[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20090430091746.GA2839@elte.hu>
Date: Thu, 30 Apr 2009 11:17:46 +0200
From: Ingo Molnar <mingo@...e.hu>
To: Thomas Renninger <trenn@...e.de>
Cc: djwong@...ibm.com, Len Brown <lenb@...nel.org>,
Matthew Garrett <mjg59@...f.ucam.org>,
linux-kernel <linux-kernel@...r.kernel.org>,
cpufreq@...r.kernel.org, linux-acpi@...r.kernel.org
Subject: Re: [PATCH v2] acpi: Fix regression where _PPC is not read at boot
even when ignore_ppc=0
* Thomas Renninger <trenn@...e.de> wrote:
> On Wednesday 29 April 2009 11:43:12 pm Darrick J. Wong wrote:
> > On Wed, Apr 29, 2009 at 03:19:28PM +0200, Thomas Renninger wrote:
> > > Ingo, did you do a BIOS update in between?
> > > The DSDT and SSDTs you added do not contain a _PPC function.
> > > These ACPI tables definitely do not suffer from an initial _PPC
> > > evaluation.
> >
> > P-state tables are probably loaded from separate SSDT tables "CPU0IST" and
> > "CPU1IST". You can download 'em with something like this:
> >
> > # dmesg | grep IST
> > ACPI: SSDT BFD30566, 01FB (r1 PmRef CPU0IST 3000 INTL 20030224)
> > ^addr ^length
> >
> > # acpidump --addr 0xBFD30566 --length 0x1FB > cpu0ist
> > # iasl -d cpu0ist
> >
> > Then paste cpu0ist.dsl somewhere.
>
> Oops forgot about the tables loaded at runtime, I take back the definitely :)
> Ingo, can you also do:
> acpidump --addr 0x7F6E609E --length 0x1ED >cpu0ist
> acpidump --addr 0x7F6E628B --length 0x94 >cpu1ist
> acpidump --addr 0x7F6E5DE1 --length 0x238 >cpu0cst
> acpidump --addr 0x7F6E6019 --length 0x85 >cpu1cst
sure - attached.
Ingo
Download attachment "cpu0ist" of type "application/octet-stream" (493 bytes)
Download attachment "cpu1ist" of type "application/octet-stream" (148 bytes)
Download attachment "cpu0cst" of type "application/octet-stream" (568 bytes)
Download attachment "cpu1cst" of type "application/octet-stream" (133 bytes)
Powered by blists - more mailing lists