From 0d6526f82c3cdefcca47f73f5fc08dc6f335eac6 Mon Sep 17 00:00:00 2001 From: Tim Chen Date: Mon, 21 Sep 2026 17:37:22 -0700 Subject: sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling bug 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 Suggested-by: Chen Yu Signed-off-by: Tim Chen Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Kayra Cizmeci Cc: # v7.2.x Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com --- block/blk-mq-debugfs.h | 91 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 91 insertions(+) create mode 100644 block/blk-mq-debugfs.h (limited to 'block/blk-mq-debugfs.h') diff --git a/block/blk-mq-debugfs.h b/block/blk-mq-debugfs.h new file mode 100644 index 000000000..49bb1aaa8 --- /dev/null +++ b/block/blk-mq-debugfs.h @@ -0,0 +1,91 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef INT_BLK_MQ_DEBUGFS_H +#define INT_BLK_MQ_DEBUGFS_H + +#ifdef CONFIG_BLK_DEBUG_FS + +#include + +struct blk_mq_hw_ctx; + +struct blk_mq_debugfs_attr { + const char *name; + umode_t mode; + int (*show)(void *, struct seq_file *); + ssize_t (*write)(void *, const char __user *, size_t, loff_t *); + /* Set either .show or .seq_ops. */ + const struct seq_operations *seq_ops; +}; + +int __blk_mq_debugfs_rq_show(struct seq_file *m, struct request *rq); +int blk_mq_debugfs_rq_show(struct seq_file *m, void *v); + +void blk_mq_debugfs_register(struct request_queue *q); +void blk_mq_debugfs_register_hctx(struct request_queue *q, + struct blk_mq_hw_ctx *hctx); +void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx); +void blk_mq_debugfs_register_hctxs(struct request_queue *q); +void blk_mq_debugfs_unregister_hctxs(struct request_queue *q); + +void blk_mq_debugfs_register_sched(struct request_queue *q); +void blk_mq_debugfs_unregister_sched(struct request_queue *q); +void blk_mq_debugfs_register_sched_hctx(struct request_queue *q, + struct blk_mq_hw_ctx *hctx); +void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hctx); + +void blk_mq_debugfs_register_rq_qos(struct request_queue *q); +#else +static inline void blk_mq_debugfs_register(struct request_queue *q) +{ +} + +static inline void blk_mq_debugfs_register_hctx(struct request_queue *q, + struct blk_mq_hw_ctx *hctx) +{ +} + +static inline void blk_mq_debugfs_unregister_hctx(struct blk_mq_hw_ctx *hctx) +{ +} + +static inline void blk_mq_debugfs_register_hctxs(struct request_queue *q) +{ +} + +static inline void blk_mq_debugfs_unregister_hctxs(struct request_queue *q) +{ +} + +static inline void blk_mq_debugfs_register_sched(struct request_queue *q) +{ +} + +static inline void blk_mq_debugfs_unregister_sched(struct request_queue *q) +{ +} + +static inline void blk_mq_debugfs_register_sched_hctx(struct request_queue *q, + struct blk_mq_hw_ctx *hctx) +{ +} + +static inline void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hctx) +{ +} + +static inline void blk_mq_debugfs_register_rq_qos(struct request_queue *q) +{ +} + +#endif + +#if defined(CONFIG_BLK_DEV_ZONED) && defined(CONFIG_BLK_DEBUG_FS) +int queue_zone_wplugs_show(void *data, struct seq_file *m); +#else +static inline int queue_zone_wplugs_show(void *data, struct seq_file *m) +{ + return 0; +} +#endif + +#endif -- cgit v1.3.1