Re: [PATCH TEST] timekeeping: Use data_race() and READ_ONCE() in ktime_get_real_seconds()⁠

Re: [PATCH TEST] timekeeping: Use data_race() and READ_ONCE() in ktime_get_real_seconds()⁠

[syzbot]

> #syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git

This crash does not have a reproducer. I cannot test it.

> master
>
> diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
> index 4fd3c70c1..b2e5a87da 100644
> --- a/kernel/time/timekeeping.c
> +++ b/kernel/time/timekeeping.c
> @@ -1153,7 +1153,7 @@ time64_t ktime_get_real_seconds(void)
> time64_t ktime_get_real_seconds(void)
> {
> #if BITS_PER_LONG == 64
> - return tk_core.timekeeper.xtime_sec;
> + return data_race(READ_ONCE(tk_core.timekeeper.xtime_sec));
> #else
> unsigned int seq;
> time64_t seconds;