Messages in this thread |  | | Date | Thu, 17 Jan 2013 13:58:55 +0100 | From | Alexander Holler <> | Subject | Re: [PATCH] rtc: hid-sensor-time: Add missing spin_lock_init |
| |
Am 17.01.2013 02:58, schrieb Thiago Farina: > On Wed, Jan 16, 2013 at 11:49 PM, Axel Lin <axel.lin@ingics.com> wrote: >> Signed-off-by: Axel Lin <axel.lin@ingics.com> >> --- >> drivers/rtc/rtc-hid-sensor-time.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/drivers/rtc/rtc-hid-sensor-time.c b/drivers/rtc/rtc-hid-sensor-time.c >> index 0438c9e..31c5728 100644 >> --- a/drivers/rtc/rtc-hid-sensor-time.c >> +++ b/drivers/rtc/rtc-hid-sensor-time.c >> @@ -225,6 +225,7 @@ static int hid_time_probe(struct platform_device *pdev) >> >> platform_set_drvdata(pdev, time_state); >> >> + spin_lock_init(&time_state->lock_last_time); > Can you explain in the commit message why it is missing? > > For people not familiar with this code that isn't obvious. >
Every spinlock must be initialized once (to SPIN_LOCK_UNLOCKED or by using spin_lock_init()). On most architectures SPIN_LOCK_UNLOCKED is just zero, therefor the kzalloc of the time_state does it (which likely is the case why I haven't spotted any error without the initialization), but that doesn't isn't true for all architectures.
Regards,
Alexander
|  |