From: Menglong Dong <menglong8.dong@gmail.com>
Date: Thu, 21 Aug 2025 09:06:08 +0000 (+0800)
Subject: bpf: use rcu_read_lock_dont_migrate() for bpf_prog_run_array_cg()
X-Git-Url: https://www.infradead.org/git/?a=commitdiff_plain;h=427a36bb5504e0fb33398a4ccd523fce95514d83;p=users%2Fhch%2Fmisc.git

bpf: use rcu_read_lock_dont_migrate() for bpf_prog_run_array_cg()

Use rcu_read_lock_dont_migrate() and rcu_read_unlock_migrate() in
bpf_prog_run_array_cg to obtain better performance when PREEMPT_RCU is
not enabled.

Signed-off-by: Menglong Dong <dongml2@chinatelecom.cn>
Link: https://lore.kernel.org/r/20250821090609.42508-7-dongml2@chinatelecom.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 180b630279b9..9912c7b9a266 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -71,8 +71,7 @@ bpf_prog_run_array_cg(const struct cgroup_bpf *cgrp,
 	u32 func_ret;
 
 	run_ctx.retval = retval;
-	migrate_disable();
-	rcu_read_lock();
+	rcu_read_lock_dont_migrate();
 	array = rcu_dereference(cgrp->effective[atype]);
 	item = &array->items[0];
 	old_run_ctx = bpf_set_run_ctx(&run_ctx.run_ctx);
@@ -88,8 +87,7 @@ bpf_prog_run_array_cg(const struct cgroup_bpf *cgrp,
 		item++;
 	}
 	bpf_reset_run_ctx(old_run_ctx);
-	rcu_read_unlock();
-	migrate_enable();
+	rcu_read_unlock_migrate();
 	return run_ctx.retval;
 }