Re: [dpdk-dev] [PATCH] mem: fix allocation failure on non-NUMA kernel

["Burakov, Anatoly" <anatoly.burakov@intel.com>]

On 17-Sep-20 3:08 PM, Nick Connolly wrote:
> Excellent - thanks - I'll amend the patch.
> 
> On 17/09/2020 15:07, Burakov, Anatoly wrote:
>> On 17-Sep-20 2:05 PM, Nick Connolly wrote:
>>> Hi Anatoly,
>>>
>>> Thanks.  My recollection is that all of the NUMA configuration flags 
>>> were set to 'n'.
>>>
>>> Regards,
>>> Nick
>>>
>>> On 17/09/2020 13:57, Burakov, Anatoly wrote:
>>>> On 17-Sep-20 1:29 PM, Nick Connolly wrote:
>>>>> Hi Anatoly,
>>>>>
>>>>> Thanks for the response.  You are asking a good question - here's 
>>>>> what I know:
>>>>>
>>>>> The issue arose on a single socket system, running WSL2 (full Linux 
>>>>> kernel running as a lightweight VM under Windows).
>>>>> The default kernel in this environment is built with CONFIG_NUMA=n 
>>>>> which means get_mempolicy() returns an error.
>>>>> This causes the check to ensure that the allocated memory is 
>>>>> associated with the correct socket to fail.
>>>>>
>>>>> The change is to skip the allocation check if check_numa() 
>>>>> indicates that NUMA-aware memory is not supported.
>>>>>
>>>>> Researching the meaning of CONFIG_NUMA, I found 
>>>>> https://cateee.net/lkddb/web-lkddb/NUMA.html which says:
>>>>>> Enable NUMA (Non-Uniform Memory Access) support.
>>>>>> The kernel will try to allocate memory used by a CPU on the local 
>>>>>> memory controller of the CPU and add some more NUMA awareness to 
>>>>>> the kernel.
>>>>>
>>>>> Clearly CONFIG_NUMA enables memory awareness, but there's no 
>>>>> indication in the description whether information about the NUMA 
>>>>> physical architecture is 'hidden', or whether it is still exposed 
>>>>> through /sys/devices/system/node* (which is used by the rte 
>>>>> initialisation code to determine how many sockets there are). 
>>>>> Unfortunately, I don't have ready access to a multi-socket Linux 
>>>>> system that I can test this out on, so I took the conservative 
>>>>> approach that it may be possible to have CONFIG_NUMA disabled, but 
>>>>> the kernel still report more than one node, and coded the change to 
>>>>> generate a debug message if this occurs.
>>>>>
>>>>> Do you know whether CONFIG_NUMA turns off all knowledge about the 
>>>>> hardware architecture?  If it does, then I agree that the test for 
>>>>> rte_socket_count() serves no purpose and should be removed.
>>>>>
>>>>
>>>> I have a system with a custom compiled kernel, i can recompile it 
>>>> without this flag and test this. I'll report back with results :)
>>>>
>>>
>>
>> With CONFIG_NUMA set to 'n':
>>
>> [root@xxx ~]# find /sys -name "node*"
>> /sys/kernel/software_nodes/node0
>> [root@xxx ~]#
>>
>> This is confirmed by running DPDK on that machine - i can see all 
>> cores from all sockets, but they're all appearing on socket 0. So, 
>> yes, that check isn't necessary :)
>>
> 

I would also add a comment explaining why we're checking for NUMA 
support when NUMA support is defined at compiled time.

-- 
Thanks,
Anatoly