lkml.org 
[lkml]   [2010]   [Nov]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[patch v2 4/4] taskstats: Export "cdata_wait" CPU times with taskstats
From: Michael Holzheu <holzheu@linux.vnet.ibm.com>

Version 2
---------
* Use thread_group_leader() instead of (tsk->pid == tsk->tgid)
* Use cdata_wait instead of cdata_acct
* I left the struct signal locking in the patch, because I think
that in order to get consistent data this is necessary. See also
do_task_stat() in fs/proc/array.c. One problem is that we
report wrong values (zero) for cdata, if lock_task_sighand()
fails. This will be the same behavior as for /proc/<pid>/stat.
But maybe we should somehow return to userspace that the
information is not correct. Any ideas?

Description
-----------
With this patch the cumulative CPU time is added to "struct taskstats".
The CPU time is only returned for the thread group leader.

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
---
include/linux/taskstats.h | 7 ++++++-
kernel/tsacct.c | 7 +++++++
2 files changed, 13 insertions(+), 1 deletion(-)

--- a/include/linux/taskstats.h
+++ b/include/linux/taskstats.h
@@ -33,7 +33,7 @@
*/


-#define TASKSTATS_VERSION 7
+#define TASKSTATS_VERSION 8
#define TS_COMM_LEN 32 /* should be >= TASK_COMM_LEN
* in linux/sched.h */

@@ -163,6 +163,11 @@ struct taskstats {
/* Delay waiting for memory reclaim */
__u64 freepages_count;
__u64 freepages_delay_total;
+ /* version 7 ends here */
+
+ /* Cumulative CPU time of dead children */
+ __u64 ac_cutime; /* User CPU time [usec] */
+ __u64 ac_cstime; /* System CPU time [usec] */
};


--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -30,6 +30,7 @@ void bacct_add_tsk(struct taskstats *sta
{
const struct cred *tcred;
struct timespec uptime, ts;
+ unsigned long flags;
u64 ac_etime;

BUILD_BUG_ON(TS_COMM_LEN < TASK_COMM_LEN);
@@ -71,6 +72,12 @@ void bacct_add_tsk(struct taskstats *sta
stats->ac_majflt = tsk->maj_flt;

strncpy(stats->ac_comm, tsk->comm, sizeof(stats->ac_comm));
+ if (thread_group_leader(tsk) && lock_task_sighand(tsk, &flags)) {
+ struct cdata *cd = &tsk->signal->cdata_wait;
+ stats->ac_cutime = cputime_to_usecs(cd->utime);
+ stats->ac_cstime = cputime_to_usecs(cd->stime);
+ unlock_task_sighand(tsk, &flags);
+ }
}




\
 
 \ /
  Last update: 2010-11-29 17:47    [W:0.086 / U:11.960 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site