linux-kernel - [PATCH v2 08/11] sched_ext: Add an event, BYPASS

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20250126101614.232388-9-changwoo@igalia.com>
Date: Sun, 26 Jan 2025 19:16:11 +0900
From: Changwoo Min <changwoo@...lia.com>
To: tj@...nel.org,
	void@...ifault.com,
	arighi@...dia.com
Cc: kernel-dev@...lia.com,
	linux-kernel@...r.kernel.org,
	Changwoo Min <changwoo@...lia.com>
Subject: [PATCH v2 08/11] sched_ext: Add an event, BYPASS_DISPATCH

Add a core event, BYPASS_DISPATCH, which represents how many tasks
have been dispatched in the bypass mode.

__scx_add_event() is used since the caller holds an rq lock,
so the preemption has already been disabled.

Signed-off-by: Changwoo Min <changwoo@...lia.com>
---
 kernel/sched/ext.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index bb363809e484..9a680774c1fc 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -1486,6 +1486,11 @@ struct scx_event_stats {
 	 */
 	u64		ENQ_SKIP_EXITING;
 
+	/*
+	 * The number of tasks dispatched in the bypassing mode.
+	 */
+	u64		BYPASS_DISPATCH;
+
 	/*
 	 * The number of times the bypassing mode has been activated.
 	 */
@@ -2941,6 +2946,8 @@ static int balance_one(struct rq *rq, struct task_struct *prev)
 	return false;
 
 has_tasks:
+	if (scx_rq_bypassing(rq))
+		__scx_add_event(BYPASS_DISPATCH, 1);
 	rq->scx.flags &= ~SCX_RQ_IN_BALANCE;
 	return true;
 }
@@ -5524,6 +5531,7 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len)
 	scx_dump_event(s, &events, DISPATCH_LOCAL_DSQ_OFFLINE);
 	scx_dump_event(s, &events, DISPATCH_KEEP_LAST);
 	scx_dump_event(s, &events, ENQ_SKIP_EXITING);
+	scx_dump_event(s, &events, BYPASS_DISPATCH);
 	scx_dump_event(s, &events, BYPASS_ACTIVATE);
 
 	if (seq_buf_has_overflowed(&s) && dump_len >= sizeof(trunc_marker))
@@ -7856,6 +7864,7 @@ __bpf_kfunc void scx_bpf_event_stats(struct scx_event_stats *events,
 		scx_agg_event(&e_sys, e_cpu, SELECT_CPU_FALLBACK);
 		scx_agg_event(&e_sys, e_cpu, DISPATCH_LOCAL_DSQ_OFFLINE);
 		scx_agg_event(&e_sys, e_cpu, DISPATCH_KEEP_LAST);
+		scx_agg_event(&e_sys, e_cpu, BYPASS_DISPATCH);
 		scx_agg_event(&e_sys, e_cpu, BYPASS_ACTIVATE);
 	}
 
-- 
2.48.1