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, 191 insertions, 0 deletions

diff --git a/include/uapi/rdma/mlx4-abi.h b/include/uapi/rdma/mlx4-abi.h
new file mode 100644
index 000000000..f74557528
--- /dev/null
+++ b/include/uapi/rdma/mlx4-abi.h
@@ -0,0 +1,191 @@
+/* SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR Linux-OpenIB) */
+/*
+ * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef MLX4_ABI_USER_H
+#define MLX4_ABI_USER_H
+
+#include <linux/types.h>
+
+/*
+ * Increment this value if any changes that break userspace ABI
+ * compatibility are made.
+ */
+
+#define MLX4_IB_UVERBS_NO_DEV_CAPS_ABI_VERSION	3
+#define MLX4_IB_UVERBS_ABI_VERSION		4
+
+/*
+ * Make sure that all structs defined in this file remain laid out so
+ * that they pack the same way on 32-bit and 64-bit architectures (to
+ * avoid incompatibility between 32-bit userspace and 64-bit kernels).
+ * In particular do not use pointer types -- pass pointers in __u64
+ * instead.
+ */
+
+struct mlx4_ib_alloc_ucontext_resp_v3 {
+	__u32	qp_tab_size;
+	__u16	bf_reg_size;
+	__u16	bf_regs_per_page;
+};
+
+enum {
+	MLX4_USER_DEV_CAP_LARGE_CQE	= 1L << 0,
+};
+
+struct mlx4_ib_alloc_ucontext_resp {
+	__u32	dev_caps;
+	__u32	qp_tab_size;
+	__u16	bf_reg_size;
+	__u16	bf_regs_per_page;
+	__u32	cqe_size;
+};
+
+struct mlx4_ib_alloc_pd_resp {
+	__u32	pdn;
+	__u32	reserved;
+};
+
+struct mlx4_ib_create_cq {
+	__aligned_u64 buf_addr;
+	__aligned_u64 db_addr;
+};
+
+struct mlx4_ib_create_cq_resp {
+	__u32	cqn;
+	__u32	reserved;
+};
+
+struct mlx4_ib_resize_cq {
+	__aligned_u64 buf_addr;
+};
+
+struct mlx4_ib_create_srq {
+	__aligned_u64 buf_addr;
+	__aligned_u64 db_addr;
+};
+
+struct mlx4_ib_create_srq_resp {
+	__u32	srqn;
+	__u32	reserved;
+};
+
+struct mlx4_ib_create_qp_rss {
+	__aligned_u64 rx_hash_fields_mask; /* Use  enum mlx4_ib_rx_hash_fields */
+	__u8    rx_hash_function; /* Use enum mlx4_ib_rx_hash_function_flags */
+	__u8    reserved[7];
+	__u8    rx_hash_key[40];
+	__u32   comp_mask;
+	__u32   reserved1;
+};
+
+struct mlx4_ib_create_qp {
+	__aligned_u64 buf_addr;
+	__aligned_u64 db_addr;
+	__u8	log_sq_bb_count;
+	__u8	log_sq_stride;
+	__u8	sq_no_prefetch;
+	__u8	reserved;
+	__u32	inl_recv_sz;
+};
+
+struct mlx4_ib_create_wq {
+	__aligned_u64 buf_addr;
+	__aligned_u64 db_addr;
+	__u8	log_range_size;
+	__u8	reserved[3];
+	__u32   comp_mask;
+};
+
+struct mlx4_ib_modify_wq {
+	__u32	comp_mask;
+	__u32	reserved;
+};
+
+struct mlx4_ib_create_rwq_ind_tbl_resp {
+	__u32	response_length;
+	__u32	reserved;
+};
+
+/* RX Hash function flags */
+enum mlx4_ib_rx_hash_function_flags {
+	MLX4_IB_RX_HASH_FUNC_TOEPLITZ	= 1 << 0,
+};
+
+/*
+ * RX Hash flags, these flags allows to set which incoming packet's field should
+ * participates in RX Hash. Each flag represent certain packet's field,
+ * when the flag is set the field that is represented by the flag will
+ * participate in RX Hash calculation.
+ */
+enum mlx4_ib_rx_hash_fields {
+	MLX4_IB_RX_HASH_SRC_IPV4	= 1 << 0,
+	MLX4_IB_RX_HASH_DST_IPV4	= 1 << 1,
+	MLX4_IB_RX_HASH_SRC_IPV6	= 1 << 2,
+	MLX4_IB_RX_HASH_DST_IPV6	= 1 << 3,
+	MLX4_IB_RX_HASH_SRC_PORT_TCP	= 1 << 4,
+	MLX4_IB_RX_HASH_DST_PORT_TCP	= 1 << 5,
+	MLX4_IB_RX_HASH_SRC_PORT_UDP	= 1 << 6,
+	MLX4_IB_RX_HASH_DST_PORT_UDP	= 1 << 7,
+	MLX4_IB_RX_HASH_INNER		= 1ULL << 31,
+};
+
+struct mlx4_ib_rss_caps {
+	__aligned_u64 rx_hash_fields_mask; /* enum mlx4_ib_rx_hash_fields */
+	__u8 rx_hash_function; /* enum mlx4_ib_rx_hash_function_flags */
+	__u8 reserved[7];
+};
+
+enum query_device_resp_mask {
+	MLX4_IB_QUERY_DEV_RESP_MASK_CORE_CLOCK_OFFSET = 1UL << 0,
+};
+
+struct mlx4_ib_tso_caps {
+	__u32 max_tso; /* Maximum tso payload size in bytes */
+	/* Corresponding bit will be set if qp type from
+	 * 'enum ib_qp_type' is supported.
+	 */
+	__u32 supported_qpts;
+};
+
+struct mlx4_uverbs_ex_query_device_resp {
+	__u32			comp_mask;
+	__u32			response_length;
+	__aligned_u64		hca_core_clock_offset;
+	__u32			max_inl_recv_sz;
+	__u32			reserved;
+	struct mlx4_ib_rss_caps	rss_caps;
+	struct mlx4_ib_tso_caps tso_caps;
+};
+
+#endif /* MLX4_ABI_USER_H */