[PATCH v3 01/29] bpf: replace deprecated SMP barriers with C11 fences
Marat Khalili
qm2k21 at gmail.com
Mon Oct 12 20:52:35 CEST 2026
On 09/10/2026 07:05, Wathsala Vithanage wrote:
>> /*
>> * Marks given callback as used by datapath.
>> */
>> static __rte_always_inline void
>> bpf_eth_cbi_inuse(struct bpf_eth_cbi *cbi)
>> {
>> - cbi->use++;
>> - /* make sure no store/load reordering could happen */
>> - rte_smp_mb();
>> + rte_atomic_store_explicit(&cbi->use,
>> + rte_atomic_load_explicit(&cbi->use,
>> rte_memory_order_relaxed) + 1,
>> + rte_memory_order_relaxed);
>> +
>> + /* full barrier: count must be visible before cb is read */
>> + rte_atomic_thread_fence(rte_memory_order_seq_cst);
NAK: optimizing this increment makes little sense when more careful
analysis shows that it can be simply removed.
I highly recommend keeping this patchset dedicated to removing
rte_smp_XX barriers and moving other bpf_pkt.c improvements elsewhere.
More information about the dev
mailing list