From: Leon Romanovsky <leon@kernel.org> To: "David S . Miller" <davem@davemloft.net>, Jakub Kicinski <kuba@kernel.org> Cc: Leon Romanovsky <leonro@nvidia.com>, linux-kernel@vger.kernel.org, netdev@vger.kernel.org Subject: [PATCH net-next] netdevsim: Forbid devlink reload when adding or deleting ports Date: Thu, 5 Aug 2021 14:02:45 +0300 [thread overview] Message-ID: <53cd1a28dd34ced9fb4c39885c6e13523e97d62c.1628161323.git.leonro@nvidia.com> (raw) From: Leon Romanovsky <leonro@nvidia.com> In order to remove complexity in devlink core related to devlink_reload_enable/disable, let's rewrite new_port/del_port logic to rely on internal to netdevsim lcok. We should protect only reload_down flow because it destroys nsim_dev, which is needed for nsim_dev_port_add/nsim_dev_port_del to hold port_list_lock. Signed-off-by: Leon Romanovsky <leonro@nvidia.com> --- drivers/net/netdevsim/bus.c | 16 ++++------------ drivers/net/netdevsim/dev.c | 7 +++++++ 2 files changed, 11 insertions(+), 12 deletions(-) diff --git a/drivers/net/netdevsim/bus.c b/drivers/net/netdevsim/bus.c index ff01e5bdc72e..a29ec264119d 100644 --- a/drivers/net/netdevsim/bus.c +++ b/drivers/net/netdevsim/bus.c @@ -183,8 +183,6 @@ new_port_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct nsim_bus_dev *nsim_bus_dev = to_nsim_bus_dev(dev); - struct nsim_dev *nsim_dev = dev_get_drvdata(dev); - struct devlink *devlink; unsigned int port_index; int ret; @@ -195,12 +193,10 @@ new_port_store(struct device *dev, struct device_attribute *attr, if (ret) return ret; - devlink = priv_to_devlink(nsim_dev); + if (!mutex_trylock(&nsim_bus_dev->nsim_bus_reload_lock)) + return -EBUSY; - mutex_lock(&nsim_bus_dev->nsim_bus_reload_lock); - devlink_reload_disable(devlink); ret = nsim_dev_port_add(nsim_bus_dev, NSIM_DEV_PORT_TYPE_PF, port_index); - devlink_reload_enable(devlink); mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return ret ? ret : count; } @@ -212,8 +208,6 @@ del_port_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct nsim_bus_dev *nsim_bus_dev = to_nsim_bus_dev(dev); - struct nsim_dev *nsim_dev = dev_get_drvdata(dev); - struct devlink *devlink; unsigned int port_index; int ret; @@ -224,12 +218,10 @@ del_port_store(struct device *dev, struct device_attribute *attr, if (ret) return ret; - devlink = priv_to_devlink(nsim_dev); + if (!mutex_trylock(&nsim_bus_dev->nsim_bus_reload_lock)) + return -EBUSY; - mutex_lock(&nsim_bus_dev->nsim_bus_reload_lock); - devlink_reload_disable(devlink); ret = nsim_dev_port_del(nsim_bus_dev, NSIM_DEV_PORT_TYPE_PF, port_index); - devlink_reload_enable(devlink); mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return ret ? ret : count; } diff --git a/drivers/net/netdevsim/dev.c b/drivers/net/netdevsim/dev.c index d538a39d4225..ff5714209b86 100644 --- a/drivers/net/netdevsim/dev.c +++ b/drivers/net/netdevsim/dev.c @@ -864,16 +864,23 @@ static int nsim_dev_reload_down(struct devlink *devlink, bool netns_change, struct netlink_ext_ack *extack) { struct nsim_dev *nsim_dev = devlink_priv(devlink); + struct nsim_bus_dev *nsim_bus_dev; + + nsim_bus_dev = nsim_dev->nsim_bus_dev; + if (!mutex_trylock(&nsim_bus_dev->nsim_bus_reload_lock)) + return -EOPNOTSUPP; if (nsim_dev->dont_allow_reload) { /* For testing purposes, user set debugfs dont_allow_reload * value to true. So forbid it. */ NL_SET_ERR_MSG_MOD(extack, "User forbid the reload for testing purposes"); + mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return -EOPNOTSUPP; } nsim_dev_reload_destroy(nsim_dev); + mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return 0; } -- 2.31.1
WARNING: multiple messages have this Message-ID (diff)
From: Leon Romanovsky <leon@kernel.org> To: "David S . Miller" <davem@davemloft.net>, Jakub Kicinski <kuba@kernel.org> Cc: Leon Romanovsky <leonro@nvidia.com>, linux-kernel@vger.kernel.org, netdev@vger.kernel.org Subject: [PATCH net-next v1] netdevsim: Forbid devlink reload when adding or deleting ports Date: Thu, 5 Aug 2021 14:05:41 +0300 [thread overview] Message-ID: <53cd1a28dd34ced9fb4c39885c6e13523e97d62c.1628161323.git.leonro@nvidia.com> (raw) Message-ID: <20210805110541.5vJK4VuKOozm7NbU1Fan7Y2FxjUiUTsdXYyGeCMAllA@z> (raw) From: Leon Romanovsky <leonro@nvidia.com> In order to remove complexity in devlink core related to devlink_reload_enable/disable, let's rewrite new_port/del_port logic to rely on internal to netdevsim lock. We should protect only reload_down flow because it destroys nsim_dev, which is needed for nsim_dev_port_add/nsim_dev_port_del to hold port_list_lock. Signed-off-by: Leon Romanovsky <leonro@nvidia.com> --- Changelog: v1: * fixed typo in the commit message v0: https://lore.kernel.org/netdev/53cd1a28dd34ced9fb4c39885c6e13523e97d62c.1628161323.git.leonro@nvidia.com/T/#u --- drivers/net/netdevsim/bus.c | 16 ++++------------ drivers/net/netdevsim/dev.c | 7 +++++++ 2 files changed, 11 insertions(+), 12 deletions(-) diff --git a/drivers/net/netdevsim/bus.c b/drivers/net/netdevsim/bus.c index ff01e5bdc72e..a29ec264119d 100644 --- a/drivers/net/netdevsim/bus.c +++ b/drivers/net/netdevsim/bus.c @@ -183,8 +183,6 @@ new_port_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct nsim_bus_dev *nsim_bus_dev = to_nsim_bus_dev(dev); - struct nsim_dev *nsim_dev = dev_get_drvdata(dev); - struct devlink *devlink; unsigned int port_index; int ret; @@ -195,12 +193,10 @@ new_port_store(struct device *dev, struct device_attribute *attr, if (ret) return ret; - devlink = priv_to_devlink(nsim_dev); + if (!mutex_trylock(&nsim_bus_dev->nsim_bus_reload_lock)) + return -EBUSY; - mutex_lock(&nsim_bus_dev->nsim_bus_reload_lock); - devlink_reload_disable(devlink); ret = nsim_dev_port_add(nsim_bus_dev, NSIM_DEV_PORT_TYPE_PF, port_index); - devlink_reload_enable(devlink); mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return ret ? ret : count; } @@ -212,8 +208,6 @@ del_port_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct nsim_bus_dev *nsim_bus_dev = to_nsim_bus_dev(dev); - struct nsim_dev *nsim_dev = dev_get_drvdata(dev); - struct devlink *devlink; unsigned int port_index; int ret; @@ -224,12 +218,10 @@ del_port_store(struct device *dev, struct device_attribute *attr, if (ret) return ret; - devlink = priv_to_devlink(nsim_dev); + if (!mutex_trylock(&nsim_bus_dev->nsim_bus_reload_lock)) + return -EBUSY; - mutex_lock(&nsim_bus_dev->nsim_bus_reload_lock); - devlink_reload_disable(devlink); ret = nsim_dev_port_del(nsim_bus_dev, NSIM_DEV_PORT_TYPE_PF, port_index); - devlink_reload_enable(devlink); mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return ret ? ret : count; } diff --git a/drivers/net/netdevsim/dev.c b/drivers/net/netdevsim/dev.c index d538a39d4225..ff5714209b86 100644 --- a/drivers/net/netdevsim/dev.c +++ b/drivers/net/netdevsim/dev.c @@ -864,16 +864,23 @@ static int nsim_dev_reload_down(struct devlink *devlink, bool netns_change, struct netlink_ext_ack *extack) { struct nsim_dev *nsim_dev = devlink_priv(devlink); + struct nsim_bus_dev *nsim_bus_dev; + + nsim_bus_dev = nsim_dev->nsim_bus_dev; + if (!mutex_trylock(&nsim_bus_dev->nsim_bus_reload_lock)) + return -EOPNOTSUPP; if (nsim_dev->dont_allow_reload) { /* For testing purposes, user set debugfs dont_allow_reload * value to true. So forbid it. */ NL_SET_ERR_MSG_MOD(extack, "User forbid the reload for testing purposes"); + mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return -EOPNOTSUPP; } nsim_dev_reload_destroy(nsim_dev); + mutex_unlock(&nsim_bus_dev->nsim_bus_reload_lock); return 0; } -- 2.31.1
next reply other threads:[~2021-08-05 11:02 UTC|newest] Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-08-05 11:02 Leon Romanovsky [this message] 2021-08-05 11:05 ` [PATCH net-next v1] netdevsim: Forbid devlink reload when adding or deleting ports Leon Romanovsky 2021-08-05 12:40 ` [PATCH net-next] " patchwork-bot+netdevbpf 2021-08-05 13:15 ` [PATCH net-next v1] " Jakub Kicinski 2021-08-05 13:51 ` Leon Romanovsky 2021-08-05 14:23 ` Jakub Kicinski 2021-08-05 14:33 ` Leon Romanovsky 2021-08-05 15:27 ` Jakub Kicinski 2021-08-05 17:35 ` Leon Romanovsky 2021-08-05 18:02 ` Leon Romanovsky 2021-08-05 19:12 ` Jakub Kicinski 2021-08-06 11:19 ` Leon Romanovsky
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=53cd1a28dd34ced9fb4c39885c6e13523e97d62c.1628161323.git.leonro@nvidia.com \ --to=leon@kernel.org \ --cc=davem@davemloft.net \ --cc=kuba@kernel.org \ --cc=leonro@nvidia.com \ --cc=linux-kernel@vger.kernel.org \ --cc=netdev@vger.kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).