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/crypto/utils.h | 73 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) create mode 100644 include/crypto/utils.h (limited to 'include/crypto/utils.h') diff --git a/include/crypto/utils.h b/include/crypto/utils.h new file mode 100644 index 000000000..2594f4577 --- /dev/null +++ b/include/crypto/utils.h @@ -0,0 +1,73 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Cryptographic utilities + * + * Copyright (c) 2023 Herbert Xu + */ +#ifndef _CRYPTO_UTILS_H +#define _CRYPTO_UTILS_H + +#include +#include +#include + +void __crypto_xor(u8 *dst, const u8 *src1, const u8 *src2, unsigned int size); + +static inline void crypto_xor(u8 *dst, const u8 *src, unsigned int size) +{ + if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && + __builtin_constant_p(size) && + (size % sizeof(unsigned long)) == 0) { + unsigned long *d = (unsigned long *)dst; + unsigned long *s = (unsigned long *)src; + unsigned long l; + + while (size > 0) { + l = get_unaligned(d) ^ get_unaligned(s++); + put_unaligned(l, d++); + size -= sizeof(unsigned long); + } + } else { + __crypto_xor(dst, dst, src, size); + } +} + +static inline void crypto_xor_cpy(u8 *dst, const u8 *src1, const u8 *src2, + unsigned int size) +{ + if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && + __builtin_constant_p(size) && + (size % sizeof(unsigned long)) == 0) { + unsigned long *d = (unsigned long *)dst; + unsigned long *s1 = (unsigned long *)src1; + unsigned long *s2 = (unsigned long *)src2; + unsigned long l; + + while (size > 0) { + l = get_unaligned(s1++) ^ get_unaligned(s2++); + put_unaligned(l, d++); + size -= sizeof(unsigned long); + } + } else { + __crypto_xor(dst, src1, src2, size); + } +} + +noinline unsigned long __crypto_memneq(const void *a, const void *b, size_t size); + +/** + * crypto_memneq - Compare two areas of memory without leaking + * timing information. + * + * @a: One area of memory + * @b: Another area of memory + * @size: The size of the area. + * + * Returns 0 when data is equal, 1 otherwise. + */ +static inline int crypto_memneq(const void *a, const void *b, size_t size) +{ + return __crypto_memneq(a, b, size) != 0UL ? 1 : 0; +} + +#endif /* _CRYPTO_UTILS_H */ -- cgit v1.3.1