[PATCH 000/227] use NUMA node terminology across the tree

[fengchengwen]

On 11/4/2026 12:16 AM, Aaron Conole wrote:
> Chengwen Feng <chengwen.feng at linux.dev> writes:
> 
>> From: Chengwen Feng <fengchengwen at huawei.com>
>>
>> The DPDK tree conflates CPU sockets with NUMA memory nodes: public
>> APIs, structure fields and macros are named after "socket" although
>> they actually identify a NUMA node (e.g. rte_socket_id(), the
>> socket_id field of rte_memzone, SOCKET_ID_ANY). This naming misleads
>> application developers and pollutes vendor code that targets NUMA
>> topologies.
>>
>> This series renames the terminology tree-wide:
> 
> Do not do this again.
> 
> I have set the robot from considering patches from you for the next 227
> hours.  One for each hour that a patch-by-patch test would take.  Do not
> repost this series again and expect the github robot to run for it.
> Please work with existing maintainers to come up with a plan for each
> area rename.
> 
> CI resources are shared, and in the case of the github robot, they are
> shared among OVS, OVN, and DPDK.  Additionally, a 227 patch series is
> completely unreviewable, and completely inconsiderate of reviewers time.
> Anything above 15 patches is, truly - who can keep that much details in
> mind.

Thanks for your feedback. This 227-patch set exposes the full scope of
the NUMA terminology cleanup, as a starting point for discussion and
planning. The scope is indeed large, and this rework direction was
previously discussed with Thomas and Konstantin.

Could you please lift the robot block for this posting?
I will wait for community consensus before splitting the changes into
proper batches for subsequent review.

I apologize for the heavy load imposed on shared CI resources.

Regards,
Chengwen

> 
>