From nobody Sun Oct 5 00:10:06 2025 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id AA9752E338D for ; Mon, 11 Aug 2025 21:22:05 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754947327; cv=none; b=X9pHRlWJuAjpNJyps0t1u49KknsW1wEUd22wsc8ba3Nm3CEL9JsespNcd1SZt6N3AyOt3ljARkXx+m7Ezd4YmaDjveGHI45+/6yVYqhhIui/RIadduYaX+anMcwpgQssrG18gG4DbBizWhzEicNcYIarpXFBSaQDqyQKgoqBL7U= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754947327; c=relaxed/simple; bh=rs1rz9seRhodTia6WJevCMyrbdJdVy+CPiy7SqP4Obo=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=orfqd3OgRq73B5qeB0XGoMCM5RCXDirk82yAGr0HtaBI9HRdoYM9vSmD5WUNRXSGC6eSTqFVekXRbX+CFi/tjQQHr7DspvZAu6DAF7h78qbwBB/24YN33zJKWqi/gkcoLuXXG9traxWruk4/5p6unSPmzKFr4A4NdEwYgEXc0Y0= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 0474F1E8D; Mon, 11 Aug 2025 14:21:57 -0700 (PDT) Received: from e127648.arm.com (unknown [10.57.55.59]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 1FACD3F63F; Mon, 11 Aug 2025 14:22:01 -0700 (PDT) From: Christian Loehle To: tj@kernel.org, arighi@nvidia.com, void@manifault.com Cc: linux-kernel@vger.kernel.org, sched-ext@lists.linux.dev, changwoo@igalia.com, hodgesd@meta.com, mingo@redhat.com, peterz@infradead.org, jake@hillion.co.uk, Christian Loehle Subject: [PATCH v4 1/3] sched_ext: Introduce scx_bpf_cpu_rq_locked() Date: Mon, 11 Aug 2025 22:21:48 +0100 Message-Id: <20250811212150.85759-2-christian.loehle@arm.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20250811212150.85759-1-christian.loehle@arm.com> References: <20250811212150.85759-1-christian.loehle@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Most fields in scx_bpf_cpu_rq() assume that its rq_lock is held. Furthermore they become meaningless without rq lock, too. Make a safer version of scx_bpf_cpu_rq() that only returns a rq if we hold rq lock of that rq. Also mark the new scx_bpf_cpu_rq_locked() as returning NULL. Signed-off-by: Christian Loehle --- kernel/sched/ext.c | 22 ++++++++++++++++++++++ tools/sched_ext/include/scx/common.bpf.h | 1 + 2 files changed, 23 insertions(+) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 7dedc9a16281..14706c36ca83 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -7426,6 +7426,27 @@ __bpf_kfunc struct rq *scx_bpf_cpu_rq(s32 cpu) return cpu_rq(cpu); } =20 +/** + * scx_bpf_cpu_rq_locked - Fetch the locked rq of a CPU + * @cpu: CPU of the rq + */ +__bpf_kfunc struct rq *scx_bpf_cpu_rq_locked(s32 cpu) +{ + struct rq *rq; + + if (!kf_cpu_valid(cpu, NULL)) + return NULL; + + preempt_disable(); + rq =3D cpu_rq(cpu); + if (rq !=3D scx_locked_rq()) { + scx_kf_error("Accessing not locked rq %d", cpu); + rq =3D NULL; + } + preempt_enable(); + return rq; +} + /** * scx_bpf_task_cgroup - Return the sched cgroup of a task * @p: task of interest @@ -7590,6 +7611,7 @@ BTF_ID_FLAGS(func, scx_bpf_put_cpumask, KF_RELEASE) BTF_ID_FLAGS(func, scx_bpf_task_running, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_task_cpu, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_cpu_rq) +BTF_ID_FLAGS(func, scx_bpf_cpu_rq_locked, KF_RET_NULL) #ifdef CONFIG_CGROUP_SCHED BTF_ID_FLAGS(func, scx_bpf_task_cgroup, KF_RCU | KF_ACQUIRE) #endif diff --git a/tools/sched_ext/include/scx/common.bpf.h b/tools/sched_ext/inc= lude/scx/common.bpf.h index d4e21558e982..7451491347ed 100644 --- a/tools/sched_ext/include/scx/common.bpf.h +++ b/tools/sched_ext/include/scx/common.bpf.h @@ -91,6 +91,7 @@ s32 scx_bpf_pick_any_cpu(const cpumask_t *cpus_allowed, u= 64 flags) __ksym; bool scx_bpf_task_running(const struct task_struct *p) __ksym; s32 scx_bpf_task_cpu(const struct task_struct *p) __ksym; struct rq *scx_bpf_cpu_rq(s32 cpu) __ksym; +struct rq *scx_bpf_cpu_rq_locked(s32 cpu) __ksym; struct cgroup *scx_bpf_task_cgroup(struct task_struct *p) __ksym __weak; u64 scx_bpf_now(void) __ksym __weak; void scx_bpf_events(struct scx_event_stats *events, size_t events__sz) __k= sym __weak; --=20 2.34.1 From nobody Sun Oct 5 00:10:06 2025 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id C09192E2EE4 for ; Mon, 11 Aug 2025 21:22:12 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754947334; cv=none; b=gLp+XS1FVtZDxIy/CCg7CFk1gEYPfhZfVUuiJ+Iz+NE4FT5covLUkr7+oRug/+b2BY1FJIHpFqbCZ+sojxE8gCfm26ZPsj29KzPQhCtlGJE/8SRx7JeZ/D3cBv+fmOQWUztikGry2fh5u+8DbUuRY5ueHNyPsHBqQkA6eTASxUc= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754947334; c=relaxed/simple; bh=5nx1DrxNB3Sn0esWx/qM3ys/wu10WGO/lfN3dbQwdxo=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=kuHauwDP//3m8Z4G3MH6TWq8yu60vxHMepyV3jlVeFgLbXAbeXPBMpE6L1nRNuSwkXaqGgIBW47irEdxuxi0wdiG7qJgPBam8zkgkktdHveASI+8wVDIYDebl2u7I6hx79W+M2nNl064NICiHu8GwTTTXa/uiDEYuAnqi7ifSjM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id EF837204C; Mon, 11 Aug 2025 14:22:03 -0700 (PDT) Received: from e127648.arm.com (unknown [10.57.55.59]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 7BEB73F63F; Mon, 11 Aug 2025 14:22:09 -0700 (PDT) From: Christian Loehle To: tj@kernel.org, arighi@nvidia.com, void@manifault.com Cc: linux-kernel@vger.kernel.org, sched-ext@lists.linux.dev, changwoo@igalia.com, hodgesd@meta.com, mingo@redhat.com, peterz@infradead.org, jake@hillion.co.uk, Christian Loehle Subject: [PATCH v4 2/3] sched_ext: Provide scx_bpf_task_acquire_remote_curr() Date: Mon, 11 Aug 2025 22:21:49 +0100 Message-Id: <20250811212150.85759-3-christian.loehle@arm.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20250811212150.85759-1-christian.loehle@arm.com> References: <20250811212150.85759-1-christian.loehle@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Provide scx_bpf_task_acquire_remote_curr() as a way for scx schedulers to check the curr task of a remote rq without assuming its lock is held. Many scx schedulers make use of scx_bpf_cpu_rq() to check a remote curr (e.g. to see if it should be preempted). This is problematic because scx_bpf_cpu_rq() provides access to all fields of struct rq, most of which aren't safe to use without holding the associated rq lock. Signed-off-by: Christian Loehle --- kernel/sched/ext.c | 24 ++++++++++++++++++++++++ tools/sched_ext/include/scx/common.bpf.h | 1 + 2 files changed, 25 insertions(+) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 14706c36ca83..ded4ace36090 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -7447,6 +7447,29 @@ __bpf_kfunc struct rq *scx_bpf_cpu_rq_locked(s32 cpu) return rq; } =20 +/** + * scx_bpf_task_acquire_remote_curr - Fetch the curr task of a rq without + * acquiring its rq lock + * @cpu: CPU of the rq + * + * Increments the refcount of the task_struct which needs to be released u= sing + * bpf_task_release(). + */ +__bpf_kfunc struct task_struct *scx_bpf_task_acquire_remote_curr(s32 cpu) +{ + struct task_struct *p; + + if (!kf_cpu_valid(cpu, NULL)) + return NULL; + + rcu_read_lock(); + p =3D cpu_rq(cpu)->curr; + if (p) + p =3D refcount_inc_not_zero(&p->rcu_users) ? p : NULL; + rcu_read_unlock(); + return p; +} + /** * scx_bpf_task_cgroup - Return the sched cgroup of a task * @p: task of interest @@ -7612,6 +7635,7 @@ BTF_ID_FLAGS(func, scx_bpf_task_running, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_task_cpu, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_cpu_rq) BTF_ID_FLAGS(func, scx_bpf_cpu_rq_locked, KF_RET_NULL) +BTF_ID_FLAGS(func, scx_bpf_task_acquire_remote_curr, KF_RET_NULL | KF_ACQU= IRE) #ifdef CONFIG_CGROUP_SCHED BTF_ID_FLAGS(func, scx_bpf_task_cgroup, KF_RCU | KF_ACQUIRE) #endif diff --git a/tools/sched_ext/include/scx/common.bpf.h b/tools/sched_ext/inc= lude/scx/common.bpf.h index 7451491347ed..2105cd0ee178 100644 --- a/tools/sched_ext/include/scx/common.bpf.h +++ b/tools/sched_ext/include/scx/common.bpf.h @@ -92,6 +92,7 @@ bool scx_bpf_task_running(const struct task_struct *p) __= ksym; s32 scx_bpf_task_cpu(const struct task_struct *p) __ksym; struct rq *scx_bpf_cpu_rq(s32 cpu) __ksym; struct rq *scx_bpf_cpu_rq_locked(s32 cpu) __ksym; +struct task_struct *scx_bpf_task_acquire_remote_curr(s32 cpu) __ksym; struct cgroup *scx_bpf_task_cgroup(struct task_struct *p) __ksym __weak; u64 scx_bpf_now(void) __ksym __weak; void scx_bpf_events(struct scx_event_stats *events, size_t events__sz) __k= sym __weak; --=20 2.34.1 From nobody Sun Oct 5 00:10:06 2025 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id D39E92DE6F5 for ; Mon, 11 Aug 2025 21:22:20 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754947342; cv=none; b=hp3g3FvzfvIgR43uzCiv0ndjrMydJjNG+1fEnIa7bgO+rBMpJsx4nHkPoA6Kq6eCU5KwVDiUMcJJE3cYvCMCUOcHGqH6Yt8FClfCkiz7/Z1enIyepeQCxZhuYeY+iuNQUJOCV+yrxLjDsqQdK31ZI2dY8Hmf4YvDHLCJ23bgqcA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1754947342; c=relaxed/simple; bh=48ai1//QqI2vc2O3Mziq7Er7rYM94ChyM3stR4M79ZY=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=r2HrKa1OL2HUkcic0OQ4Xut5Pbd+BNsEx8G+daX7ab4MrnYlBtN4Q/DD8TTbSTlLS3auabpHBplflX/fOf/CDOHJmFbQJPXJHEqj/q//dtN91evDzsfCzun1urd+q6l02v5eb3HalEuLPBCr1R8aFxxIoXZQ+RYQvx1zTbAGOP0= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 590101A32; Mon, 11 Aug 2025 14:22:12 -0700 (PDT) Received: from e127648.arm.com (unknown [10.57.55.59]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id DCCB93F63F; Mon, 11 Aug 2025 14:22:17 -0700 (PDT) From: Christian Loehle To: tj@kernel.org, arighi@nvidia.com, void@manifault.com Cc: linux-kernel@vger.kernel.org, sched-ext@lists.linux.dev, changwoo@igalia.com, hodgesd@meta.com, mingo@redhat.com, peterz@infradead.org, jake@hillion.co.uk, Christian Loehle Subject: [PATCH v4 3/3] sched_ext: deprecation warn for scx_bpf_cpu_rq() Date: Mon, 11 Aug 2025 22:21:50 +0100 Message-Id: <20250811212150.85759-4-christian.loehle@arm.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20250811212150.85759-1-christian.loehle@arm.com> References: <20250811212150.85759-1-christian.loehle@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" scx_bpf_cpu_rq() works on an unlocked rq which generally isn't safe. For the common use-cases scx_bpf_cpu_rq_locked() and scx_bpf_task_acquire_remote_curr() work, so add a deprecation warning to scx_bpf_cpu_rq() so it can eventually be removed. Signed-off-by: Christian Loehle --- kernel/sched/ext.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index ded4ace36090..7d2d88e8dd59 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -7423,6 +7423,9 @@ __bpf_kfunc struct rq *scx_bpf_cpu_rq(s32 cpu) if (!kf_cpu_valid(cpu, NULL)) return NULL; =20 + pr_warn_once("%s() is deprecated in favor of scx_bpf_cpu_rq_locked() or " + "scx_bpf_task_acquire_remote_curr() for unlocked remote curr\n", + __func__); return cpu_rq(cpu); } =20 --=20 2.34.1