From: Michael Holzheu <holzheu@linux.vnet.ibm.com>
To: Oleg Nesterov <oleg@redhat.com>,
Shailabh Nagar <nagar1234@in.ibm.com>,
Andrew Morton <akpm@linux-foundation.org>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
John stultz <johnstul@us.ibm.com>,
Thomas Gleixner <tglx@linutronix.de>,
Balbir Singh <balbir@linux.vnet.ibm.com>,
Martin Schwidefsky <schwidefsky@de.ibm.com>,
Heiko Carstens <heiko.carstens@de.ibm.com>,
Roland McGrath <roland@redhat.com>,
Valdis.Kletnieks@vt.edu
Cc: linux-kernel@vger.kernel.org, linux-s390@vger.kernel.org
Subject: [patch v2 3/4] taskstats: Introduce kernel.full_cdata sysctl
Date: Mon, 29 Nov 2010 17:42:40 +0100 [thread overview]
Message-ID: <20101129164435.557525646@linux.vnet.ibm.com> (raw)
In-Reply-To: 20101129164237.522034198@linux.vnet.ibm.com
[-- Attachment #1: 03-taskstats-top-improve-ctime-account-cdata_acct.patch --]
[-- Type: text/plain, Size: 4304 bytes --]
From: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Version 2
---------
* Implement sysctl instead of adding parallel cdata_acct
* Call __account_cdata() in __exit_signal() without tsk siglock
Description
-----------
Currently the cumulative time accounting in Linux is not complete.
Due to POSIX POSIX.1-2001, the CPU time of processes is not accounted
to the cumulative time of the parents, if the parents ignore SIGCHLD
or have set SA_NOCLDWAIT. This behaviour has the major drawback that
it is not possible to calculate all consumed CPU time of a system by
looking at the current tasks. CPU time can be lost.
This patch adds a new sysctl "kernel.full_cdata" that allows to switch
between the POSIX behavior and complete cumulative accounting.
The default is the POSIX semantics.
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
---
Documentation/sysctl/kernel.txt | 12 ++++++++++++
kernel/exit.c | 13 +++++++++----
kernel/sysctl.c | 10 ++++++++++
3 files changed, 31 insertions(+), 4 deletions(-)
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -247,6 +247,18 @@ see the hostname(1) man page.
==============================================================
+full_cdata:
+
+With "full_cdata = 0" (default) cumulative resource accounting
+under Linux is done according to POSIX.1-2001: The resource
+counters of processes are not accounted to the cumulative counters
+of their parents, if the parents ignore SIGCHLD or have set
+SA_NOCLDWAIT. With "full_cdata = 1" it is possible to enforce
+that all dead processes without exception are accounted to their
+parents.
+
+==============================================================
+
hotplug:
Path for the hotplug policy agent.
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -57,6 +57,8 @@
#include <asm/pgtable.h>
#include <asm/mmu_context.h>
+int full_cdata_enabled;
+
static void exit_mm(struct task_struct * tsk);
static void __unhash_process(struct task_struct *p, bool group_dead)
@@ -77,7 +79,7 @@ static void __unhash_process(struct task
static void __account_cdata(struct task_struct *p)
{
struct cdata *cd, *pcd, *tcd;
- unsigned long maxrss;
+ unsigned long maxrss, flags;
cputime_t tgutime, tgstime;
/*
@@ -104,7 +106,7 @@ static void __account_cdata(struct task_
tcd = &p->signal->cdata_threads;
cd = &p->signal->cdata_wait;
- spin_lock_irq(&p->real_parent->sighand->siglock);
+ spin_lock_irqsave(&p->real_parent->sighand->siglock, flags);
pcd->utime =
cputime_add(pcd->utime,
cputime_add(tgutime,
@@ -137,7 +139,7 @@ static void __account_cdata(struct task_
pcd->maxrss = maxrss;
task_io_accounting_add(&p->real_parent->signal->ioac, &p->ioac);
task_io_accounting_add(&p->real_parent->signal->ioac, &p->signal->ioac);
- spin_unlock_irq(&p->real_parent->sighand->siglock);
+ spin_unlock_irqrestore(&p->real_parent->sighand->siglock, flags);
}
/*
@@ -150,6 +152,9 @@ static void __exit_signal(struct task_st
struct sighand_struct *sighand;
struct tty_struct *uninitialized_var(tty);
+ if (group_dead && full_cdata_enabled)
+ __account_cdata(tsk);
+
sighand = rcu_dereference_check(tsk->sighand,
rcu_read_lock_held() ||
lockdep_tasklist_lock_is_held());
@@ -1292,7 +1297,7 @@ static int wait_task_zombie(struct wait_
* It can be ptraced but not reparented, check
* !task_detached() to filter out sub-threads.
*/
- if (likely(!traced) && likely(!task_detached(p)))
+ if (likely(!traced) && likely(!task_detached(p)) && !full_cdata_enabled)
__account_cdata(p);
/*
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -85,6 +85,7 @@
#if defined(CONFIG_SYSCTL)
/* External variables not in a header file. */
+extern int full_cdata_enabled;
extern int sysctl_overcommit_memory;
extern int sysctl_overcommit_ratio;
extern int max_threads;
@@ -963,6 +964,15 @@ static struct ctl_table kern_table[] = {
.proc_handler = proc_dointvec,
},
#endif
+ {
+ .procname = "full_cdata",
+ .data = &full_cdata_enabled,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = proc_dointvec_minmax,
+ .extra1 = &zero,
+ .extra2 = &one,
+ },
/*
* NOTE: do not add new entries to this table unless you have read
* Documentation/sysctl/ctl_unnumbered.txt
next prev parent reply other threads:[~2010-11-29 16:42 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-29 16:42 [patch v2 0/4] taskstats: Improve cumulative time accounting Michael Holzheu
2010-11-29 16:42 ` [patch v2 1/4] taskstats: Introduce "struct cdata" Michael Holzheu
2010-11-29 16:42 ` [patch v2 2/4] taskstats: Introduce __account_cdata() function Michael Holzheu
2010-11-29 16:42 ` Michael Holzheu [this message]
2010-11-29 16:42 ` [patch v2 4/4] taskstats: Export "cdata_wait" CPU times with taskstats Michael Holzheu
2010-12-01 18:51 ` Oleg Nesterov
2010-12-02 16:34 ` Michael Holzheu
2010-12-06 9:06 ` Balbir Singh
2010-12-08 20:23 ` Oleg Nesterov
2010-12-10 13:26 ` Michael Holzheu
[not found] ` <20101211173931.GA8084@redhat.com>
2010-12-13 13:05 ` Michael Holzheu
2010-12-13 13:20 ` Michael Holzheu
2010-12-13 14:33 ` Oleg Nesterov
2010-12-13 16:42 ` Michael Holzheu
2010-12-03 7:33 ` Balbir Singh
2010-12-06 12:37 ` Michael Holzheu
2010-12-06 15:15 ` Balbir Singh
2010-12-07 10:45 ` Michael Holzheu
2010-12-08 20:39 ` Oleg Nesterov
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=20101129164435.557525646@linux.vnet.ibm.com \
--to=holzheu@linux.vnet.ibm.com \
--cc=Valdis.Kletnieks@vt.edu \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=balbir@linux.vnet.ibm.com \
--cc=heiko.carstens@de.ibm.com \
--cc=johnstul@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=nagar1234@in.ibm.com \
--cc=oleg@redhat.com \
--cc=roland@redhat.com \
--cc=schwidefsky@de.ibm.com \
--cc=tglx@linutronix.de \
/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;
as well as URLs for NNTP newsgroup(s).