team: fix hang in team_mode_get()
Subsystems (3)
PATCH v3 1/1 team: fix hang in team_mode_get()
When team mode is changed or set, the team_mode_get() is called to check whether the mode module is inserted or not. If the mode module is not inserted, it calls the request_module().
In the request_module(), it creates a child process, which is the "modprobe" process and waits for the done of the child process. At this point, the following locks were used. down_read(&cb_lock()); by genl_rcv()
genl_lock(); by genl_rcv_msc()
rtnl_lock(); by team_nl_cmd_options_set()
mutex_lock(&team->lock); by team_nl_team_get()Concurrently, the team module could be removed by rmmod or "modprobe -r" The __exit function of team module is team_module_exit(), which calls team_nl_fini() and it tries to acquire following locks.
down_write(&cb_lock);
genl_lock();Because of the genl_lock() and cb_lock, this process can't be finished earlier than request_module() routine.
The problem secenario.
CPU0 CPU1team_mode_get
request_module()
modprobe -r team_mode_roundrobin
team <--(B)
modprobe team <--(A)
team_mode_roundrobin
By request_module(), the "modprobe team_mode_roundrobin" command will be executed. At this point, the modprobe process will decide that the team module should be inserted before team_mode_roundrobin. Because the team module is being removed.
By the module infrastructure, the same module insert/remove operations can't be executed concurrently.
So, (A) waits for (B) but (B) also waits for (A) because of locks. So that the hang occurs at this point.
Test commands:
while :
do
teamd -d &
killall teamd &
modprobe -rv team_mode_roundrobin &
done
The approach of this patch is to hold the reference count of the team module if the team module is compiled as a module. If the reference count of the team module is not zero while request_module() is being called, the team module will not be removed at that moment.
So that the above scenario could not occur.
---
v2 -> v3:
- Remove unrelated change
v1 -> v2:
- Remove unnecessary #if statement and 'put' variable
drivers/net/team/team.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/net/team/team.c +4 -0
@@ -468,6 +468,9 @@ static const struct team_mode *team_mode_get(const char *kind)struct team_mode_item *mitem;const struct team_mode *mode =NULL ; +if (!try_module_get (THIS_MODULE )) +return NULL ; +spin_lock (&mode_list_lock); mitem =__find_mode (kind);if (!mitem) {
@@ -483,6 +486,7 @@ static const struct team_mode *team_mode_get(const char *kind) }spin_unlock (&mode_list_lock); +module_put (THIS_MODULE );return mode; }
Signature · 1 line
2.17.1
Quoted 46 lines · Mon, Apr 20, 2020 at 05:01:33PM CEST, [email protected] wrote:
When team mode is changed or set, the
team_mode_get()is called to check whether the mode module is inserted or not. If the mode module is not inserted, it calls therequest_module().
In therequest_module(), it creates a child process, which is the "modprobe" process and waits for the done of the child process. At this point, the following locks were used. down_read(&cb_lock()); bygenl_rcv()
genl_lock(); bygenl_rcv_msc()rtnl_lock(); by team_nl_cmd_options_set() mutex_lock(&team->lock); by team_nl_team_get()Concurrently, the team module could be removed by rmmod or "modprobe -r" The
__exitfunction of team module isteam_module_exit(), which callsteam_nl_fini()and it tries to acquire following locks.down_write(&cb_lock); genl_lock();Because of the
genl_lock()andcb_lock, this process can't be finished earlier thanrequest_module()routine.The problem secenario.
CPU0 CPU1team_mode_get
request_module()
modprobe -rteam_mode_roundrobin
team <--(B)
modprobe team <--(A)
team_mode_roundrobinBy
request_module(), the "modprobeteam_mode_roundrobin" command will be executed. At this point, the modprobe process will decide that the team module should be inserted beforeteam_mode_roundrobin. Because the team module is being removed.By the module infrastructure, the same module insert/remove operations can't be executed concurrently.
So, (A) waits for (B) but (B) also waits for (A) because of locks. So that the hang occurs at this point.Test commands:
while :
do
teamd -d &
killall teamd &
modprobe -rvteam_mode_roundrobin&
done
Quoted 8 lines
The approach of this patch is to hold the reference count of the team module if the team module is compiled as a module. If the reference count of the team module is not zero while
request_module()is being called, the team module will not be removed at that moment.
So that the above scenario could not occur.Fixes 3d249d4ca7d0 ("net: introduce ethernet teaming device")Signed-off-by Taehee Yoo <[email protected]>
David Miller
maintainerFrom: Taehee Yoo <[email protected]>
Date: Mon, 20 Apr 2020 15:01:33 +0000
Quoted 46 lines
When team mode is changed or set, the
team_mode_get()is called to check whether the mode module is inserted or not. If the mode module is not inserted, it calls therequest_module().
In therequest_module(), it creates a child process, which is the "modprobe" process and waits for the done of the child process. At this point, the following locks were used. down_read(&cb_lock()); bygenl_rcv()genl_lock(); by genl_rcv_msc() rtnl_lock(); by team_nl_cmd_options_set() mutex_lock(&team->lock); by team_nl_team_get()Concurrently, the team module could be removed by rmmod or "modprobe -r" The
__exitfunction of team module isteam_module_exit(), which callsteam_nl_fini()and it tries to acquire following locks.down_write(&cb_lock); genl_lock();Because of the
genl_lock()andcb_lock, this process can't be finished earlier thanrequest_module()routine.The problem secenario.
CPU0 CPU1team_mode_get
request_module()
modprobe -rteam_mode_roundrobin
team <--(B)
modprobe team <--(A)
team_mode_roundrobinBy
request_module(), the "modprobeteam_mode_roundrobin" command will be executed. At this point, the modprobe process will decide that the team module should be inserted beforeteam_mode_roundrobin. Because the team module is being removed.By the module infrastructure, the same module insert/remove operations can't be executed concurrently.
So, (A) waits for (B) but (B) also waits for (A) because of locks. So that the hang occurs at this point.Test commands:
while :
do
teamd -d &
killall teamd &
modprobe -rvteam_mode_roundrobin&
done
Quoted 8 lines
The approach of this patch is to hold the reference count of the team module if the team module is compiled as a module. If the reference count of the team module is not zero while
request_module()is being called, the team module will not be removed at that moment.
So that the above scenario could not occur.Fixes 3d249d4ca7d0 ("net: introduce ethernet teaming device")Signed-off-by Taehee Yoo <[email protected]>
Applied and queued up for -stable, thanks.