Re: [PATCH bpf-next v11 04/10] libbpf: Probe percpu data feature

On 7/8/26 01:05, [email protected] wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Medium] The newly added feature probe `FEAT_PERCPU_DATA` is never used to make capability decisions.
> --
> 
> --- Patch [4]: [PATCH bpf-next v11 04/10] libbpf: Probe percpu data feature ---
> commit 925264284b89c241360ff401bd914f9c19997613
> Author: Leon Hwang <[email protected]>
> 
> libbpf: Probe percpu data feature
> 
> This commit adds a feature probe to detect if the kernel supports global
> percpu data. It aims to allow libbpf to fail predictably when this feature
> is unavailable on older kernels.
> 
>> diff --git a/tools/lib/bpf/features.c b/tools/lib/bpf/features.c
>> index b7e388f99d0bb..ef9581c113035 100644
>> --- a/tools/lib/bpf/features.c
>> +++ b/tools/lib/bpf/features.c
> [ ... ]
>> @@ -707,6 +739,9 @@ static struct kern_feature_desc {
>>  	[FEAT_BPF_SYSCALL_COMMON_ATTRS] = {
>>  		"BPF syscall common attributes support", probe_bpf_syscall_common_attrs,
>>  	},
>> +	[FEAT_PERCPU_DATA] = {
>> +		"kernel supports percpu data", probe_kern_percpu_data,
>> +	},
>>  };
> 
> [Severity: Medium]
> Are there any callers that actually query this new feature probe?
> 
> The commit message notes that libbpf can make capability decisions early
> and fail predictably when global percpu data is unavailable. However,
> FEAT_PERCPU_DATA appears to only be defined and registered, without being
> queried in the loader code.
> 
> Is there a missing update to the loader, perhaps using kernel_supports(),
> that checks this flag before attempting to load percpu global data?
> 

No. The loader checks this flag in the next commit. The selftests also
check this flag to decide whether to run.

Thanks,
Leon