As discussed in the stream @krisnova I put some more thought into those buffer sizes. Since the per-CPU buffers of the BPF_PERF_EVENT_ARRAY BPF map are mmaped it made sense to align those buffers with pages.
Also added some comments in an attempt to make this code a bit more readable.
As discussed in the stream @krisnova I put some more thought into those buffer sizes. Since the per-CPU buffers of the BPF_PERF_EVENT_ARRAY BPF map are mmaped it made sense to align those buffers with pages.
Also added some comments in an attempt to make this code a bit more readable.