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 --- include/asm-generic/atomic.h | 133 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 133 insertions(+) create mode 100644 include/asm-generic/atomic.h (limited to 'include/asm-generic/atomic.h') diff --git a/include/asm-generic/atomic.h b/include/asm-generic/atomic.h new file mode 100644 index 000000000..22142c71d --- /dev/null +++ b/include/asm-generic/atomic.h @@ -0,0 +1,133 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Generic C implementation of atomic counter operations. Do not include in + * machine independent code. + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ +#ifndef __ASM_GENERIC_ATOMIC_H +#define __ASM_GENERIC_ATOMIC_H + +#include +#include + +#ifdef CONFIG_SMP + +/* we can build all atomic primitives from cmpxchg */ + +#define ATOMIC_OP(op, c_op) \ +static inline void generic_atomic_##op(int i, atomic_t *v) \ +{ \ + int c, old; \ + \ + c = v->counter; \ + while ((old = arch_cmpxchg(&v->counter, c, c c_op i)) != c) \ + c = old; \ +} + +#define ATOMIC_OP_RETURN(op, c_op) \ +static inline int generic_atomic_##op##_return(int i, atomic_t *v) \ +{ \ + int c, old; \ + \ + c = v->counter; \ + while ((old = arch_cmpxchg(&v->counter, c, c c_op i)) != c) \ + c = old; \ + \ + return c c_op i; \ +} + +#define ATOMIC_FETCH_OP(op, c_op) \ +static inline int generic_atomic_fetch_##op(int i, atomic_t *v) \ +{ \ + int c, old; \ + \ + c = v->counter; \ + while ((old = arch_cmpxchg(&v->counter, c, c c_op i)) != c) \ + c = old; \ + \ + return c; \ +} + +#else + +#include + +#define ATOMIC_OP(op, c_op) \ +static inline void generic_atomic_##op(int i, atomic_t *v) \ +{ \ + unsigned long flags; \ + \ + raw_local_irq_save(flags); \ + v->counter = v->counter c_op i; \ + raw_local_irq_restore(flags); \ +} + +#define ATOMIC_OP_RETURN(op, c_op) \ +static inline int generic_atomic_##op##_return(int i, atomic_t *v) \ +{ \ + unsigned long flags; \ + int ret; \ + \ + raw_local_irq_save(flags); \ + ret = (v->counter = v->counter c_op i); \ + raw_local_irq_restore(flags); \ + \ + return ret; \ +} + +#define ATOMIC_FETCH_OP(op, c_op) \ +static inline int generic_atomic_fetch_##op(int i, atomic_t *v) \ +{ \ + unsigned long flags; \ + int ret; \ + \ + raw_local_irq_save(flags); \ + ret = v->counter; \ + v->counter = v->counter c_op i; \ + raw_local_irq_restore(flags); \ + \ + return ret; \ +} + +#endif /* CONFIG_SMP */ + +ATOMIC_OP_RETURN(add, +) +ATOMIC_OP_RETURN(sub, -) + +ATOMIC_FETCH_OP(add, +) +ATOMIC_FETCH_OP(sub, -) +ATOMIC_FETCH_OP(and, &) +ATOMIC_FETCH_OP(or, |) +ATOMIC_FETCH_OP(xor, ^) + +ATOMIC_OP(add, +) +ATOMIC_OP(sub, -) +ATOMIC_OP(and, &) +ATOMIC_OP(or, |) +ATOMIC_OP(xor, ^) + +#undef ATOMIC_FETCH_OP +#undef ATOMIC_OP_RETURN +#undef ATOMIC_OP + +#define arch_atomic_add_return generic_atomic_add_return +#define arch_atomic_sub_return generic_atomic_sub_return + +#define arch_atomic_fetch_add generic_atomic_fetch_add +#define arch_atomic_fetch_sub generic_atomic_fetch_sub +#define arch_atomic_fetch_and generic_atomic_fetch_and +#define arch_atomic_fetch_or generic_atomic_fetch_or +#define arch_atomic_fetch_xor generic_atomic_fetch_xor + +#define arch_atomic_add generic_atomic_add +#define arch_atomic_sub generic_atomic_sub +#define arch_atomic_and generic_atomic_and +#define arch_atomic_or generic_atomic_or +#define arch_atomic_xor generic_atomic_xor + +#define arch_atomic_read(v) READ_ONCE((v)->counter) +#define arch_atomic_set(v, i) WRITE_ONCE(((v)->counter), (i)) + +#endif /* __ASM_GENERIC_ATOMIC_H */ -- cgit v1.3.1