From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-wm1-f49.google.com (mail-wm1-f49.google.com [209.85.128.49]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 439AF25392C for ; Mon, 10 Aug 2026 07:18:33 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.128.49 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786346315; cv=none; b=fpnryKaOJoiomUWwFmxv2dtfPyt4Szp2zSGDi/NcHdQcnPdNMdhrmQwysxsVq0a2eJdYVtRo0UnfdWg6XfnRcWTMr62AYOqgjKwqAAW5HPY70CPGJK91/nJckT0bfpeu8crGo/X6wXpoo25TCeVnMy9oPIk7gkPSkiqixqcw1Cg= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786346315; c=relaxed/simple; bh=dZmmtYnK5ZfoMK7fkd8gezFvAqLjWQZtfAN+1DdmHNk=; h=Message-ID:Date:MIME-Version:Subject:To:Cc:References:From: In-Reply-To:Content-Type; b=eIkD0av4s47Ql1ESxpN69/22CxkykWLoYTqyM5gPGOXcz35OWFBWKdwWGA9Hp53p3zXq9Y0dqPNgspemmrfSzqWNoDlsEoSGjq0oBqGTIht7Ju/R37pUEmDoKiJwc598BrsL386goMFbc70l+qhOfTzEvVhHYDemY2jKulSVyrI= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=blackwall.org; spf=none smtp.mailfrom=blackwall.org; dkim=pass (2048-bit key) header.d=blackwall.org header.i=@blackwall.org header.b=neRaPry5; arc=none smtp.client-ip=209.85.128.49 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=blackwall.org Authentication-Results: smtp.subspace.kernel.org; spf=none smtp.mailfrom=blackwall.org Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=blackwall.org header.i=@blackwall.org header.b="neRaPry5" Received: by mail-wm1-f49.google.com with SMTP id 5b1f17b1804b1-4954a9e8490so21743565e9.1 for ; Mon, 10 Aug 2026 00:18:33 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=blackwall.org; s=google; t=1786346311; x=1786951111; darn=vger.kernel.org; h=content-transfer-encoding:content-type:in-reply-to:from:references :cc:to:content-language:subject:user-agent:mime-version:date :message-id:from:to:cc:subject:date:message-id:reply-to:content-type; bh=OZw6DBqMzFtZ7sE+YyKqzNiJoc28YODNR4pO+dPyYsE=; b=neRaPry5NCnIYxH5/cdJeLdAlASlW6NG4RlUnG0HMzNzn+cH1QYRlvVP+Nu3Qqvfog wdxUL3kBAmr0HXKi0KUI3wIOd8+CI/sNxruZar8ompEJY5XSdVokITAAGVc/JuSwD5Oa m7ZjQ5kOtwdNw+irFimS5EdOn9HP6aDM9OPG4K7iifDIx6xx/d49V49mtkwwh8/xLLma 2mJbaZxZYnFLJQm8J7N1XtUpbNcEiqTskFT1zYloplWbgPovbJbGa7gMZ5CykjrL9x12 8f/Q40hTBS4E74IOrNbARybQsZTLS3+7uNZu6wrQG+L5GXO2K59VhAPmdOUYa722xp0U D47Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1786346311; x=1786951111; h=content-transfer-encoding:content-type:in-reply-to:from:references :cc:to:content-language:subject:user-agent:mime-version:date :message-id:x-gm-gg:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to:content-type; bh=OZw6DBqMzFtZ7sE+YyKqzNiJoc28YODNR4pO+dPyYsE=; b=a8TeFFK7ufhqzYGXponpS6YqNbCCmMhRVNwlDIl73lfrvWS5nDiZfVBAmjTH5Of5ve pJHPckfCshL91Msp0ixNyUg5hrzDyTQ137y9mAYCdohMnXZVRxOckuNzCZiI0thOuolE GaMtCB5/wNKKfIs8HLkYRjpuG6llNLro+LB6T8N5EGatkrY+I5abiflTPfNFlqsaSAaM SXv+dScos1+OZtJV2srnCX4Q2QCYfjEzaohQ6Y0p9X9GHgyVRuf/e6eY2hxcFvBZFpmT qL7ittnn6EDdct+FBf3kr91fyqYA/EYuBQPy5Ueva1jaGn9JkMIQkplfYE8l4xbA+lFK ANCA== X-Forwarded-Encrypted: i=1; AHgh+RpnY+eWRsUEwjWykg1ajItiv69zrHL9jBFe0YT8jiQjS3mY9v/7r92kb/1cEQsbSK26vPy861b6hZ/w9h0=@vger.kernel.org X-Gm-Message-State: AOJu0YyMhhkquWXSA/V89mHVGD/B3MW/6RIRlnj+xbCcdpdxKp+WdwQF 7EJ05F3fHtIL4rUzTZKm4zDcCF2RT7B1XL1O7ZZW0aVXKoAjHr06mg5FEIM/3sliflc= X-Gm-Gg: AR+sD100UQAMl3ZGXFDdvnX3VO15l9pRSpiQ2G77HMUVSEb5O1WW0O3uNHd09RFzA+L mg/CPXmi6eo10mkpQTa426zJZs6xsD8YvRTUgJ0esRJTvAFuJbLTdoy7NYK4R0D/Ex4qOsDrFoi KCq37TfXYCr+KeRqyxw72rCg8if6oc+yZGwMHhOGHLxCFsP3huy4KSekpeXjIxp9pmQcRBcUH+w oOU1xKWBTjObEX6Zc+ZlGRFJDxfgmARMQsTOVHxpzMqSefG90Aj4If10N6m5LFv+SPMf1UHMFx9 iCiCkKkA35I2z1GhuudClalVVz4d7djcXdcppzkYIiMthQZaJG50Ua1kn7tvk71fF3UNe1kz5/o ha6d1a0QVBWorFCMSopyh1wduDqK8+1qMU9YSIcbrpkUsG14BcKLGSG0qgFRSLi7dyLxYU4dal9 ijlETtPla69ldpckcsw8KalruXZNdLcdJkSwzpMQX9A2z0wArwC+rcDUyfAc05b5rjlXqsugIWT 9mywxyC3OcKMuGcTE4= X-Received: by 2002:a05:600c:4e8e:b0:497:ff5a:38b9 with SMTP id 5b1f17b1804b1-4996247e640mr186732625e9.9.1786346311317; Mon, 10 Aug 2026 00:18:31 -0700 (PDT) Received: from [192.168.0.161] (78-154-15-182.ip.btc-net.bg. [78.154.15.182]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-4995428f294sm383337555e9.15.2026.08.10.00.18.29 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Mon, 10 Aug 2026 00:18:30 -0700 (PDT) Message-ID: <9cff59e3-247f-463f-8d43-5138eefc18e3@blackwall.org> Date: Mon, 10 Aug 2026 10:18:29 +0300 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH] net: convert netdev_nl_sock bindings list to xarray Content-Language: en-US, bg To: Shivaji Kant , netdev@vger.kernel.org, davem@davemloft.net, edumazet@google.com, kuba@kernel.org, pabeni@redhat.com Cc: praan@google.com, Mina Almasry , Simon Horman , Bobby Eshleman , Stanislav Fomichev , Dragos Tatulea , Kees Cook , David Carlier , Yue Haibing , Daniel Borkmann , David Wei , Maoyi Xie , linux-kernel@vger.kernel.org References: <20260810065759.3735934-1-shivajikant@google.com> From: Nikolay Aleksandrov In-Reply-To: <20260810065759.3735934-1-shivajikant@google.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit On 10/08/2026 09:57, Shivaji Kant wrote: > netdev_nl_sock previously used a struct list_head bindings to keep > track of active netdev netlink bindings, protected by a per-socket > struct mutex lock (priv->lock). > > Since list modifications and iterations are not concurrency-safe, > priv->lock was introduced to serialize operations on priv->bindings. > However, xarray manages its own internal locking (xa_lock) for store, > erase, and iteration operations. > > Convert bindings in struct netdev_nl_sock from struct list_head to > struct xarray and remove priv->lock. This simplifies the code by > removing explicit mutex locking in netdev_nl_bind_rx_doit(), > netdev_nl_bind_tx_doit(), and socket initialization/teardown functions. > > Reviewed-by: Mina Almasry > Signed-off-by: Shivaji Kant > --- > include/net/netdev_netlink.h | 5 ++--- > net/core/devmem.c | 13 ++++++++++--- > net/core/devmem.h | 4 ++-- > net/core/netdev-genl.c | 25 ++++++------------------- > 4 files changed, 20 insertions(+), 27 deletions(-) > This patch should target net-next. > diff --git a/include/net/netdev_netlink.h b/include/net/netdev_netlink.h > index 075962dbe743..45a4a482d42e 100644 > --- a/include/net/netdev_netlink.h > +++ b/include/net/netdev_netlink.h > @@ -2,11 +2,10 @@ > #ifndef __NET_NETDEV_NETLINK_H > #define __NET_NETDEV_NETLINK_H > > -#include > +#include > > struct netdev_nl_sock { > - struct mutex lock; > - struct list_head bindings; > + struct xarray bindings; > }; > > #endif /* __NET_NETDEV_NETLINK_H */ > diff --git a/net/core/devmem.c b/net/core/devmem.c > index 957d6b96216b..52fda18b80ef 100644 > --- a/net/core/devmem.c > +++ b/net/core/devmem.c > @@ -134,8 +134,8 @@ void net_devmem_unbind_dmabuf(struct net_devmem_dmabuf_binding *binding) > */ > synchronize_net(); > > - if (binding->list.next) > - list_del(&binding->list); > + if (binding->sock_priv) > + xa_erase(&binding->sock_priv->bindings, binding->id); > > xa_for_each(&binding->bound_rxqs, xa_idx, rxq) { > const struct pp_memory_provider_params mp_params = { > @@ -193,6 +193,7 @@ net_devmem_bind_dmabuf(struct net_device *dev, void *vdev, > struct dma_buf *dmabuf; > unsigned int sg_idx, i; > unsigned long virtual; > + void *res; > int err; > > if (!dma_dev) { > @@ -325,7 +326,13 @@ net_devmem_bind_dmabuf(struct net_device *dev, void *vdev, > if (err < 0) > goto err_free_chunks; > > - list_add(&binding->list, &priv->bindings); > + binding->sock_priv = priv; > + res = xa_store(&priv->bindings, binding->id, binding, GFP_KERNEL); > + if (xa_is_err(res)) { > + err = xa_err(res); > + xa_erase(&net_devmem_dmabuf_bindings, binding->id); > + goto err_free_chunks; This error cleans up after the previous xa_alloc_cyclic call itself instead of following the style of the rest of the error cleanups. I'd move the xa_erase() into its own err_ label and goto there. > + } > > return binding; > > diff --git a/net/core/devmem.h b/net/core/devmem.h > index 3852a56036cb..bf77e5a9a8fe 100644 > --- a/net/core/devmem.h > +++ b/net/core/devmem.h > @@ -49,10 +49,10 @@ struct net_devmem_dmabuf_binding { > */ > struct percpu_ref ref; > > - /* The list of bindings currently active. Used for netlink to notify us > + /* The socket priv this binding belongs to. Used for netlink to notify us > * of the user dropping the bind. > */ > - struct list_head list; > + struct netdev_nl_sock *sock_priv; > > /* rxq's this binding is active on. */ > struct xarray bound_rxqs; > diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c > index c15d8d4ca1f8..d3f3a91a6d93 100644 > --- a/net/core/netdev-genl.c > +++ b/net/core/netdev-genl.c > @@ -1044,13 +1044,11 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info) > goto err_genlmsg_free; > } > > - mutex_lock(&priv->lock); > - > err = 0; > netdev = netdev_get_by_index_lock(genl_info_net(info), ifindex); > if (!netdev) { > err = -ENODEV; > - goto err_unlock_sock; > + goto err_genlmsg_free; > } > if (!netif_device_present(netdev)) > err = -ENODEV; > @@ -1102,8 +1100,6 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info) > > netdev_unlock(netdev); > > - mutex_unlock(&priv->lock); > - > return err < 0 ? err : 0; > > err_unbind: > @@ -1112,8 +1108,6 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info) > bitmap_free(rxq_bitmap); > err_unlock: > netdev_unlock(netdev); > -err_unlock_sock: > - mutex_unlock(&priv->lock); > err_genlmsg_free: > nlmsg_free(rsp); > return err; > @@ -1185,12 +1179,10 @@ int netdev_nl_bind_tx_doit(struct sk_buff *skb, struct genl_info *info) > goto err_genlmsg_free; > } > > - mutex_lock(&priv->lock); > - > netdev = netdev_get_by_index_lock(genl_info_net(info), ifindex); > if (!netdev) { > err = -ENODEV; > - goto err_unlock_sock; > + goto err_genlmsg_free; > } > > if (!netif_device_present(netdev)) { > @@ -1233,7 +1225,6 @@ int netdev_nl_bind_tx_doit(struct sk_buff *skb, struct genl_info *info) > if (bind_dev != netdev) > netdev_unlock(bind_dev); > netdev_unlock(netdev); > - mutex_unlock(&priv->lock); > > return genlmsg_reply(rsp, info); > > @@ -1242,8 +1233,6 @@ int netdev_nl_bind_tx_doit(struct sk_buff *skb, struct genl_info *info) > netdev_unlock(bind_dev); > err_unlock_netdev: > netdev_unlock(netdev); > -err_unlock_sock: > - mutex_unlock(&priv->lock); > err_genlmsg_free: > nlmsg_free(rsp); > return err; > @@ -1418,19 +1407,17 @@ int netdev_nl_queue_create_doit(struct sk_buff *skb, struct genl_info *info) > > void netdev_nl_sock_priv_init(struct netdev_nl_sock *priv) > { > - INIT_LIST_HEAD(&priv->bindings); > - mutex_init(&priv->lock); > + xa_init(&priv->bindings); > } > > void netdev_nl_sock_priv_destroy(struct netdev_nl_sock *priv) > { > struct net_devmem_dmabuf_binding *binding; > - struct net_devmem_dmabuf_binding *temp; > netdevice_tracker dev_tracker; > struct net_device *dev; > + unsigned long xa_idx; > > - mutex_lock(&priv->lock); > - list_for_each_entry_safe(binding, temp, &priv->bindings, list) { > + xa_for_each(&priv->bindings, xa_idx, binding) { > mutex_lock(&binding->lock); > dev = binding->dev; > if (!dev) { > @@ -1446,7 +1433,7 @@ void netdev_nl_sock_priv_destroy(struct netdev_nl_sock *priv) > netdev_unlock(dev); > netdev_put(dev, &dev_tracker); > } > - mutex_unlock(&priv->lock); > + xa_destroy(&priv->bindings); > } > > static int netdev_genl_netdevice_event(struct notifier_block *nb,