the Linux scheduler is starving a number of workloads. So default
to more agressive idle-balancing. This hurts lmbench context-switching
numbers (which was the main reason we sucked at idle-balancing for
such a long time) but the lmbench numbers are fine once the system is
minimally utilized.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
.cache_nice_tries = 0, \
.busy_idx = 0, \
.idle_idx = 0, \
- .newidle_idx = 1, \
+ .newidle_idx = 0, \
.wake_idx = 0, \
.forkexec_idx = 0, \
.flags = SD_LOAD_BALANCE \
.imbalance_pct = 125, \
.cache_nice_tries = 1, \
.busy_idx = 2, \
- .idle_idx = 1, \
- .newidle_idx = 2, \
+ .idle_idx = 0, \
+ .newidle_idx = 0, \
.wake_idx = 1, \
.forkexec_idx = 1, \
.flags = SD_LOAD_BALANCE \
| SD_BALANCE_NEWIDLE \
| SD_BALANCE_EXEC \
| SD_WAKE_AFFINE \
+ | SD_WAKE_IDLE \
| SD_SHARE_PKG_RESOURCES\
| BALANCE_FOR_MC_POWER, \
.last_balance = jiffies, \
.imbalance_pct = 125, \
.cache_nice_tries = 1, \
.busy_idx = 2, \
- .idle_idx = 1, \
- .newidle_idx = 2, \
+ .idle_idx = 0, \
+ .newidle_idx = 0, \
.wake_idx = 1, \
.forkexec_idx = 1, \
.flags = SD_LOAD_BALANCE \
| SD_BALANCE_NEWIDLE \
| SD_BALANCE_EXEC \
| SD_WAKE_AFFINE \
+ | SD_WAKE_IDLE \
| BALANCE_FOR_PKG_POWER,\
.last_balance = jiffies, \
.balance_interval = 1, \