[PATCH 4.4 066/115] target: fix race during implicit transition work flushes

From: Greg Kroah-Hartman
Date: Mon Dec 18 2017 - 13:02:52 EST

4.4-stable review patch. If anyone has any objections, please let me know.


From: Mike Christie <mchristi@xxxxxxxxxx>

[ Upstream commit 760bf578edf8122f2503a3a6a3f4b0de3b6ce0bb ]

This fixes the following races:

1. core_alua_do_transition_tg_pt could have read
tg_pt_gp_alua_access_state and gone into this if chunk:

if (!explicit &&
atomic_read(&tg_pt_gp->tg_pt_gp_alua_access_state) ==

and then core_alua_do_transition_tg_pt_work could update the
state. core_alua_do_transition_tg_pt would then only set
tg_pt_gp_alua_pending_state and the tg_pt_gp_alua_access_state would
not get updated with the second calls state.

2. core_alua_do_transition_tg_pt could be setting
tg_pt_gp_transition_complete while the tg_pt_gp_transition_work
is already completing. core_alua_do_transition_tg_pt then waits on the
completion that will never be called.

To handle these issues, we just call flush_work which will return when
core_alua_do_transition_tg_pt_work has completed so there is no need
to do the complete/wait. And, if core_alua_do_transition_tg_pt_work
was running, instead of trying to sneak in the state change, we just
schedule up another core_alua_do_transition_tg_pt_work call.

Note that this does not handle a possible race where there are multiple
threads call core_alua_do_transition_tg_pt at the same time. I think
we need a mutex in target_tg_pt_gp_alua_access_state_store.

Signed-off-by: Mike Christie <mchristi@xxxxxxxxxx>
Signed-off-by: Nicholas Bellinger <nab@xxxxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <alexander.levin@xxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
drivers/target/target_core_alua.c | 10 +---------
1 file changed, 1 insertion(+), 9 deletions(-)

--- a/drivers/target/target_core_alua.c
+++ b/drivers/target/target_core_alua.c
@@ -1073,16 +1073,8 @@ static int core_alua_do_transition_tg_pt
* Flush any pending transitions
- if (!explicit && atomic_read(&tg_pt_gp->tg_pt_gp_alua_access_state) ==
- /* Just in case */
- tg_pt_gp->tg_pt_gp_alua_pending_state = new_state;
- tg_pt_gp->tg_pt_gp_transition_complete = &wait;
+ if (!explicit)
- wait_for_completion(&wait);
- tg_pt_gp->tg_pt_gp_transition_complete = NULL;
- return 0;
- }

* Save the old primary ALUA access state, and set the current state