From patchwork Fri May 22 02:23:42 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Martin KaFai Lau X-Patchwork-Id: 1295815 X-Patchwork-Delegate: bpf@iogearbox.net Return-Path: X-Original-To: patchwork-incoming-netdev@ozlabs.org Delivered-To: patchwork-incoming-netdev@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=netdev-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=pass (p=quarantine 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=FhIfD7z3; dkim-atps=neutral Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by ozlabs.org (Postfix) with ESMTP id 49Sqzz2vrBz9sT1 for ; Fri, 22 May 2020 12:23:59 +1000 (AEST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727851AbgEVCXt (ORCPT ); Thu, 21 May 2020 22:23:49 -0400 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:46190 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727794AbgEVCXs (ORCPT ); Thu, 21 May 2020 22:23:48 -0400 Received: from pps.filterd (m0044012.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.16.0.42/8.16.0.42) with SMTP id 04M2LM53003737 for ; Thu, 21 May 2020 19:23:47 -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=nDfUsQsACtAzt8gfblfmYubzrh49rUze0IvmFCrIFjg=; b=FhIfD7z39W3teLNHlvPElsYI5hX9HqQG5D3XhndYGFforVNIkDAsYRznnM1HnWTZUX63 b01AGGwAoniTT/CtiGsxbjnct/2d8Q8VxUtFlwI7mGFaBUqjkMyCN1uOu+HU5JspO0dB aSiaRrrBpLzMIo6/3A6Th2Ps09X5RIPe14A= Received: from mail.thefacebook.com ([163.114.132.120]) by mx0a-00082601.pphosted.com with ESMTP id 3152xu2nf6-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Thu, 21 May 2020 19:23:47 -0700 Received: from intmgw003.08.frc2.facebook.com (2620:10d:c085:108::4) by mail.thefacebook.com (2620:10d:c085:21d::7) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.1979.3; Thu, 21 May 2020 19:23:46 -0700 Received: by devbig005.ftw2.facebook.com (Postfix, from userid 6611) id E970929455B1; Thu, 21 May 2020 19:23:42 -0700 (PDT) Smtp-Origin-Hostprefix: devbig From: Martin KaFai Lau Smtp-Origin-Hostname: devbig005.ftw2.facebook.com To: CC: Alexei Starovoitov , Daniel Borkmann , , , Andrey Ignatov Smtp-Origin-Cluster: ftw2c04 Subject: [PATCH v2 bpf-next 1/3] bpf: Consolidate inner-map-compatible properties into bpf_types.h Date: Thu, 21 May 2020 19:23:42 -0700 Message-ID: <20200522022342.899756-1-kafai@fb.com> X-Mailer: git-send-email 2.24.1 In-Reply-To: <20200522022336.899416-1-kafai@fb.com> References: <20200522022336.899416-1-kafai@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.216,18.0.676 definitions=2020-05-21_16:2020-05-21,2020-05-21 signatures=0 X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 suspectscore=13 cotscore=-2147483648 priorityscore=1501 mlxscore=0 malwarescore=0 adultscore=0 mlxlogscore=999 lowpriorityscore=0 phishscore=0 clxscore=1015 bulkscore=0 spamscore=0 impostorscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2004280000 definitions=main-2005220017 X-FB-Internal: deliver Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org This patch adds a "properties" member to "struct bpf_map". The properties of individual map-type is tagged in the "BPF_MAP_TYPE_FL" in bpf_types.h. The original "BPF_MAP_TYPE" is defined to "BPF_MAP_TYPE_FL(..., 0)". It will be less error prone when a map's properties is decided at the same place as the new map type is added in bpf_types.h. That will help to avoid mistake like missing modification in other source files like the map_in_map.c here or other source files in the future. Cc: Andrey Ignatov Signed-off-by: Martin KaFai Lau --- include/linux/bpf.h | 8 ++++++-- include/linux/bpf_types.h | 19 +++++++++++++++---- kernel/bpf/btf.c | 4 ++-- kernel/bpf/map_in_map.c | 9 ++------- kernel/bpf/syscall.c | 25 +++++++++++++++++++------ kernel/bpf/verifier.c | 4 ++-- 6 files changed, 46 insertions(+), 23 deletions(-) diff --git a/include/linux/bpf.h b/include/linux/bpf.h index efe8836b5c48..f947d899aa46 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -97,6 +97,9 @@ struct bpf_map_memory { struct user_struct *user; }; +/* Cannot be used as an inner map */ +#define BPF_MAP_NO_INNER_MAP (1 << 0) + struct bpf_map { /* The first two cachelines with read-mostly members of which some * are also accessed in fast-path (e.g. ops, max_entries). @@ -120,6 +123,7 @@ struct bpf_map { struct bpf_map_memory memory; char name[BPF_OBJ_NAME_LEN]; u32 btf_vmlinux_value_type_id; + u32 properties; bool bypass_spec_v1; bool frozen; /* write-once; write-protected by freeze_mutex */ /* 22 bytes hole */ @@ -1037,12 +1041,12 @@ extern const struct file_operations bpf_iter_fops; #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \ extern const struct bpf_prog_ops _name ## _prog_ops; \ extern const struct bpf_verifier_ops _name ## _verifier_ops; -#define BPF_MAP_TYPE(_id, _ops) \ +#define BPF_MAP_TYPE_FL(_id, _ops, properties) \ extern const struct bpf_map_ops _ops; #define BPF_LINK_TYPE(_id, _name) #include #undef BPF_PROG_TYPE -#undef BPF_MAP_TYPE +#undef BPF_MAP_TYPE_FL #undef BPF_LINK_TYPE extern const struct bpf_prog_ops bpf_offload_prog_ops; diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h index 29d22752fc87..3f32702c9bf4 100644 --- a/include/linux/bpf_types.h +++ b/include/linux/bpf_types.h @@ -76,16 +76,25 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_LSM, lsm, #endif /* CONFIG_BPF_LSM */ #endif +#define BPF_MAP_TYPE(x, y) BPF_MAP_TYPE_FL(x, y, 0) + BPF_MAP_TYPE(BPF_MAP_TYPE_ARRAY, array_map_ops) BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_ARRAY, percpu_array_map_ops) -BPF_MAP_TYPE(BPF_MAP_TYPE_PROG_ARRAY, prog_array_map_ops) +/* prog_array->aux->{type,jited} is a runtime binding. + * Doing static check alone in the verifier is not enough, + * so BPF_MAP_NO_INNTER_MAP is needed. + */ +BPF_MAP_TYPE_FL(BPF_MAP_TYPE_PROG_ARRAY, prog_array_map_ops, + BPF_MAP_NO_INNER_MAP) BPF_MAP_TYPE(BPF_MAP_TYPE_PERF_EVENT_ARRAY, perf_event_array_map_ops) #ifdef CONFIG_CGROUPS BPF_MAP_TYPE(BPF_MAP_TYPE_CGROUP_ARRAY, cgroup_array_map_ops) #endif #ifdef CONFIG_CGROUP_BPF -BPF_MAP_TYPE(BPF_MAP_TYPE_CGROUP_STORAGE, cgroup_storage_map_ops) -BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE, cgroup_storage_map_ops) +BPF_MAP_TYPE_FL(BPF_MAP_TYPE_CGROUP_STORAGE, cgroup_storage_map_ops, + BPF_MAP_NO_INNER_MAP) +BPF_MAP_TYPE_FL(BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE, cgroup_storage_map_ops, + BPF_MAP_NO_INNER_MAP) #endif BPF_MAP_TYPE(BPF_MAP_TYPE_HASH, htab_map_ops) BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_HASH, htab_percpu_map_ops) @@ -116,8 +125,10 @@ BPF_MAP_TYPE(BPF_MAP_TYPE_REUSEPORT_SOCKARRAY, reuseport_array_ops) BPF_MAP_TYPE(BPF_MAP_TYPE_QUEUE, queue_map_ops) BPF_MAP_TYPE(BPF_MAP_TYPE_STACK, stack_map_ops) #if defined(CONFIG_BPF_JIT) -BPF_MAP_TYPE(BPF_MAP_TYPE_STRUCT_OPS, bpf_struct_ops_map_ops) +BPF_MAP_TYPE_FL(BPF_MAP_TYPE_STRUCT_OPS, bpf_struct_ops_map_ops, + BPF_MAP_NO_INNER_MAP) #endif +#undef BPF_MAP_TYPE BPF_LINK_TYPE(BPF_LINK_TYPE_RAW_TRACEPOINT, raw_tracepoint) BPF_LINK_TYPE(BPF_LINK_TYPE_TRACING, tracing) diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 58c9af1d4808..f527a120eef3 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -3481,7 +3481,7 @@ extern char __weak __start_BTF[]; extern char __weak __stop_BTF[]; extern struct btf *btf_vmlinux; -#define BPF_MAP_TYPE(_id, _ops) +#define BPF_MAP_TYPE_FL(_id, _ops, properties) #define BPF_LINK_TYPE(_id, _name) static union { struct bpf_ctx_convert { @@ -3508,7 +3508,7 @@ static u8 bpf_ctx_convert_map[] = { #undef BPF_PROG_TYPE 0, /* avoid empty array */ }; -#undef BPF_MAP_TYPE +#undef BPF_MAP_TYPE_FL #undef BPF_LINK_TYPE static const struct btf_member * diff --git a/kernel/bpf/map_in_map.c b/kernel/bpf/map_in_map.c index 17738c93bec8..d965a1d328a9 100644 --- a/kernel/bpf/map_in_map.c +++ b/kernel/bpf/map_in_map.c @@ -17,13 +17,7 @@ struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd) if (IS_ERR(inner_map)) return inner_map; - /* prog_array->aux->{type,jited} is a runtime binding. - * Doing static check alone in the verifier is not enough. - */ - if (inner_map->map_type == BPF_MAP_TYPE_PROG_ARRAY || - inner_map->map_type == BPF_MAP_TYPE_CGROUP_STORAGE || - inner_map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE || - inner_map->map_type == BPF_MAP_TYPE_STRUCT_OPS) { + if (inner_map->properties & BPF_MAP_NO_INNER_MAP) { fdput(f); return ERR_PTR(-ENOTSUPP); } @@ -56,6 +50,7 @@ struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd) inner_map_meta->map_flags = inner_map->map_flags; inner_map_meta->max_entries = inner_map->max_entries; inner_map_meta->spin_lock_off = inner_map->spin_lock_off; + inner_map_meta->properties = inner_map->properties; /* Misc members not needed in bpf_map_meta_equal() check. */ inner_map_meta->ops = inner_map->ops; diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 431241c74614..02ec14e3e19a 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -49,12 +49,23 @@ int sysctl_unprivileged_bpf_disabled __read_mostly; static const struct bpf_map_ops * const bpf_map_types[] = { #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) -#define BPF_MAP_TYPE(_id, _ops) \ +#define BPF_MAP_TYPE_FL(_id, _ops, properties) \ [_id] = &_ops, #define BPF_LINK_TYPE(_id, _name) #include #undef BPF_PROG_TYPE -#undef BPF_MAP_TYPE +#undef BPF_MAP_TYPE_FL +#undef BPF_LINK_TYPE +}; + +static const u32 bpf_map_properties[] = { +#define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) +#define BPF_MAP_TYPE_FL(_id, _ops, properties) \ + [_id] = properties, +#define BPF_LINK_TYPE(_id, _name) +#include +#undef BPF_PROG_TYPE +#undef BPF_MAP_TYPE_FL #undef BPF_LINK_TYPE }; @@ -131,6 +142,8 @@ static struct bpf_map *find_and_alloc_map(union bpf_attr *attr) return map; map->ops = ops; map->map_type = type; + map->properties = bpf_map_properties[type]; + return map; } @@ -1551,11 +1564,11 @@ static int map_freeze(const union bpf_attr *attr) static const struct bpf_prog_ops * const bpf_prog_types[] = { #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \ [_id] = & _name ## _prog_ops, -#define BPF_MAP_TYPE(_id, _ops) +#define BPF_MAP_TYPE_FL(_id, _ops, properties) #define BPF_LINK_TYPE(_id, _name) #include #undef BPF_PROG_TYPE -#undef BPF_MAP_TYPE +#undef BPF_MAP_TYPE_FL #undef BPF_LINK_TYPE }; @@ -2333,14 +2346,14 @@ static int bpf_link_release(struct inode *inode, struct file *filp) #ifdef CONFIG_PROC_FS #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) -#define BPF_MAP_TYPE(_id, _ops) +#define BPF_MAP_TYPE_FL(_id, _ops, properties) #define BPF_LINK_TYPE(_id, _name) [_id] = #_name, static const char *bpf_link_type_strs[] = { [BPF_LINK_TYPE_UNSPEC] = "", #include }; #undef BPF_PROG_TYPE -#undef BPF_MAP_TYPE +#undef BPF_MAP_TYPE_FL #undef BPF_LINK_TYPE static void bpf_link_show_fdinfo(struct seq_file *m, struct file *filp) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 2ed8351f47a4..1743e070d08f 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -27,11 +27,11 @@ static const struct bpf_verifier_ops * const bpf_verifier_ops[] = { #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \ [_id] = & _name ## _verifier_ops, -#define BPF_MAP_TYPE(_id, _ops) +#define BPF_MAP_TYPE_FL(_id, _ops, properties) #define BPF_LINK_TYPE(_id, _name) #include #undef BPF_PROG_TYPE -#undef BPF_MAP_TYPE +#undef BPF_MAP_TYPE_FL #undef BPF_LINK_TYPE }; From patchwork Fri May 22 02:23:49 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Martin KaFai Lau X-Patchwork-Id: 1295819 X-Patchwork-Delegate: bpf@iogearbox.net Return-Path: X-Original-To: patchwork-incoming-netdev@ozlabs.org Delivered-To: patchwork-incoming-netdev@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=netdev-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=pass (p=quarantine 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=LAmLSMcB; dkim-atps=neutral Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by ozlabs.org (Postfix) with ESMTP id 49Sr0N39Xxz9sSw for ; Fri, 22 May 2020 12:24:20 +1000 (AEST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727880AbgEVCYS (ORCPT ); Thu, 21 May 2020 22:24:18 -0400 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:6272 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727050AbgEVCYR (ORCPT ); Thu, 21 May 2020 22:24:17 -0400 Received: from pps.filterd (m0044012.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.16.0.42/8.16.0.42) with SMTP id 04M2LLZR003705 for ; Thu, 21 May 2020 19:24:17 -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=p9yZPdW2gni/uc5W3qY2pPfvdzlqbhy/hev0rUupsZU=; b=LAmLSMcBjxuwlZ6x52F8nFyD3HFhKLEreF9aW+KmVzVzjw7XzoguEC+rNYxiktMBDtKd cereCH3XaLFPmxyjxo97OkAoN+oRSeGCo1k9F/6CSCbg/G8Yst9vEG7QSP/tVPBkUppO bVCDuAcFW2FAbDtRr+PUHh1X9yhY2aqv0Fo= Received: from mail.thefacebook.com ([163.114.132.120]) by mx0a-00082601.pphosted.com with ESMTP id 3152xu2nkv-2 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Thu, 21 May 2020 19:24:17 -0700 Received: from intmgw002.08.frc2.facebook.com (2620:10d:c085:208::f) by mail.thefacebook.com (2620:10d:c085:11d::7) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.1979.3; Thu, 21 May 2020 19:23:55 -0700 Received: by devbig005.ftw2.facebook.com (Postfix, from userid 6611) id 3626D29455B1; Thu, 21 May 2020 19:23:49 -0700 (PDT) Smtp-Origin-Hostprefix: devbig From: Martin KaFai Lau Smtp-Origin-Hostname: devbig005.ftw2.facebook.com To: CC: Alexei Starovoitov , Daniel Borkmann , , , Andrey Ignatov Smtp-Origin-Cluster: ftw2c04 Subject: [PATCH v2 bpf-next 2/3] bpf: Relax the max_entries check for inner map Date: Thu, 21 May 2020 19:23:49 -0700 Message-ID: <20200522022349.900034-1-kafai@fb.com> X-Mailer: git-send-email 2.24.1 In-Reply-To: <20200522022336.899416-1-kafai@fb.com> References: <20200522022336.899416-1-kafai@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.216,18.0.676 definitions=2020-05-21_16:2020-05-21,2020-05-21 signatures=0 X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 suspectscore=13 cotscore=-2147483648 priorityscore=1501 mlxscore=0 malwarescore=0 adultscore=0 mlxlogscore=676 lowpriorityscore=0 phishscore=0 clxscore=1015 bulkscore=0 spamscore=0 impostorscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2004280000 definitions=main-2005220017 X-FB-Internal: deliver Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org This patch relaxes the max_entries check for most of the inner map types during an update to the outer map. The max_entries of those map types are only used in runtime. By doing this, an inner map with different size can be updated to the outer map in runtime. People has a use case that starts with a smaller inner map first and then replaces it with a larger inner map later when it is needed. The max_entries of arraymap and xskmap are used statically in verification time to generate the inline code, so they are excluded in this patch. Cc: Andrey Ignatov Signed-off-by: Martin KaFai Lau --- include/linux/bpf.h | 12 ++++++++++++ include/linux/bpf_types.h | 6 ++++-- kernel/bpf/map_in_map.c | 3 ++- 3 files changed, 18 insertions(+), 3 deletions(-) diff --git a/include/linux/bpf.h b/include/linux/bpf.h index f947d899aa46..1839ef9aca02 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -99,6 +99,18 @@ struct bpf_map_memory { /* Cannot be used as an inner map */ #define BPF_MAP_NO_INNER_MAP (1 << 0) +/* When a prog has used map-in-map, the verifier requires + * an inner-map as a template to verify the access operations + * on the outer and inner map. For some inner map-types, + * the verifier uses the inner_map's max_entries statically + * (e.g. to generate inline code). If this verification + * time usage on max_entries applies to an inner map-type, + * during runtime, only the inner map with the same + * max_entries can be updated to this outer map. + * + * Please see bpf_map_meta_equal() for details. + */ +#define BPF_MAP_NO_DYNAMIC_INNER_MAP_SIZE (1 << 1) struct bpf_map { /* The first two cachelines with read-mostly members of which some diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h index 3f32702c9bf4..85861722b7f3 100644 --- a/include/linux/bpf_types.h +++ b/include/linux/bpf_types.h @@ -78,7 +78,8 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_LSM, lsm, #define BPF_MAP_TYPE(x, y) BPF_MAP_TYPE_FL(x, y, 0) -BPF_MAP_TYPE(BPF_MAP_TYPE_ARRAY, array_map_ops) +BPF_MAP_TYPE_FL(BPF_MAP_TYPE_ARRAY, array_map_ops, + BPF_MAP_NO_DYNAMIC_INNER_MAP_SIZE) BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_ARRAY, percpu_array_map_ops) /* prog_array->aux->{type,jited} is a runtime binding. * Doing static check alone in the verifier is not enough, @@ -116,7 +117,8 @@ BPF_MAP_TYPE(BPF_MAP_TYPE_SOCKHASH, sock_hash_ops) #endif BPF_MAP_TYPE(BPF_MAP_TYPE_CPUMAP, cpu_map_ops) #if defined(CONFIG_XDP_SOCKETS) -BPF_MAP_TYPE(BPF_MAP_TYPE_XSKMAP, xsk_map_ops) +BPF_MAP_TYPE_FL(BPF_MAP_TYPE_XSKMAP, xsk_map_ops, + BPF_MAP_NO_DYNAMIC_INNER_MAP_SIZE) #endif #ifdef CONFIG_INET BPF_MAP_TYPE(BPF_MAP_TYPE_REUSEPORT_SOCKARRAY, reuseport_array_ops) diff --git a/kernel/bpf/map_in_map.c b/kernel/bpf/map_in_map.c index d965a1d328a9..b296fe8af1ad 100644 --- a/kernel/bpf/map_in_map.c +++ b/kernel/bpf/map_in_map.c @@ -77,7 +77,8 @@ bool bpf_map_meta_equal(const struct bpf_map *meta0, meta0->key_size == meta1->key_size && meta0->value_size == meta1->value_size && meta0->map_flags == meta1->map_flags && - meta0->max_entries == meta1->max_entries; + (meta0->max_entries == meta1->max_entries || + !(meta0->properties & BPF_MAP_NO_DYNAMIC_INNER_MAP_SIZE)); } void *bpf_map_fd_get_ptr(struct bpf_map *map, From patchwork Fri May 22 02:23:55 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Martin KaFai Lau X-Patchwork-Id: 1295818 X-Patchwork-Delegate: bpf@iogearbox.net Return-Path: X-Original-To: patchwork-incoming-netdev@ozlabs.org Delivered-To: patchwork-incoming-netdev@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=netdev-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=pass (p=quarantine 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=BwVrVZnb; dkim-atps=neutral Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by ozlabs.org (Postfix) with ESMTP id 49Sr0D5kGmz9sT1 for ; Fri, 22 May 2020 12:24:12 +1000 (AEST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727854AbgEVCYI (ORCPT ); Thu, 21 May 2020 22:24:08 -0400 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:25672 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727050AbgEVCYI (ORCPT ); Thu, 21 May 2020 22:24:08 -0400 Received: from pps.filterd (m0109334.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.16.0.42/8.16.0.42) with SMTP id 04M2O7Mp018088 for ; Thu, 21 May 2020 19:24:07 -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=gU+TNATQ6pS6u5lCv+LO1PcwVc8Wrajrb96jRxaRnMs=; b=BwVrVZnb+ren9zjhCTf6vka3t/uirAaX0rXx7sm0w+UOBQKISgoN5cLf8M1FTzy1ybw3 A9J5ov5JwmFxYxNxdTQN4Bgqu/h7IuFdeX6BwThjFBPf6OHe/vHZSZnVIQ2C2PMwlLUG NI2XKnyZJNx6ZiNM8mvNRNF7fn8okRdCTpI= Received: from mail.thefacebook.com ([163.114.132.120]) by mx0a-00082601.pphosted.com with ESMTP id 314jubn1sn-4 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Thu, 21 May 2020 19:24:07 -0700 Received: from intmgw002.03.ash8.facebook.com (2620:10d:c085:208::f) by mail.thefacebook.com (2620:10d:c085:11d::6) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.1979.3; Thu, 21 May 2020 19:23:57 -0700 Received: by devbig005.ftw2.facebook.com (Postfix, from userid 6611) id 74E8829455B1; Thu, 21 May 2020 19:23:55 -0700 (PDT) Smtp-Origin-Hostprefix: devbig From: Martin KaFai Lau Smtp-Origin-Hostname: devbig005.ftw2.facebook.com To: CC: Alexei Starovoitov , Daniel Borkmann , , , Andrey Ignatov Smtp-Origin-Cluster: ftw2c04 Subject: [PATCH v2 bpf-next 3/3] bpf: selftests: Add test for different inner map size Date: Thu, 21 May 2020 19:23:55 -0700 Message-ID: <20200522022355.900283-1-kafai@fb.com> X-Mailer: git-send-email 2.24.1 In-Reply-To: <20200522022336.899416-1-kafai@fb.com> References: <20200522022336.899416-1-kafai@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.216,18.0.676 definitions=2020-05-21_16:2020-05-21,2020-05-21 signatures=0 X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 priorityscore=1501 lowpriorityscore=0 malwarescore=0 mlxscore=0 spamscore=0 cotscore=-2147483648 mlxlogscore=674 clxscore=1015 bulkscore=0 adultscore=0 impostorscore=0 suspectscore=13 phishscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2004280000 definitions=main-2005220018 X-FB-Internal: deliver Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org This patch tests the inner map size can be different for reuseport_sockarray but has to be the same for arraymap. Cc: Andrey Ignatov Signed-off-by: Martin KaFai Lau --- .../selftests/bpf/prog_tests/btf_map_in_map.c | 12 +++++++ .../selftests/bpf/progs/test_btf_map_in_map.c | 31 +++++++++++++++++++ 2 files changed, 43 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/btf_map_in_map.c b/tools/testing/selftests/bpf/prog_tests/btf_map_in_map.c index f7ee8fa377ad..7ae767d15608 100644 --- a/tools/testing/selftests/bpf/prog_tests/btf_map_in_map.c +++ b/tools/testing/selftests/bpf/prog_tests/btf_map_in_map.c @@ -44,6 +44,18 @@ void test_btf_map_in_map(void) bpf_map_lookup_elem(bpf_map__fd(skel->maps.inner_map2), &key, &val); CHECK(val != 3, "inner2", "got %d != exp %d\n", val, 3); + val = bpf_map__fd(skel->maps.sockarr_sz2); + err = bpf_map_update_elem(bpf_map__fd(skel->maps.outer_sockarr_sz1), + &key, &val, 0); + CHECK(err, "outer_sockarr inner map size check", + "cannot use an inner_map with different size\n"); + + val = bpf_map__fd(skel->maps.inner_map_sz2); + err = bpf_map_update_elem(bpf_map__fd(skel->maps.outer_arr), &key, + &val, 0); + CHECK(!err, "outer_arr inner map size check", + "incorrectly updated with an inner_map in different size\n"); + cleanup: test_btf_map_in_map__destroy(skel); } diff --git a/tools/testing/selftests/bpf/progs/test_btf_map_in_map.c b/tools/testing/selftests/bpf/progs/test_btf_map_in_map.c index e5093796be97..0b8e04a817f6 100644 --- a/tools/testing/selftests/bpf/progs/test_btf_map_in_map.c +++ b/tools/testing/selftests/bpf/progs/test_btf_map_in_map.c @@ -11,6 +11,13 @@ struct inner_map { } inner_map1 SEC(".maps"), inner_map2 SEC(".maps"); +struct inner_map_sz2 { + __uint(type, BPF_MAP_TYPE_ARRAY); + __uint(max_entries, 2); + __type(key, int); + __type(value, int); +} inner_map_sz2 SEC(".maps"); + struct outer_arr { __uint(type, BPF_MAP_TYPE_ARRAY_OF_MAPS); __uint(max_entries, 3); @@ -50,6 +57,30 @@ struct outer_hash { }, }; +struct sockarr_sz1 { + __uint(type, BPF_MAP_TYPE_REUSEPORT_SOCKARRAY); + __uint(max_entries, 1); + __type(key, int); + __type(value, int); +} sockarr_sz1 SEC(".maps"); + +struct sockarr_sz2 { + __uint(type, BPF_MAP_TYPE_REUSEPORT_SOCKARRAY); + __uint(max_entries, 2); + __type(key, int); + __type(value, int); +} sockarr_sz2 SEC(".maps"); + +struct outer_sockarr_sz1 { + __uint(type, BPF_MAP_TYPE_ARRAY_OF_MAPS); + __uint(max_entries, 1); + __uint(key_size, sizeof(int)); + __uint(value_size, sizeof(int)); + __array(values, struct sockarr_sz1); +} outer_sockarr_sz1 SEC(".maps") = { + .values = { (void *)&sockarr_sz1 }, +}; + int input = 0; SEC("raw_tp/sys_enter")