sched/balancing: Rename find_busiest_queue() => sched_balance_find_src_rq()

The find_busiest_queue() naming has two small quirks:

 - Scheduler functions that deal with runqueues usually have a rq_ prefix
   or _rq postfix, but this function has neither.

 - Plus the 'busiest' qualifier to this function was historically
   correct, but has become somewhat of a misnomer: in quite a few
   cases we will not pick the busiest runqueue - but the best
   (possible) runqueue we can balance tasks from. So name it a
   bit more neutrally, similar to the 'src/dst' nomenclature
   we are already using when moving tasks between runqueues.

To fix both quirks, and to standardize scheduler load-balancing
function names on the sched_balance_() prefix, rename the
function to sched_balance_find_src_rq().

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://lore.kernel.org/r/20240308111819.1101550-7-mingo@kernel.org
This commit is contained in:
Ingo Molnar 2023-10-23 13:04:12 +02:00
parent 4c3e509ea9
commit f1cd2e2e79

View File

@ -10959,9 +10959,9 @@ out_balanced:
} }
/* /*
* find_busiest_queue - find the busiest runqueue among the CPUs in the group. * sched_balance_find_src_rq - find the busiest runqueue among the CPUs in the group.
*/ */
static struct rq *find_busiest_queue(struct lb_env *env, static struct rq *sched_balance_find_src_rq(struct lb_env *env,
struct sched_group *group) struct sched_group *group)
{ {
struct rq *busiest = NULL, *rq; struct rq *busiest = NULL, *rq;
@ -11280,7 +11280,7 @@ redo:
goto out_balanced; goto out_balanced;
} }
busiest = find_busiest_queue(&env, group); busiest = sched_balance_find_src_rq(&env, group);
if (!busiest) { if (!busiest) {
schedstat_inc(sd->lb_nobusyq[idle]); schedstat_inc(sd->lb_nobusyq[idle]);
goto out_balanced; goto out_balanced;