MFC r336855

Fix the long term ULE load balancer so that it actually works.  The
initial call to sched_balance() during startup is meant to initialize
balance_ticks, but does not actually do that since smp_started is
still zero at that time.  Since balance_ticks does not get set,
there are no further calls to sched_balance().  Fix this by setting
balance_ticks in sched_initticks() since we know the value of
balance_interval at that time, and eliminate the useless startup
call to sched_balance().  We don't need to randomize the intial
value of balance_ticks.

Since there is now only one call to sched_balance(), we can hoist
the tests at the top of this function out to the caller and avoid
the overhead of the function call when running a SMP kernel on UP
hardware.

PR:		223914
Reviewed by:	kib
This commit is contained in:
Don Lewis 2018-08-12 03:22:28 +00:00
parent b986bd1470
commit ec65747f54

View File

@ -882,9 +882,6 @@ sched_balance(void)
{
struct tdq *tdq;
if (smp_started == 0 || rebalance == 0)
return;
balance_ticks = max(balance_interval / 2, 1) +
(sched_random() % balance_interval);
tdq = TDQ_SELF();
@ -1408,7 +1405,6 @@ sched_setup_smp(void)
panic("Can't find cpu group for %d\n", i);
}
balance_tdq = TDQ_SELF();
sched_balance();
}
#endif
@ -1469,6 +1465,7 @@ sched_initticks(void *dummy)
* what realstathz is.
*/
balance_interval = realstathz;
balance_ticks = balance_interval;
affinity = SCHED_AFFINITY_DEFAULT;
#endif
if (sched_idlespinthresh < 0)
@ -2384,7 +2381,7 @@ sched_clock(struct thread *td)
/*
* We run the long term load balancer infrequently on the first cpu.
*/
if (balance_tdq == tdq) {
if (balance_tdq == tdq && smp_started != 0 && rebalance != 0) {
if (balance_ticks && --balance_ticks == 0)
sched_balance();
}