Re: [PATCH] mm: page_alloc: don't allocate page from memoryless nodes

[David Hildenbrand <david@redhat.com>]

On 14.02.23 11:26, Qi Zheng wrote:
> 
> 
> On 2023/2/14 17:43, Mike Rapoport wrote:
>> On Tue, Feb 14, 2023 at 10:17:03AM +0100, David Hildenbrand wrote:
>>> On 14.02.23 09:42, Vlastimil Babka wrote:
>>>> On 2/13/23 12:00, Qi Zheng wrote:
>>>>>
>>>>>
>>>>> On 2023/2/13 16:47, Vlastimil Babka wrote:
>>>>>> On 2/12/23 12:03, Qi Zheng wrote:
>>>>>>> In x86, numa_register_memblks() is only interested in
>>>>>>> those nodes which have enough memory, so it skips over
>>>>>>> all nodes with memory below NODE_MIN_SIZE (treated as
>>>>>>> a memoryless node). Later on, we will initialize these
>>>>>>> memoryless nodes (allocate pgdat in free_area_init()
>>>>>>> and build zonelist etc), and will online these nodes
>>>>>>> in init_cpu_to_node() and init_gi_nodes().
>>>>>>>
>>>>>>> After boot, these memoryless nodes are in N_ONLINE
>>>>>>> state but not in N_MEMORY state. But we can still allocate
>>>>>>> pages from these memoryless nodes.
>>>>>>>
>>>>>>> In SLUB, we only process nodes in the N_MEMORY state,
>>>>>>> such as allocating their struct kmem_cache_node. So if
>>>>>>> we allocate a page from the memoryless node above to
>>>>>>> SLUB, the struct kmem_cache_node of the node corresponding
>>>>>>> to this page is NULL, which will cause panic.
>>>>>>>
>>>>>>> For example, if we use qemu to start a two numa node kernel,
>>>>>>> one of the nodes has 2M memory (less than NODE_MIN_SIZE),
>>>>>>> and the other node has 2G, then we will encounter the
>>>>>>> following panic:
>>>>>>>
>>>>>>> [    0.149844] BUG: kernel NULL pointer dereference, address: 0000000000000000
>>>>>>> [    0.150783] #PF: supervisor write access in kernel mode
>>>>>>> [    0.151488] #PF: error_code(0x0002) - not-present page
>>>>>>> <...>
>>>>>>> [    0.156056] RIP: 0010:_raw_spin_lock_irqsave+0x22/0x40
>>>>>>> <...>
>>>>>>> [    0.169781] Call Trace:
>>>>>>> [    0.170159]  <TASK>
>>>>>>> [    0.170448]  deactivate_slab+0x187/0x3c0
>>>>>>> [    0.171031]  ? bootstrap+0x1b/0x10e
>>>>>>> [    0.171559]  ? preempt_count_sub+0x9/0xa0
>>>>>>> [    0.172145]  ? kmem_cache_alloc+0x12c/0x440
>>>>>>> [    0.172735]  ? bootstrap+0x1b/0x10e
>>>>>>> [    0.173236]  bootstrap+0x6b/0x10e
>>>>>>> [    0.173720]  kmem_cache_init+0x10a/0x188
>>>>>>> [    0.174240]  start_kernel+0x415/0x6ac
>>>>>>> [    0.174738]  secondary_startup_64_no_verify+0xe0/0xeb
>>>>>>> [    0.175417]  </TASK>
>>>>>>> [    0.175713] Modules linked in:
>>>>>>> [    0.176117] CR2: 0000000000000000
>>>>>>>
>>>>>>> In addition, we can also encountered this panic in the actual
>>>>>>> production environment. We set up a 2c2g container with two
>>>>>>> numa nodes, and then reserved 128M for kdump, and then we
>>>>>>> can encountered the above panic in the kdump kernel.
>>>>>>>
>>>>>>> To fix it, we can filter memoryless nodes when allocating
>>>>>>> pages.
>>>>>>>
>>>>>>> Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
>>>>>>> Reported-by: Teng Hu <huteng.ht@bytedance.com>
>>>>>>
>>>>>> Well AFAIK the key mechanism to only allocate from "good" nodes is the
>>>>>> zonelist, we shouldn't need to start putting extra checks like this. So it
>>>>>> seems to me that the code building the zonelists should take the
>>>>>> NODE_MIN_SIZE constraint in mind.
>>>>>
>>>>> Indeed. How about the following patch:
>>>>
>>>> +Cc also David, forgot earlier.
>>>>
>>>> Looks good to me, at least.
>>>>
>>>>> @@ -6382,8 +6378,11 @@ int find_next_best_node(int node, nodemask_t
>>>>> *used_node_mask)
>>>>>             int min_val = INT_MAX;
>>>>>             int best_node = NUMA_NO_NODE;
>>>>>
>>>>> -       /* Use the local node if we haven't already */
>>>>> -       if (!node_isset(node, *used_node_mask)) {
>>>>> +       /*
>>>>> +        * Use the local node if we haven't already. But for memoryless
>>>>> local
>>>>> +        * node, we should skip it and fallback to other nodes.
>>>>> +        */
>>>>> +       if (!node_isset(node, *used_node_mask) && node_state(node,
>>>>> N_MEMORY)) {
>>>>>                     node_set(node, *used_node_mask);
>>>>>                     return node;
>>>>>             }
>>>>>
>>>>> For memoryless node, we skip it and fallback to other nodes when
>>>>> build its zonelists.
>>>>>
>>>>> Say we have node0 and node1, and node0 is memoryless, then:
>>>>>
>>>>> [    0.102400] Fallback order for Node 0: 1
>>>>> [    0.102931] Fallback order for Node 1: 1
>>>>>
>>>>> In this way, we will not allocate pages from memoryless node0.
>>>>>
>>>
>>> In offline_pages(), we'll first build_all_zonelists() to then
>>> node_states_clear_node()->node_clear_state(node, N_MEMORY);
>>>
>>> So at least on the offlining path, we wouldn't detect it properly yet I
>>> assume, and build a zonelist that contains a now-memory-less node?
>>
>> Another question is what happens if a new memory is plugged into a node
>> that had < NODE_MIN_SIZE of memory and after hotplug it stops being
>> "memoryless".
> 
> When going online and offline a memory will re-call
> build_all_zonelists() to re-establish the zonelists (the zonelist of
> itself and other nodes). So it can stop being "memoryless"
> automatically.
> 
> But in online_pages(), did not see the check of < NODE_MIN_SIZE.

TBH, this is the first time I hear of NODE_MIN_SIZE and it seems to be a 
pretty x86 specific thing.

Are we sure we want to get NODE_MIN_SIZE involved?

-- 
Thanks,

David / dhildenb