(I pruned the CC list, hopefully I didn't leave out anybody who cares.) Johannes Nixdorf via Bridge <bridge@xxxxxxxxxxxxxxxxxxxxxxxxxx> writes: > Support setting the FDB limit through ip link. The arguments is: > - fdb_max_learned_entries: A 32-bit unsigned integer specifying the > maximum number of learned FDB entries, with 0 > disabling the limit. ... > Signed-off-by: Johannes Nixdorf <jnixdorf-oss@xxxxxx> Code looks good to me. A couple points though: - The corresponding kernel changes have not yet been merged, have they? This patch should obviously only be merged after that happens. - The UAPI changes should not be part of the patch, the maintainers will update themselves. - The names fdb_n_learned_entries, fdb_max_learned_entries... they are somewhat unwieldy IMHO. Just for consideration, I don't feel strongly about this: Your kconfig option is named BRIDGE_DEFAULT_FDB_MAX_LEARNED, and that makes sense to me, because yeah, given the word "learned" in context of FDB, "entries" is the obvious continuation, so why mention it explicitly. Consider following suit with the other source artifacts -- the attribute names, struct fields, keywords in this patch. "fdb_n_learned", "fdb_max_learned" is IMHO just as understandable and will be easier to type.
