Re: [BUG? -rc7] SMP: Just one CPU activated: P4 3GHz HT
Il Mon, Apr 23, 2007 at 10:11:52AM +0200, Miguel Ojeda ha scritto: > On 4/22/07, Luca Tettamanti <[EMAIL PROTECTED]> wrote: > >Probably the two siblings are enumerated only in ACPI tables. If you > >disable ACPI the kernel won't be aware of the second "core". > > > >Luca > >-- > >"Su cio` di cui non si puo` parlare e` bene tacere". > > Ludwig Wittgenstein > > > > On 4/23/07, Robert Hancock <[EMAIL PROTECTED]> wrote: > > > >You didn't enable ACPI, it's needed for almost all systems to detect HT > >and also for many systems to detect multi-cores as well. > > > >Aside from that, in general I would say that on any modern x86 system > >ACPI should always be enabled. In many cases it seems the BIOS code is > >not tested much without ACPI anymore, so going without ACPI can be > >problematic. > > > >-- > >Robert Hancock Saskatoon, SK, Canada > >To email, remove "nospam" from [EMAIL PROTECTED] > >Home Page: http://www.roberthancock.com/ > > > > > > Thanks all of you. It seems the problem is ACPI, however, should not > it be noted in some way? Maybe at Documentation/smp.txt, maybe at the > help section at CONFIG_SMP, maybe a "depends on ..." > > Or maybe CONFIG_SMP should enable some bits of the ACPI code needed to > detect such additional cores. "depends on" is too strong. There exists architecture without ACPI but with SMP support; even in the x86 world a number of systems still support both MPS and ACPI. Mentioning it in the help text should be enough. Roman, are you in charge for the KConfig files? --- SMP support in modern systems often relies on ACPI being active; add a couple of lines in the help text for CONFIG_SMP Signed-Off-By: Luca Tettamanti <[EMAIL PROTECTED]> --- arch/i386/Kconfig |3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 53d6237..48ef899 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -111,6 +111,9 @@ config SMP Y to "Enhanced Real Time Clock Support", below. The "Advanced Power Management" code will be disabled if you say Y here. + Note that ACPI is often required to support newer multi-core CPUs + and HT on Intel P4 processors. + See also the , , and the SMP-HOWTO available at Luca -- "Vorrei morire ucciso dagli agi. Vorrei che di me si dicesse: ``Com'� morto?'' ``Gli � scoppiato il portafogli''" -- Marcello Marchesi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [BUG? -rc7] SMP: Just one CPU activated: P4 3GHz HT
On 4/22/07, Luca Tettamanti <[EMAIL PROTECTED]> wrote: Probably the two siblings are enumerated only in ACPI tables. If you disable ACPI the kernel won't be aware of the second "core". Luca -- "Su cio` di cui non si puo` parlare e` bene tacere". Ludwig Wittgenstein On 4/23/07, Robert Hancock <[EMAIL PROTECTED]> wrote: You didn't enable ACPI, it's needed for almost all systems to detect HT and also for many systems to detect multi-cores as well. Aside from that, in general I would say that on any modern x86 system ACPI should always be enabled. In many cases it seems the BIOS code is not tested much without ACPI anymore, so going without ACPI can be problematic. -- Robert Hancock Saskatoon, SK, Canada To email, remove "nospam" from [EMAIL PROTECTED] Home Page: http://www.roberthancock.com/ Thanks all of you. It seems the problem is ACPI, however, should not it be noted in some way? Maybe at Documentation/smp.txt, maybe at the help section at CONFIG_SMP, maybe a "depends on ..." Or maybe CONFIG_SMP should enable some bits of the ACPI code needed to detect such additional cores. -- Miguel Ojeda http://maxextreme.googlepages.com/index.htm - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [BUG? -rc7] SMP: Just one CPU activated: P4 3GHz HT
Miguel Ojeda wrote: Hi all, I have a ASUS P4P800 Deluxe, P4 3GHz HT, 1GB RAM and testing -rc7 I noticed I just got 1 CPU. I checked my .config but I do not see anything bad. Also I read Documentation/smp.txt. Just for being sure this is not a bug, I'm posting it. Here you have .config and dmesg. You didn't enable ACPI, it's needed for almost all systems to detect HT and also for many systems to detect multi-cores as well. Aside from that, in general I would say that on any modern x86 system ACPI should always be enabled. In many cases it seems the BIOS code is not tested much without ACPI anymore, so going without ACPI can be problematic. -- Robert Hancock Saskatoon, SK, Canada To email, remove "nospam" from [EMAIL PROTECTED] Home Page: http://www.roberthancock.com/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [BUG? -rc7] SMP: Just one CPU activated: P4 3GHz HT
Miguel Ojeda <[EMAIL PROTECTED]> ha scritto: > On 4/22/07, oliver pinter <[EMAIL PROTECTED]> wrote: >> I have CC the kernel > > Did you? > >> and the result is: >> >> your'e config : failed smp >> >> your'e config + power managment : ok >> > > Interesting, can anyone tell me why does not SMP work without Power > Managment? Should it be that way? Probably the two siblings are enumerated only in ACPI tables. If you disable ACPI the kernel won't be aware of the second "core". Luca -- "Su cio` di cui non si puo` parlare e` bene tacere". Ludwig Wittgenstein - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [BUG? -rc7] SMP: Just one CPU activated: P4 3GHz HT
On 4/22/07, oliver pinter <[EMAIL PROTECTED]> wrote: I have CC the kernel Did you? and the result is: your'e config : failed smp your'e config + power managment : ok Interesting, can anyone tell me why does not SMP work without Power Managment? Should it be that way? -- Miguel Ojeda http://maxextreme.googlepages.com/index.htm - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [BUG? -rc7] SMP: Just one CPU activated: P4 3GHz HT
On 4/22/07, oliver pinter <[EMAIL PROTECTED]> wrote: Hello I have ASUS P4C800, P4 3GHz HT (Nortwood), 512RAM and I have nothing errors in my cpu. in bios have you HT option enabled Sure, and the hardware is fine as other kernels and other systems (as WinXP) work fine. or im =<2.6.21-rc7 functioniert? ? The only big difference I see is that you did not set CONFIG_PREEMPT. However, that should not matter. Any other general options are almost the same. (please CC linux-kernel) thanks, -- Miguel Ojeda http://maxextreme.googlepages.com/index.htm - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/