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 --- net/ethtool/phc_vclocks.c | 94 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 94 insertions(+) create mode 100644 net/ethtool/phc_vclocks.c (limited to 'net/ethtool/phc_vclocks.c') diff --git a/net/ethtool/phc_vclocks.c b/net/ethtool/phc_vclocks.c new file mode 100644 index 000000000..15146e38a --- /dev/null +++ b/net/ethtool/phc_vclocks.c @@ -0,0 +1,94 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright 2021 NXP + */ +#include "common.h" +#include "netlink.h" + +struct phc_vclocks_req_info { + struct ethnl_req_info base; +}; + +struct phc_vclocks_reply_data { + struct ethnl_reply_data base; + int num; + int *index; +}; + +#define PHC_VCLOCKS_REPDATA(__reply_base) \ + container_of(__reply_base, struct phc_vclocks_reply_data, base) + +const struct nla_policy ethnl_phc_vclocks_get_policy[] = { + [ETHTOOL_A_PHC_VCLOCKS_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy), +}; + +static int phc_vclocks_prepare_data(const struct ethnl_req_info *req_base, + struct ethnl_reply_data *reply_base, + const struct genl_info *info) +{ + struct phc_vclocks_reply_data *data = PHC_VCLOCKS_REPDATA(reply_base); + struct net_device *dev = reply_base->dev; + int ret; + + ret = ethnl_ops_begin(dev); + if (ret < 0) + return ret; + data->num = ethtool_get_phc_vclocks(dev, &data->index); + ethnl_ops_complete(dev); + + return ret; +} + +static int phc_vclocks_reply_size(const struct ethnl_req_info *req_base, + const struct ethnl_reply_data *reply_base) +{ + const struct phc_vclocks_reply_data *data = + PHC_VCLOCKS_REPDATA(reply_base); + int len = 0; + + if (data->num > 0) { + len += nla_total_size(sizeof(u32)); + len += nla_total_size(sizeof(s32) * data->num); + } + + return len; +} + +static int phc_vclocks_fill_reply(struct sk_buff *skb, + const struct ethnl_req_info *req_base, + const struct ethnl_reply_data *reply_base) +{ + const struct phc_vclocks_reply_data *data = + PHC_VCLOCKS_REPDATA(reply_base); + + if (data->num <= 0) + return 0; + + if (nla_put_u32(skb, ETHTOOL_A_PHC_VCLOCKS_NUM, data->num) || + nla_put(skb, ETHTOOL_A_PHC_VCLOCKS_INDEX, + sizeof(s32) * data->num, data->index)) + return -EMSGSIZE; + + return 0; +} + +static void phc_vclocks_cleanup_data(struct ethnl_reply_data *reply_base) +{ + const struct phc_vclocks_reply_data *data = + PHC_VCLOCKS_REPDATA(reply_base); + + kfree(data->index); +} + +const struct ethnl_request_ops ethnl_phc_vclocks_request_ops = { + .request_cmd = ETHTOOL_MSG_PHC_VCLOCKS_GET, + .reply_cmd = ETHTOOL_MSG_PHC_VCLOCKS_GET_REPLY, + .hdr_attr = ETHTOOL_A_PHC_VCLOCKS_HEADER, + .req_info_size = sizeof(struct phc_vclocks_req_info), + .reply_data_size = sizeof(struct phc_vclocks_reply_data), + + .prepare_data = phc_vclocks_prepare_data, + .reply_size = phc_vclocks_reply_size, + .fill_reply = phc_vclocks_fill_reply, + .cleanup_data = phc_vclocks_cleanup_data, +}; -- cgit v1.3.1