On Mon, Nov 04, 2013 at 06:34:23PM +0100, Frederic Weisbecker wrote:
> On Sun, Oct 20, 2013 at 01:10:06PM +0200, Andreas Mohr wrote:
> > In case of high update traffic of parts guarded by rq->iowait_lock
> > (is that a relevant case?),
> >
> > it might be useful to merely grab all relevant values i
On Sun, Oct 20, 2013 at 01:10:06PM +0200, Andreas Mohr wrote:
> Hi,
>
>
> +u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time)
> +{
> + ktime_t iowait, delta = { .tv64 = 0 };
> + struct rq *rq = cpu_rq(cpu);
> + ktime_t now = ktime_get();
> + unsigned int seq;
> +
> + d
Hi,
+u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time)
+{
+ ktime_t iowait, delta = { .tv64 = 0 };
+ struct rq *rq = cpu_rq(cpu);
+ ktime_t now = ktime_get();
+ unsigned int seq;
+
+ do {
+ seq = read_seqbegin(&rq->iowait_lock);
+
On Sat, Oct 19, 2013 at 06:05:17PM +0200, Peter Zijlstra wrote:
> On Sat, Oct 19, 2013 at 06:02:37PM +0200, Frederic Weisbecker wrote:
> > > I just had a look at delayacct; wth wrote that crap; that too uses
> > > gtod.
> >
> > I can't find where it does that. kernel/delayacct.c doesn't seem to at
On Sat, Oct 19, 2013 at 06:02:37PM +0200, Frederic Weisbecker wrote:
> > I just had a look at delayacct; wth wrote that crap; that too uses
> > gtod.
>
> I can't find where it does that. kernel/delayacct.c doesn't seem to at least.
Look for do_posix_clock_monotonic_gettime() ;-)
--
To unsubscribe
On Sat, Oct 19, 2013 at 05:35:35PM +0200, Peter Zijlstra wrote:
> On Sat, Oct 19, 2013 at 05:17:20PM +0200, Frederic Weisbecker wrote:
> > +u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time)
> > +{
> > + ktime_t iowait, delta = { .tv64 = 0 };
> > + struct rq *rq = cpu_rq(cpu);
> > + k
On Sat, Oct 19, 2013 at 05:17:20PM +0200, Frederic Weisbecker wrote:
> +u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time)
> +{
> + ktime_t iowait, delta = { .tv64 = 0 };
> + struct rq *rq = cpu_rq(cpu);
> + ktime_t now = ktime_get();
> + unsigned int seq;
> +
> + do {
>
The iowait time accounting has awkward semantics. It is performed per
CPU. The iowait time of a CPU starts when a task sleeps on IO on
this CPU and stops when that task later wakes up on any CPU.
This assumes that a sleeping task is still assigned to the CPU
it was running on last until it wakes u
8 matches
Mail list logo
