linux-kernel - Re: [PATCH: 2.6.18.1] delayacct: cpu_count in taskstats updated correctly

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]

Message-ID: <Pine.LNX.4.63.0610270545000.21448@honbori.ifi.uio.no>
Date:	Fri, 27 Oct 2006 06:47:12 +0200 (CEST)
From:	Martin Tostrup Setek <martitse@...dent.matnat.uio.no>
To:	David Rientjes <rientjes@...washington.edu>
cc:	linux-kernel@...r.kernel.org
Subject: Re: [PATCH: 2.6.18.1] delayacct: cpu_count in taskstats updated
 correctly

On Thu, 26 Oct 2006, David Rientjes wrote:
> On Fri, 27 Oct 2006, Martin Tostrup Setek wrote:
>> from: Martin T. Setek <martitse@....uio.no>
>> cpu_count in struct taskstats should be the same as the corresponding (third)
>> value found in /proc/<pid>/schedstat
>
> I disagree in favor of Documentation/accounting/taskstats-struct.txt.
> cpu_count is the number of delay values recorded, so accumulating them is
> appropriate.

Perhaps I'm mistaken, but the code accumulates the value it 
finds in sched_info.pcnt in the task_struct. Now, in sched.h I found this:

struct sched_info {
 	/* cumulative counters */
 	unsigned long	cpu_time,	/* time spent on the cpu */
 			run_delay,	/* time spent waiting on a runqueue */
 			pcnt;		/* # of timeslices run on this cpu */

The comment says that these counters are cumulative... The code that 
updates them (sched.c: sched_info_arrive()), does accumulate them.

In include/linux/taskstats.h, I found:

          * xxx_count is the number of delay values recorded
          * xxx_delay_total is the corresponding cumulative delay in nanoseconds

I interpret these comments as saying that:

cpu_delay should be the total number of nanoseconds a task has been 
waiting in a runqueue for a CPU, and cpu_count is equal to the number of 
times the task got the CPU (or waited for it). If so, then the code 
updates taskstats.cpu_delay_total incorrectly too (which my patch didn't 
fix).

If not, then the comments in taskstats.h are very confusing....

Martin
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/