From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Piel Subject: Re: [PATCH] (1/3) cpufreq_ondemand - 01_ignore-nice.diff Date: Mon, 21 Feb 2005 13:58:34 +0100 Message-ID: <4219DAFA.2040103@tremplin-utc.net> References: <20050220131518.GB4495@inskipp> <4218B768.8020401@tremplin-utc.net> <4219B603.8050403@tremplin-utc.net> <20050221102756.GA4404@inskipp> Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable In-Reply-To: <20050221102756.GA4404@inskipp> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: cpufreq-bounces@lists.linux.org.uk Errors-To: cpufreq-bounces+glkc-cpufreq=gmane.org@lists.linux.org.uk Content-Type: text/plain; charset="iso-8859-1"; format="flowed" To: Alexander Clouter Cc: davej@redhat.com, cpufreq@zenii.linux.org.uk, linux@dominikbrodowski.de Alexander Clouter a =C3=A9crit : > On Feb 21, Eric Piel wrote: >=20 >>Here is the patch, it does exactly the same functionality but it's=20 >>completly static and therefore much smaller. Just tested on my computer= ,=20 >>it does the trick. >> >=20 > I'm happy for this to go ahead as I will want it on always by default,=20 > however you pretty much kill the folk who prefer to 'nice' tasks to pre= vent=20 > interuption to other applications rather than using it to say "I do not= care=20 > how long this task takes". >=20 > I know on our proxy server there are a few rebuilds which we run at a n= ice'd > level, not because we do not care how long they take (we want the rebui= lds to > finish as soon as possible) but rather so that squid's performance is n= ot > affected. >=20 That's very unusual cases. You cannot take care of all unusual case with=20 a governor that tries to be as simple as possible. Probably in the case=20 you cite either you want to use a userspace governor which can be much=20 more complex and sofisticated or maybe just: renice -5 $(pidof) squid Eric