Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: Memory hotplug softlock issue
To:     Baoquan He <bhe@redhat.com>
Cc:     linux-mm@kvack.org, linux-kernel@vger.kernel.org, mhocko@suse.com,
        akpm@linux-foundation.org, aarcange@redhat.com
References: <20181114070909.GB2653@MiWiFi-R3L-srv>
 <5a6c6d6b-ebcd-8bfa-d6e0-4312bfe86586@redhat.com>
 <20181114090042.GD2653@MiWiFi-R3L-srv>
From:   David Hildenbrand <david@redhat.com>
Openpgp: preference=signencrypt
Autocrypt: addr=david@redhat.com; prefer-encrypt=mutual; keydata=
 xsFNBFXLn5EBEAC+zYvAFJxCBY9Tr1xZgcESmxVNI/0ffzE/ZQOiHJl6mGkmA1R7/uUpiCjJ
 dBrn+lhhOYjjNefFQou6478faXE6o2AhmebqT4KiQoUQFV4R7y1KMEKoSyy8hQaK1umALTdL
 QZLQMzNE74ap+GDK0wnacPQFpcG1AE9RMq3aeErY5tujekBS32jfC/7AnH7I0v1v1TbbK3Gp
 XNeiN4QroO+5qaSr0ID2sz5jtBLRb15RMre27E1ImpaIv2Jw8NJgW0k/D1RyKCwaTsgRdwuK
 Kx/Y91XuSBdz0uOyU/S8kM1+ag0wvsGlpBVxRR/xw/E8M7TEwuCZQArqqTCmkG6HGcXFT0V9
 PXFNNgV5jXMQRwU0O/ztJIQqsE5LsUomE//bLwzj9IVsaQpKDqW6TAPjcdBDPLHvriq7kGjt
 WhVhdl0qEYB8lkBEU7V2Yb+SYhmhpDrti9Fq1EsmhiHSkxJcGREoMK/63r9WLZYI3+4W2rAc
 UucZa4OT27U5ZISjNg3Ev0rxU5UH2/pT4wJCfxwocmqaRr6UYmrtZmND89X0KigoFD/XSeVv
 jwBRNjPAubK9/k5NoRrYqztM9W6sJqrH8+UWZ1Idd/DdmogJh0gNC0+N42Za9yBRURfIdKSb
 B3JfpUqcWwE7vUaYrHG1nw54pLUoPG6sAA7Mehl3nd4pZUALHwARAQABzSREYXZpZCBIaWxk
 ZW5icmFuZCA8ZGF2aWRAcmVkaGF0LmNvbT7CwX4EEwECACgFAljj9eoCGwMFCQlmAYAGCwkI
 BwMCBhUIAgkKCwQWAgMBAh4BAheAAAoJEE3eEPcA/4Na5IIP/3T/FIQMxIfNzZshIq687qgG
 8UbspuE/YSUDdv7r5szYTK6KPTlqN8NAcSfheywbuYD9A4ZeSBWD3/NAVUdrCaRP2IvFyELj
 xoMvfJccbq45BxzgEspg/bVahNbyuBpLBVjVWwRtFCUEXkyazksSv8pdTMAs9IucChvFmmq3
 jJ2vlaz9lYt/lxN246fIVceckPMiUveimngvXZw21VOAhfQ+/sofXF8JCFv2mFcBDoa7eYob
 s0FLpmqFaeNRHAlzMWgSsP80qx5nWWEvRLdKWi533N2vC/EyunN3HcBwVrXH4hxRBMco3jvM
 m8VKLKao9wKj82qSivUnkPIwsAGNPdFoPbgghCQiBjBe6A75Z2xHFrzo7t1jg7nQfIyNC7ez
 MZBJ59sqA9EDMEJPlLNIeJmqslXPjmMFnE7Mby/+335WJYDulsRybN+W5rLT5aMvhC6x6POK
 z55fMNKrMASCzBJum2Fwjf/VnuGRYkhKCqqZ8gJ3OvmR50tInDV2jZ1DQgc3i550T5JDpToh
 dPBxZocIhzg+MBSRDXcJmHOx/7nQm3iQ6iLuwmXsRC6f5FbFefk9EjuTKcLMvBsEx+2DEx0E
 UnmJ4hVg7u1PQ+2Oy+Lh/opK/BDiqlQ8Pz2jiXv5xkECvr/3Sv59hlOCZMOaiLTTjtOIU7Tq
 7ut6OL64oAq+zsFNBFXLn5EBEADn1959INH2cwYJv0tsxf5MUCghCj/CA/lc/LMthqQ773ga
 uB9mN+F1rE9cyyXb6jyOGn+GUjMbnq1o121Vm0+neKHUCBtHyseBfDXHA6m4B3mUTWo13nid
 0e4AM71r0DS8+KYh6zvweLX/LL5kQS9GQeT+QNroXcC1NzWbitts6TZ+IrPOwT1hfB4WNC+X
 2n4AzDqp3+ILiVST2DT4VBc11Gz6jijpC/KI5Al8ZDhRwG47LUiuQmt3yqrmN63V9wzaPhC+
 xbwIsNZlLUvuRnmBPkTJwwrFRZvwu5GPHNndBjVpAfaSTOfppyKBTccu2AXJXWAE1Xjh6GOC
 8mlFjZwLxWFqdPHR1n2aPVgoiTLk34LR/bXO+e0GpzFXT7enwyvFFFyAS0Nk1q/7EChPcbRb
 hJqEBpRNZemxmg55zC3GLvgLKd5A09MOM2BrMea+l0FUR+PuTenh2YmnmLRTro6eZ/qYwWkC
 u8FFIw4pT0OUDMyLgi+GI1aMpVogTZJ70FgV0pUAlpmrzk/bLbRkF3TwgucpyPtcpmQtTkWS
 gDS50QG9DR/1As3LLLcNkwJBZzBG6PWbvcOyrwMQUF1nl4SSPV0LLH63+BrrHasfJzxKXzqg
 rW28CTAE2x8qi7e/6M/+XXhrsMYG+uaViM7n2je3qKe7ofum3s4vq7oFCPsOgwARAQABwsFl
 BBgBAgAPBQJVy5+RAhsMBQkJZgGAAAoJEE3eEPcA/4NagOsP/jPoIBb/iXVbM+fmSHOjEshl
 KMwEl/m5iLj3iHnHPVLBUWrXPdS7iQijJA/VLxjnFknhaS60hkUNWexDMxVVP/6lbOrs4bDZ
 NEWDMktAeqJaFtxackPszlcpRVkAs6Msn9tu8hlvB517pyUgvuD7ZS9gGOMmYwFQDyytpepo
 YApVV00P0u3AaE0Cj/o71STqGJKZxcVhPaZ+LR+UCBZOyKfEyq+ZN311VpOJZ1IvTExf+S/5
 lqnciDtbO3I4Wq0ArLX1gs1q1XlXLaVaA3yVqeC8E7kOchDNinD3hJS4OX0e1gdsx/e6COvy
 qNg5aL5n0Kl4fcVqM0LdIhsubVs4eiNCa5XMSYpXmVi3HAuFyg9dN+x8thSwI836FoMASwOl
 C7tHsTjnSGufB+D7F7ZBT61BffNBBIm1KdMxcxqLUVXpBQHHlGkbwI+3Ye+nE6HmZH7IwLwV
 W+Ajl7oYF+jeKaH4DZFtgLYGLtZ1LDwKPjX7VAsa4Yx7S5+EBAaZGxK510MjIx6SGrZWBrrV
 TEvdV00F2MnQoeXKzD7O4WFbL55hhyGgfWTHwZ457iN9SgYi1JLPqWkZB0JRXIEtjd4JEQcx
 +8Umfre0Xt4713VxMygW0PnQt5aSQdMD58jHFxTk092mU+yIHj5LeYgvwSgZN4airXk5yRXl
 SE+xAvmumFBY
Organization: Red Hat GmbH
Message-ID: <8c03f925-8ca4-688c-569a-a7a449612782@redhat.com>
Date:   Wed, 14 Nov 2018 10:25:57 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.3.0
MIME-Version: 1.0
In-Reply-To: <20181114090042.GD2653@MiWiFi-R3L-srv>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 14.11.18 10:00, Baoquan He wrote:
> Hi David,
> 
> On 11/14/18 at 09:18am, David Hildenbrand wrote:
>> Code seems to be waiting for the mem_hotplug_lock in read.
>> We hold mem_hotplug_lock in write whenever we online/offline/add/remove
>> memory. There are two ways to trigger offlining of memory:
>>
>> 1. Offlining via "cat offline > /sys/devices/system/memory/memory0/state"
>>
>> This always properly took the mem_hotplug_lock. Nothing changed
>>
>> 2. Offlining via "cat 0 > /sys/devices/system/memory/memory0/online"
>>
>> This didn't take the mem_hotplug_lock and I fixed that for this release.
>>
>> So if you were testing with 1., you should have seen the same error
>> before this release (unless there is something else now broken in this
>> release).
> 
> Thanks a lot for looking into this.
> 
> I triggered sysrq+t to check threads' state. You can see that we use
> firmware to trigger ACPI event to go to acpi_bus_offline(), it truly
> didn't take mem_hotplug_lock() and has taken it with your fix in
> commit 381eab4a6ee mm/memory_hotplug: fix online/offline_pages called w.o. mem_hotplug_lock
> 
> [  +0.007062] Workqueue: kacpi_hotplug acpi_hotplug_work_fn
> [  +0.005398] Call Trace:
> [  +0.002476]  ? page_vma_mapped_walk+0x307/0x710
> [  +0.004538]  ? page_remove_rmap+0xa2/0x340
> [  +0.004104]  ? ptep_clear_flush+0x54/0x60
> [  +0.004027]  ? enqueue_entity+0x11c/0x620
> [  +0.005904]  ? schedule+0x28/0x80
> [  +0.003336]  ? rmap_walk_file+0xf9/0x270
> [  +0.003940]  ? try_to_unmap+0x9c/0xf0
> [  +0.003695]  ? migrate_pages+0x2b0/0xb90
> [  +0.003959]  ? try_offline_node+0x160/0x160
> [  +0.004214]  ? __offline_pages+0x6ce/0x8e0
> [  +0.004134]  ? memory_subsys_offline+0x40/0x60
> [  +0.004474]  ? device_offline+0x81/0xb0
> [  +0.003867]  ? acpi_bus_offline+0xdb/0x140
> [  +0.004117]  ? acpi_device_hotplug+0x21c/0x460
> [  +0.004458]  ? acpi_hotplug_work_fn+0x1a/0x30
> [  +0.004372]  ? process_one_work+0x1a1/0x3a0
> [  +0.004195]  ? worker_thread+0x30/0x380
> [  +0.003851]  ? drain_workqueue+0x120/0x120
> [  +0.004117]  ? kthread+0x112/0x130
> [  +0.003411]  ? kthread_park+0x80/0x80
> [  +0.005325]  ? ret_from_fork+0x35/0x40
> 

Yes, this is indeed another code path that was fixed (and I didn't
actually realize it ;) ). Thanks for the callchain. Before my fix
hotplug still would have never succeeded (offline_pages would have
silently looped forever) as far as I can tell.

> 
>>
>>
>> The real question is, however, why offlining of the last block doesn't
>> succeed. In __offline_pages() we basically have an endless loop (while
>> holding the mem_hotplug_lock in write). Now I consider this piece of
>> code very problematic (we should automatically fail after X
>> attempts/after X seconds, we should not ignore -ENOMEM), and we've had
>> other BUGs whereby we would run into an endless loop here (e.g. related
>> to hugepages I guess).
> 
> Hmm, even though memory hotplug stalled there, there are still much
> memory. E.g in this system, it has 8 nodes and each node has 64 GB
> memory, it's 512 GB in all. Now I run "stress -m 200" to trigger 200
> processes to malloc then free 256 MB contiously, and it's eating 50 GB
> in all. In theory, it still has much memory for migrating to.

Maybe a NUMA issue? But I am just making wild guesses. Maybe it is not
-ENOMEM but just some other migration condition that is not properly
handled (see Michals reply).

> 
>>
>> You mentioned memory pressure, if our host is under memory pressure we
>> can easily trigger running into an endless loop there, because we
>> basically ignore -ENOMEM e.g. when we cannot get a page to migrate some
>> memory to be offlined. I assume this is the case here.
>> do_migrate_range() could be the bad boy if it keeps failing forever and
>> we keep retrying.
> 
> Not sure what other people think about this. If failed the memory removing
> when still much free memory left, I worry customer will complain.

Indeed, we have to look into this.

> 
> Yeah, it stoped at do_migrate_range() when try to migrate the last
> memory block. And each time it's the last memory block which can't be
> offlined and hang.

It would be interesting to see which error message we keep getting.

> 
> If any message or information needed, I can provide.
> 
> Thanks
> Baoquan
> 


-- 

Thanks,

David / dhildenb