sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling buggrafted

alb_break_llc() decides whether to break LLC preference during active
load balance. It does so by testing that every runnable fair task on the
source rq prefers its LLC:

	env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable

But the two counters cover different sets. nr_pref_llc_running is updated
in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued,
so it follows queued tasks. h_nr_runnable is updated in set_delayed()/
clear_delayed() and drops delay-dequeued tasks.

So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted
in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks,
alb_break_llc() returns false, and active balance is free to pull a task
off its preferred LLC. Active balance only moves runnable tasks, and this
is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB
skips the per-task test in can_migrate_task(). The runnable set is the one
we want.

Fix it on the counter side. A task should be counted in
nr_pref_llc_running exactly while it is both queued on its preferred LLC
(pref_llc_queued) and runnable (!sched_delayed). Define that membership
once in task_pref_llc_runnable(), and adjust the counter only through
pref_llc_running_inc()/pref_llc_running_dec() from the four sites that
change either input: account_llc_enqueue(), account_llc_dequeue(),
set_delayed() and clear_delayed(). Gating every update on the same
predicate keeps the delay, wake and dequeue paths from double-counting
or underflowing; see the comments at those sites for the ordering.

nr_llc_running and sd->llc_counts are not touched and stay on queued
semantics.

Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC")
Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/
Reported-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Suggested-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kayra Cizmeci <kayracizmeci@gmail.com>
Cc: <stable@kernel.org> # v7.2.x
Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com

1 files changed, 70 insertions, 0 deletions

diff --git a/tools/perf/util/parse-sublevel-options.c b/tools/perf/util/parse-sublevel-options.c
new file mode 100644
index 000000000..a841d17ff
--- /dev/null
+++ b/tools/perf/util/parse-sublevel-options.c
@@ -0,0 +1,70 @@
+#include <stdlib.h>
+#include <stdint.h>
+#include <string.h>
+#include <stdio.h>
+
+#include "util/debug.h"
+#include "util/parse-sublevel-options.h"
+
+static int parse_one_sublevel_option(const char *str,
+				     struct sublevel_option *opts)
+{
+	struct sublevel_option *opt = opts;
+	char *vstr, *s = strdup(str);
+	int v = 1;
+
+	if (!s) {
+		pr_err("no memory\n");
+		return -1;
+	}
+
+	vstr = strchr(s, '=');
+	if (vstr)
+		*vstr++ = 0;
+
+	while (opt->name) {
+		if (!strcmp(s, opt->name))
+			break;
+		opt++;
+	}
+
+	if (!opt->name) {
+		pr_err("Unknown option name '%s'\n", s);
+		free(s);
+		return -1;
+	}
+
+	if (vstr)
+		v = atoi(vstr);
+
+	*opt->value_ptr = v;
+	free(s);
+	return 0;
+}
+
+/* parse options like --foo a=<n>,b,c... */
+int perf_parse_sublevel_options(const char *str, struct sublevel_option *opts)
+{
+	char *s = strdup(str);
+	char *p = NULL;
+	int ret;
+
+	if (!s) {
+		pr_err("no memory\n");
+		return -1;
+	}
+
+	p = strtok(s, ",");
+	while (p) {
+		ret = parse_one_sublevel_option(p, opts);
+		if (ret) {
+			free(s);
+			return ret;
+		}
+
+		p = strtok(NULL, ",");
+	}
+
+	free(s);
+	return 0;
+}