The releasing of ACPI_MTX_EVENTS in the middle of
acpi_remove_gpe_handler() in order to call
acpi_os_wait_events_complete() is a mistake, because it creates a
window for undesirable concurrent changes to be made while
acpi_os_wait_events_complete() is being executed. At the same
time it doesn't make any difference to call
acpi_os_wait_events_complete() before acquiring ACPI_MTX_EVENTS for
the first time, which allows us to avoid the race condition.
Signed-off-by: Rafael J. Wysocki <rjw@xxxxxxx>
---
drivers/acpi/acpica/evxface.c | 13 ++++---------
1 file changed, 4 insertions(+), 9 deletions(-)
Index: linux-2.6/drivers/acpi/acpica/evxface.c
===================================================================
--- linux-2.6.orig/drivers/acpi/acpica/evxface.c
+++ linux-2.6/drivers/acpi/acpica/evxface.c
@@ -842,6 +842,10 @@ acpi_remove_gpe_handler(acpi_handle gpe_
return_ACPI_STATUS(AE_BAD_PARAMETER);
}
+ /* Make sure all deferred tasks are completed */
+
+ acpi_os_wait_events_complete(NULL);
+
status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
if (ACPI_FAILURE(status)) {
return_ACPI_STATUS(status);
@@ -870,15 +874,6 @@ acpi_remove_gpe_handler(acpi_handle gpe_
goto unlock_and_exit;
}
- /* Make sure all deferred tasks are completed */
-
- (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
- acpi_os_wait_events_complete(NULL);
- status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
- if (ACPI_FAILURE(status)) {
- return_ACPI_STATUS(status);
- }
-
/* Remove the handler */
flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
