From patchwork Wed Oct 28 17:46:01 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paolo Abeni X-Patchwork-Id: 11863317 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 B9F8261C for ; Wed, 28 Oct 2020 21:48:28 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 9223224724 for ; Wed, 28 Oct 2020 21:48:28 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="EcNJ1+7J" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727798AbgJ1VsX (ORCPT ); Wed, 28 Oct 2020 17:48:23 -0400 Received: from us-smtp-delivery-124.mimecast.com ([216.205.24.124]:22704 "EHLO us-smtp-delivery-124.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727745AbgJ1VsP (ORCPT ); Wed, 28 Oct 2020 17:48:15 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1603921693; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=WnT8yDT00QyRsUEZ7gp5zTCd68WeeplEqh2AFWI8M0Q=; b=EcNJ1+7JAVXNRy6zyZJdJq6Gzi/neB9ZBC1IZvsV3IzGWnXvjlKmEJm6tRlLeMvWw22kXZ Askn1PKvw7M25ZspAqglkYJRw4gtBkcFXfHdAHOOuKL2CapWCBPoIaDoNr5p1cZGj6a9LS zbZBOabc5lXoNub0Ltcax/KwmAx0Cwg= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-350-hjv0Nrk9NZSQqCrpIjV5fA-1; Wed, 28 Oct 2020 13:46:55 -0400 X-MC-Unique: hjv0Nrk9NZSQqCrpIjV5fA-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 48008106B26F; Wed, 28 Oct 2020 17:46:35 +0000 (UTC) Received: from gerbillo.redhat.com (ovpn-115-68.ams2.redhat.com [10.36.115.68]) by smtp.corp.redhat.com (Postfix) with ESMTP id 48FDC6198D; Wed, 28 Oct 2020 17:46:33 +0000 (UTC) From: Paolo Abeni To: netdev@vger.kernel.org Cc: Jonathan Corbet , "David S. Miller" , Jakub Kicinski , Shuah Khan , linux-doc@vger.kernel.org, linux-kselftest@vger.kernel.org, Marcelo Tosatti Subject: [PATCH net-next 1/3] net/sysctl: factor-out netdev_rx_queue_set_rps_mask() helper Date: Wed, 28 Oct 2020 18:46:01 +0100 Message-Id: <30af9047229612189e454d4279cb669fe01db833.1603906564.git.pabeni@redhat.com> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Precedence: bulk List-ID: X-Mailing-List: linux-kselftest@vger.kernel.org Will simplify the following patch. No functional change intended. Signed-off-by: Paolo Abeni --- net/core/net-sysfs.c | 66 ++++++++++++++++++++++++-------------------- 1 file changed, 36 insertions(+), 30 deletions(-) diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 94fff0700bdd..b57426707216 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -737,42 +737,18 @@ static ssize_t show_rps_map(struct netdev_rx_queue *queue, char *buf) return len < PAGE_SIZE ? len : -EINVAL; } -static ssize_t store_rps_map(struct netdev_rx_queue *queue, - const char *buf, size_t len) +static int netdev_rx_queue_set_rps_mask(struct netdev_rx_queue *queue, + cpumask_var_t mask) { - struct rps_map *old_map, *map; - cpumask_var_t mask; - int err, cpu, i, hk_flags; static DEFINE_MUTEX(rps_map_mutex); - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (!alloc_cpumask_var(&mask, GFP_KERNEL)) - return -ENOMEM; - - err = bitmap_parse(buf, len, cpumask_bits(mask), nr_cpumask_bits); - if (err) { - free_cpumask_var(mask); - return err; - } - - if (!cpumask_empty(mask)) { - hk_flags = HK_FLAG_DOMAIN | HK_FLAG_WQ; - cpumask_and(mask, mask, housekeeping_cpumask(hk_flags)); - if (cpumask_empty(mask)) { - free_cpumask_var(mask); - return -EINVAL; - } - } + struct rps_map *old_map, *map; + int cpu, i; map = kzalloc(max_t(unsigned int, RPS_MAP_SIZE(cpumask_weight(mask)), L1_CACHE_BYTES), GFP_KERNEL); - if (!map) { - free_cpumask_var(mask); + if (!map) return -ENOMEM; - } i = 0; for_each_cpu_and(cpu, mask, cpu_online_mask) @@ -799,9 +775,39 @@ static ssize_t store_rps_map(struct netdev_rx_queue *queue, if (old_map) kfree_rcu(old_map, rcu); + return 0; +} + +static ssize_t store_rps_map(struct netdev_rx_queue *queue, + const char *buf, size_t len) +{ + cpumask_var_t mask; + int err, hk_flags; + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) + return -ENOMEM; + + err = bitmap_parse(buf, len, cpumask_bits(mask), nr_cpumask_bits); + if (err) + goto out; + if (!cpumask_empty(mask)) { + hk_flags = HK_FLAG_DOMAIN | HK_FLAG_WQ; + cpumask_and(mask, mask, housekeeping_cpumask(hk_flags)); + if (cpumask_empty(mask)) { + err = -EINVAL; + goto out; + } + } + + err = netdev_rx_queue_set_rps_mask(queue, mask); + +out: free_cpumask_var(mask); - return len; + return err ? : len; } static ssize_t show_rps_dev_flow_table_cnt(struct netdev_rx_queue *queue, From patchwork Wed Oct 28 17:46:02 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paolo Abeni X-Patchwork-Id: 11863319 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 A077161C for ; Wed, 28 Oct 2020 21:48:30 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 76A9D2473C for ; Wed, 28 Oct 2020 21:48:30 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="FuMmOga7" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727779AbgJ1VsX (ORCPT ); Wed, 28 Oct 2020 17:48:23 -0400 Received: from us-smtp-delivery-124.mimecast.com ([216.205.24.124]:28326 "EHLO us-smtp-delivery-124.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727738AbgJ1VsP (ORCPT ); Wed, 28 Oct 2020 17:48:15 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1603921693; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=PEE231hPUo3ICeopUFgljqFJfbF/lI2sunK5lJ1xv+0=; b=FuMmOga7NqRpBg2Ux+usq7Wa/9LwdBoRt7EOITMS/qzbEI5jYNRjE96+vyeuCSzDVXCXjE zMTW8Zg9hg7OUkUn3uBEnHiO1ZNxXuGY7DE5FNnzd2KZUD1RWpoaLrbrlxq829nBNcS6bx 9+qqmFQUjf2Vo9wM0Sb/g0T/3CJaahU= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-248-1r-j0238PgmTTE84SwwMUg-1; Wed, 28 Oct 2020 13:46:57 -0400 X-MC-Unique: 1r-j0238PgmTTE84SwwMUg-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 903991020913; Wed, 28 Oct 2020 17:46:37 +0000 (UTC) Received: from gerbillo.redhat.com (ovpn-115-68.ams2.redhat.com [10.36.115.68]) by smtp.corp.redhat.com (Postfix) with ESMTP id 9E85F6198D; Wed, 28 Oct 2020 17:46:35 +0000 (UTC) From: Paolo Abeni To: netdev@vger.kernel.org Cc: Jonathan Corbet , "David S. Miller" , Jakub Kicinski , Shuah Khan , linux-doc@vger.kernel.org, linux-kselftest@vger.kernel.org, Marcelo Tosatti Subject: [PATCH net-next 2/3] net/core: introduce default_rps_mask netns attribute Date: Wed, 28 Oct 2020 18:46:02 +0100 Message-Id: <9e86568c264696dbe0fd44b2a8662bd233e2c3e8.1603906564.git.pabeni@redhat.com> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Precedence: bulk List-ID: X-Mailing-List: linux-kselftest@vger.kernel.org If RPS is enabled, this allows configuring a default rps mask, which is effective since receive queue creation time. A default RPS mask allows the system admin to ensure proper isolation, avoiding races at network namespace or device creation time. The default RPS mask is initially empty, and can be modified via a newly added sysctl entry. Signed-off-by: Paolo Abeni --- Documentation/admin-guide/sysctl/net.rst | 6 +++ net/core/net-sysfs.c | 9 ++++ net/core/sysctl_net_core.c | 58 ++++++++++++++++++++++++ 3 files changed, 73 insertions(+) diff --git a/Documentation/admin-guide/sysctl/net.rst b/Documentation/admin-guide/sysctl/net.rst index 57fd6ce68fe0..818cb2030a8b 100644 --- a/Documentation/admin-guide/sysctl/net.rst +++ b/Documentation/admin-guide/sysctl/net.rst @@ -211,6 +211,12 @@ rmem_max The maximum receive socket buffer size in bytes. +rps_default_mask +---------------- + +The default RPS CPU mask used on newly created network devices. An empty +mask means RPS disabled by default. + tstamp_allow_data ----------------- Allow processes to receive tx timestamps looped together with the original diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index b57426707216..991b668f8a12 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -715,6 +715,8 @@ static const struct sysfs_ops rx_queue_sysfs_ops = { }; #ifdef CONFIG_RPS +extern struct cpumask rps_default_mask; + static ssize_t show_rps_map(struct netdev_rx_queue *queue, char *buf) { struct rps_map *map; @@ -983,6 +985,13 @@ static int rx_queue_add_kobject(struct net_device *dev, int index) goto err; } +#if IS_ENABLED(CONFIG_RPS) && IS_ENABLED(CONFIG_SYSCTL) + if (!cpumask_empty(&rps_default_mask)) { + error = netdev_rx_queue_set_rps_mask(queue, &rps_default_mask); + if (error) + goto err; + } +#endif kobject_uevent(kobj, KOBJ_ADD); return error; diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index d86d8d11cfe4..13451ac88a74 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -46,6 +47,54 @@ int sysctl_devconf_inherit_init_net __read_mostly; EXPORT_SYMBOL(sysctl_devconf_inherit_init_net); #ifdef CONFIG_RPS +struct cpumask rps_default_mask; + +static int rps_default_mask_sysctl(struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + int len, err = 0; + + rtnl_lock(); + if (write) { + err = cpumask_parse(buffer, &rps_default_mask); + if (err) + goto done; + + if (!cpumask_empty(&rps_default_mask)) { + int hk_flags = HK_FLAG_DOMAIN | HK_FLAG_WQ; + cpumask_and(&rps_default_mask, &rps_default_mask, + housekeeping_cpumask(hk_flags)); + if (cpumask_empty(&rps_default_mask)) { + err = -EINVAL; + goto done; + } + } + } else { + char kbuf[128]; + + if (*ppos || !*lenp) { + *lenp = 0; + goto done; + } + + len = min(sizeof(kbuf) - 1, *lenp); + len = scnprintf(kbuf, len, "%*pb", cpumask_pr_args(&rps_default_mask)); + if (!len) { + *lenp = 0; + goto done; + } + if (len < *lenp) + kbuf[len++] = '\n'; + memcpy(buffer, kbuf, len); + *lenp = len; + *ppos += len; + } + +done: + rtnl_unlock(); + return err; +} + static int rps_sock_flow_sysctl(struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { @@ -466,6 +515,11 @@ static struct ctl_table net_core_table[] = { .mode = 0644, .proc_handler = rps_sock_flow_sysctl }, + { + .procname = "rps_default_mask", + .mode = 0644, + .proc_handler = rps_default_mask_sysctl + }, #endif #ifdef CONFIG_NET_FLOW_LIMIT { @@ -648,6 +702,10 @@ static __net_initdata struct pernet_operations sysctl_core_ops = { static __init int sysctl_core_init(void) { +#if IS_ENABLED(CONFIG_RPS) + cpumask_copy(&rps_default_mask, cpu_none_mask); +#endif + register_net_sysctl(&init_net, "net/core", net_core_table); return register_pernet_subsys(&sysctl_core_ops); } From patchwork Wed Oct 28 17:46:03 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paolo Abeni X-Patchwork-Id: 11864893 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 34B9814B2 for ; Thu, 29 Oct 2020 02:02:07 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 11EF220796 for ; Thu, 29 Oct 2020 02:02:07 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="GFAlwCKG" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727486AbgJ1VtO (ORCPT ); Wed, 28 Oct 2020 17:49:14 -0400 Received: from us-smtp-delivery-124.mimecast.com ([63.128.21.124]:43167 "EHLO us-smtp-delivery-124.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727461AbgJ1VtM (ORCPT ); Wed, 28 Oct 2020 17:49:12 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1603921750; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=vRfli5Z5DUhcuyRlIgm4cbCOglHgREI9yrOIn+85HoI=; b=GFAlwCKGoTzx9mf6V63YhHPBFIrZ3ARut8oZMHBjIy+5c3kkNJhQcTbvr0XxUep4aE63Js FbLzmeaQLAxhSKnHkOQH7fwL6Ps94G94qnDXTLD3qSo0jDNonOVC+Bf8mtxD7S4Dz+R93A WJtXZ2MgmJFj+G8rKi3LyYSYpWcXWS8= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-306-VDuhqbblOD6oapEgFdZfoQ-1; Wed, 28 Oct 2020 13:47:02 -0400 X-MC-Unique: VDuhqbblOD6oapEgFdZfoQ-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 00D1F18A0727; Wed, 28 Oct 2020 17:46:40 +0000 (UTC) Received: from gerbillo.redhat.com (ovpn-115-68.ams2.redhat.com [10.36.115.68]) by smtp.corp.redhat.com (Postfix) with ESMTP id EB78F6198D; Wed, 28 Oct 2020 17:46:37 +0000 (UTC) From: Paolo Abeni To: netdev@vger.kernel.org Cc: Jonathan Corbet , "David S. Miller" , Jakub Kicinski , Shuah Khan , linux-doc@vger.kernel.org, linux-kselftest@vger.kernel.org, Marcelo Tosatti Subject: [PATCH net-next 3/3] self-tests: introduce self-tests for RPS default mask Date: Wed, 28 Oct 2020 18:46:03 +0100 Message-Id: <27a3dd523e015ba27ba5666ad727ff5d935ce3a8.1603906564.git.pabeni@redhat.com> In-Reply-To: References: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 Precedence: bulk List-ID: X-Mailing-List: linux-kselftest@vger.kernel.org Ensure that RPS default mask changes take place on all newly created netns/devices and don't affect existing ones. Signed-off-by: Paolo Abeni --- tools/testing/selftests/net/Makefile | 1 + tools/testing/selftests/net/config | 3 + .../testing/selftests/net/rps_default_mask.sh | 57 +++++++++++++++++++ 3 files changed, 61 insertions(+) create mode 100644 tools/testing/selftests/net/rps_default_mask.sh diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile index ef352477cac6..2531ec3e5027 100644 --- a/tools/testing/selftests/net/Makefile +++ b/tools/testing/selftests/net/Makefile @@ -21,6 +21,7 @@ TEST_PROGS += rxtimestamp.sh TEST_PROGS += devlink_port_split.py TEST_PROGS += drop_monitor_tests.sh TEST_PROGS += vrf_route_leaking.sh +TEST_PROGS += rps_default_mask.sh TEST_PROGS_EXTENDED := in_netns.sh TEST_GEN_FILES = socket nettest TEST_GEN_FILES += psock_fanout psock_tpacket msg_zerocopy reuseport_addr_any diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config index 4d5df8e1eee7..5d467364f082 100644 --- a/tools/testing/selftests/net/config +++ b/tools/testing/selftests/net/config @@ -34,3 +34,6 @@ CONFIG_TRACEPOINTS=y CONFIG_NET_DROP_MONITOR=m CONFIG_NETDEVSIM=m CONFIG_NET_FOU=m +CONFIG_RPS=y +CONFIG_SYSFS=y +CONFIG_PROC_SYSCTL=y diff --git a/tools/testing/selftests/net/rps_default_mask.sh b/tools/testing/selftests/net/rps_default_mask.sh new file mode 100644 index 000000000000..c81c0ac7ddfe --- /dev/null +++ b/tools/testing/selftests/net/rps_default_mask.sh @@ -0,0 +1,57 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 + +readonly ksft_skip=4 +readonly cpus=$(nproc) +ret=0 + +[ $cpus -gt 2 ] || exit $ksft_skip + +readonly INITIAL_RPS_DEFAULT_MASK=$(cat /proc/sys/net/core/rps_default_mask) +readonly NETNS="ns-$(mktemp -u XXXXXX)" + +setup() { + ip netns add "${NETNS}" + ip -netns "${NETNS}" link set lo up +} + +cleanup() { + echo $INITIAL_RPS_DEFAULT_MASK > /proc/sys/net/core/rps_default_mask + ip netns del $NETNS +} + +chk_rps() { + local rps_mask expected_rps_mask=$3 + local dev_name=$2 + local msg=$1 + + rps_mask=$(ip netns exec $NETNS cat /sys/class/net/$dev_name/queues/rx-0/rps_cpus) + printf "%-60s" "$msg" + if [ $rps_mask -eq $expected_rps_mask ]; then + echo "[ ok ]" + else + echo "[fail] expected $expected_rps_mask found $rps_mask" + ret=1 + fi +} + +trap cleanup EXIT + +echo 0 > /proc/sys/net/core/rps_default_mask +setup +chk_rps "empty rps_default_mask" lo 0 +cleanup + +echo 1 > /proc/sys/net/core/rps_default_mask +setup +chk_rps "non zero rps_default_mask" lo 1 + +echo 3 > /proc/sys/net/core/rps_default_mask +chk_rps "changing rps_default_mask dont affect existing netns" lo 1 + +ip -n $NETNS link add type veth +ip -n $NETNS link set dev veth0 up +ip -n $NETNS link set dev veth1 up +chk_rps "changing rps_default_mask affect newly created devices" veth0 3 +chk_rps "changing rps_default_mask affect newly created devices[II]" veth1 3 +exit $ret