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, 75 insertions, 0 deletions

diff --git a/drivers/scsi/scsi_debugfs.c b/drivers/scsi/scsi_debugfs.c
new file mode 100644
index 000000000..eb52e39f3
--- /dev/null
+++ b/drivers/scsi/scsi_debugfs.c
@@ -0,0 +1,75 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bitops.h>
+#include <linux/cleanup.h>
+#include <linux/seq_file.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_dbg.h>
+#include <scsi/scsi_host.h>
+#include "scsi_debugfs.h"
+
+#define SCSI_CMD_FLAG_NAME(name)[const_ilog2(SCMD_##name)] = #name
+static const char *const scsi_cmd_flags[] = {
+	SCSI_CMD_FLAG_NAME(TAGGED),
+	SCSI_CMD_FLAG_NAME(INITIALIZED),
+	SCSI_CMD_FLAG_NAME(LAST),
+};
+#undef SCSI_CMD_FLAG_NAME
+
+static int scsi_flags_show(struct seq_file *m, const unsigned long flags,
+			   const char *const *flag_name, int flag_name_count)
+{
+	bool sep = false;
+	int i;
+
+	for_each_set_bit(i, &flags, BITS_PER_LONG) {
+		if (sep)
+			seq_puts(m, "|");
+		sep = true;
+		if (i < flag_name_count && flag_name[i])
+			seq_puts(m, flag_name[i]);
+		else
+			seq_printf(m, "%d", i);
+	}
+	return 0;
+}
+
+static const char *scsi_cmd_list_info(struct scsi_cmnd *cmd)
+{
+	struct Scsi_Host *shost = cmd->device->host;
+	struct scsi_cmnd *cmd2;
+
+	guard(spinlock_irq)(shost->host_lock);
+
+	list_for_each_entry(cmd2, &shost->eh_abort_list, eh_entry)
+		if (cmd == cmd2)
+			return "on eh_abort_list";
+
+	list_for_each_entry(cmd2, &shost->eh_cmd_q, eh_entry)
+		if (cmd == cmd2)
+			return "on eh_cmd_q";
+
+	return NULL;
+}
+
+void scsi_show_rq(struct seq_file *m, struct request *rq)
+{
+	struct scsi_cmnd *cmd = blk_mq_rq_to_pdu(rq);
+	int alloc_ms = jiffies_to_msecs(jiffies - cmd->jiffies_at_alloc);
+	int timeout_ms = jiffies_to_msecs(rq->timeout);
+	char buf[80] = "(?)";
+
+	if (cmd->flags & SCMD_INITIALIZED) {
+		const char *list_info = scsi_cmd_list_info(cmd);
+
+		__scsi_format_command(buf, sizeof(buf), cmd->cmnd, cmd->cmd_len);
+		seq_printf(m, ", .cmd=%s, .retries=%d, .allowed=%d, .result = %#x%s%s",
+			   buf, cmd->retries, cmd->allowed, cmd->result,
+			   list_info ? ", " : "", list_info ? : "");
+		seq_printf(m, ", .timeout=%d.%03d, allocated %d.%03d s ago",
+			   timeout_ms / 1000, timeout_ms % 1000,
+			   alloc_ms / 1000, alloc_ms % 1000);
+	}
+	seq_printf(m, ", .flags=");
+	scsi_flags_show(m, cmd->flags, scsi_cmd_flags,
+			ARRAY_SIZE(scsi_cmd_flags));
+}