From patchwork Wed Sep 30 21:28:23 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Song Liu X-Patchwork-Id: 11810285 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id EA547618 for ; Wed, 30 Sep 2020 21:28:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id C88132075F for ; Wed, 30 Sep 2020 21:28:39 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=fb.com header.i=@fb.com header.b="DelnAIe1" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1730178AbgI3V2j (ORCPT ); Wed, 30 Sep 2020 17:28:39 -0400 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:42086 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728660AbgI3V2j (ORCPT ); Wed, 30 Sep 2020 17:28:39 -0400 Received: from pps.filterd (m0109333.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.16.0.42/8.16.0.42) with SMTP id 08ULPlHB014740 for ; Wed, 30 Sep 2020 14:28:39 -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=6o+g6CvLKaTKQuIHDUh/KcMdLPvg7ltHmTb/OOFUgJM=; b=DelnAIe1lN5giDjfnukB22YjXqbamaSGrxVJemYfvwAzWSyNIXJFnNcf6IgM5UA6+8cD 38tS7W6LJgNNWc4wI1aQzK3BQHmG8Hft5vR8f6f/3k+CMLS68rVKHB5BP+b/Fpqxs7Lj 57qRO47xeycNmL9LH4vn2LVw9dhgWqt6cnk= Received: from mail.thefacebook.com ([163.114.132.120]) by mx0a-00082601.pphosted.com with ESMTP id 33vwu39mye-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Wed, 30 Sep 2020 14:28:38 -0700 Received: from intmgw002.08.frc2.facebook.com (2620:10d:c085:108::8) by mail.thefacebook.com (2620:10d:c085:21d::6) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.1979.3; Wed, 30 Sep 2020 14:28:38 -0700 Received: by devbig006.ftw2.facebook.com (Postfix, from userid 4523) id 5C31262E585F; Wed, 30 Sep 2020 14:28:35 -0700 (PDT) From: Song Liu To: , CC: , , , , , Song Liu Subject: [PATCH v4 bpf-next 1/2] bpf: introduce BPF_F_PRESERVE_ELEMS for perf event array Date: Wed, 30 Sep 2020 14:28:23 -0700 Message-ID: <20200930212824.1418683-2-songliubraving@fb.com> X-Mailer: git-send-email 2.24.1 In-Reply-To: <20200930212824.1418683-1-songliubraving@fb.com> References: <20200930212824.1418683-1-songliubraving@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=2020-09-30_13:2020-09-30,2020-09-30 signatures=0 X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 clxscore=1015 phishscore=0 mlxlogscore=999 spamscore=0 suspectscore=2 impostorscore=0 mlxscore=0 malwarescore=0 bulkscore=0 adultscore=0 lowpriorityscore=0 priorityscore=1501 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2006250000 definitions=main-2009300174 X-FB-Internal: deliver Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org Currently, perf event in perf event array is removed from the array when the map fd used to add the event is closed. This behavior makes it difficult to the share perf events with perf event array. Introduce perf event map that keeps the perf event open with a new flag BPF_F_PRESERVE_ELEMS. With this flag set, perf events in the array are not removed when the original map fd is closed. Instead, the perf event will stay in the map until 1) it is explicitly removed from the array; or 2) the array is freed. Signed-off-by: Song Liu --- include/uapi/linux/bpf.h | 3 +++ kernel/bpf/arraymap.c | 19 +++++++++++++++++-- tools/include/uapi/linux/bpf.h | 3 +++ 3 files changed, 23 insertions(+), 2 deletions(-) diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 1f17c6752debb..4f556cfcbfbee 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -414,6 +414,9 @@ enum { /* Enable memory-mapping BPF map */ BPF_F_MMAPABLE = (1U << 10), + +/* Share perf_event among processes */ + BPF_F_PRESERVE_ELEMS = (1U << 11), }; /* Flags for BPF_PROG_QUERY. */ diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index e5fd31268ae02..bd777dd6f9677 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -15,7 +15,8 @@ #include "map_in_map.h" #define ARRAY_CREATE_FLAG_MASK \ - (BPF_F_NUMA_NODE | BPF_F_MMAPABLE | BPF_F_ACCESS_MASK) + (BPF_F_NUMA_NODE | BPF_F_MMAPABLE | BPF_F_ACCESS_MASK | \ + BPF_F_PRESERVE_ELEMS) static void bpf_array_free_percpu(struct bpf_array *array) { @@ -64,6 +65,10 @@ int array_map_alloc_check(union bpf_attr *attr) attr->map_flags & BPF_F_MMAPABLE) return -EINVAL; + if (attr->map_type != BPF_MAP_TYPE_PERF_EVENT_ARRAY && + attr->map_flags & BPF_F_PRESERVE_ELEMS) + return -EINVAL; + if (attr->value_size > KMALLOC_MAX_SIZE) /* if value_size is bigger, the user space won't be able to * access the elements. @@ -1134,6 +1139,9 @@ static void perf_event_fd_array_release(struct bpf_map *map, struct bpf_event_entry *ee; int i; + if (map->map_flags & BPF_F_PRESERVE_ELEMS) + return; + rcu_read_lock(); for (i = 0; i < array->map.max_entries; i++) { ee = READ_ONCE(array->ptrs[i]); @@ -1143,12 +1151,19 @@ static void perf_event_fd_array_release(struct bpf_map *map, rcu_read_unlock(); } +static void perf_event_fd_array_map_free(struct bpf_map *map) +{ + if (map->map_flags & BPF_F_PRESERVE_ELEMS) + bpf_fd_array_map_clear(map); + fd_array_map_free(map); +} + static int perf_event_array_map_btf_id; const struct bpf_map_ops perf_event_array_map_ops = { .map_meta_equal = bpf_map_meta_equal, .map_alloc_check = fd_array_map_alloc_check, .map_alloc = array_map_alloc, - .map_free = fd_array_map_free, + .map_free = perf_event_fd_array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = fd_array_map_lookup_elem, .map_delete_elem = fd_array_map_delete_elem, diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 1f17c6752debb..4f556cfcbfbee 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -414,6 +414,9 @@ enum { /* Enable memory-mapping BPF map */ BPF_F_MMAPABLE = (1U << 10), + +/* Share perf_event among processes */ + BPF_F_PRESERVE_ELEMS = (1U << 11), }; /* Flags for BPF_PROG_QUERY. */