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 --- drivers/md/dm-zero.c | 81 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 81 insertions(+) create mode 100644 drivers/md/dm-zero.c (limited to 'drivers/md/dm-zero.c') diff --git a/drivers/md/dm-zero.c b/drivers/md/dm-zero.c new file mode 100644 index 000000000..9a0bb623e --- /dev/null +++ b/drivers/md/dm-zero.c @@ -0,0 +1,81 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2003 Jana Saout + * + * This file is released under the GPL. + */ + +#include + +#include +#include +#include + +#define DM_MSG_PREFIX "zero" + +/* + * Construct a dummy mapping that only returns zeros + */ +static int zero_ctr(struct dm_target *ti, unsigned int argc, char **argv) +{ + if (argc != 0) { + ti->error = "No arguments required"; + return -EINVAL; + } + + /* + * Silently drop discards, avoiding -EOPNOTSUPP. + */ + ti->num_discard_bios = 1; + ti->discards_supported = true; + + return 0; +} + +/* + * Return zeros only on reads + */ +static int zero_map(struct dm_target *ti, struct bio *bio) +{ + switch (bio_op(bio)) { + case REQ_OP_READ: + if (bio->bi_opf & REQ_RAHEAD) { + /* readahead of null bytes only wastes buffer cache */ + return DM_MAPIO_KILL; + } + zero_fill_bio(bio); + break; + case REQ_OP_WRITE: + case REQ_OP_DISCARD: + /* writes get silently dropped */ + break; + default: + return DM_MAPIO_KILL; + } + + bio_endio(bio); + + /* accepted bio, don't make new request */ + return DM_MAPIO_SUBMITTED; +} + +static void zero_io_hints(struct dm_target *ti, struct queue_limits *limits) +{ + limits->max_hw_discard_sectors = UINT_MAX; + limits->discard_granularity = 512; +} + +static struct target_type zero_target = { + .name = "zero", + .version = {1, 2, 0}, + .features = DM_TARGET_NOWAIT, + .module = THIS_MODULE, + .ctr = zero_ctr, + .map = zero_map, + .io_hints = zero_io_hints, +}; +module_dm(zero); + +MODULE_AUTHOR("Jana Saout "); +MODULE_DESCRIPTION(DM_NAME " dummy target returning zeros"); +MODULE_LICENSE("GPL"); -- cgit v1.3.1