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, 93 insertions, 0 deletions

diff --git a/arch/powerpc/boot/of.c b/arch/powerpc/boot/of.c
new file mode 100644
index 000000000..2fbd4ae60
--- /dev/null
+++ b/arch/powerpc/boot/of.c
@@ -0,0 +1,93 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) Paul Mackerras 1997.
+ */
+#include <stdarg.h>
+#include <stddef.h>
+#include "types.h"
+#include "elf.h"
+#include "string.h"
+#include "stdio.h"
+#include "page.h"
+#include "ops.h"
+
+#include "of.h"
+
+/* Value picked to match that used by yaboot */
+#define PROG_START	0x01400000	/* only used on 64-bit systems */
+#define RAM_END		(512<<20)	/* Fixme: use OF */
+#define	ONE_MB		0x100000
+
+
+
+static unsigned long claim_base;
+
+void epapr_platform_init(unsigned long r3, unsigned long r4, unsigned long r5,
+			 unsigned long r6, unsigned long r7);
+
+static void *of_try_claim(unsigned long size)
+{
+	unsigned long addr = 0;
+
+	if (claim_base == 0)
+		claim_base = _ALIGN_UP((unsigned long)_end, ONE_MB);
+
+	for(; claim_base < RAM_END; claim_base += ONE_MB) {
+#ifdef DEBUG
+		printf("    trying: 0x%08lx\n\r", claim_base);
+#endif
+		addr = (unsigned long) of_claim(claim_base, size, 0);
+		if (addr != PROM_ERROR)
+			break;
+	}
+	if (addr == 0)
+		return NULL;
+	claim_base = PAGE_ALIGN(claim_base + size);
+	return (void *)addr;
+}
+
+static void of_image_hdr(const void *hdr)
+{
+	const Elf64_Ehdr *elf64 = hdr;
+
+	if (elf64->e_ident[EI_CLASS] == ELFCLASS64) {
+		/*
+		 * Maintain a "magic" minimum address. This keeps some older
+		 * firmware platforms running.
+		 */
+		if (claim_base < PROG_START)
+			claim_base = PROG_START;
+	}
+}
+
+static void of_platform_init(unsigned long a1, unsigned long a2, void *promptr)
+{
+	platform_ops.image_hdr = of_image_hdr;
+	platform_ops.malloc = of_try_claim;
+	platform_ops.exit = of_exit;
+	platform_ops.vmlinux_alloc = of_vmlinux_alloc;
+
+	dt_ops.finddevice = of_finddevice;
+	dt_ops.getprop = of_getprop;
+	dt_ops.setprop = of_setprop;
+
+	of_console_init();
+
+	of_init(promptr);
+	loader_info.promptr = promptr;
+	if (a1 && a2 && a2 != 0xdeadbeef) {
+		loader_info.initrd_addr = a1;
+		loader_info.initrd_size = a2;
+	}
+}
+
+void platform_init(unsigned long r3, unsigned long r4, unsigned long r5,
+		   unsigned long r6, unsigned long r7)
+{
+	/* Detect OF vs. ePAPR boot */
+	if (r5)
+		of_platform_init(r3, r4, (void *)r5);
+	else
+		epapr_platform_init(r3, r4, r5, r6, r7);
+}
+