Received-SPF: Pass (protection.outlook.com: domain of amd.com designates
 165.204.84.17 as permitted sender) receiver=protection.outlook.com;
 client-ip=165.204.84.17; helo=SATLEXMB04.amd.com; pr=C
Subject: [RFC v2 PATCH 1/7] x86/resctrl: Add multiple tasks to the resctrl
 group at once
From:   Babu Moger <babu.moger@amd.com>
To:     <fenghua.yu@intel.com>, <reinette.chatre@intel.com>
CC:     <babu.moger@amd.com>, <tglx@linutronix.de>, <mingo@redhat.com>,
        <bp@alien8.de>, <dave.hansen@linux.intel.com>, <x86@kernel.org>,
        <hpa@zytor.com>, <corbet@lwn.net>, <linux-kernel@vger.kernel.org>,
        <linux-doc@vger.kernel.org>, <eranian@google.com>,
        <peternewman@google.com>
Date:   Thu, 2 Feb 2023 15:46:54 -0600
Message-ID: <167537441417.647488.18261538242584692230.stgit@bmoger-ubuntu>
In-Reply-To: <167537433143.647488.9641864719195184123.stgit@bmoger-ubuntu>
References: <167537433143.647488.9641864719195184123.stgit@bmoger-ubuntu>
User-Agent: StGit/1.1.dev103+g5369f4c
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 02 Feb 2023 21:46:56.3935
 (UTC)
X-MS-Exchange-CrossTenant-Network-Message-Id: a13a78c9-2064-438e-ec87-08db05670159
X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d
X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=3dd8961f-e488-4e60-8e11-a82d994e183d;Ip=[165.204.84.17];Helo=[SATLEXMB04.amd.com]
X-MS-Exchange-CrossTenant-AuthSource: CO1PEPF00001A63.namprd05.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Anonymous
X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem
X-MS-Exchange-Transport-CrossTenantHeadersStamped: IA1PR12MB6355
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

The resctrl task assignment for MONITOR or CONTROL group needs to be=0A=
done one at a time. For example:=0A=
=0A=
  $mount -t resctrl resctrl /sys/fs/resctrl/=0A=
  $mkdir /sys/fs/resctrl/clos1=0A=
  $echo 123 > /sys/fs/resctrl/clos1/tasks=0A=
  $echo 456 > /sys/fs/resctrl/clos1/tasks=0A=
  $echo 789 > /sys/fs/resctrl/clos1/tasks=0A=
=0A=
This is not user-friendly when dealing with hundreds of tasks.=0A=
=0A=
Improve the user experience by supporting the multiple task assignment=0A=
in one command with the tasks separated by commas. For example:=0A=
=0A=
  $echo 123,456,789 > /sys/fs/resctrl/clos1/tasks=0A=
=0A=
Signed-off-by: Babu Moger <babu.moger@amd.com>=0A=
---=0A=
 Documentation/x86/resctrl.rst          |    9 +++++++--=0A=
 arch/x86/kernel/cpu/resctrl/rdtgroup.c |   24 +++++++++++++++++++++++-=0A=
 2 files changed, 30 insertions(+), 3 deletions(-)=0A=
=0A=
diff --git a/Documentation/x86/resctrl.rst b/Documentation/x86/resctrl.rst=
=0A=
index 058257dc56c8..58b76fc75cb7 100644=0A=
--- a/Documentation/x86/resctrl.rst=0A=
+++ b/Documentation/x86/resctrl.rst=0A=
@@ -292,13 +292,18 @@ All groups contain the following files:=0A=
 "tasks":=0A=
 	Reading this file shows the list of all tasks that belong to=0A=
 	this group. Writing a task id to the file will add a task to the=0A=
-	group. If the group is a CTRL_MON group the task is removed from=0A=
+	group. Multiple tasks can be assigned together in one command by=0A=
+	inputting the tasks separated by commas. Tasks will be assigned=0A=
+	sequentially in the order it is provided. Failure while assigning=0A=
+	the tasks will be aborted immediately and tasks next in the=0A=
+	sequence will not be assigned. Users may need to retry them again.=0A=
+=0A=
+	If the group is a CTRL_MON group the task is removed from=0A=
 	whichever previous CTRL_MON group owned the task and also from=0A=
 	any MON group that owned the task. If the group is a MON group,=0A=
 	then the task must already belong to the CTRL_MON parent of this=0A=
 	group. The task is removed from any previous MON group.=0A=
 =0A=
-=0A=
 "cpus":=0A=
 	Reading this file shows a bitmask of the logical CPUs owned by=0A=
 	this group. Writing a mask to this file will add and remove=0A=
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/r=
esctrl/rdtgroup.c=0A=
index e2c1599d1b37..13b7c5f3a27c 100644=0A=
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c=0A=
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c=0A=
@@ -683,16 +683,34 @@ static ssize_t rdtgroup_tasks_write(struct kernfs_ope=
n_file *of,=0A=
 				    char *buf, size_t nbytes, loff_t off)=0A=
 {=0A=
 	struct rdtgroup *rdtgrp;=0A=
+	char *pid_str;=0A=
 	int ret =3D 0;=0A=
 	pid_t pid;=0A=
 =0A=
-	if (kstrtoint(strstrip(buf), 0, &pid) || pid < 0)=0A=
+	/* Valid input requires a trailing newline */=0A=
+	if (nbytes =3D=3D 0 || buf[nbytes - 1] !=3D '\n')=0A=
 		return -EINVAL;=0A=
+=0A=
+	buf[nbytes - 1] =3D '\0';=0A=
+=0A=
 	rdtgrp =3D rdtgroup_kn_lock_live(of->kn);=0A=
 	if (!rdtgrp) {=0A=
 		rdtgroup_kn_unlock(of->kn);=0A=
 		return -ENOENT;=0A=
 	}=0A=
+=0A=
+next:=0A=
+	if (!buf || buf[0] =3D=3D '\0')=0A=
+		goto unlock;=0A=
+=0A=
+	pid_str =3D strim(strsep(&buf, ","));=0A=
+=0A=
+	if (kstrtoint(pid_str, 0, &pid) || pid < 0) {=0A=
+		rdt_last_cmd_puts("Invalid pid value\n");=0A=
+		ret =3D -EINVAL;=0A=
+		goto unlock;=0A=
+	}=0A=
+=0A=
 	rdt_last_cmd_clear();=0A=
 =0A=
 	if (rdtgrp->mode =3D=3D RDT_MODE_PSEUDO_LOCKED ||=0A=
@@ -703,6 +721,10 @@ static ssize_t rdtgroup_tasks_write(struct kernfs_open=
_file *of,=0A=
 	}=0A=
 =0A=
 	ret =3D rdtgroup_move_task(pid, rdtgrp, of);=0A=
+	if (ret)=0A=
+		goto unlock;=0A=
+	else=0A=
+		goto next;=0A=
 =0A=
 unlock:=0A=
 	rdtgroup_kn_unlock(of->kn);=0A=
=0A=