Ivanov" <iivanov@...e.de>
Cc:     jpoimboe@...hat.com, peterz@...radead.org, chenzhongjin@...wei.com,
        broonie@...nel.org, nobuta.keiya@...itsu.com,
        sjitindarsingh@...il.com, catalin.marinas@....com, will@...nel.org,
        jamorris@...ux.microsoft.com, linux-arm-kernel@...ts.infradead.org,
        live-patching@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH v2 00/20] arm64: livepatch: Use ORC for dynamic frame
 pointer validation



On 6/15/22 08:37, Mark Rutland wrote:
> On Wed, Jun 15, 2022 at 03:18:44PM +0300, Ivan T. Ivanov wrote:
>> I have run following [1] livepatch tests on kernel build from your repository.
>> Overall results looks good, but when I run klp_tc_13.shI there is something which
>> I still can not understand completely. It is because of kaslr.
>>
>> [   36.817617] livepatch: enabling patch 'klp_tc_13_livepatch'
>> [   36.819602] branch_imm_common: offset out of range
>> [   36.820113] branch_imm_common: offset out of range
>> [   36.820643] ------------[ ftrace bug ]------------
>> [   36.821172] ftrace failed to modify
>> [   36.821173] [<ffffdde931176804>] orig_do_read_active_livepatch_id+0x4/0xa8 [klp_test_support_mod]
>> [   36.822465]  actual:   e4:01:00:94
>> [   36.822821] Updating ftrace call site to call a different ftrace function
>> [   36.823537] ftrace record flags: e4000002
>> [   36.823953]  (2) R
>> [   36.823953]  expected tramp: ffffdde96882e224
>> [   36.824619] ------------[ cut here ]------------
>> [   36.825125] WARNING: CPU: 0 PID: 950 at kernel/trace/ftrace.c:2085 ftrace_bug+0x98/0x280
>> [   36.826027] Modules linked in: klp_tc_13_livepatch(OK+) klp_test_support_mod(O) crct10dif_ce
>> [   36.826943] CPU: 0 PID: 950 Comm: insmod Tainted: G           O  K   5.18.0-rc1-00020-g1ffee6fdcfda #39
>> [   36.827987] Hardware name: linux,dummy-virt (DT)
>> [   36.828546] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
>> [   36.829348] pc : ftrace_bug+0x98/0x280
>> [   36.829790] lr : ftrace_bug+0x228/0x280
>> [   36.830224] sp : ffff8000084038e0
>> [   36.830620] x29: ffff8000084038e0 x28: ffff00000485a920 x27: ffffdde931176804
>> [   36.831419] x26: ffffdde93117d1a0 x25: ffff00000485a900 x24: ffffdde96aea1000
>> [   36.832226] x23: 0000000000000000 x22: 0000000000000001 x21: ffffdde96a124da0
>> [   36.833024] x20: ffff0000045620f0 x19: ffffdde96b54a358 x18: ffffffffffffffff
>> [   36.833818] x17: 5b20386178302f34 x16: 78302b64695f6863 x15: ffffdde96a3078f8
>> [   36.834621] x14: 0000000000000000 x13: 3432326532383836 x12: ffffdde96ae9b3d8
>> [   36.835425] x11: 0000000000000001 x10: 0000000000000001 x9 : ffffdde96891305c
>> [   36.836221] x8 : c0000000ffffefff x7 : 0000000000017fe8 x6 : ffffdde96ae83398
>> [   36.837023] x5 : 0000000000057fa8 x4 : 0000000000000000 x3 : 00000000ffffefff
>> [   36.837836] x2 : e32439832ffb9700 x1 : 0000000000000000 x0 : 0000000000000022
>> [   36.838635] Call trace:
>> [   36.838923]  ftrace_bug+0x98/0x280
>> [   36.839319]  ftrace_replace_code+0xa0/0xb8
>> [   36.839768]  ftrace_modify_all_code+0xc0/0x160
>> [   36.840273]  arch_ftrace_update_code+0x14/0x20
>> [   36.840780]  ftrace_run_update_code+0x24/0x78
>> [   36.841283]  ftrace_startup_enable+0x50/0x60
>> [   36.841781]  ftrace_startup+0xb4/0x178
>> [   36.842214]  register_ftrace_function+0x68/0x88
>> [   36.842738]  klp_patch_object+0x1c8/0x330
>> [   36.843196]  klp_enable_patch+0x468/0x828
> 
> IIUC that splat specifically is due to ftrace_modify_call() missing module PLT lookups.
> 
> That should be fixed by:
> 
>   https://lore.kernel.org/all/20220614080944.1349146-3-mark.rutland@arm.com/
> 
> I have not looked at the rest of this series (yet).
> 
> Thanks,
> Mark.

Thank you so much, Mark!

I will include this in my patch series until it is upstream.

Madhavan