Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1753556AbdHXToo (ORCPT <rfc822;w@1wt.eu>);
        Thu, 24 Aug 2017 15:44:44 -0400
Received: from mail-vk0-f46.google.com ([209.85.213.46]:34793 "EHLO
        mail-vk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1752751AbdHXTon (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Thu, 24 Aug 2017 15:44:43 -0400
MIME-Version: 1.0
From: Sriram V <vshrirama@gmail.com>
Date: Fri, 25 Aug 2017 01:14:41 +0530
Message-ID: <CAH9_wRPWDKnoAt1adbJ1TtZuzwvS+RSZTjyKerZWp5Jf-F16zg@mail.gmail.com>
Subject: Question on debugging Alignment trap
To: linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 2261
Lines: 61

Hi,

I am debugging a random alignment trap in a 3.12 Kernel.
I am not sure what could cause the alighment trap. Has anyone seen
this issue before


[147278.696759] Alignment trap: not handling instruction e1901f9f at
[<c010ad10>]
[147278.714884] Unhandled fault: alignment exception (0x001) at 0xffffff7f
[147278.725921] Internal error: : 1 [#1] PREEMPT SMP ARM
[147278.731321] Modules linked in: omaplfb(O) pvrsrvkm(O)
pruss_remoteproc remoteproc virtio virtio_ring
[147278.741197] CPU: 0 PID: 21706 Comm: zygote Tainted: G           O
3.12.20+ #1
[147278.748893] task: ea86ebc0 ti: ec6cc000 task.ti: ec6cc000
[147278.754749] PC is at __destroy_inode+0xfc/0x14c
[147278.759676] LR is at _raw_spin_unlock+0x30/0x50
[147278.764596] pc : [<c010ad14>]    lr : [<c06d8918>]    psr: a00f0013
[147278.764596] sp : ec6cdd40  ip : ec6cdcf0  fp : ec6cdd54
[147278.777007] r10: dc0e77e0  r9 : e13b0c90  r8 : dc0e7778
[147278.782663] r7 : c06ed8f4  r6 : c06ed8f4  r5 : ece67b80  r4 : ece67b28
[147278.789705] r3 : 00000000  r2 : ec6cdcf0  r1 : 00000000  r0 : ffffff7f
[147278.796750] Flags: NzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM
Segment user


Disassembly of the vmlinux shows that it might come from inside atomic.h

c010ad08 <__destroy_inode+0xf0> e3a03000        mov     r3, #0
c010ad0c <__destroy_inode+0xf4> ee073fba        mcr     15, 0, r3,
cr7, cr10, {5}
/home/build/kernel/arch/arm/include/asm/atomic.h:100

        __asm__ __volatile__("@ atomic_sub_return\n"
c010ad10 <__destroy_inode+0xf8> e1901f9f        ldrex   r1, [r0]
c010ad14 <__destroy_inode+0xfc> e2411001        sub     r1, r1, #1
c010ad18 <__destroy_inode+0x100> e1802f91       strex   r2, r1, [r0]
c010ad1c <__destroy_inode+0x104> e3320000       teq     r2, #0
c010ad20 <__destroy_inode+0x108> 1afffffa       bne     c010ad10
<__destroy_inode+0xf8>
/home/build/kernel/arch/arm/include/asm/atomic.h:110
"       bne     1b"
        : "=&r" (result), "=&r" (tmp), "+Qo" (v->counter)
        : "r" (&v->counter), "Ir" (i)
        : "cc");

        smp_mb();
c010ad24 <__destroy_inode+0x10c> ee073fba       mcr     15, 0, r3,
cr7, cr10, {5}
posix_acl_release():


Not sure what the issue here is. I would need some help in
understanding what the problem.


Thanks for your help.

Regards,
Sriram