Re: [PATCH bpf-next v11 05/10] libbpf: Add support for global percpu data

On 2026/8/7 10:36, Leon Hwang wrote:
> On 7/8/26 00:57, [email protected] wrote:
>> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
>> - [Low] Dead code in percpu global data feature check
>> --
>>
>> --- Patch [5]: [PATCH bpf-next v11 05/10] libbpf: Add support for global percpu data ---
>> commit ee97f8a8e25b5e9337024df3dd6f3a1aae4f6cdc
>> Author: Leon Hwang <[email protected]>
>>
>> libbpf: Add support for global percpu data
>>
>> This commit adds support for global percpu data in libbpf by introducing a 
>> new ".percpu" section. It updates the lightweight skeleton loader to use 
>> BPF_F_ALL_CPUS when updating the percpu_array map and marks the mmaped data 
>> as read-only after population.
>>
>>> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
>>> index 514e4e9daa825..9f79a8e850fdb 100644
>>> --- a/tools/lib/bpf/libbpf.c
>>> +++ b/tools/lib/bpf/libbpf.c
>>
>> [ ... ]
>>
>>> @@ -5293,18 +5326,26 @@ static int
>>>  bpf_object__populate_internal_map(struct bpf_object *obj, struct bpf_map *map)
>>>  {
>>>  	enum libbpf_map_type map_type = map->libbpf_type;
>>> +	bool is_percpu = map_type == LIBBPF_MAP_PERCPU;
>>> +	const __u64 update_flags = is_percpu ? BPF_F_ALL_CPUS : 0;
>>>  	int err, zero = 0;
>>>  	size_t mmap_sz;
>>>  
>>> +	if (is_percpu && !obj->gen_loader && !kernel_supports(obj, FEAT_PERCPU_DATA)) {
>>> +		pr_warn("map '%s': kernel does not support percpu data.\n",
>>> +			bpf_map__name(map));
>>> +		return -EOPNOTSUPP;
>>> +	}
>>> +
>>
>> [Severity: Low]
>> Is this check reachable? 
>>
>> Looking at bpf_object__create_maps(), if the kernel does not support
>> FEAT_PERCPU_DATA, map->autocreate is set to false and the map processing
>> loop continues early:
>>
>> bpf_object__create_maps() {
>>     ...
>>     if (bpf_map__is_internal(map) && bpf_map__type(map) == BPF_MAP_TYPE_PERCPU_ARRAY &&
>>         !kernel_supports(obj, FEAT_PERCPU_DATA))
>>         map->autocreate = false;
>>
>>     if (!map->autocreate) {
>>         pr_debug("map '%s': skipped auto-creating...\n", map->name);
>>         continue;
>>     }
>>     ...
>> }
>>
>> Since bpf_object__populate_internal_map() is called later in this same
>> loop (after the continue), wouldn't the !kernel_supports() condition
>> already be filtered out, making this warning dead code?
>>
> 
> True.
> 
> Will drop this warning.
> 
> And, update 'bpf_map__type(map) == BPF_MAP_TYPE_PERCPU_ARRAY' to
> 'map->libbpf_type == LIBBPF_MAP_PERCPU' in bpf_object__create_maps().
> 


Hi Andrii,

Could you apply this series to bpf-next tree? This series seems good
enough to be merged.

Then, I'll post a separate patch to drop the warning, and update the iff
condition.

Thanks,
Leon