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, 90 insertions, 0 deletions

diff --git a/drivers/firmware/imx/rm.c b/drivers/firmware/imx/rm.c
new file mode 100644
index 000000000..d492b99e1
--- /dev/null
+++ b/drivers/firmware/imx/rm.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Copyright 2020 NXP
+ *
+ * File containing client-side RPC functions for the RM service. These
+ * function are ported to clients that communicate to the SC.
+ */
+
+#include <linux/firmware/imx/svc/rm.h>
+
+struct imx_sc_msg_rm_rsrc_owned {
+	struct imx_sc_rpc_msg hdr;
+	u16 resource;
+} __packed __aligned(4);
+
+/*
+ * This function check @resource is owned by current partition or not
+ *
+ * @param[in]     ipc         IPC handle
+ * @param[in]     resource    resource the control is associated with
+ *
+ * @return Returns 0 for not owned and 1 for owned.
+ */
+bool imx_sc_rm_is_resource_owned(struct imx_sc_ipc *ipc, u16 resource)
+{
+	struct imx_sc_msg_rm_rsrc_owned msg;
+	struct imx_sc_rpc_msg *hdr = &msg.hdr;
+
+	hdr->ver = IMX_SC_RPC_VERSION;
+	hdr->svc = IMX_SC_RPC_SVC_RM;
+	hdr->func = IMX_SC_RM_FUNC_IS_RESOURCE_OWNED;
+	hdr->size = 2;
+
+	msg.resource = resource;
+
+	/*
+	 * SCU firmware only returns value 0 or 1
+	 * for resource owned check which means not owned or owned.
+	 * So it is always successful.
+	 */
+	imx_scu_call_rpc(ipc, &msg, true);
+
+	return hdr->func;
+}
+EXPORT_SYMBOL(imx_sc_rm_is_resource_owned);
+
+struct imx_sc_msg_rm_get_resource_owner {
+	struct imx_sc_rpc_msg hdr;
+	union {
+		struct {
+			u16 resource;
+		} req;
+		struct {
+			u8 val;
+		} resp;
+	} data;
+} __packed __aligned(4);
+
+/*
+ * This function get @resource partition number
+ *
+ * @param[in]     ipc         IPC handle
+ * @param[in]     resource    resource the control is associated with
+ * @param[out]    pt          pointer to return the partition number
+ *
+ * @return Returns 0 for success and < 0 for errors.
+ */
+int imx_sc_rm_get_resource_owner(struct imx_sc_ipc *ipc, u16 resource, u8 *pt)
+{
+	struct imx_sc_msg_rm_get_resource_owner msg;
+	struct imx_sc_rpc_msg *hdr = &msg.hdr;
+	int ret;
+
+	hdr->ver = IMX_SC_RPC_VERSION;
+	hdr->svc = IMX_SC_RPC_SVC_RM;
+	hdr->func = IMX_SC_RM_FUNC_GET_RESOURCE_OWNER;
+	hdr->size = 2;
+
+	msg.data.req.resource = resource;
+
+	ret = imx_scu_call_rpc(ipc, &msg, true);
+	if (ret)
+		return ret;
+
+	if (pt)
+		*pt = msg.data.resp.val;
+
+	return 0;
+}
+EXPORT_SYMBOL(imx_sc_rm_get_resource_owner);