linux-kernel - Re: [PATCH] taskstats: Add e/u/stime for TGID command

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 6 Mar 2017 14:40:17 -0800
From:   Andrew Morton <akpm@...ux-foundation.org>
To:     Zhang Xiao <xiao.zhang@...driver.com>
Cc:     <bsingharora@...il.com>, <linux-kernel@...r.kernel.org>,
        Oleg Nesterov <oleg@...hat.com>
Subject: Re: [PATCH] taskstats: Add e/u/stime for TGID command

On Fri, 3 Mar 2017 10:33:44 +0800 Zhang Xiao <xiao.zhang@...driver.com> wrote:

> Add elapsed time, user CPU time and system CPU time to
> thread group status request.
> 
> --- a/kernel/taskstats.c
> +++ b/kernel/taskstats.c
> @@ -210,6 +210,7 @@ static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats)
>  	struct task_struct *tsk, *first;
>  	unsigned long flags;
>  	int rc = -ESRCH;
> +	u64 delta, utime, stime;
>  
>  	/*
>  	 * Add additional stats from live tasks except zombie thread group
> @@ -238,6 +239,16 @@ static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats)
>  		 */
>  		delayacct_add_tsk(stats, tsk);
>  
> +		/* calculate task elapsed time in nsec */
> +		delta = ktime_get_ns() - tsk->start_time;
> +		/* Convert to micro seconds */
> +		do_div(delta, NSEC_PER_USEC);
> +		stats->ac_etime += delta;
> +
> +		task_cputime(tsk, &utime, &stime);
> +		stats->ac_utime += div_u64(utime, NSEC_PER_USEC);
> +		stats->ac_stime += div_u64(stime, NSEC_PER_USEC);
> +
>  		stats->nvcsw += tsk->nvcsw;
>  		stats->nivcsw += tsk->nivcsw;
>  	} while_each_thread(first, tsk);

hm, OK, we were just leaving these at zero.  Seems sane to me.

It would be more efficient and perhaps more deterministic to sample
ktime just once?

--- a/kernel/taskstats.c~taskstats-add-e-u-stime-for-tgid-command-fix
+++ a/kernel/taskstats.c
@@ -211,6 +211,7 @@ static int fill_stats_for_tgid(pid_t tgi
 	unsigned long flags;
 	int rc = -ESRCH;
 	u64 delta, utime, stime;
+	u64 start_time;
 
 	/*
 	 * Add additional stats from live tasks except zombie thread group
@@ -228,6 +229,7 @@ static int fill_stats_for_tgid(pid_t tgi
 		memset(stats, 0, sizeof(*stats));
 
 	tsk = first;
+	start_time = ktime_get_ns();
 	do {
 		if (tsk->exit_state)
 			continue;
@@ -240,7 +242,7 @@ static int fill_stats_for_tgid(pid_t tgi
 		delayacct_add_tsk(stats, tsk);
 
 		/* calculate task elapsed time in nsec */
-		delta = ktime_get_ns() - tsk->start_time;
+		delta = start_time - tsk->start_time;
 		/* Convert to micro seconds */
 		do_div(delta, NSEC_PER_USEC);
 		stats->ac_etime += delta;
_