6.12-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stefan Wahren wahrenst@gmx.net
[ Upstream commit 3db89bc6d973e2bcaa852f6409c98c228f39a926 ]
In case vchiq_platform_conn_state_changed() is never called or fails before driver removal, ka_thread won't be a valid pointer to a task_struct. So do the necessary checks before calling kthread_stop to avoid a crash.
Fixes: 863a756aaf49 ("staging: vc04_services: vchiq_core: Stop kthreads on vchiq module unload") Signed-off-by: Stefan Wahren wahrenst@gmx.net Link: https://lore.kernel.org/r/20250309125014.37166-3-wahrenst@gmx.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c index ecf6e9635a10b..97787002080a1 100644 --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c @@ -1786,7 +1786,8 @@ static void vchiq_remove(struct platform_device *pdev) kthread_stop(mgmt->state.slot_handler_thread);
arm_state = vchiq_platform_get_arm_state(&mgmt->state); - kthread_stop(arm_state->ka_thread); + if (!IS_ERR_OR_NULL(arm_state->ka_thread)) + kthread_stop(arm_state->ka_thread); }
static struct platform_driver vchiq_driver = {