[linux-yocto] [PATCH] Revert mm/vmstat.c: fix vmstat_update() preemption BUG

[He Zhe]

On 2018年07月25日 04:07, Bruce Ashfield wrote:
> On 2018-07-23 10:07 PM, He Zhe wrote:
>>
>>
>> On 2018年07月24日 03:21, Bruce Ashfield wrote:
>>> Which version/branches is this for ?
>>>
>>
>> This is for linux-yocto rt branches. linux-yocto-4.12 does not need this
>> fix, since the commit, see below, causing the issue have not been included.
>
> Rather than me guessing. Which versions are you looking at ? I've got
> active -stable updates for many of the linux-yocto trees, so any number
> could be impacted.

I'm looking at
git://git.yoctoproject.org/linux-yocto
v4.14/standard/preempt-rt/base
v4.14/standard/preempt-rt/intel

I want to fix
http://git.yoctoproject.org/cgit/cgit.cgi/linux-yocto/commit/?h=v4.14/standard/preempt-rt/base&id=08e9dbd5184e4e059adf1cc77b5dc08eca314a77
by porting
https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-stable-rt.git/commit/?h=v4.14-rt&id=ac7768148a10fe7e934a7a510cc067883163f115


Thanks,
Zhe


>
> Bruce
>
>>
>> git://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-stable-rt.git
>>
>> commit 08e9dbd5184e4e059adf1cc77b5dc08eca314a77
>> Author:     Steven J. Hill <steven.hill at cavium.com>
>> AuthorDate: Wed Mar 28 16:01:09 2018 -0700
>> Commit:     Greg Kroah-Hartman <gregkh at linuxfoundation.org>
>> CommitDate: Wed May 30 07:52:21 2018 +0200
>>
>>      mm/vmstat.c: fix vmstat_update() preemption BUG
>>
>>
>> Thanks,
>> Zhe
>>
>>> Given the author, I'm assuming it is for the -rt branches and also
>>> for 4.12 .. but I'd like a confirmation of that.
>>>
>>> Bruce
>>>
>>> On 2018-07-20 4:13 AM, zhe.he at windriver.com wrote:
>>>> From: Sebastian Andrzej Siewior <bigeasy at linutronix.de>
>>>>
>>>> commit 97731753d44d5efcb95b994dc952c0e8195b3e96 upstream
>>>>
>>>> This patch reverts commit c7f26ccfb2c3 ("mm/vmstat.c: fix
>>>> vmstat_update() preemption BUG").
>>>> Steven saw a "using smp_processor_id() in preemptible" message and
>>>> added a preempt_disable() section around it to keep it quiet. This is
>>>> not the right thing to do it does not fix the real problem.
>>>>
>>>> vmstat_update() is invoked by a kworker on a specific CPU. This worker
>>>> it bound to this CPU. The name of the worker was "kworker/1:1" so it
>>>> should have been a worker which was bound to CPU1. A worker which can
>>>> run on any CPU would have a `u' before the first digit.
>>>>
>>>> smp_processor_id() can be used in a preempt-enabled region as long as
>>>> the task is bound to a single CPU which is the case here. If it could
>>>> run on an arbitrary CPU then this is the problem we have an should seek
>>>> to resolve.
>>>> Not only this smp_processor_id() must not be migrated to another CPU but
>>>> also refresh_cpu_vm_stats() which might access wrong per-CPU variables.
>>>> Not to mention that other code relies on the fact that such a worker
>>>> runs on one specific CPU only.
>>>>
>>>> Therefore I revert that commit and we should look instead what broke the
>>>> affinity mask of the kworker.
>>>>
>>>> Cc: Steven J. Hill <steven.hill at cavium.com>
>>>> Cc: Tejun Heo <htejun at gmail.com>
>>>> Cc: Andrew Morton <akpm at linux-foundation.org>
>>>> Signed-off-by: Sebastian Andrzej Siewior <bigeasy at linutronix.de>
>>>> Signed-off-by: Steven Rostedt (VMware) <rostedt at goodmis.org>
>>>>
>>>> Backport this to fix following serious crash for preempt-rt kernel
>>>>
>>>> BUG: scheduling while atomic: kworker/3:1/91/0x00000002
>>>> Preemption disabled at:
>>>> [<ffffffff8437c37d>] vmstat_update+0x2d/0xb0
>>>> CPU: 3 PID: 91 Comm: kworker/3:1 Not tainted 4.14.48-rt29-yocto-preempt-rt #1
>>>> Hardware name: Intel Corporation Broadwell Client platform/Basking Ridge, BIOS BDW-E2R1.86C.0118.R01.1503110618 03/11/2015
>>>> Workqueue: mm_percpu_wq vmstat_update
>>>> Call Trace:
>>>>    dump_stack+0x4f/0x6b
>>>>    ? vmstat_update+0x2d/0xb0
>>>>    __schedule_bug.cold.24+0x7d/0x9a
>>>>    __schedule+0x45a/0x6c0
>>>>    ? task_blocks_on_rt_mutex+0x173/0x300
>>>>    schedule+0x3d/0xd0
>>>>    rt_spin_lock_slowlock_locked+0x118/0x2a0
>>>>    rt_spin_lock_slowlock+0x48/0x60
>>>>    rt_spin_lock+0x3f/0x50
>>>>    queue_delayed_work_on+0x63/0x100
>>>>    vmstat_update+0x6e/0xb0
>>>>    process_one_work+0x1c3/0x430
>>>>    worker_thread+0x32/0x440
>>>>    kthread+0x127/0x140
>>>>    ? process_one_work+0x430/0x430
>>>>    ? kthread_create_on_node+0x40/0x40
>>>>    ret_from_fork+0x35/0x40
>>>>
>>>> Signed-off-by: He Zhe <zhe.he at windriver.com>
>>>> ---
>>>>    mm/vmstat.c | 2 --
>>>>    1 file changed, 2 deletions(-)
>>>>
>>>> diff --git a/mm/vmstat.c b/mm/vmstat.c
>>>> index 84fd6eb..0d17b8f 100644
>>>> --- a/mm/vmstat.c
>>>> +++ b/mm/vmstat.c
>>>> @@ -1782,11 +1782,9 @@ static void vmstat_update(struct work_struct *w)
>>>>             * to occur in the future. Keep on running the
>>>>             * update worker thread.
>>>>             */
>>>> -        preempt_disable();
>>>>            queue_delayed_work_on(smp_processor_id(), mm_percpu_wq,
>>>>                    this_cpu_ptr(&vmstat_work),
>>>>                    round_jiffies_relative(sysctl_stat_interval));
>>>> -        preempt_enable();
>>>>        }
>>>>    }
>>>>   
>>>
>>>
>>
>
>