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, 74 insertions, 0 deletions

diff --git a/mm/ioremap.c b/mm/ioremap.c
new file mode 100644
index 000000000..c36dd9f62
--- /dev/null
+++ b/mm/ioremap.c
@@ -0,0 +1,74 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Re-map IO memory to kernel address space so that we can access it.
+ * This is needed for high PCI addresses that aren't mapped in the
+ * 640k-1MB IO memory area on PC's
+ *
+ * (C) Copyright 1995 1996 Linus Torvalds
+ */
+#include <linux/vmalloc.h>
+#include <linux/mm.h>
+#include <linux/io.h>
+#include <linux/export.h>
+#include <linux/ioremap.h>
+
+void __iomem *generic_ioremap_prot(phys_addr_t phys_addr, size_t size,
+				   pgprot_t prot)
+{
+	unsigned long offset, vaddr;
+	phys_addr_t last_addr;
+	struct vm_struct *area;
+
+	/* An early platform driver might end up here */
+	if (WARN_ON_ONCE(!slab_is_available()))
+		return NULL;
+
+	/* Disallow wrap-around or zero size */
+	last_addr = phys_addr + size - 1;
+	if (!size || last_addr < phys_addr)
+		return NULL;
+
+	/* Page-align mappings */
+	offset = phys_addr & (~PAGE_MASK);
+	phys_addr -= offset;
+	size = PAGE_ALIGN(size + offset);
+
+	area = __get_vm_area_caller(size, VM_IOREMAP, IOREMAP_START,
+				    IOREMAP_END, __builtin_return_address(0));
+	if (!area)
+		return NULL;
+	vaddr = (unsigned long)area->addr;
+	area->phys_addr = phys_addr;
+
+	if (ioremap_page_range(vaddr, vaddr + size, phys_addr, prot)) {
+		free_vm_area(area);
+		return NULL;
+	}
+
+	return (void __iomem *)(vaddr + offset);
+}
+
+#ifndef ioremap_prot
+void __iomem *ioremap_prot(phys_addr_t phys_addr, size_t size,
+			   pgprot_t prot)
+{
+	return generic_ioremap_prot(phys_addr, size, prot);
+}
+EXPORT_SYMBOL(ioremap_prot);
+#endif
+
+void generic_iounmap(volatile void __iomem *addr)
+{
+	void *vaddr = (void *)((unsigned long)addr & PAGE_MASK);
+
+	if (is_ioremap_addr(vaddr))
+		vunmap(vaddr);
+}
+
+#ifndef iounmap
+void iounmap(volatile void __iomem *addr)
+{
+	generic_iounmap(addr);
+}
+EXPORT_SYMBOL(iounmap);
+#endif