Skip to content

bpf: hash: optimization

Ming Lei lkml netdev 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest6 patches 11 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/6bpf: hash: use atomic countkernel/bpf/hashtab.c+6 -60 replies
2/6hlist: prepare for supporting bit spinlockinclude/linux/rculist.h+55 -00 replies
3/6bpf: hash: move select_bucket() out of htab's spinlockkernel/bpf/hashtab.c+2 -40 replies
4/6bpf: hash: convert per-hashtable lock into per-bucket bit spinlockkernel/bpf/hashtab.c+26 -125 replies
5/6bpf: hash: avoid to call kmalloc() in eBPF progkernel/bpf/hashtab.c+167 -376 replies
6/6bpf: hash: reorganize 'struct htab_elem'kernel/bpf/hashtab.c+12 -70 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [PATCH 0/6] bpf: hash: optimization

Hi,

This patchset tries to optimize ebpf hash map, and follows the ideas:

1) 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, so this patch converts the lock into per-bucket bit spinlock.

2) kmalloc() is called in htab_map_update_elem() for allocating element, together with one global counter for tracking how many elementes have been allocated. kmalloc is often a bit slow, and the global counter doesn't scale well. This patch pre-allocates one element pool and uses percpu ida for runtime element allocation/free, and the global counter is removed too with this approach.

With this patchset, looks the performance penalty from eBPF decreased a lot, see the following test:

1) run 'tools/biolatency' of bcc before running block test;

2) run fio to test block throught over /dev/nullb0, (randread, 16jobs, libaio, 4k bs) and the test box is one 24cores(dual sockets) VM server:

- without patchset: 607K IOPS
- with this patchset: 1332K IOPS
- without running eBPF prog: 1492K IOPS

 include/linux/rculist.h |  55 +++++++++++
 kernel/bpf/hashtab.c    | 247 ++++++++++++++++++++++++++++++++++++++----------
 2 files changed, 252 insertions(+), 50 deletions(-)

Thanks,
Ming