tick-sched: fix idle and iowait sleeptime accounting vs CPU hotplug
Subsystems (2)
Heiko Carstens
maintainerPATCH 1/1 tick-sched: fix idle and iowait sleeptime accounting vs CPU hotplug
When offlining and onlining CPUs the overall reported idle and iowait times as reported by /proc/stat jump backward and forward:
Quoted 1 line
cat
/proc/stat
cpu 132 0 176 225249 47 6 6 21 0 0
cpu0 80 0 115 112575 33 3 4 18 0 0
cpu1 52 0 60 112673 13 3 1 2 0 0
Quoted 2 lines
chcpu -d 1
cat/proc/stat
cpu 133 0 177 226681 47 6 6 21 0 0
cpu0 80 0 116 113387 33 3 4 18 0 0
Quoted 2 lines
chcpu -e 1
cat/proc/stat
cpu 133 0 178 114431 33 6 6 21 0 0 <---- jump backward cpu0 80 0 116 114247 33 3 4 18 0 0
cpu1 52 0 61 183 0 3 1 2 0 0 <---- idle + iowait start with 0
Quoted 2 lines
chcpu -d 1
cat/proc/stat
cpu 133 0 178 228956 47 6 6 21 0 0 <---- jump forward cpu0 81 0 117 114929 33 3 4 18 0 0
Reason for this is that get_idle_time() in fs/proc/stat.c has different sources for both values depending on if a CPU is online or offline:
- if a CPU is online the values may be taken from its per cpu tick_cpu_sched structure
- if a CPU is offline the values are taken from its per cpu cpustat structure
The problem is that the per cpu tick_cpu_sched structure is set to zero on CPU offline. See tick_cancel_sched_timer() in kernel/time/tick-sched.c.
Therefore when a CPU is brought offline and online afterwards both its idle and iowait sleeptime will be zero, causing a jump backward in total system idle and iowait sleeptime. In a similar way if a CPU is then brought offline again the total idle and iowait sleeptimes will jump forward.
It looks like this behavior was introduced with commit 4b0c0f294f60 ("tick: Cleanup NOHZ per cpu data on cpu down").
This was only noticed now on s390, since we switched to generic idle time reporting with commit be76ea614460 ("s390/idle: remove arch_cpu_idle_time() and corresponding code").
Fix this by preserving the values of idle_sleeptime and iowait_sleeptime members of the per-cpu tick_sched structure on CPU hotplug.
--- kernel/time/tick-sched.c | 5 +++++ 1 file changed, 5 insertions(+)
kernel/time/tick-sched.c +5 -0
@@ -1576,13 +1576,18 @@ void tick_setup_sched_timer(void)void tick_cancel_sched_timer (int cpu) {struct tick_sched *ts = &per_cpu (tick_cpu_sched, cpu); +ktime_t idle_sleeptime, iowait_sleeptime;# ifdef CONFIG_HIGH_RES_TIMERS if (ts->sched_timer.base)hrtimer_cancel (&ts->sched_timer);# endif + idle_sleeptime = ts->idle_sleeptime; + iowait_sleeptime = ts->iowait_sleeptime;memset (ts,0 ,sizeof (*ts)); + ts->idle_sleeptime = idle_sleeptime; + ts->iowait_sleeptime = iowait_sleeptime; }#endif
Signature · 1 line
2.40.1
Quoted 68 lines · Heiko Carstens
When offlining and onlining CPUs the overall reported idle and iowait times as reported by
/proc/statjump backward and forward:Quoted 1 line
cat
/proc/statcpu 132 0 176 225249 47 6 6 21 0 0
cpu0 80 0 115 112575 33 3 4 18 0 0
cpu1 52 0 60 112673 13 3 1 2 0 0Quoted 2 lines
chcpu -d 1
cat/proc/statcpu 133 0 177 226681 47 6 6 21 0 0
cpu0 80 0 116 113387 33 3 4 18 0 0Quoted 2 lines
chcpu -e 1
cat/proc/statcpu 133 0 178 114431 33 6 6 21 0 0 <---- jump backward cpu0 80 0 116 114247 33 3 4 18 0 0
cpu1 52 0 61 183 0 3 1 2 0 0 <---- idle + iowait start with 0Quoted 2 lines
chcpu -d 1
cat/proc/statcpu 133 0 178 228956 47 6 6 21 0 0 <---- jump forward cpu0 81 0 117 114929 33 3 4 18 0 0
Reason for this is that
get_idle_time()infs/proc/stat.chas different sources for both values depending on if a CPU is online or offline:- if a CPU is online the values may be taken from its per cpu
tick_cpu_schedstructure- if a CPU is offline the values are taken from its per cpu cpustat structure
The problem is that the per cpu
tick_cpu_schedstructure is set to zero on CPU offline. Seetick_cancel_sched_timer()inkernel/time/tick-sched.c.Therefore when a CPU is brought offline and online afterwards both its idle and iowait sleeptime will be zero, causing a jump backward in total system idle and iowait sleeptime. In a similar way if a CPU is then brought offline again the total idle and iowait sleeptimes will jump forward.
It looks like this behavior was introduced with commit
4b0c0f294f60("tick: Cleanup NOHZ per cpu data on cpu down").This was only noticed now on s390, since we switched to generic idle time reporting with commit
be76ea614460("s390/idle: removearch_cpu_idle_time()and corresponding code").Fix this by preserving the values of
idle_sleeptimeandiowait_sleeptimemembers of the per-cputick_schedstructure on CPU hotplug.Fixes 4b0c0f294f60 ("tick: Cleanup NOHZ per cpu data on cpu down")Reported-by Gerald Schaefer <[email protected]>Signed-off-by Heiko Carstens <[email protected]>--- kernel/time/tick-sched.c | 5 +++++ 1 file changed, 5 insertions(+)kernel/time/tick-sched.c +1
-0@@ -1576,13 +1576,18 @@ void tick_setup_sched_timer(void)void tick_cancel_sched_timer (int cpu) {struct tick_sched *ts = &per_cpu (tick_cpu_sched, cpu); +ktime_t idle_sleeptime, iowait_sleeptime;
kernel/time/tick-sched.c void tick_setup_sched_timer(void) · line 1590
Quoted 10 lines
kernel/time/tick-sched.c +4
-0# ifdef CONFIG_HIGH_RES_TIMERS if (ts->sched_timer.base)hrtimer_cancel (&ts->sched_timer);# endif + idle_sleeptime = ts->idle_sleeptime; + iowait_sleeptime = ts->iowait_sleeptime;memset (ts,0 ,sizeof (*ts)); + ts->idle_sleeptime = idle_sleeptime; + ts->iowait_sleeptime = iowait_sleeptime;
And this is safe because it is in global stop machine. So we are guaranteed that nobody sees the transitionning state. In the worst case ts->idle_sleeptime_seq is observed as changed to 0 in read_seqcount_retry() and the values are simply fetched again.
This makes me think that we should always use cpustat[CPUTIME_IDLE] instead of maintaining this separate ts->idle_sleeptime field. kcpustat even has a seqcount that would make ts->idle_sleeptime_seq obsolete. Then the tick based idle accounting could disappear on nohz, along with a few hacks. Instead of that we are currently maintaining two different idle accounting that are roughly the same.
But anyway this is all a different story, just mumbling to myself for the next nohz cleanups.
Thanks!
[tip: timers/core] tick-sched: Fix idle and iowait sleeptime accounting vs CPU hotplug
In reply to Heiko CarstensThe following commit has been merged into the timers/core branch of tip:
The following commit has been merged into the timers/core branch of tip:
Commit-ID: 71fee48fb772ac4f6cfa63dbebc5629de8b4cc09 Gitweb: https://git.kernel.org/tip/71fee48fb772ac4f6cfa63dbebc5629de8b4cc09 Author: Heiko Carstens <[email protected]> AuthorDate: Mon, 15 Jan 2024 17:35:55 +01:00 Committer: Thomas Gleixner <[email protected]> CommitterDate: Fri, 19 Jan 2024 16:40:38 +01:00
tick-sched: Fix idle and iowait sleeptime accounting vs CPU hotplug
When offlining and onlining CPUs the overall reported idle and iowait times as reported by /proc/stat jump backward and forward:
cpu 132 0 176 225249 47 6 6 21 0 0
cpu0 80 0 115 112575 33 3 4 18 0 0
cpu1 52 0 60 112673 13 3 1 2 0 0
cpu 133 0 177 226681 47 6 6 21 0 0
cpu0 80 0 116 113387 33 3 4 18 0 0
cpu 133 0 178 114431 33 6 6 21 0 0 <---- jump backward cpu0 80 0 116 114247 33 3 4 18 0 0
cpu1 52 0 61 183 0 3 1 2 0 0 <---- idle + iowait start with 0
cpu 133 0 178 228956 47 6 6 21 0 0 <---- jump forward cpu0 81 0 117 114929 33 3 4 18 0 0
Reason for this is that get_idle_time() in fs/proc/stat.c has different sources for both values depending on if a CPU is online or offline:
- if a CPU is online the values may be taken from its per cpu tick_cpu_sched structure
- if a CPU is offline the values are taken from its per cpu cpustat structure
The problem is that the per cpu tick_cpu_sched structure is set to zero on CPU offline. See tick_cancel_sched_timer() in kernel/time/tick-sched.c.
Therefore when a CPU is brought offline and online afterwards both its idle and iowait sleeptime will be zero, causing a jump backward in total system idle and iowait sleeptime. In a similar way if a CPU is then brought offline again the total idle and iowait sleeptimes will jump forward.
It looks like this behavior was introduced with commit 4b0c0f294f60 ("tick: Cleanup NOHZ per cpu data on cpu down").
This was only noticed now on s390, since we switched to generic idle time reporting with commit be76ea614460 ("s390/idle: remove arch_cpu_idle_time() and corresponding code").
Fix this by preserving the values of idle_sleeptime and iowait_sleeptime members of the per-cpu tick_sched structure on CPU hotplug.
--- kernel/time/tick-sched.c | 5 +++++ 1 file changed, 5 insertions(+)
kernel/time/tick-sched.c +5 -0
@@ -1576,13 +1576,18 @@ void tick_setup_sched_timer(void)void tick_cancel_sched_timer (int cpu) {struct tick_sched *ts = &per_cpu (tick_cpu_sched, cpu); +ktime_t idle_sleeptime, iowait_sleeptime;# ifdef CONFIG_HIGH_RES_TIMERS if (ts->sched_timer.base)hrtimer_cancel (&ts->sched_timer);# endif + idle_sleeptime = ts->idle_sleeptime; + iowait_sleeptime = ts->iowait_sleeptime;memset (ts,0 ,sizeof (*ts)); + ts->idle_sleeptime = idle_sleeptime; + ts->iowait_sleeptime = iowait_sleeptime; }#endif
kernel/time/tick-sched.c void tick_setup_sched_timer(void) · line 1591
Quoted 8 lines · Heiko Carstens
kernel/time/tick-sched.c +4
-0+ idle_sleeptime = ts->idle_sleeptime; + iowait_sleeptime = ts->iowait_sleeptime;memset (ts,0 ,sizeof (*ts)); + ts->idle_sleeptime = idle_sleeptime; + ts->iowait_sleeptime = iowait_sleeptime; }
Should idle_calls and idle_sleeps be preserved and restored too?
Seems like if we preserve theidle_sleeptime, and wish to compute the average sleep time per sleep, we will need to know the value of idle_sleeps that's also preserved across CPU offline/online.
Tim
Frederic Weisbecker
maintainerQuoted 13 lines · Tim Chen
kernel/time/tick-sched.c void tick_setup_sched_timer(void) · line 1591
Quoted 8 lines · Heiko Carstens
kernel/time/tick-sched.c +4
-0+ idle_sleeptime = ts->idle_sleeptime; + iowait_sleeptime = ts->iowait_sleeptime;memset (ts,0 ,sizeof (*ts)); + ts->idle_sleeptime = idle_sleeptime; + ts->iowait_sleeptime = iowait_sleeptime; }Should
idle_callsandidle_sleepsbe preserved and restored too?
Quoted 4 lines
Seems like if we preserve the
idle_sleeptime, and wish to compute the average sleep time per sleep, we will need to know the value ofidle_sleepsthat's also preserved across CPU offline/online.
I guess those can be saved as well. Would you like to send the patch?
Thanks.
Quoted 2 lines
Tim
Quoted 14 lines · Frederic Weisbecker
Quoted 13 lines · Tim Chen
kernel/time/tick-sched.c void tick_setup_sched_timer(void) · line 1591
Quoted 8 lines · Heiko Carstens
kernel/time/tick-sched.c +4
-0+ idle_sleeptime = ts->idle_sleeptime; + iowait_sleeptime = ts->iowait_sleeptime;memset (ts,0 ,sizeof (*ts)); + ts->idle_sleeptime = idle_sleeptime; + ts->iowait_sleeptime = iowait_sleeptime; }Should
idle_callsandidle_sleepsbe preserved and restored too?
Quoted 7 lines
Quoted 4 lines
Seems like if we preserve the
idle_sleeptime, and wish to compute the average sleep time per sleep, we will need to know the value ofidle_sleepsthat's also preserved across CPU offline/online.I guess those can be saved as well. Would you like to send the patch?
Okay, sent the patch in a separate email.
Tim