Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 16 Apr 2005 20:10:19 +0100 (BST)
From:      Robert Watson <rwatson@FreeBSD.org>
To:        =?iso-8859-1?q?Dag-Erling_Sm=F8rgrav?= <des@des.no>
Cc:        cvs-all@FreeBSD.org
Subject:   Re: cvs commit: src/usr.bin/top machine.c
Message-ID:  <20050416200918.E69456@fledge.watson.org>
In-Reply-To: <863btq4k9m.fsf@xps.des.no>
References:  <200504161543.j3GFhclO075103@repoman.freebsd.org> <86acnyd2k7.fsf@xps.des.no> <20050416191436.G68941@fledge.watson.org> <863btq4k9m.fsf@xps.des.no>

next in thread | previous in thread | raw e-mail | index | archive | help
  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--0-1474543579-1113678619=:69456
Content-Type: TEXT/PLAIN; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: QUOTED-PRINTABLE


On Sat, 16 Apr 2005, Dag-Erling Sm=F8rgrav wrote:

> Robert Watson <rwatson@FreeBSD.org> writes:
>> The 'C' column is quite useful when tuning SMP systems, and something
>> I frequently look at when debugging performance issues in 5.x and
>> 6.x.
>
> The C column is pointless when you're not showing individual threads,=20
> since different threads in the same process might be running=20
> simultaneously on different CPUs.

With many current applications, you get processes, in which case there's=20
still a benefit.  I agree that there's ambiguity in the threaded but=20
non-H case.  In the work I'm doing, I'm primarily interested in the 'C'=20
column in the context of "-S", because I want to see where the kernel and=
=20
user processes land -- i.e., where the ithreads and netisr end up, and how=
=20
well they load balance.

Robert N M Watson
--0-1474543579-1113678619=:69456--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050416200918.E69456>