diff options
| author | Ingo Molnar <mingo@kernel.org> | 2016-03-08 11:09:53 +0100 |
|---|---|---|
| committer | Ingo Molnar <mingo@kernel.org> | 2016-03-08 11:09:53 +0100 |
| commit | 6436257b491cc0d456c39330dfc22126148d5ed7 (patch) | |
| tree | 4e3f4a1da9f88e429c95b7ea5661390fc611e836 | |
| parent | 69365272332a5cbef778c97b2337b4f79c509293 (diff) | |
time/timekeeping: Work around false positive GCC warning
Newer GCC versions trigger the following warning:
kernel/time/timekeeping.c: In function ‘get_device_system_crosststamp’:
kernel/time/timekeeping.c:987:5: warning: ‘clock_was_set_seq’ may be used uninitialized in this function [-Wmaybe-uninitialized]
if (discontinuity) {
^
kernel/time/timekeeping.c:1045:15: note: ‘clock_was_set_seq’ was declared here
unsigned int clock_was_set_seq;
^
GCC clearly is unable to recognize that the 'do_interp' boolean tracks
the initialization status of 'clock_was_set_seq'.
The GCC version used was:
gcc version 5.3.1 20151207 (Red Hat 5.3.1-2) (GCC)
Work it around by initializing clock_was_set_seq to 0. Compilers that
are able to recognize the code flow will eliminate the unnecessary
initialization.
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
| -rw-r--r-- | kernel/time/timekeeping.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 931b0b1a71e9..9c629bbed572 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -1042,7 +1042,7 @@ int get_device_system_crosststamp(int (*get_time_fn) struct system_counterval_t system_counterval; struct timekeeper *tk = &tk_core.timekeeper; cycle_t cycles, now, interval_start; - unsigned int clock_was_set_seq; + unsigned int clock_was_set_seq = 0; ktime_t base_real, base_raw; s64 nsec_real, nsec_raw; u8 cs_was_changed_seq; |