Linux commit 27727df240c7 ("Avoid taking lock in NMI path with CONFIG_DEBUG_TIMEKEEPING"), changed the logic to open-code the timekeeping_get_ns() function, but forgot to include the unit conversion from cycles to nanoseconds, breaking the function's output, which impacts LTTng.
We expected Linux commit 58bfea9532 "timekeeping: Fix __ktime_get_fast_ns() regression" to make its way into stable kernels promptly, but it appears new stable kernel releases were done before the fix was cherry-picked from the master branch. We therefore need to bump the version ranges for the work-around in lttng-modules. Signed-off-by: Mathieu Desnoyers <mathieu.desnoy...@efficios.com> CC: Greg Kroah-Hartman <gre...@linuxfoundation.org> CC: John Stultz <john.stu...@linaro.org> --- wrapper/trace-clock.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/wrapper/trace-clock.h b/wrapper/trace-clock.h index 14d41af..496fec4 100644 --- a/wrapper/trace-clock.h +++ b/wrapper/trace-clock.h @@ -52,10 +52,10 @@ extern struct lttng_trace_clock *lttng_trace_clock; * CONFIG_DEBUG_TIMEKEEPING") introduces a buggy ktime_get_mono_fast_ns(). * This is fixed by patch "timekeeping: Fix __ktime_get_fast_ns() regression". */ -#if (LTTNG_KERNEL_RANGE(4,8,0, 4,8,1) \ - || LTTNG_KERNEL_RANGE(4,7,4, 4,7,7) \ - || LTTNG_KERNEL_RANGE(4,4,20, 4,4,24) \ - || LTTNG_KERNEL_RANGE(4,1,32, 4,1,34)) +#if (LTTNG_KERNEL_RANGE(4,8,0, 4,8,2) \ + || LTTNG_KERNEL_RANGE(4,7,4, 4,7,8) \ + || LTTNG_KERNEL_RANGE(4,4,20, 4,4,25) \ + || LTTNG_KERNEL_RANGE(4,1,32, 4,1,35)) #define LTTNG_CLOCK_NMI_SAFE_BROKEN #endif -- 2.1.4 _______________________________________________ lttng-dev mailing list lttng-dev@lists.lttng.org https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev