[PATCH v2] sched/cache: Fix a thread aggregation conflict when there is one runnable task

[Chen Yu <yu.c.chen@intel.com>]

Problem Statement:
On systems with SD_ASYM_PACKING set at the PKG domain level (e.g.,
with ITMT enabled), the sched_asym() check in sched_balance_find_src_rq()
prevents pulling a single task from a higher-priority CPU to a
lower-priority CPU. This blocks migrate_llc_task migrations where a task
wants to move to its preferred LLC for cache locality.

For example, CPU 0 (highest priority) has a task preferring LLC3,
but the load balancer on CPU 120 (LLC3) cannot pull it because:
sched_asym(PKG_sd, cpu=0, dst_cpu=120) and nr_running == 1
The task remains stuck in the wrong LLC indefinitely.

Proposal:
Fix this by exempting migrate_llc_task from the sched_asym filter.

The adjacent SD_ASYM_CPUCAPACITY filter was left unchanged. In theory
it should have been handled in the same way. SD_ASYM_CPUCAPACITY is mostly
used in the hybrid CPU situations (e.g. P-core and E-core on some Intel
client CPUs). For those, the difference in CPU performance is pretty
significant and it is unlikely to get as much performance back from cache
co-location vs the CPU capacity. Leave the SD_ASYM_CPUCAPACITY code as is
till a workload and platform show difference otherwise.

Fixes: e4c9a4cb244a ("sched/cache: Add migrate_llc_task migration type for cache-aware balancing")
Reported-by: Yi Lai <yi1.lai@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Tested-by: Yi Lai <yi1.lai@intel.com>
Reviewed-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
---
v1->v2:
   Revise commit log to include the reason why SD_ASYM_CPUCAPACITY is not
   touched. (Zhan Xusheng)
---
 kernel/sched/fair.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index d78467ec6ee1..e6e6fa84c9e1 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -12986,8 +12986,12 @@ static struct rq *sched_balance_find_src_rq(struct lb_env *env,
 		 *
 		 * If balancing between cores, let lower priority CPUs help
 		 * SMT cores with more than one busy sibling.
+		 *
+		 * For migrate_llc_task, skip this check: cache locality
+		 * outweighs the asym priority.
 		 */
-		if (sched_asym(env->sd, i, env->dst_cpu) && nr_running == 1)
+		if (sched_asym(env->sd, i, env->dst_cpu) && nr_running == 1 &&
+		    env->migration_type != migrate_llc_task)
 			continue;
 
 		switch (env->migration_type) {
-- 
2.25.1