PM / QoS: avoid calling cancel_delayed_work_sync() during early boot

of_clk_init() ends up calling into pm_qos_update_request() very early
during boot where irq is expected to stay disabled.
pm_qos_update_request() uses cancel_delayed_work_sync() which
correctly assumes that irq is enabled on invocation and
unconditionally disables and re-enables it.

Gate cancel_delayed_work_sync() invocation with kevented_up() to avoid
enabling irq unexpectedly during early boot.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-tested-by: Qiao Zhou <qiaozhou@asrmicro.com>
Link: http://lkml.kernel.org/r/d2501c4c-8e7b-bea3-1b01-000b36b5dfe9@asrmicro.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

authored by Tejun Heo and committed by Rafael J. Wysocki c86d06ba c6935931

+10 -1
+10 -1
kernel/power/qos.c
··· 482 return; 483 } 484 485 - cancel_delayed_work_sync(&req->work); 486 __pm_qos_update_request(req, new_value); 487 } 488 EXPORT_SYMBOL_GPL(pm_qos_update_request);
··· 482 return; 483 } 484 485 + /* 486 + * This function may be called very early during boot, for example, 487 + * from of_clk_init(), where irq needs to stay disabled. 488 + * cancel_delayed_work_sync() assumes that irq is enabled on 489 + * invocation and re-enables it on return. Avoid calling it until 490 + * workqueue is initialized. 491 + */ 492 + if (keventd_up()) 493 + cancel_delayed_work_sync(&req->work); 494 + 495 __pm_qos_update_request(req, new_value); 496 } 497 EXPORT_SYMBOL_GPL(pm_qos_update_request);