From mboxrd@z Thu Jan 1 00:00:00 1970 From: Carsten Schiers Subject: Re: AW: Re: AW: Re: AW: Re: Xen 3.4 strange behaviour as compared to Xen 3.3.1 Date: Sun, 31 May 2009 22:39:05 +0200 Message-ID: <4A22EAE9.5050504@schiers.de> References: Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1"; format="flowed" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: Content-Disposition: inline List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: "keir.fraser" , xen-devel List-Id: xen-devel@lists.xenproject.org By the way: in the forum of my favorite distro (c't Server), I found=20 another guy with exactly the same problem. 3.3.1 working, 3.4 no stepping up. BR, Carsten. Carsten Schiers schrieb: > No it's the kernel of Xen 3.4.0 with the same config. I will try later=20 > with the=20 > older kernel, which is the one of 06.04.2009 (I have not realy=20 > understood the > tagging or versioning of the Xen kernel). The other combination would b= e=20 > Xen 3.3.1 > with the latest kernel. That's how we narrow it down to hypervisor vs.=20 > kernel. > > As I use cpufreq=3Ddom0-kernel and enable cpufreq modules in=20 > drivers/cpufreq/Kconfig, > I guess it should use the hypercall for idle time. But to double check,= =20 > I generated > load in dom0. No difference. It simply doesn't step upwards. The kernel= =20 > with cpufreq > debugging is enabled. I will also try it, maybe it's loging something=20 > usefull. > > BR, > Carsten. > > -----Urspr=FCngliche Nachricht----- > Von: Keir Fraser [mailto:keir.fraser@eu.citrix.com]=20 > Gesendet: Samstag, 30. Mai 2009 11:30 > An: Carsten Schiers; xen-devel > Betreff: Re: AW: Re: AW: Re: [Xen-devel] Xen 3.4 strange behaviour as=20 > compared to Xen 3.3.1 > > Are you using the exact same dom0 kernel as before, when it worked? One > theory would be that dom0 is looking at its own idle stats, and it=20 > probably > is pretty idle. So then it steps down the CPUs and keeps them down. Whe= n=20 > you > work the CPUs, are you working dom0 hard? > > -- Keir > > On 30/05/2009 07:59, "Carsten Schiers" wrote: > > =20 >> Sorry, it's lowest frequency (1.0 out of 1.0, 1.8, 2.0, and 2.1 GHz). >> When booting, >> the CPU will be at 2.1 GHz, when switching the governor from=20 >> =20 > performance > =20 >> to ondemand, >> this one will set it to 1.0 GHz, where it's sort of fixed. I can set=20 >> =20 > it > =20 >> manually with >> cpufreq-set when switching to userspace governor, though. >> >> So my guess is that the ondemand governor donesn't get te right >> information about idle >> time though the according hypercall. I recompiled with CPUFREQ DEBUG >> set. But as said >> earlier, I don't have that much knowledge about how to debug kernels. >> >> BR, >> Carsten. >> >> -----Urspr=FCngliche Nachricht----- >> Von: Keir Fraser [mailto:keir.fraser@eu.citrix.com] >> Gesendet: Freitag, 29. Mai 2009 23:31 >> An: Carsten Schiers; xen-devel >> Betreff: Re: AW: Re: [Xen-devel] Xen 3.4 strange behaviour as compared >> to Xen 3.3.1 >> >> Is lowest p-state lowest or highest frequency/voltage? >> >> -- Keir >> >> On 29/05/2009 17:47, "Carsten Schiers" wrote: >> >> =20 >>> Thanks Keir, I tried it out but there is no difference. And by the >>> =20 >> way: >> =20 >>> it >>> is in lowest p-state and doesn't come up, even if under heavy load. >>> =20 >> Hmm. >> =20 >>> BR, >>> Carsten. >>> >>> -----Urspr=FCngliche Nachricht----- >>> Von: Keir Fraser [mailto:keir.fraser@eu.citrix.com] >>> Gesendet: Freitag, 29. Mai 2009 17:02 >>> An: Carsten Schiers; xen-devel >>> Betreff: Re: [Xen-devel] Xen 3.4 strange behaviour as compared to Xen >>> 3.3.1 >>> >>> On 29/05/2009 15:03, "Carsten Schiers" wrote: >>> >>> =20 >>>> - as already reported, ondemand manager in dom0-kernel doesn't >>>> =20 >> step, >> =20 >>>> manual >>>> setting works so it seems a bit like the communication between >>>> =20 >>> dom0 >>> =20 >>>> and >>>> hypervisor regarding idle time is not working >>>> =20 >>> Could be an interaction with C-state support, preferring deep sleep=20 >>> =20 > to > =20 >>> running at lower voltage/frequency? You could try no-cpuidle on Xen's >>> command line at boot time and see if that changes things. >>> >>> -- Keir >>> >>> =20 >>>> - all beside one domu use Xen 3.4.0 kernel, the one who uses it's >>>> customized >>>> kernel won't start up as first domu. It simply hangs and this >>>> prevents also >>>> all other domus (I all auto start them, no save/restore) don't >>>> =20 >>> come >>> =20 >>>> up. When >>>> I start the chain with a different one and this (with the >>>> =20 >>> different >>> =20 >>>> kernel) >>>> is started as #2 or #3, not problem >>>> >>>> - one domu is for vdr with three dvb pci cards passed trough. This >>>> one, when >>>> started as the first one, will cause xentop to show 20% load. >>>> =20 >> When >> =20 >>>> restarted >>>> or started as #2, the load is like with 3.3.1 at roughly 3-5%. >>>> =20 >>> >>> >>> =20 >> >> >> =20 > > > > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel > =20