Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 16 Aug 2004 13:40:12 -0500
From:      Dan Nelson <dnelson@allantgroup.com>
To:        Slawek Zak <zaks@prioris.mini.pw.edu.pl>
Cc:        freebsd-current@freebsd.org
Subject:   Re: Deadlocks with recent SMP current
Message-ID:  <20040816184012.GK73391@dan.emsphone.com>
In-Reply-To: <86y8kfqcp8.fsf@thirst.unx.era.pl>
References:  <20040813121208.M31181@cvs.imp.ch> <86y8kfqcp8.fsf@thirst.unx.era.pl>

next in thread | previous in thread | raw e-mail | index | archive | help
In the last episode (Aug 16), Slawek Zak said:
> Another thing is bothering me. In top I see:
> 
> CPU states: 16.2% user,  0.0% nice,  2.5% system,  1.0% interrupt, 80.3% idle
> Mem: 254M Active, 605M Inact, 254M Wired, 16K Cache, 112M Buf, 2622M Free
> Swap: 4096M Total, 4096M Free
> 
>   PID USERNAME PRI NICE   SIZE    RES STATE  C   TIME   WCPU    CPU COMMAND
> 37033 mysql     20    0   265M   178M kserel 0 610:59 124.37% 124.37% mysqld
> 
> MySQLd is running with libpthread and can go up to 400% without
> significantly afecting the idle percentage shown by top. I don't see
> how I could fit 400% into 20% on those two suckers, therefore I ask
> :)

The CPU accounting for KSE threads is a bit inaccurate; new threads
inherit the CPU usage of their parent, so if a program does a lot of
computation, then spawns a bunch of threads, you would end up with 50
threads each with 90% cpu.  Top then totals them all up :)  I don't know
if it's possible to accurately display per-thread CPU usage in a M:N
threading scheme.

-- 
	Dan Nelson
	dnelson@allantgroup.com



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040816184012.GK73391>