LSK Boot test failed: fvp_base_cortex-a57x4-a53x4 at cpuidle_driver_ref

Bug #1304028 reported by Naresh Kamboju
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Linaro Stable Kernel (LSK)
Fix Committed
Undecided
Unassigned

Bug Description

Boot test failed:
- linaro-android_armv8-lsk-pre-merge-ci on device rtsm_fvp_base_cortex-a57x4-a53x4
https://validation.linaro.org/scheduler/job/119144/log_file#L_9_184

- FVP Base models
- ARMv8-LSK-Android-pre-merge-ci build

Kernel back trace:
---------------------------
[ 14.514316] binder: 1744:1744 ioctl c0186201 fffef9c0 returned -22
 [ 14.514479] binder: 1744:1744 ioctl c0186201 fffef8e8 returned -22
 [ 14.615182] Unable to handle kernel NULL pointer dereference at virtual address 00000010
 [ 14.615238] pgd = ffffffc87f1d9000
 [ 14.615282] [00000010] *pgd=0000000000000000
 [ 14.615336] Internal error: Oops: 96000005 [#1] SMP
 [ 14.615377] Modules linked in: gator
 [ 14.615457] CPU: 1 PID: 1754 Comm: keystore Not tainted 3.10.36-00191-g14a31bf #1
 [ 14.615526] task: ffffffc87f445500 ti: ffffffc87ed5c000 task.ti: ffffffc87ed5c000
 [ 14.615597] PC is at cpuidle_driver_ref+0x24/0x4c
 [ 14.615660] LR is at cpuidle_driver_ref+0x18/0x4c

...

[ 14.625129] [<ffffffc0004536dc>] cpuidle_driver_ref+0x24/0x4c
 [ 14.625201] [<ffffffc0000cef50>] hmp_idle_pull+0x354/0x4b0
 [ 14.625274] [<ffffffc0000d2b5c>] idle_balance+0x14c/0x188
 [ 14.625351] [<ffffffc00058cd54>] __schedule+0x63c/0x76c
 [ 14.625426] [<ffffffc00058cea8>] schedule+0x24/0x68
 [ 14.625502] [<ffffffc00058d150>] io_schedule+0x5c/0x90
 [ 14.625574] [<ffffffc000127194>] sleep_on_page_killable+0x8/0x48
 [ 14.625650] [<ffffffc00058b3e4>] __wait_on_bit+0x98/0xd0
 [ 14.625726] [<ffffffc000129638>] wait_on_page_bit_killable+0x78/0x8c
 [ 14.625803] [<ffffffc0001296f4>] __lock_page_or_retry+0xa8/0xcc
 [ 14.625879] [<ffffffc000129a98>] filemap_fault+0x380/0x3a8
 [ 14.625951] [<ffffffc0001464dc>] __do_fault+0x68/0x538
 [ 14.626026] [<ffffffc000148ff0>] handle_pte_fault+0x6c/0x840
 [ 14.626103] [<ffffffc00014a778>] handle_mm_fault+0xec/0x180
 [ 14.626176] [<ffffffc0000912f0>] do_page_fault+0x208/0x35c
 [ 14.626243] [<ffffffc000081118>] do_mem_abort+0x38/0x9c
 [ 14.626303] Exception stack(0xffffffc87ed5fe30 to 0xffffffc87ed5ff50)

Linux kernel Version:
Linux version 3.10.36-00191-g14a31bf (jenkins-build@ip-10-40-85-170) (gcc version 4.8.3 20131111 (prerelease) (Linaro GCC 4.8-2013.11) ) #1 SMP Fri Apr 4 22:19:16 UTC 2014

Android Build:
https://android-build.linaro.org/builds/~linaro-android/armv8-lsk-pre-merge-ci/#build=62

Tags: armv8 lsk
Revision history for this message
Naresh Kamboju (naresh-kamboju) wrote :

linaro-android_armv8-lsk-pre-merge-ci_lava_job log

Revision history for this message
Alex Shi (alex-shi) wrote :

Naresh,

I don't know the last commit for this bug, no commit '14a31bf' exists in current lsk-test branch.
And can not reproduce this bug on latest lsk-test branch:

commit 92a58eb83599bbab42364d3a11ec9672411a6257
Merge: b31c695 270cfc4
Author: Mark Brown <email address hidden>
Date: Fri Apr 4 21:39:29 2014 +0100

    Merge branch 'lsk-v3.10-pinctrl' into linux-linaro-lsk

Can you find out and confirm which commit is bad? and how to reproduce this bug?

Thanks!

Revision history for this message
Naresh Kamboju (naresh-kamboju) wrote :

As per the pinned manifest from the build.
http://snapshots.linaro.org/android/~linaro-android/armv8-lsk-pre-merge-ci/62/

Linux tree, branch and commit.
<project groups="vexpress-lsk" name="kernel/linux-linaro-stable" path="kernel/linaro/vexpress-lsk" remote="linaro-other" revision="14a31bfd6672fe6b6c9b152cdafa71863f937504" upstream="linux-linaro-lsk-android-test"/>

Revision history for this message
Naresh Kamboju (naresh-kamboju) wrote :
description: updated
Revision history for this message
Tixy (Jon Medhurst) (tixy) wrote :

This is cased by cpuidle_driver_ref() not coping with there being no driver present. This was fixed in linus 3.12 by commit 3b9c10e98021e1f92e6f8c7ce1778b86ba68db10 (cpuidle: Check the result of cpuidle_get_driver() against NULL)

After cherry-picking that, the ARMv8 base model boots OK for me.

Revision history for this message
Mark Brown (broonie) wrote :

I've pulled a revert of the entire set of HMP updates from Tixy following additional discussion.

Will propose the above fix for LTS anyway, and if it doesn't get in pick it for LSK - Tixy, please nag me if this doesn't actually happen.

Changed in linaro-stable-kernel:
status: New → Fix Committed
tags: added: armv8 lsk
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.