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, 142 insertions, 0 deletions

diff --git a/drivers/clk/ti/interface.c b/drivers/clk/ti/interface.c
new file mode 100644
index 000000000..039c8075e
--- /dev/null
+++ b/drivers/clk/ti/interface.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * OMAP interface clock support
+ *
+ * Copyright (C) 2013 Texas Instruments, Inc.
+ *
+ * Tero Kristo <t-kristo@ti.com>
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/slab.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/clk/ti.h>
+#include "clock.h"
+
+#undef pr_fmt
+#define pr_fmt(fmt) "%s: " fmt, __func__
+
+static const struct clk_ops ti_interface_clk_ops = {
+	.init		= &omap2_init_clk_clkdm,
+	.enable		= &omap2_dflt_clk_enable,
+	.disable	= &omap2_dflt_clk_disable,
+	.is_enabled	= &omap2_dflt_clk_is_enabled,
+};
+
+static struct clk *_register_interface(struct device_node *node,
+				       const char *name,
+				       const char *parent_name,
+				       struct clk_omap_reg *reg, u8 bit_idx,
+				       const struct clk_hw_omap_ops *ops)
+{
+	struct clk_init_data init = { NULL };
+	struct clk_hw_omap *clk_hw;
+	struct clk *clk;
+
+	clk_hw = kzalloc_obj(*clk_hw);
+	if (!clk_hw)
+		return ERR_PTR(-ENOMEM);
+
+	clk_hw->hw.init = &init;
+	clk_hw->ops = ops;
+	memcpy(&clk_hw->enable_reg, reg, sizeof(*reg));
+	clk_hw->enable_bit = bit_idx;
+
+	init.name = name;
+	init.ops = &ti_interface_clk_ops;
+	init.flags = 0;
+
+	init.num_parents = 1;
+	init.parent_names = &parent_name;
+
+	clk = of_ti_clk_register_omap_hw(node, &clk_hw->hw, name);
+
+	if (IS_ERR(clk))
+		kfree(clk_hw);
+
+	return clk;
+}
+
+static void __init _of_ti_interface_clk_setup(struct device_node *node,
+					      const struct clk_hw_omap_ops *ops)
+{
+	struct clk *clk;
+	const char *parent_name;
+	struct clk_omap_reg reg;
+	u8 enable_bit = 0;
+	const char *name;
+
+	if (ti_clk_get_reg_addr(node, 0, &reg))
+		return;
+
+	enable_bit = reg.bit;
+
+	parent_name = of_clk_get_parent_name(node, 0);
+	if (!parent_name) {
+		pr_err("%pOFn must have a parent\n", node);
+		return;
+	}
+
+	name = ti_dt_clk_name(node);
+	clk = _register_interface(node, name, parent_name, &reg,
+				  enable_bit, ops);
+
+	if (!IS_ERR(clk))
+		of_clk_add_provider(node, of_clk_src_simple_get, clk);
+}
+
+static void __init of_ti_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node, &clkhwops_iclk_wait);
+}
+CLK_OF_DECLARE(ti_interface_clk, "ti,omap3-interface-clock",
+	       of_ti_interface_clk_setup);
+
+static void __init of_ti_no_wait_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node, &clkhwops_iclk);
+}
+CLK_OF_DECLARE(ti_no_wait_interface_clk, "ti,omap3-no-wait-interface-clock",
+	       of_ti_no_wait_interface_clk_setup);
+
+#ifdef CONFIG_ARCH_OMAP3
+static void __init of_ti_hsotgusb_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node,
+				   &clkhwops_omap3430es2_iclk_hsotgusb_wait);
+}
+CLK_OF_DECLARE(ti_hsotgusb_interface_clk, "ti,omap3-hsotgusb-interface-clock",
+	       of_ti_hsotgusb_interface_clk_setup);
+
+static void __init of_ti_dss_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node,
+				   &clkhwops_omap3430es2_iclk_dss_usbhost_wait);
+}
+CLK_OF_DECLARE(ti_dss_interface_clk, "ti,omap3-dss-interface-clock",
+	       of_ti_dss_interface_clk_setup);
+
+static void __init of_ti_ssi_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node, &clkhwops_omap3430es2_iclk_ssi_wait);
+}
+CLK_OF_DECLARE(ti_ssi_interface_clk, "ti,omap3-ssi-interface-clock",
+	       of_ti_ssi_interface_clk_setup);
+
+static void __init of_ti_am35xx_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node, &clkhwops_am35xx_ipss_wait);
+}
+CLK_OF_DECLARE(ti_am35xx_interface_clk, "ti,am35xx-interface-clock",
+	       of_ti_am35xx_interface_clk_setup);
+#endif
+
+#ifdef CONFIG_SOC_OMAP2430
+static void __init of_ti_omap2430_interface_clk_setup(struct device_node *node)
+{
+	_of_ti_interface_clk_setup(node, &clkhwops_omap2430_i2chs_wait);
+}
+CLK_OF_DECLARE(ti_omap2430_interface_clk, "ti,omap2430-interface-clock",
+	       of_ti_omap2430_interface_clk_setup);
+#endif