From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965056AbbLOLVU (ORCPT ); Tue, 15 Dec 2015 06:21:20 -0500 Received: from mail-qg0-f43.google.com ([209.85.192.43]:35406 "EHLO mail-qg0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964936AbbLOLVS (ORCPT ); Tue, 15 Dec 2015 06:21:18 -0500 From: Ming Lei To: linux-kernel@vger.kernel.org, Alexei Starovoitov Cc: "David S. Miller" , netdev@vger.kernel.org Subject: [PATCH 0/6] bpf: hash: optimization Date: Tue, 15 Dec 2015 19:20:58 +0800 Message-Id: <1450178464-27721-1-git-send-email-tom.leiming@gmail.com> X-Mailer: git-send-email 1.9.1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 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