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 --- fs/proc/kmsg.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 63 insertions(+) create mode 100644 fs/proc/kmsg.c (limited to 'fs/proc/kmsg.c') diff --git a/fs/proc/kmsg.c b/fs/proc/kmsg.c new file mode 100644 index 000000000..2fc92a13f --- /dev/null +++ b/fs/proc/kmsg.c @@ -0,0 +1,63 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * linux/fs/proc/kmsg.c + * + * Copyright (C) 1992 by Linus Torvalds + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +static int kmsg_open(struct inode * inode, struct file * file) +{ + return do_syslog(SYSLOG_ACTION_OPEN, NULL, 0, SYSLOG_FROM_PROC); +} + +static int kmsg_release(struct inode * inode, struct file * file) +{ + (void) do_syslog(SYSLOG_ACTION_CLOSE, NULL, 0, SYSLOG_FROM_PROC); + return 0; +} + +static ssize_t kmsg_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + if ((file->f_flags & O_NONBLOCK) && + !do_syslog(SYSLOG_ACTION_SIZE_UNREAD, NULL, 0, SYSLOG_FROM_PROC)) + return -EAGAIN; + return do_syslog(SYSLOG_ACTION_READ, buf, count, SYSLOG_FROM_PROC); +} + +static __poll_t kmsg_poll(struct file *file, poll_table *wait) +{ + poll_wait(file, &log_wait, wait); + if (do_syslog(SYSLOG_ACTION_SIZE_UNREAD, NULL, 0, SYSLOG_FROM_PROC)) + return EPOLLIN | EPOLLRDNORM; + return 0; +} + + +static const struct proc_ops kmsg_proc_ops = { + .proc_flags = PROC_ENTRY_PERMANENT, + .proc_read = kmsg_read, + .proc_poll = kmsg_poll, + .proc_open = kmsg_open, + .proc_release = kmsg_release, + .proc_lseek = generic_file_llseek, +}; + +static int __init proc_kmsg_init(void) +{ + proc_create("kmsg", S_IRUSR, NULL, &kmsg_proc_ops); + return 0; +} +fs_initcall(proc_kmsg_init); -- cgit v1.3.1