From: Corey Minyard <minyard@acm.org>
To: Amol Grover <frextrite@gmail.com>
Cc: Arnd Bergmann <arnd@arndb.de>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
openipmi-developer@lists.sourceforge.net,
linux-kernel@vger.kernel.org,
linux-kernel-mentees@lists.linuxfoundation.org,
John Garry <john.garry@huawei.com>,
Joel Fernandes <joel@joelfernandes.org>,
Madhuparna Bhowmik <madhuparnabhowmik10@gmail.com>,
"Paul E . Mckenney" <paulmck@kernel.org>
Subject: Re: [PATCH] ipmi: Fix RCU list lockdep debugging warnings
Date: Fri, 28 Feb 2020 11:44:44 -0600 [thread overview]
Message-ID: <20200228174444.GE3840@minyard.net> (raw)
In-Reply-To: <20200228081731.18149-1-frextrite@gmail.com>
On Fri, Feb 28, 2020 at 01:47:33PM +0530, Amol Grover wrote:
> It is completely safe to traverse ipmi_interfaces and
> intf->users under SRCU read lock using list_for_each_entry_rcu().
> Tell lockdep about it as well else it will show false-positive
> warnings as the one below.
Thanks, this is queued for 5.7, and I hadded John Garry's test by.
-corey
>
> Fixes the following false-positive warning and others that may follow.
>
> [ 29.772408] =============================
> [ 29.776863] WARNING: suspicious RCU usage
> [ 29.780915] 5.6.0-rc3-00001-g907305ae6618-dirty #1755 Not tainted
> [ 29.787046] -----------------------------
> [ 29.791100] drivers/char/ipmi/ipmi_msghandler.c:744 RCU-list traversed in
> non-reader section!!
>
> Reported-by: John Garry <john.garry@huawei.com>
> Suggested-by: Paul E. McKenney <paulmck@kernel.org>
> Signed-off-by: Amol Grover <frextrite@gmail.com>
> ---
> drivers/char/ipmi/ipmi_msghandler.c | 18 ++++++++++++------
> 1 file changed, 12 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c
> index cad9563f8f48..d202022c69de 100644
> --- a/drivers/char/ipmi/ipmi_msghandler.c
> +++ b/drivers/char/ipmi/ipmi_msghandler.c
> @@ -741,7 +741,8 @@ int ipmi_smi_watcher_register(struct ipmi_smi_watcher *watcher)
> list_add(&watcher->link, &smi_watchers);
>
> index = srcu_read_lock(&ipmi_interfaces_srcu);
> - list_for_each_entry_rcu(intf, &ipmi_interfaces, link) {
> + list_for_each_entry_rcu(intf, &ipmi_interfaces, link,
> + srcu_read_lock_held(&ipmi_interfaces_srcu)) {
> int intf_num = READ_ONCE(intf->intf_num);
>
> if (intf_num == -1)
> @@ -1188,7 +1189,8 @@ int ipmi_create_user(unsigned int if_num,
> return -ENOMEM;
>
> index = srcu_read_lock(&ipmi_interfaces_srcu);
> - list_for_each_entry_rcu(intf, &ipmi_interfaces, link) {
> + list_for_each_entry_rcu(intf, &ipmi_interfaces, link,
> + srcu_read_lock_held(&ipmi_interfaces_srcu)) {
> if (intf->intf_num == if_num)
> goto found;
> }
> @@ -1241,7 +1243,8 @@ int ipmi_get_smi_info(int if_num, struct ipmi_smi_info *data)
> struct ipmi_smi *intf;
>
> index = srcu_read_lock(&ipmi_interfaces_srcu);
> - list_for_each_entry_rcu(intf, &ipmi_interfaces, link) {
> + list_for_each_entry_rcu(intf, &ipmi_interfaces, link,
> + srcu_read_lock_held(&ipmi_interfaces_srcu)) {
> if (intf->intf_num == if_num)
> goto found;
> }
> @@ -4098,7 +4101,8 @@ static int handle_read_event_rsp(struct ipmi_smi *intf,
> * getting events.
> */
> index = srcu_read_lock(&intf->users_srcu);
> - list_for_each_entry_rcu(user, &intf->users, link) {
> + list_for_each_entry_rcu(user, &intf->users, link,
> + srcu_read_lock_held(&intf->users_srcu)) {
> if (!user->gets_events)
> continue;
>
> @@ -4453,7 +4457,8 @@ static void handle_new_recv_msgs(struct ipmi_smi *intf)
> int index;
>
> index = srcu_read_lock(&intf->users_srcu);
> - list_for_each_entry_rcu(user, &intf->users, link) {
> + list_for_each_entry_rcu(user, &intf->users, link,
> + srcu_read_lock_held(&intf->users_srcu)) {
> if (user->handler->ipmi_watchdog_pretimeout)
> user->handler->ipmi_watchdog_pretimeout(
> user->handler_data);
> @@ -4746,7 +4751,8 @@ static void ipmi_timeout(struct timer_list *unused)
> return;
>
> index = srcu_read_lock(&ipmi_interfaces_srcu);
> - list_for_each_entry_rcu(intf, &ipmi_interfaces, link) {
> + list_for_each_entry_rcu(intf, &ipmi_interfaces, link,
> + srcu_read_lock_held(&ipmi_interfaces_srcu)) {
> if (atomic_read(&intf->event_waiters)) {
> intf->ticks_to_req_ev--;
> if (intf->ticks_to_req_ev == 0) {
> --
> 2.25.0
>
prev parent reply other threads:[~2020-02-28 17:44 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-28 8:17 [PATCH] ipmi: Fix RCU list lockdep debugging warnings Amol Grover
2020-02-28 11:49 ` John Garry
2020-02-28 17:44 ` Corey Minyard [this message]
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=20200228174444.GE3840@minyard.net \
--to=minyard@acm.org \
--cc=arnd@arndb.de \
--cc=frextrite@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=joel@joelfernandes.org \
--cc=john.garry@huawei.com \
--cc=linux-kernel-mentees@lists.linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=madhuparnabhowmik10@gmail.com \
--cc=openipmi-developer@lists.sourceforge.net \
--cc=paulmck@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: link
Be 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