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/kvm/iodev.h | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 58 insertions(+) create mode 100644 include/kvm/iodev.h (limited to 'include/kvm/iodev.h') diff --git a/include/kvm/iodev.h b/include/kvm/iodev.h new file mode 100644 index 000000000..56619e332 --- /dev/null +++ b/include/kvm/iodev.h @@ -0,0 +1,58 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#ifndef __KVM_IODEV_H__ +#define __KVM_IODEV_H__ + +#include +#include + +struct kvm_io_device; +struct kvm_vcpu; + +/** + * kvm_io_device_ops are called under kvm slots_lock. + * read and write handlers return 0 if the transaction has been handled, + * or non-zero to have it passed to the next device. + **/ +struct kvm_io_device_ops { + int (*read)(struct kvm_vcpu *vcpu, + struct kvm_io_device *this, + gpa_t addr, + int len, + void *val); + int (*write)(struct kvm_vcpu *vcpu, + struct kvm_io_device *this, + gpa_t addr, + int len, + const void *val); + void (*destructor)(struct kvm_io_device *this); +}; + + +struct kvm_io_device { + const struct kvm_io_device_ops *ops; +}; + +static inline void kvm_iodevice_init(struct kvm_io_device *dev, + const struct kvm_io_device_ops *ops) +{ + dev->ops = ops; +} + +static inline int kvm_iodevice_read(struct kvm_vcpu *vcpu, + struct kvm_io_device *dev, gpa_t addr, + int l, void *v) +{ + return dev->ops->read ? dev->ops->read(vcpu, dev, addr, l, v) + : -EOPNOTSUPP; +} + +static inline int kvm_iodevice_write(struct kvm_vcpu *vcpu, + struct kvm_io_device *dev, gpa_t addr, + int l, const void *v) +{ + return dev->ops->write ? dev->ops->write(vcpu, dev, addr, l, v) + : -EOPNOTSUPP; +} + +#endif /* __KVM_IODEV_H__ */ -- cgit v1.3.1