From patchwork Fri Sep 2 21:10:48 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Alexei Starovoitov X-Patchwork-Id: 12964669 X-Patchwork-Delegate: bpf@iogearbox.net Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5457AC6FA82 for ; Fri, 2 Sep 2022 21:11:29 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230501AbiIBVL1 (ORCPT ); Fri, 2 Sep 2022 17:11:27 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:53472 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230500AbiIBVL0 (ORCPT ); Fri, 2 Sep 2022 17:11:26 -0400 Received: from mail-pf1-x42f.google.com (mail-pf1-x42f.google.com [IPv6:2607:f8b0:4864:20::42f]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 11A2FD9D75 for ; Fri, 2 Sep 2022 14:11:24 -0700 (PDT) Received: by mail-pf1-x42f.google.com with SMTP id t129so3072420pfb.6 for ; Fri, 02 Sep 2022 14:11:24 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date; bh=YkiR1pR4pf3gc+NcW9hCzitUfw6uraWiWMrmU6Iy0/0=; b=FaIIYGQgFP3Xkdw5zMo/3PUWqN+uoPYtIgxds76CsSEVcDaIIT7rig2+TOOvru+wmZ t1hS3Va3+GJkfVsGewHGjNvWRmMWjYGXYIThZimEpDoVvgnhB/zbwQe1oGCDmeYPQpKL vVsCZ5hfQzNp85QoaKA51Ef7yi+Esy+dACwgQtYoMoUZGfRRxiLFOGQAdkN9kSvkVdXb UpEbKJJtXi5gTlAcLPSn9XJAJV5gBE6cSH2NDe3hYAvTeYfBfaiHn6duYCwPJDPoUxVz 3UBOXB+NtRM1KEy4fJaQMxnTRYroZ4irsiwiSQjwPeteKEK/Ezli6giWt2wBzHqRMOLs qfSA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date; bh=YkiR1pR4pf3gc+NcW9hCzitUfw6uraWiWMrmU6Iy0/0=; b=pZqyQHtqv8wIxUj5jx80LXovmIy9J1RIMN42k4yN8haFImDqlfB6GjzuFxqTaSLOt8 87dbn24d/o3/CINwljvGs32Y4K2dh5Ic6WiCeSIIMr+g+2nuIneK+GtkXJHG9EBR+QQH JAUV9iOvXDWG/HYxmN8HXF58QyHmRxWb+v5ouAi4DbjqvOMUj6EX5ifKHFtSRc9x66Nj UgW9MHqbLOHfZdUOHrsmPwQpu3d6ZEF2YRICZD9xI64nFtBZXlJR61tsGv1RYfJZttEp 3VbtfwHONFMcNoXo918rgJavdQJ+AbzjCZySQB8VaGct1cUoasYBwqlfOEni3yo/HAaI DqvA== X-Gm-Message-State: ACgBeo1qpddANuh+ylj5eK0oB4VAEY4I/kuR+eqWXz/WxGqy0X5XUzwC KAqzy3B1fpNtUrEp79AWR7k= X-Google-Smtp-Source: AA6agR5l3VmyLc3SLS6tKg66MjVA1vEHfr+VeW3GoIsLpZR90P7fjzurUE5ZvL8SsP9MOyCuxc0NQg== X-Received: by 2002:aa7:9430:0:b0:537:ab14:6cd8 with SMTP id y16-20020aa79430000000b00537ab146cd8mr37505123pfo.29.1662153084179; Fri, 02 Sep 2022 14:11:24 -0700 (PDT) Received: from localhost.localdomain ([2620:10d:c090:500::c978]) by smtp.gmail.com with ESMTPSA id 35-20020a630c63000000b00433c26caf45sm356021pgm.73.2022.09.02.14.11.22 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Fri, 02 Sep 2022 14:11:23 -0700 (PDT) From: Alexei Starovoitov To: davem@davemloft.net Cc: daniel@iogearbox.net, andrii@kernel.org, tj@kernel.org, memxor@gmail.com, delyank@fb.com, linux-mm@kvack.org, bpf@vger.kernel.org, kernel-team@fb.com Subject: [PATCH v6 bpf-next 06/16] bpf: Optimize element count in non-preallocated hash map. Date: Fri, 2 Sep 2022 14:10:48 -0700 Message-Id: <20220902211058.60789-7-alexei.starovoitov@gmail.com> X-Mailer: git-send-email 2.36.1 In-Reply-To: <20220902211058.60789-1-alexei.starovoitov@gmail.com> References: <20220902211058.60789-1-alexei.starovoitov@gmail.com> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org X-Patchwork-Delegate: bpf@iogearbox.net From: Alexei Starovoitov The atomic_inc/dec might cause extreme cache line bouncing when multiple cpus access the same bpf map. Based on specified max_entries for the hash map calculate when percpu_counter becomes faster than atomic_t and use it for such maps. For example samples/bpf/map_perf_test is using hash map with max_entries 1000. On a system with 16 cpus the 'map_perf_test 4' shows 14k events per second using atomic_t. On a system with 15 cpus it shows 100k events per second using percpu. map_perf_test is an extreme case where all cpus colliding on atomic_t which causes extreme cache bouncing. Note that the slow path of percpu_counter is 5k events per secound vs 14k for atomic, so the heuristic is necessary. See comment in the code why the heuristic is based on num_online_cpus(). Acked-by: Kumar Kartikeya Dwivedi Acked-by: Andrii Nakryiko Signed-off-by: Alexei Starovoitov --- kernel/bpf/hashtab.c | 70 +++++++++++++++++++++++++++++++++++++++----- 1 file changed, 62 insertions(+), 8 deletions(-) diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index 508e64351f87..36aa16dc43ad 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -101,7 +101,12 @@ struct bpf_htab { struct bpf_lru lru; }; struct htab_elem *__percpu *extra_elems; - atomic_t count; /* number of elements in this hashtable */ + /* number of elements in non-preallocated hashtable are kept + * in either pcount or count + */ + struct percpu_counter pcount; + atomic_t count; + bool use_percpu_counter; u32 n_buckets; /* number of hash buckets */ u32 elem_size; /* size of each element in bytes */ u32 hashrnd; @@ -565,6 +570,29 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) htab_init_buckets(htab); +/* compute_batch_value() computes batch value as num_online_cpus() * 2 + * and __percpu_counter_compare() needs + * htab->max_entries - cur_number_of_elems to be more than batch * num_online_cpus() + * for percpu_counter to be faster than atomic_t. In practice the average bpf + * hash map size is 10k, which means that a system with 64 cpus will fill + * hashmap to 20% of 10k before percpu_counter becomes ineffective. Therefore + * define our own batch count as 32 then 10k hash map can be filled up to 80%: + * 10k - 8k > 32 _batch_ * 64 _cpus_ + * and __percpu_counter_compare() will still be fast. At that point hash map + * collisions will dominate its performance anyway. Assume that hash map filled + * to 50+% isn't going to be O(1) and use the following formula to choose + * between percpu_counter and atomic_t. + */ +#define PERCPU_COUNTER_BATCH 32 + if (attr->max_entries / 2 > num_online_cpus() * PERCPU_COUNTER_BATCH) + htab->use_percpu_counter = true; + + if (htab->use_percpu_counter) { + err = percpu_counter_init(&htab->pcount, 0, GFP_KERNEL); + if (err) + goto free_map_locked; + } + if (prealloc) { err = prealloc_init(htab); if (err) @@ -891,6 +919,31 @@ static void htab_put_fd_value(struct bpf_htab *htab, struct htab_elem *l) } } +static bool is_map_full(struct bpf_htab *htab) +{ + if (htab->use_percpu_counter) + return __percpu_counter_compare(&htab->pcount, htab->map.max_entries, + PERCPU_COUNTER_BATCH) >= 0; + return atomic_read(&htab->count) >= htab->map.max_entries; +} + +static void inc_elem_count(struct bpf_htab *htab) +{ + if (htab->use_percpu_counter) + percpu_counter_add_batch(&htab->pcount, 1, PERCPU_COUNTER_BATCH); + else + atomic_inc(&htab->count); +} + +static void dec_elem_count(struct bpf_htab *htab) +{ + if (htab->use_percpu_counter) + percpu_counter_add_batch(&htab->pcount, -1, PERCPU_COUNTER_BATCH); + else + atomic_dec(&htab->count); +} + + static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l) { htab_put_fd_value(htab, l); @@ -899,7 +952,7 @@ static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l) check_and_free_fields(htab, l); __pcpu_freelist_push(&htab->freelist, &l->fnode); } else { - atomic_dec(&htab->count); + dec_elem_count(htab); l->htab = htab; call_rcu(&l->rcu, htab_elem_free_rcu); } @@ -983,16 +1036,15 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, l_new = container_of(l, struct htab_elem, fnode); } } else { - if (atomic_inc_return(&htab->count) > htab->map.max_entries) - if (!old_elem) { + if (is_map_full(htab)) + if (!old_elem) /* when map is full and update() is replacing * old element, it's ok to allocate, since * old element will be freed immediately. * Otherwise return an error */ - l_new = ERR_PTR(-E2BIG); - goto dec_count; - } + return ERR_PTR(-E2BIG); + inc_elem_count(htab); l_new = bpf_mem_cache_alloc(&htab->ma); if (!l_new) { l_new = ERR_PTR(-ENOMEM); @@ -1034,7 +1086,7 @@ static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, l_new->hash = hash; return l_new; dec_count: - atomic_dec(&htab->count); + dec_elem_count(htab); return l_new; } @@ -1513,6 +1565,8 @@ static void htab_map_free(struct bpf_map *map) free_percpu(htab->extra_elems); bpf_map_area_free(htab->buckets); bpf_mem_alloc_destroy(&htab->ma); + if (htab->use_percpu_counter) + percpu_counter_destroy(&htab->pcount); for (i = 0; i < HASHTAB_MAP_LOCK_COUNT; i++) free_percpu(htab->map_locked[i]); lockdep_unregister_key(&htab->lockdep_key);