[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YAA8qyBUAurgCeEz@blackbook>
Date: Thu, 14 Jan 2021 13:44:27 +0100
From: Michal Koutný <mkoutny@...e.com>
To: Hao Lee <haolee.swjtu@...il.com>
Cc: tj@...nel.org, lizefan@...wei.com, hannes@...xchg.org,
cgroups@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] cgroup: Remove unnecessary call to strstrip()
Hello.
On Sun, Jan 03, 2021 at 02:50:01AM +0000, Hao Lee <haolee.swjtu@...il.com> wrote:
> The string buf will be stripped in cgroup_procs_write_start() before it
> is converted to int, so remove this unnecessary call to strstrip().
Good catch, Hao.
Perhaps the code be then simplified a bit
-- >8 --
From: =?UTF-8?q?Michal=20Koutn=C3=BD?= <mkoutny@...e.com>
Date: Thu, 14 Jan 2021 13:23:39 +0100
Subject: [PATCH] cgroup: cgroup.{procs,threads} factor out common parts
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
The functions cgroup_threads_start and cgroup_procs_start are almost
identical. In order to reduce duplication, factor out the common code in
similar fashion we already do for other threadgroup/task functions. No
functional changes are intended.
Suggested-by: Hao Lee <haolee.swjtu@...il.com>
Signed-off-by: Michal Koutný <mkoutny@...e.com>
---
kernel/cgroup/cgroup.c | 55 +++++++++++-------------------------------
1 file changed, 14 insertions(+), 41 deletions(-)
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 613845769103c..f6279df507f4b 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -4726,8 +4726,8 @@ static int cgroup_attach_permissions(struct cgroup *src_cgrp,
return ret;
}
-static ssize_t cgroup_procs_write(struct kernfs_open_file *of,
- char *buf, size_t nbytes, loff_t off)
+static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf,
+ bool threadgroup)
{
struct cgroup *src_cgrp, *dst_cgrp;
struct task_struct *task;
@@ -4738,7 +4738,7 @@ static ssize_t cgroup_procs_write(struct kernfs_open_file *of,
if (!dst_cgrp)
return -ENODEV;
- task = cgroup_procs_write_start(buf, true, &locked);
+ task = cgroup_procs_write_start(buf, threadgroup, &locked);
ret = PTR_ERR_OR_ZERO(task);
if (ret)
goto out_unlock;
@@ -4748,19 +4748,26 @@ static ssize_t cgroup_procs_write(struct kernfs_open_file *of,
src_cgrp = task_cgroup_from_root(task, &cgrp_dfl_root);
spin_unlock_irq(&css_set_lock);
+ /* process and thread migrations follow same delegation rule */
ret = cgroup_attach_permissions(src_cgrp, dst_cgrp,
- of->file->f_path.dentry->d_sb, true);
+ of->file->f_path.dentry->d_sb, threadgroup);
if (ret)
goto out_finish;
- ret = cgroup_attach_task(dst_cgrp, task, true);
+ ret = cgroup_attach_task(dst_cgrp, task, threadgroup);
out_finish:
cgroup_procs_write_finish(task, locked);
out_unlock:
cgroup_kn_unlock(of->kn);
- return ret ?: nbytes;
+ return ret;
+}
+
+static ssize_t cgroup_procs_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes, loff_t off)
+{
+ return __cgroup_procs_write(of, buf, true) ?: nbytes;
}
static void *cgroup_threads_start(struct seq_file *s, loff_t *pos)
@@ -4771,41 +4778,7 @@ static void *cgroup_threads_start(struct seq_file *s, loff_t *pos)
static ssize_t cgroup_threads_write(struct kernfs_open_file *of,
char *buf, size_t nbytes, loff_t off)
{
- struct cgroup *src_cgrp, *dst_cgrp;
- struct task_struct *task;
- ssize_t ret;
- bool locked;
-
- buf = strstrip(buf);
-
- dst_cgrp = cgroup_kn_lock_live(of->kn, false);
- if (!dst_cgrp)
- return -ENODEV;
-
- task = cgroup_procs_write_start(buf, false, &locked);
- ret = PTR_ERR_OR_ZERO(task);
- if (ret)
- goto out_unlock;
-
- /* find the source cgroup */
- spin_lock_irq(&css_set_lock);
- src_cgrp = task_cgroup_from_root(task, &cgrp_dfl_root);
- spin_unlock_irq(&css_set_lock);
-
- /* thread migrations follow the cgroup.procs delegation rule */
- ret = cgroup_attach_permissions(src_cgrp, dst_cgrp,
- of->file->f_path.dentry->d_sb, false);
- if (ret)
- goto out_finish;
-
- ret = cgroup_attach_task(dst_cgrp, task, false);
-
-out_finish:
- cgroup_procs_write_finish(task, locked);
-out_unlock:
- cgroup_kn_unlock(of->kn);
-
- return ret ?: nbytes;
+ return __cgroup_procs_write(of, buf, false) ?: nbytes;
}
/* cgroup core interface files for the default hierarchy */
--
2.29.2
Powered by blists - more mailing lists