linux/kernel/bpf
tom.leiming@gmail.com 688ecfe602 bpf: hash: use per-bucket spinlock
Both htab_map_update_elem() and htab_map_delete_elem() can be
called from eBPF program, and they may be in kernel hot path,
so it isn't efficient to use a per-hashtable lock in this two
helpers.

The per-hashtable spinlock is used for protecting bucket's
hlist, and per-bucket lock is just enough. This patch converts
the per-hashtable lock into per-bucket spinlock, so that
contention can be decreased a lot.

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-12-29 15:13:44 -05:00
..
arraymap.c bpf: fix allocation warnings in bpf maps and integer overflow 2015-12-02 23:36:00 -05:00
core.c bpf: move clearing of A/X into classic to eBPF migration prologue 2015-12-18 16:04:51 -05:00
hashtab.c bpf: hash: use per-bucket spinlock 2015-12-29 15:13:44 -05:00
helpers.c bpf: split state from prandom_u32() and consolidate {c, e}BPF prngs 2015-10-08 05:26:39 -07:00
inode.c bpf, inode: allow for rename and link ops 2015-12-12 18:44:23 -05:00
Makefile bpf: add support for persistent maps/progs 2015-11-02 22:48:39 -05:00
syscall.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2015-12-03 21:09:12 -05:00
verifier.c bpf: fix clearing on persistent program array maps 2015-11-25 12:14:09 -05:00