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, 156 insertions, 0 deletions

diff --git a/block/partitions/atari.c b/block/partitions/atari.c
new file mode 100644
index 000000000..2438d1448
--- /dev/null
+++ b/block/partitions/atari.c
@@ -0,0 +1,156 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  fs/partitions/atari.c
+ *
+ *  Code extracted from drivers/block/genhd.c
+ *
+ *  Copyright (C) 1991-1998  Linus Torvalds
+ *  Re-organised Feb 1998 Russell King
+ */
+
+#include <linux/ctype.h>
+#include "check.h"
+#include "atari.h"
+
+/* ++guenther: this should be settable by the user ("make config")?.
+ */
+#define ICD_PARTS
+
+/* check if a partition entry looks valid -- Atari format is assumed if at
+   least one of the primary entries is ok this way */
+#define	VALID_PARTITION(pi,hdsiz)					     \
+    (((pi)->flg & 1) &&							     \
+     isalnum((pi)->id[0]) && isalnum((pi)->id[1]) && isalnum((pi)->id[2]) && \
+     be32_to_cpu((pi)->st) <= (hdsiz) &&				     \
+     be32_to_cpu((pi)->st) + be32_to_cpu((pi)->siz) <= (hdsiz))
+
+static inline int OK_id(char *s)
+{
+	return  memcmp (s, "GEM", 3) == 0 || memcmp (s, "BGM", 3) == 0 ||
+		memcmp (s, "LNX", 3) == 0 || memcmp (s, "SWP", 3) == 0 ||
+		memcmp (s, "RAW", 3) == 0 ;
+}
+
+int atari_partition(struct parsed_partitions *state)
+{
+	Sector sect;
+	struct rootsector *rs;
+	struct partition_info *pi;
+	u32 extensect;
+	u32 hd_size;
+	int slot;
+#ifdef ICD_PARTS
+	int part_fmt = 0; /* 0:unknown, 1:AHDI, 2:ICD/Supra */
+#endif
+
+	/*
+	 * ATARI partition scheme supports 512 lba only.  If this is not
+	 * the case, bail early to avoid miscalculating hd_size.
+	 */
+	if (queue_logical_block_size(state->disk->queue) != 512)
+		return 0;
+
+	rs = read_part_sector(state, 0, &sect);
+	if (!rs)
+		return -1;
+
+	/* Verify this is an Atari rootsector: */
+	hd_size = get_capacity(state->disk);
+	if (!VALID_PARTITION(&rs->part[0], hd_size) &&
+	    !VALID_PARTITION(&rs->part[1], hd_size) &&
+	    !VALID_PARTITION(&rs->part[2], hd_size) &&
+	    !VALID_PARTITION(&rs->part[3], hd_size)) {
+		/*
+		 * if there's no valid primary partition, assume that no Atari
+		 * format partition table (there's no reliable magic or the like
+	         * :-()
+		 */
+		put_dev_sector(sect);
+		return 0;
+	}
+
+	pi = &rs->part[0];
+	seq_buf_puts(&state->pp_buf, " AHDI");
+	for (slot = 1; pi < &rs->part[4] && slot < state->limit; slot++, pi++) {
+		struct rootsector *xrs;
+		Sector sect2;
+		ulong partsect;
+
+		if ( !(pi->flg & 1) )
+			continue;
+		/* active partition */
+		if (memcmp (pi->id, "XGM", 3) != 0) {
+			/* we don't care about other id's */
+			put_partition (state, slot, be32_to_cpu(pi->st),
+					be32_to_cpu(pi->siz));
+			continue;
+		}
+		/* extension partition */
+#ifdef ICD_PARTS
+		part_fmt = 1;
+#endif
+		seq_buf_puts(&state->pp_buf, " XGM<");
+		partsect = extensect = be32_to_cpu(pi->st);
+		while (1) {
+			xrs = read_part_sector(state, partsect, &sect2);
+			if (!xrs) {
+				printk (" block %ld read failed\n", partsect);
+				put_dev_sector(sect);
+				return -1;
+			}
+
+			/* ++roman: sanity check: bit 0 of flg field must be set */
+			if (!(xrs->part[0].flg & 1)) {
+				printk( "\nFirst sub-partition in extended partition is not valid!\n" );
+				put_dev_sector(sect2);
+				break;
+			}
+
+			put_partition(state, slot,
+				   partsect + be32_to_cpu(xrs->part[0].st),
+				   be32_to_cpu(xrs->part[0].siz));
+
+			if (!(xrs->part[1].flg & 1)) {
+				/* end of linked partition list */
+				put_dev_sector(sect2);
+				break;
+			}
+			if (memcmp( xrs->part[1].id, "XGM", 3 ) != 0) {
+				printk("\nID of extended partition is not XGM!\n");
+				put_dev_sector(sect2);
+				break;
+			}
+
+			partsect = be32_to_cpu(xrs->part[1].st) + extensect;
+			put_dev_sector(sect2);
+			if (++slot == state->limit) {
+				printk( "\nMaximum number of partitions reached!\n" );
+				break;
+			}
+		}
+		seq_buf_puts(&state->pp_buf, " >");
+	}
+#ifdef ICD_PARTS
+	if ( part_fmt!=1 ) { /* no extended partitions -> test ICD-format */
+		pi = &rs->icdpart[0];
+		/* sanity check: no ICD format if first partition invalid */
+		if (OK_id(pi->id)) {
+			seq_buf_puts(&state->pp_buf, " ICD<");
+			for (; pi < &rs->icdpart[8] && slot < state->limit; slot++, pi++) {
+				/* accept only GEM,BGM,RAW,LNX,SWP partitions */
+				if (!((pi->flg & 1) && OK_id(pi->id)))
+					continue;
+				put_partition (state, slot,
+						be32_to_cpu(pi->st),
+						be32_to_cpu(pi->siz));
+			}
+			seq_buf_puts(&state->pp_buf, " >");
+		}
+	}
+#endif
+	put_dev_sector(sect);
+
+	seq_buf_puts(&state->pp_buf, "\n");
+
+	return 1;
+}