Re: Time Problems with 2.6.23-rc1-gf695baf2

[Eric Sesterhenn / Snakebyte <snakebyte@gmx.de>]

* Michal Piotrowski (michal.k.k.piotrowski@gmail.com) wrote:
> Hi Eric,
> 
> On 26/07/07, Eric Sesterhenn / Snakebyte <snakebyte@gmx.de> wrote:
> > * Len Brown (lenb@kernel.org) wrote:
> > > > > > > [   13.506890] ACPI Exception (processor_throttling-0084): AE_NOT_FOUND, Evaluating _PTC [20070126]
> > > > > > > [   13.507101] ACPI Exception (processor_throttling-0147): AE_NOT_FOUND, Evaluating _TSS [20070126]
> > >
> > > Note that these are just noise -- new code being verbose when looking for an optional feature.
> > >
> > > The fact that hitting the power button a bunch of times
> > > to make the system move along suggests some sort of missing interrupt problem --
> > > most likely the timer itself.
> > >
> > > [   13.868574] Probing IDE interface ide0...
> > > [  387.279576] Clocksource tsc unstable (delta = 370195339890 ns)
> > >
> > > 5-minutes -- a long probe:-)
> > >
> > > > CONFIG_NO_HZ=y
> > >
> > > does CONFIG_NO_HZ=n make a difference?
> >
> > [   41.007654] EXT3 FS on hda1, internal journal
> > [  322.133656] Clocksource tsc unstable (delta = 276476174785 ns)
> > Boot went fine but the system got pretty unresponsive later, 2-3 seconds
> > delay after keypresses on an idle system and a hang during shutdown which i had to resolve by
> > pressing the power button (not to switch it of the hard way, but to keep it rebooting)
> >
> > > > CONFIG_HIGH_RES_TIMERS=y
> > >
> > > does CONFIG_HIGH_RES_TIMERS=n make a difference?
> >
> > doesnt change anything
> >
> > > does "irqpoll" make any difference?
> > > does "notsc" make any difference?
> > > does "idle=poll" make any difference?
> >
> > I tried these with the HIGH_RES_TIMERS=n, irqpoll and notsc dont change
> > a thing, idle=poll makes it boot normally
> 
> Please use git-bisect
> 
> git-bisect start
> git-bisect bad
> git-bisect good 7dcca30a32aadb0520417521b0c44f42d09fe05c


took some time, but i got a scape goat, added venkatesh to the CC list,

greetings, Eric

b8550397f1f3d4b8b2c5257c88dae25d58ed is first bad commit
commit 18eab8550397f1f3d4b8b2c5257c88dae25d58ed
Author: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Date:   Fri Jun 15 19:37:00 2007 -0400

    ACPI: Enable C3 even when PM2_control is zero
    
    On systems that do not have pm2_control_block, we cannot really use
    ARB_DISABLE before C3. We used to disable C3 totally on such systems.
    
    To be compatible with Windows, we need to enable C3 on such systems now.
    We just skip ARB_DISABLE step before entering the C3-state and assume
    hardware is handling things correctly. Also, ACPI spec is not clear
    about pm2_control is _needed_ for C3 or not.
    
    We have atleast one system that need this to enable C3.
    
    Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
    Signed-off-by: Len Brown <len.brown@intel.com>