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, 103 insertions, 0 deletions

diff --git a/arch/mips/tools/elf-entry.c b/arch/mips/tools/elf-entry.c
new file mode 100644
index 000000000..dbd14ff05
--- /dev/null
+++ b/arch/mips/tools/elf-entry.c
@@ -0,0 +1,103 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <byteswap.h>
+#include <elf.h>
+#include <endian.h>
+#include <inttypes.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#ifdef be32toh
+/* If libc provides [bl]e{32,64}toh() then we'll use them */
+#elif BYTE_ORDER == LITTLE_ENDIAN
+# define be32toh(x)	bswap_32(x)
+# define le32toh(x)	(x)
+# define be64toh(x)	bswap_64(x)
+# define le64toh(x)	(x)
+#elif BYTE_ORDER == BIG_ENDIAN
+# define be32toh(x)	(x)
+# define le32toh(x)	bswap_32(x)
+# define be64toh(x)	(x)
+# define le64toh(x)	bswap_64(x)
+#endif
+
+__attribute__((noreturn))
+static void die(const char *msg)
+{
+	fputs(msg, stderr);
+	exit(EXIT_FAILURE);
+}
+
+int main(int argc, const char *argv[])
+{
+	uint64_t entry;
+	size_t nread;
+	FILE *file;
+	union {
+		Elf32_Ehdr ehdr32;
+		Elf64_Ehdr ehdr64;
+	} hdr;
+
+	if (argc != 2)
+		die("Usage: elf-entry <elf-file>\n");
+
+	file = fopen(argv[1], "r");
+	if (!file) {
+		perror("Unable to open input file");
+		return EXIT_FAILURE;
+	}
+
+	nread = fread(&hdr, 1, sizeof(hdr), file);
+	if (nread != sizeof(hdr)) {
+		perror("Unable to read input file");
+		fclose(file);
+		return EXIT_FAILURE;
+	}
+
+	if (memcmp(hdr.ehdr32.e_ident, ELFMAG, SELFMAG)) {
+		fclose(file);
+		die("Input is not an ELF\n");
+	}
+
+	switch (hdr.ehdr32.e_ident[EI_CLASS]) {
+	case ELFCLASS32:
+		switch (hdr.ehdr32.e_ident[EI_DATA]) {
+		case ELFDATA2LSB:
+			entry = le32toh(hdr.ehdr32.e_entry);
+			break;
+		case ELFDATA2MSB:
+			entry = be32toh(hdr.ehdr32.e_entry);
+			break;
+		default:
+			fclose(file);
+			die("Invalid ELF encoding\n");
+		}
+
+		/* Sign extend to form a canonical address */
+		entry = (int64_t)(int32_t)entry;
+		break;
+
+	case ELFCLASS64:
+		switch (hdr.ehdr32.e_ident[EI_DATA]) {
+		case ELFDATA2LSB:
+			entry = le64toh(hdr.ehdr64.e_entry);
+			break;
+		case ELFDATA2MSB:
+			entry = be64toh(hdr.ehdr64.e_entry);
+			break;
+		default:
+			fclose(file);
+			die("Invalid ELF encoding\n");
+		}
+		break;
+
+	default:
+		fclose(file);
+		die("Invalid ELF class\n");
+	}
+
+	printf("0x%016" PRIx64 "\n", entry);
+	fclose(file);
+	return EXIT_SUCCESS;
+}