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, 102 insertions, 0 deletions

diff --git a/drivers/net/phy/bcm63xx.c b/drivers/net/phy/bcm63xx.c
new file mode 100644
index 000000000..b46a736a3
--- /dev/null
+++ b/drivers/net/phy/bcm63xx.c
@@ -0,0 +1,102 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ *	Driver for Broadcom 63xx SOCs integrated PHYs
+ */
+#include "bcm-phy-lib.h"
+#include <linux/module.h>
+#include <linux/phy.h>
+
+#define MII_BCM63XX_IR		0x1a	/* interrupt register */
+#define MII_BCM63XX_IR_EN	0x4000	/* global interrupt enable */
+#define MII_BCM63XX_IR_DUPLEX	0x0800	/* duplex changed */
+#define MII_BCM63XX_IR_SPEED	0x0400	/* speed changed */
+#define MII_BCM63XX_IR_LINK	0x0200	/* link changed */
+#define MII_BCM63XX_IR_GMASK	0x0100	/* global interrupt mask */
+
+MODULE_DESCRIPTION("Broadcom 63xx internal PHY driver");
+MODULE_AUTHOR("Maxime Bizon <mbizon@freebox.fr>");
+MODULE_LICENSE("GPL");
+
+static int bcm63xx_config_intr(struct phy_device *phydev)
+{
+	int reg, err;
+
+	reg = phy_read(phydev, MII_BCM63XX_IR);
+	if (reg < 0)
+		return reg;
+
+	if (phydev->interrupts == PHY_INTERRUPT_ENABLED) {
+		err = bcm_phy_ack_intr(phydev);
+		if (err)
+			return err;
+
+		reg &= ~MII_BCM63XX_IR_GMASK;
+		err = phy_write(phydev, MII_BCM63XX_IR, reg);
+	} else {
+		reg |= MII_BCM63XX_IR_GMASK;
+		err = phy_write(phydev, MII_BCM63XX_IR, reg);
+		if (err)
+			return err;
+
+		err = bcm_phy_ack_intr(phydev);
+	}
+
+	return err;
+}
+
+static int bcm63xx_config_init(struct phy_device *phydev)
+{
+	int reg, err;
+
+	/* ASYM_PAUSE bit is marked RO in datasheet, so don't cheat */
+	linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT, phydev->supported);
+
+	reg = phy_read(phydev, MII_BCM63XX_IR);
+	if (reg < 0)
+		return reg;
+
+	/* Mask interrupts globally.  */
+	reg |= MII_BCM63XX_IR_GMASK;
+	err = phy_write(phydev, MII_BCM63XX_IR, reg);
+	if (err < 0)
+		return err;
+
+	/* Unmask events we are interested in  */
+	reg = ~(MII_BCM63XX_IR_DUPLEX |
+		MII_BCM63XX_IR_SPEED |
+		MII_BCM63XX_IR_LINK) |
+		MII_BCM63XX_IR_EN;
+	return phy_write(phydev, MII_BCM63XX_IR, reg);
+}
+
+static struct phy_driver bcm63xx_driver[] = {
+{
+	.phy_id		= 0x00406000,
+	.phy_id_mask	= 0xfffffc00,
+	.name		= "Broadcom BCM63XX (1)",
+	/* PHY_BASIC_FEATURES */
+	.flags		= PHY_IS_INTERNAL,
+	.config_init	= bcm63xx_config_init,
+	.config_intr	= bcm63xx_config_intr,
+	.handle_interrupt = bcm_phy_handle_interrupt,
+}, {
+	/* same phy as above, with just a different OUI */
+	.phy_id		= 0x002bdc00,
+	.phy_id_mask	= 0xfffffc00,
+	.name		= "Broadcom BCM63XX (2)",
+	/* PHY_BASIC_FEATURES */
+	.flags		= PHY_IS_INTERNAL,
+	.config_init	= bcm63xx_config_init,
+	.config_intr	= bcm63xx_config_intr,
+	.handle_interrupt = bcm_phy_handle_interrupt,
+} };
+
+module_phy_driver(bcm63xx_driver);
+
+static const struct mdio_device_id __maybe_unused bcm63xx_tbl[] = {
+	{ 0x00406000, 0xfffffc00 },
+	{ 0x002bdc00, 0xfffffc00 },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(mdio, bcm63xx_tbl);