Messages in this thread Patch in this message | | | From | Xie XiuQi <> | Subject | [PATCH 1/2] tracing: fix uptime overflow problem | Date | Sat, 28 Jun 2014 19:10:00 +0800 |
| |
The "uptime" tracer added in: commit 8aacf017b065a805d27467843490c976835eb4a5 tracing: Add "uptime" trace clock that uses jiffies has wraparound problems when the system has been up more than 1 hour 11 minutes and 34 seconds. It converts jiffies to nanoseconds using: (u64)jiffies_to_usecs(jiffy) * 1000ULL but since jiffies_to_usecs() only returns a 32-bit value, it truncates at 2^32 microseconds. An additional problem on 32-bit systems is that the argument is "unsigned long", so fixing the return value only helps until 2^32 jiffies (49.7 days on a HZ=1000 system).
Tony provide full featured jiffies_to_nsecs() function, but can't resolve another problem that jiffies_lock is not safe in NMI context.
Now we use the lockless function __current_kernel_time() and getboottime() to calculate the uptime.
The former discussion is here: http://lkml.org/lkml/2014/4/8/525
Additional, I changed trace_clock_jiffies to trace_clock_uptime, in order to better describe its function.
Reported-by: Tony Luck <tony.luck@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: stable@vger.kernel.org # 3.10+ Signed-off-by: Xie XiuQi <xiexiuqi@huawei.com> --- include/linux/trace_clock.h | 2 +- kernel/trace/trace.c | 2 +- kernel/trace/trace_clock.c | 15 +++++++++++---- 3 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/include/linux/trace_clock.h b/include/linux/trace_clock.h index 1d7ca27..2961ac7 100644 --- a/include/linux/trace_clock.h +++ b/include/linux/trace_clock.h @@ -16,7 +16,7 @@ extern u64 notrace trace_clock_local(void); extern u64 notrace trace_clock(void); -extern u64 notrace trace_clock_jiffies(void); +extern u64 notrace trace_clock_uptime(void); extern u64 notrace trace_clock_global(void); extern u64 notrace trace_clock_counter(void); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 384ede3..867e849 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -809,7 +809,7 @@ static struct { { trace_clock_local, "local", 1 }, { trace_clock_global, "global", 1 }, { trace_clock_counter, "counter", 0 }, - { trace_clock_jiffies, "uptime", 1 }, + { trace_clock_uptime, "uptime", 1 }, { trace_clock, "perf", 1 }, ARCH_TRACE_CLOCKS }; diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 26dc348..59e4d4d 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -19,6 +19,7 @@ #include <linux/percpu.h> #include <linux/sched.h> #include <linux/ktime.h> +#include <linux/time.h> #include <linux/trace_clock.h> /* @@ -58,14 +59,20 @@ u64 notrace trace_clock(void) } /* - * trace_jiffy_clock(): Simply use jiffies as a clock counter. + * trace_clock_uptime(): Use lockless version __current_kernel_time, + * so it's safe in NMI context. */ -u64 notrace trace_clock_jiffies(void) +u64 notrace trace_clock_uptime(void) { - u64 jiffy = jiffies - INITIAL_JIFFIES; + struct timespec uptime, now, boottime; + + /* Does not take xtime_lock, so it's safe in NMI context. */ + now = __current_kernel_time(); + getboottime(&boottime); + uptime = timespec_sub(now, boottime); /* Return nsecs */ - return (u64)jiffies_to_usecs(jiffy) * 1000ULL; + return timespec_to_ns(&uptime); } /* -- 2.0.0
| |