From patchwork Wed Jul 22 18:49:52 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Yonghong Song X-Patchwork-Id: 1334100 X-Patchwork-Delegate: bpf@iogearbox.net Return-Path: X-Original-To: incoming-bpf@patchwork.ozlabs.org Delivered-To: patchwork-incoming-bpf@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=vger.kernel.org (client-ip=23.128.96.18; helo=vger.kernel.org; envelope-from=bpf-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=pass (p=reject dis=none) header.from=fb.com Authentication-Results: ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=fb.com header.i=@fb.com header.a=rsa-sha256 header.s=facebook header.b=FGms2hsL; dkim-atps=neutral Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by ozlabs.org (Postfix) with ESMTP id 4BBkzT1d1lz9sRR for ; Thu, 23 Jul 2020 04:49:57 +1000 (AEST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1732579AbgGVSt4 (ORCPT ); Wed, 22 Jul 2020 14:49:56 -0400 Received: from mx0b-00082601.pphosted.com ([67.231.153.30]:43504 "EHLO mx0b-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1732543AbgGVSt4 (ORCPT ); Wed, 22 Jul 2020 14:49:56 -0400 Received: from pps.filterd (m0109331.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.16.0.42/8.16.0.42) with SMTP id 06MIcIcM012189 for ; Wed, 22 Jul 2020 11:49:55 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-transfer-encoding : content-type; s=facebook; bh=MeSp6YJimNz4EDj5AU69yUG+jJ+MXrqRNjGW64u3JpA=; b=FGms2hsLJHNLztE3e5ZGukC7XbJFiyrFJX3FtMSdZd/lAAbBeXwXTEzGkeF1YpxtlsVJ JoohHTKK8I0zP15ZONm6ND/eTlV/hGAoimf61LuzFpscQmAv7vnECWhr/WLfgUohuTO5 J4Em2HJQ3N126i0Srjc37UNRvGxT7RY9GSA= Received: from maileast.thefacebook.com ([163.114.130.16]) by mx0a-00082601.pphosted.com with ESMTP id 32embc28gu-2 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Wed, 22 Jul 2020 11:49:55 -0700 Received: from intmgw002.08.frc2.facebook.com (2620:10d:c0a8:1b::d) by mail.thefacebook.com (2620:10d:c0a8:82::f) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.1979.3; Wed, 22 Jul 2020 11:49:54 -0700 Received: by devbig003.ftw2.facebook.com (Postfix, from userid 128203) id 3DA1B3702F09; Wed, 22 Jul 2020 11:49:52 -0700 (PDT) Smtp-Origin-Hostprefix: devbig From: Yonghong Song Smtp-Origin-Hostname: devbig003.ftw2.facebook.com To: , CC: Alexei Starovoitov , Daniel Borkmann , , Martin KaFai Lau Smtp-Origin-Cluster: ftw2c04 Subject: [PATCH bpf-next v2 06/13] bpf: implement bpf iterator for array maps Date: Wed, 22 Jul 2020 11:49:52 -0700 Message-ID: <20200722184952.3777650-1-yhs@fb.com> X-Mailer: git-send-email 2.24.1 In-Reply-To: <20200722184945.3777103-1-yhs@fb.com> References: <20200722184945.3777103-1-yhs@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.235, 18.0.687 definitions=2020-07-22_10:2020-07-22,2020-07-22 signatures=0 X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 bulkscore=0 mlxlogscore=999 impostorscore=0 clxscore=1015 priorityscore=1501 lowpriorityscore=0 adultscore=0 phishscore=0 mlxscore=0 suspectscore=8 spamscore=0 malwarescore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2006250000 definitions=main-2007220119 X-FB-Internal: deliver Sender: bpf-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org The bpf iterators for array and percpu array are implemented. Similar to hash maps, for percpu array map, bpf program will receive values from all cpus. Signed-off-by: Yonghong Song --- kernel/bpf/arraymap.c | 138 ++++++++++++++++++++++++++++++++++++++++++ kernel/bpf/map_iter.c | 6 +- 2 files changed, 142 insertions(+), 2 deletions(-) diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index c66e8273fccd..8ff419b632a6 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -487,6 +487,142 @@ static int array_map_mmap(struct bpf_map *map, struct vm_area_struct *vma) vma->vm_pgoff + pgoff); } +struct bpf_iter_seq_array_map_info { + struct bpf_map *map; + void *percpu_value_buf; + u32 index; +}; + +static void *bpf_array_map_seq_start(struct seq_file *seq, loff_t *pos) +{ + struct bpf_iter_seq_array_map_info *info = seq->private; + struct bpf_map *map = info->map; + struct bpf_array *array; + u32 index; + + if (info->index >= map->max_entries) + return NULL; + + if (*pos == 0) + ++*pos; + array = container_of(map, struct bpf_array, map); + index = info->index & array->index_mask; + if (info->percpu_value_buf) + return array->pptrs[index]; + return array->value + array->elem_size * index; +} + +static void *bpf_array_map_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct bpf_iter_seq_array_map_info *info = seq->private; + struct bpf_map *map = info->map; + struct bpf_array *array; + u32 index; + + ++*pos; + ++info->index; + if (info->index >= map->max_entries) + return NULL; + + array = container_of(map, struct bpf_array, map); + index = info->index & array->index_mask; + if (info->percpu_value_buf) + return array->pptrs[index]; + return array->value + array->elem_size * index; +} + +static int __bpf_array_map_seq_show(struct seq_file *seq, void *v) +{ + struct bpf_iter_seq_array_map_info *info = seq->private; + struct bpf_iter__bpf_map_elem ctx = {}; + struct bpf_map *map = info->map; + struct bpf_iter_meta meta; + struct bpf_prog *prog; + int off = 0, cpu = 0; + void __percpu **pptr; + u32 size; + + meta.seq = seq; + prog = bpf_iter_get_info(&meta, v == NULL); + if (!prog) + return 0; + + ctx.meta = &meta; + ctx.map = info->map; + if (v) { + ctx.key = &info->index; + + if (!info->percpu_value_buf) { + ctx.value = v; + } else { + pptr = v; + size = round_up(map->value_size, 8); + for_each_possible_cpu(cpu) { + bpf_long_memcpy(info->percpu_value_buf + off, + per_cpu_ptr(pptr, cpu), + size); + off += size; + } + ctx.value = info->percpu_value_buf; + } + } + + return bpf_iter_run_prog(prog, &ctx); +} + +static int bpf_array_map_seq_show(struct seq_file *seq, void *v) +{ + return __bpf_array_map_seq_show(seq, v); +} + +static void bpf_array_map_seq_stop(struct seq_file *seq, void *v) +{ + if (!v) + (void)__bpf_array_map_seq_show(seq, NULL); +} + +static int bpf_iter_init_array_map(void *priv_data, + struct bpf_iter_aux_info *aux) +{ + struct bpf_iter_seq_array_map_info *seq_info = priv_data; + struct bpf_map *map = aux->map; + void *value_buf; + u32 buf_size; + + if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) { + buf_size = round_up(map->value_size, 8) * num_possible_cpus(); + value_buf = kmalloc(buf_size, GFP_USER | __GFP_NOWARN); + if (!value_buf) + return -ENOMEM; + + seq_info->percpu_value_buf = value_buf; + } + + seq_info->map = map; + return 0; +} + +static void bpf_iter_fini_array_map(void *priv_data) +{ + struct bpf_iter_seq_array_map_info *seq_info = priv_data; + + kfree(seq_info->percpu_value_buf); +} + +static const struct seq_operations bpf_array_map_seq_ops = { + .start = bpf_array_map_seq_start, + .next = bpf_array_map_seq_next, + .stop = bpf_array_map_seq_stop, + .show = bpf_array_map_seq_show, +}; + +static const struct bpf_iter_seq_info iter_seq_info = { + .seq_ops = &bpf_array_map_seq_ops, + .init_seq_private = bpf_iter_init_array_map, + .fini_seq_private = bpf_iter_fini_array_map, + .seq_priv_size = sizeof(struct bpf_iter_seq_array_map_info), +}; + static int array_map_btf_id; const struct bpf_map_ops array_map_ops = { .map_alloc_check = array_map_alloc_check, @@ -506,6 +642,7 @@ const struct bpf_map_ops array_map_ops = { .map_update_batch = generic_map_update_batch, .map_btf_name = "bpf_array", .map_btf_id = &array_map_btf_id, + .iter_seq_info = &iter_seq_info, }; static int percpu_array_map_btf_id; @@ -521,6 +658,7 @@ const struct bpf_map_ops percpu_array_map_ops = { .map_check_btf = array_map_check_btf, .map_btf_name = "bpf_array", .map_btf_id = &percpu_array_map_btf_id, + .iter_seq_info = &iter_seq_info, }; static int fd_array_map_alloc_check(union bpf_attr *attr) diff --git a/kernel/bpf/map_iter.c b/kernel/bpf/map_iter.c index 6cd532f26836..5d834b9dc005 100644 --- a/kernel/bpf/map_iter.c +++ b/kernel/bpf/map_iter.c @@ -110,10 +110,12 @@ static int bpf_iter_check_map(struct bpf_prog *prog, bool is_percpu = false; if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH || - map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH) + map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH || + map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) is_percpu = true; else if (map->map_type != BPF_MAP_TYPE_HASH && - map->map_type != BPF_MAP_TYPE_LRU_HASH) + map->map_type != BPF_MAP_TYPE_LRU_HASH && + map->map_type != BPF_MAP_TYPE_ARRAY) return -EINVAL; key_acc_size = prog->aux->max_rdonly_access;