Revision 483b4ee60edbefdfbff0dd538fb81f368d9e7c0d authored by Suresh Siddha on 04 February 2009, 19:59:44 UTC, committed by Ingo Molnar on 04 February 2009, 21:31:19 UTC
Christian Borntraeger reports: > After a logical cpu offline, even on a complete idle system, there > is one cpu with full ticks. It turns out that nohz.cpu_mask has the > the offlined cpu still set. > > In select_nohz_load_balancer() we check if the system is completely > idle to turn of load balancing. We compare cpu_online_map with > nohz.cpu_mask. Since cpu_online_map is updated on cpu unplug, > but nohz.cpu_mask is not, the check fails and the scheduler believes > that we need an "idle load balancer" even on a fully idle system. > Since the ilb cpu does not deactivate the timer tick this breaks NOHZ. Fix the select_nohz_load_balancer() to not set the nohz.cpu_mask while a cpu is going offline. Reported-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
1 parent 3562612
| File | Mode | Size |
|---|---|---|
| Kconfig | -rw-r--r-- | 2.5 KB |
| Makefile | -rw-r--r-- | 821 bytes |
| apm_power.c | -rw-r--r-- | 10.1 KB |
| bq27x00_battery.c | -rw-r--r-- | 8.1 KB |
| da9030_battery.c | -rw-r--r-- | 15.8 KB |
| ds2760_battery.c | -rw-r--r-- | 12.5 KB |
| olpc_battery.c | -rw-r--r-- | 11.1 KB |
| pcf50633-charger.c | -rw-r--r-- | 8.7 KB |
| pda_power.c | -rw-r--r-- | 8.8 KB |
| pmu_battery.c | -rw-r--r-- | 5.2 KB |
| power_supply.h | -rw-r--r-- | 1.2 KB |
| power_supply_core.c | -rw-r--r-- | 4.3 KB |
| power_supply_leds.c | -rw-r--r-- | 4.0 KB |
| power_supply_sysfs.c | -rw-r--r-- | 7.2 KB |
| tosa_battery.c | -rw-r--r-- | 11.5 KB |
| wm8350_power.c | -rw-r--r-- | 15.2 KB |
| wm97xx_battery.c | -rw-r--r-- | 6.3 KB |
Computing file changes ...
