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, 141 insertions, 0 deletions

diff --git a/io_uring/kbuf.h b/io_uring/kbuf.h
new file mode 100644
index 000000000..401773e1e
--- /dev/null
+++ b/io_uring/kbuf.h
@@ -0,0 +1,141 @@
+// SPDX-License-Identifier: GPL-2.0
+#ifndef IOU_KBUF_H
+#define IOU_KBUF_H
+
+#include <uapi/linux/io_uring.h>
+#include <linux/io_uring_types.h>
+
+enum {
+	/* ring mapped provided buffers */
+	IOBL_BUF_RING	= 1,
+	/* buffers are consumed incrementally rather than always fully */
+	IOBL_INC	= 2,
+};
+
+struct io_buffer_list {
+	/*
+	 * If the IOBL_BUF_RING flag is set, then buf_ring is used. If not, then
+	 * these are classic provided buffers and ->buf_list is used.
+	 */
+	union {
+		struct list_head buf_list;
+		struct io_uring_buf_ring *buf_ring;
+	};
+	/* count of classic/legacy buffers in buffer list */
+	int nbufs;
+
+	__u16 bgid;
+
+	/* below is for ring provided buffers */
+	__u16 head;
+	__u16 mask;
+
+	__u16 flags;
+
+	/*
+	 * minimum required amount to be left to reuse an incrementally
+	 * consumed buffer. If less than this is left at consumption time,
+	 * buffer is done and head is incremented to the next buffer.
+	 */
+	__u32 min_left_sub_one;
+
+	struct io_mapped_region region;
+};
+
+struct io_buffer {
+	struct list_head list;
+	__u64 addr;
+	__u32 len;
+	__u16 bid;
+	__u16 bgid;
+};
+
+enum {
+	/* can alloc a bigger vec */
+	KBUF_MODE_EXPAND	= 1,
+	/* if bigger vec allocated, free old one */
+	KBUF_MODE_FREE		= 2,
+};
+
+struct buf_sel_arg {
+	struct iovec *iovs;
+	size_t out_len;
+	size_t max_len;
+	unsigned short nr_iovs;
+	unsigned short mode;
+	unsigned short buf_group;
+	unsigned short partial_map;
+};
+
+struct io_br_sel io_buffer_select(struct io_kiocb *req, size_t *len,
+				  unsigned buf_group, unsigned int issue_flags);
+int io_buffers_select(struct io_kiocb *req, struct buf_sel_arg *arg,
+		      struct io_br_sel *sel, unsigned int issue_flags);
+int io_buffers_peek(struct io_kiocb *req, struct buf_sel_arg *arg,
+		    struct io_br_sel *sel);
+void io_destroy_buffers(struct io_ring_ctx *ctx);
+
+int io_remove_buffers_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
+int io_provide_buffers_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
+int io_manage_buffers_legacy(struct io_kiocb *req, unsigned int issue_flags);
+
+int io_register_pbuf_ring(struct io_ring_ctx *ctx, void __user *arg);
+int io_unregister_pbuf_ring(struct io_ring_ctx *ctx, void __user *arg);
+int io_register_pbuf_status(struct io_ring_ctx *ctx, void __user *arg);
+
+bool io_kbuf_recycle_legacy(struct io_kiocb *req, unsigned issue_flags);
+void io_kbuf_drop_legacy(struct io_kiocb *req);
+
+unsigned int __io_put_kbufs(struct io_kiocb *req, struct io_buffer_list *bl,
+			    int len, int nbufs);
+bool io_kbuf_commit(struct io_kiocb *req,
+		    struct io_buffer_list *bl, int len, int nr);
+
+struct io_mapped_region *io_pbuf_get_region(struct io_ring_ctx *ctx,
+					    unsigned int bgid);
+
+static inline bool io_kbuf_recycle_ring(struct io_kiocb *req,
+					struct io_buffer_list *bl)
+{
+	if (bl) {
+		req->flags &= ~(REQ_F_BUFFER_RING|REQ_F_BUFFERS_COMMIT);
+		return true;
+	}
+	return false;
+}
+
+static inline bool io_do_buffer_select(struct io_kiocb *req)
+{
+	if (!(req->flags & REQ_F_BUFFER_SELECT))
+		return false;
+	return !(req->flags & (REQ_F_BUFFER_SELECTED|REQ_F_BUFFER_RING));
+}
+
+static inline bool io_kbuf_recycle(struct io_kiocb *req, struct io_buffer_list *bl,
+				   unsigned issue_flags)
+{
+	if (req->flags & REQ_F_BL_NO_RECYCLE)
+		return false;
+	if (req->flags & REQ_F_BUFFER_RING)
+		return io_kbuf_recycle_ring(req, bl);
+	if (req->flags & REQ_F_BUFFER_SELECTED)
+		return io_kbuf_recycle_legacy(req, issue_flags);
+	return false;
+}
+
+static inline unsigned int io_put_kbuf(struct io_kiocb *req, int len,
+				       struct io_buffer_list *bl)
+{
+	if (!(req->flags & (REQ_F_BUFFER_RING | REQ_F_BUFFER_SELECTED)))
+		return 0;
+	return __io_put_kbufs(req, bl, len, 1);
+}
+
+static inline unsigned int io_put_kbufs(struct io_kiocb *req, int len,
+					struct io_buffer_list *bl, int nbufs)
+{
+	if (!(req->flags & (REQ_F_BUFFER_RING | REQ_F_BUFFER_SELECTED)))
+		return 0;
+	return __io_put_kbufs(req, bl, len, nbufs);
+}
+#endif