public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: David Carlier <devnexen@gmail.com>
To: Tejun Heo <tj@kernel.org>, David Vernet <void@manifault.com>
Cc: linux-kernel@vger.kernel.org, David Carlier <devnexen@gmail.com>
Subject: [PATCH] tools/sched_ext: scx_userland: fix data races on shared counters
Date: Wed, 18 Feb 2026 19:22:23 +0000	[thread overview]
Message-ID: <20260218192223.8665-1-devnexen@gmail.com> (raw)

The stats thread reads nr_vruntime_enqueues, nr_vruntime_dispatches,
nr_vruntime_failed, and nr_curr_enqueued concurrently with the main
thread writing them, with no synchronization.

Use __atomic builtins with relaxed ordering for all accesses to these
counters to eliminate the data races.

Only display accuracy is affected, not scheduling correctness.

Signed-off-by: David Carlier <devnexen@gmail.com>
---
 tools/sched_ext/scx_userland.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/tools/sched_ext/scx_userland.c b/tools/sched_ext/scx_userland.c
index 504a80824f5c..3f2aba658b4a 100644
--- a/tools/sched_ext/scx_userland.c
+++ b/tools/sched_ext/scx_userland.c
@@ -157,9 +157,9 @@ static int dispatch_task(__s32 pid)
 
 	err = bpf_map_update_elem(dispatched_fd, NULL, &pid, 0);
 	if (err) {
-		nr_vruntime_failed++;
+		__atomic_add_fetch(&nr_vruntime_failed, 1, __ATOMIC_RELAXED);
 	} else {
-		nr_vruntime_dispatches++;
+		__atomic_add_fetch(&nr_vruntime_dispatches, 1, __ATOMIC_RELAXED);
 	}
 
 	return err;
@@ -202,8 +202,8 @@ static int vruntime_enqueue(const struct scx_userland_enqueued_task *bpf_task)
 		return ENOENT;
 
 	update_enqueued(curr, bpf_task);
-	nr_vruntime_enqueues++;
-	nr_curr_enqueued++;
+	__atomic_add_fetch(&nr_vruntime_enqueues, 1, __ATOMIC_RELAXED);
+	__atomic_add_fetch(&nr_curr_enqueued, 1, __ATOMIC_RELAXED);
 
 	/*
 	 * Enqueue the task in a vruntime-sorted list. A more optimal data
@@ -279,9 +279,9 @@ static void dispatch_batch(void)
 			LIST_INSERT_HEAD(&vruntime_head, task, entries);
 			break;
 		}
-		nr_curr_enqueued--;
+		__atomic_sub_fetch(&nr_curr_enqueued, 1, __ATOMIC_RELAXED);
 	}
-	skel->bss->nr_scheduled = nr_curr_enqueued;
+	skel->bss->nr_scheduled = __atomic_load_n(&nr_curr_enqueued, __ATOMIC_RELAXED);
 }
 
 static void *run_stats_printer(void *arg)
@@ -306,9 +306,9 @@ static void *run_stats_printer(void *arg)
 		printf("|-----------------------|\n");
 		printf("| VRUNTIME / USER       |\n");
 		printf("|-----------------------|\n");
-		printf("|  enq:      %10llu |\n", nr_vruntime_enqueues);
-		printf("|  disp:     %10llu |\n", nr_vruntime_dispatches);
-		printf("|  failed:   %10llu |\n", nr_vruntime_failed);
+		printf("|  enq:      %10llu |\n", __atomic_load_n(&nr_vruntime_enqueues, __ATOMIC_RELAXED));
+		printf("|  disp:     %10llu |\n", __atomic_load_n(&nr_vruntime_dispatches, __ATOMIC_RELAXED));
+		printf("|  failed:   %10llu |\n", __atomic_load_n(&nr_vruntime_failed, __ATOMIC_RELAXED));
 		printf("o-----------------------o\n");
 		printf("\n\n");
 		fflush(stdout);
@@ -376,10 +376,10 @@ static void bootstrap(char *comm)
 {
 	exit_req = 0;
 	min_vruntime = 0.0;
-	nr_vruntime_enqueues = 0;
-	nr_vruntime_dispatches = 0;
-	nr_vruntime_failed = 0;
-	nr_curr_enqueued = 0;
+	__atomic_store_n(&nr_vruntime_enqueues, 0, __ATOMIC_RELAXED);
+	__atomic_store_n(&nr_vruntime_dispatches, 0, __ATOMIC_RELAXED);
+	__atomic_store_n(&nr_vruntime_failed, 0, __ATOMIC_RELAXED);
+	__atomic_store_n(&nr_curr_enqueued, 0, __ATOMIC_RELAXED);
 	memset(tasks, 0, pid_max * sizeof(*tasks));
 	LIST_INIT(&vruntime_head);
 
-- 
2.51.0


             reply	other threads:[~2026-02-18 19:22 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-18 19:22 David Carlier [this message]
2026-02-21  3:19 ` [PATCH] tools/sched_ext: scx_userland: fix data races on shared counters Tejun Heo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260218192223.8665-1-devnexen@gmail.com \
    --to=devnexen@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=tj@kernel.org \
    --cc=void@manifault.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox