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, 90 insertions, 0 deletions

diff --git a/io_uring/nop.c b/io_uring/nop.c
new file mode 100644
index 000000000..60ab19604
--- /dev/null
+++ b/io_uring/nop.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/io_uring.h>
+
+#include <uapi/linux/io_uring.h>
+
+#include "io_uring.h"
+#include "rsrc.h"
+#include "nop.h"
+
+struct io_nop {
+	struct file     *file;
+	int             result;
+	int		fd;
+	unsigned int	flags;
+	__u64		extra1;
+	__u64		extra2;
+};
+
+#define NOP_FLAGS	(IORING_NOP_INJECT_RESULT | IORING_NOP_FIXED_FILE | \
+			 IORING_NOP_FIXED_BUFFER | IORING_NOP_FILE | \
+			 IORING_NOP_TW | IORING_NOP_CQE32)
+
+int io_nop_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_nop *nop = io_kiocb_to_cmd(req, struct io_nop);
+
+	nop->flags = READ_ONCE(sqe->nop_flags);
+	if (nop->flags & ~NOP_FLAGS)
+		return -EINVAL;
+
+	if (nop->flags & IORING_NOP_INJECT_RESULT)
+		nop->result = READ_ONCE(sqe->len);
+	else
+		nop->result = 0;
+	if (nop->flags & IORING_NOP_FILE)
+		nop->fd = READ_ONCE(sqe->fd);
+	else
+		nop->fd = -1;
+	if (nop->flags & IORING_NOP_FIXED_FILE)
+		req->flags |= REQ_F_FIXED_FILE;
+	if (nop->flags & IORING_NOP_FIXED_BUFFER)
+		req->buf_index = READ_ONCE(sqe->buf_index);
+	if (nop->flags & IORING_NOP_CQE32) {
+		struct io_ring_ctx *ctx = req->ctx;
+
+		if (!(ctx->flags & (IORING_SETUP_CQE32|IORING_SETUP_CQE_MIXED)))
+			return -EINVAL;
+		nop->extra1 = READ_ONCE(sqe->off);
+		nop->extra2 = READ_ONCE(sqe->addr);
+	}
+	return 0;
+}
+
+int io_nop(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_nop *nop = io_kiocb_to_cmd(req, struct io_nop);
+	int ret = nop->result;
+
+	if (nop->flags & IORING_NOP_FILE) {
+		if (req->flags & REQ_F_FIXED_FILE)
+			req->file = io_file_get_fixed(req, nop->fd, issue_flags);
+		else
+			req->file = io_file_get_normal(req, nop->fd);
+		if (!req->file) {
+			ret = -EBADF;
+			goto done;
+		}
+	}
+	if (nop->flags & IORING_NOP_FIXED_BUFFER) {
+		if (!io_find_buf_node(req, issue_flags))
+			ret = -EFAULT;
+	}
+done:
+	if (ret < 0)
+		req_set_fail(req);
+	if (nop->flags & IORING_NOP_CQE32)
+		io_req_set_res32(req, ret, 0, nop->extra1, nop->extra2);
+	else
+		io_req_set_res(req, ret, 0);
+	if (nop->flags & IORING_NOP_TW) {
+		req->io_task_work.func = io_req_task_complete;
+		io_req_task_work_add(req);
+		return IOU_ISSUE_SKIP_COMPLETE;
+	}
+	return IOU_COMPLETE;
+}