Re: [PATCH 2/6] bacct_add_tsk: fix unsafe and wrong parent/group_leader dereference

[Jay Lan <jlan@sgi.com>]

Oleg Nesterov wrote:
> 1. ts = timespec_sub(uptime, current->group_leader->start_time);
> 
>    It is possible that current != tsk. Probably it was supposed
>    to be 'tsk->group_leader->start_time. But why we are reading
>    group_leader's start_time ? This accounting is per thread,
>    not per procees, I changed this to 'tsk->start_time.
>    Please corect me.

This is right. Thanks!

> 
> 2. stats->ac_ppid = (tsk->parent) ? tsk->parent->pid : 0;
> 
>    tsk->parent never == NULL, and it is unsafe to dereference it.
>    Both the task and it's parent may exit after the caller unlocks
>    tasklist_lock, the memory could be unmapped (DEBUG_SLAB).
>    (And we should use ->real_parent->tgid in fact).
> 
> Q: I don't understand the 'if (thread_group_leader(tsk))' check.
> Why it is needed ?

The code was borrowed from kernel/acct.c. The CSA code was extended
from the BSD accounting, so we just borrowed the basic accounting
stuff from the code.

Thanks,
 - jay

> 
> Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
> 
> --- STATS/kernel/tsacct.c~2_par_fix	2006-10-22 18:24:03.000000000 +0400
> +++ STATS/kernel/tsacct.c	2006-10-27 01:03:26.000000000 +0400
> @@ -36,7 +36,7 @@ void bacct_add_tsk(struct taskstats *sta
>  
>  	/* calculate task elapsed time in timespec */
>  	do_posix_clock_monotonic_gettime(&uptime);
> -	ts = timespec_sub(uptime, current->group_leader->start_time);
> +	ts = timespec_sub(uptime, tsk->start_time);
>  	/* rebase elapsed time to usec */
>  	ac_etime = timespec_to_ns(&ts);
>  	do_div(ac_etime, NSEC_PER_USEC);
> @@ -58,7 +58,10 @@ void bacct_add_tsk(struct taskstats *sta
>  	stats->ac_uid	 = tsk->uid;
>  	stats->ac_gid	 = tsk->gid;
>  	stats->ac_pid	 = tsk->pid;
> -	stats->ac_ppid	 = (tsk->parent) ? tsk->parent->pid : 0;
> +	rcu_read_lock();
> +	stats->ac_ppid	 = pid_alive(tsk) ?
> +				rcu_dereference(tsk->real_parent)->tgid : 0;
> +	rcu_read_unlock();
>  	stats->ac_utime	 = cputime_to_msecs(tsk->utime) * USEC_PER_MSEC;
>  	stats->ac_stime	 = cputime_to_msecs(tsk->stime) * USEC_PER_MSEC;
>  	stats->ac_minflt = tsk->min_flt;
>