From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mahmood Naderan Subject: Re: BIOS and CPU C_states are strange Date: Sun, 21 Jun 2009 14:56:07 +0430 Message-ID: <1acba2fa0906210326l40f6fc44h7e928036b57a0160@mail.gmail.com> References: <1acba2fa0906191006i3cd5a035pf18e096fdb92e27e@mail.gmail.com> <1acba2fa0906191007l4b8d5c17id068d868e6178fb8@mail.gmail.com> <1acba2fa0906192354i3cb0f2dbp2c8abed6704db3f@mail.gmail.com> <1acba2fa0906210302x19c31974md75c72b7a171c852@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-bw0-f213.google.com ([209.85.218.213]:44814 "EHLO mail-bw0-f213.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752088AbZFUK0H convert rfc822-to-8bit (ORCPT ); Sun, 21 Jun 2009 06:26:07 -0400 Received: by bwz9 with SMTP id 9so2550763bwz.37 for ; Sun, 21 Jun 2009 03:26:08 -0700 (PDT) In-Reply-To: Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: linux-acpi@vger.kernel.org Hi, Thanks. Can I expect future versions of powertop will consider my problem or it is a specific bug related to my BIOS (my laptop is HP Pavilion DV2000)? I mean, this misreading is a bug in powertop or not? -------------------------- *Mahmood Naderan* On Sun, Jun 21, 2009 at 2:43 PM, Edward Shao wrote= : > Hi~ > powertop can download from > http://www.lesswatts.org/projects/powertop/download/powertop-1.11.tar= =2Egz > > Sorry~I forgot the name of Intel utility. > > Best Regards, > Edward > > On Sun, Jun 21, 2009 at 6:02 PM, Mahmood Naderan wrote: >> Hi, >> 1- where can I find powertop source code? >> 2- what is the name of intel utility in windows as you said? >> -------------------------- >> *Mahmood Naderan* >> >> >> >> >> On Sun, Jun 21, 2009 at 2:19 PM, Edward Shao wr= ote: >>> Hi~ >>> It's all depended on how utility to interpret raw information (both >>> information reported by CPU or BIOS). >>> You can check source code of powertop for detail information. >>> For BIOS part, powertop checks value of mwait hint register reporte= d >>> by ACPI _CST method. >>> >>> As my experience, powertop will cause a misleading sometimes, >>> especially for new Intel CPU. >>> For example: >>> BIOS engineer said they followed Intel BIOS porting guide to report= an >>> accuracy C-state (C6) information via ACPI _CST method. >>> However powertop said BIOS only reports C3. (but Intel utility unde= r >>> Windows said CPU can enter C6). >>> >>> Best Regards, >>> Edward >>> >>> On Sat, Jun 20, 2009 at 2:54 PM, Mahmood Naderan wrote: >>>> Hi, >>>>> Only simple check value of edx. >>>>> Well, unless CPU uses CPUID.5.EDX[31:20], the reserved area. >>>> >>>> You mean that powertop is not correctly reading the states? Do you >>>> mean that the BIOS reports better that the CPU ? >>>> >>>> -------------------------- >>>> *Mahmood Naderan* >>>> >>>> >>>> >>>> On Sat, Jun 20, 2009 at 7:26 AM, Edward Shao = wrote: >>>>> >>>>> Hi Mahmood, >>>>> >>>>> So far as I know, if your CPU is Intel, powertop maybe cause a mi= sleading. >>>>> Powertop uses CPUID.5.EDX to check CPU supported C states. >>>>> According Intel's spec (Intel Processor Identification and the CP= UID >>>>> Instruction (Application Note 485), a CPU only has >>>>> >>>>> C0, C1, C2, C3, C4 (prior Core i7) >>>>> C0, C1, C2, C6, C7 (Core i7 and subsequent) >>>>> >>>>> Check with source code of powertop >>>>> while (edx) { >>>>> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0if (edx&7) >>>>> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0printf("C%i ", i); >>>>> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0edx =3D edx >> 4; >>>>> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0i++; >>>>> =A0 =A0 =A0 =A0} >>>>> Only simple check value of edx. >>>>> Well, unless CPU uses CPUID.5.EDX[31:20], the reserved area. >>>>> >>>>> Ps: I does not check AMD's spec. >>>>> >>>>> Best Regards, >>>>> Edward >>>>> >>>>> >>>>> On Sat, Jun 20, 2009 at 1:07 AM, Mahmood Naderan wrote: >>>>> > Hi, >>>>> > I have run powertop with root permission with both AC and batte= r powers. The >>>>> > reported C_States are the same and strange: >>>>> > >>>>> > mahmood@magma:~$ sudo powertop >>>>> > PowerTOP 1.11 =A0 (C) 2007, 2008 Intel Corporation >>>>> > >>>>> > Collecting data for 5 seconds >>>>> > >>>>> > Your CPU supports the following C-states : C1 C2 C3 C4 C5 C6 >>>>> > Your BIOS reports the following C-states : C1 C2 C6 >>>>> > >>>>> > Where are C4 and C5? >>>>> > >>>>> > -------------------------- >>>>> > *Mahmood Naderan* >>>>> > -- >>>>> > To unsubscribe from this list: send the line "unsubscribe linux= -acpi" in >>>>> > the body of a message to majordomo@vger.kernel.org >>>>> > More majordomo info at =A0http://vger.kernel.org/majordomo-info= =2Ehtml >>>>> > >>>>> >>>>> >>>>> >>>>> -- >>>>> Best Regards, >>>>> Edward >>>> -- >>>> To unsubscribe from this list: send the line "unsubscribe linux-ac= pi" in >>>> the body of a message to majordomo@vger.kernel.org >>>> More majordomo info at =A0http://vger.kernel.org/majordomo-info.ht= ml >>>> >>> >>> >>> >>> -- >>> Best Regards, >>> Edward >>> >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-acpi= " in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at =A0http://vger.kernel.org/majordomo-info.html >> > > > > -- > Best Regards, > Edward > -- > To unsubscribe from this list: send the line "unsubscribe linux-acpi"= in > the body of a message to majordomo@vger.kernel.org > More majordomo info at =A0http://vger.kernel.org/majordomo-info.html > -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html