Boot problems with 2.6.29.1 - cannot smoothly boot on battery

Boot problems with 2.6.29.1 - cannot smoothly boot on battery

[Chris Howie]

Folks:

I've been running 2.6.29.1 for some time and reported an issue to this
mailing list a while ago about the boot process.  I believe I have now
nailed down the one factor that determines whether I see this issue.

The problem arises only when booting or halting my laptop when there
is no power plugged in, e.g. the system is running on battery.
Sometime after init begins starting services I must hold down a key on
the keyboard to convince it to continue booting, otherwise it just
hangs.  If the system is running on AC then it boots fine.  This same
problem repeats itself during the halt process if running on battery.

This also happens when resuming a suspend-to-disk image, after loading
the data from the hard drive, but before control is returned to
userspace.  The keyboard has no effect on this hang, presumably
because the kernel is not listening to the keyboard at this point.
When this happens the only solution is to hold the power button and
boot normally instead.

An issue like this is a serious usability problem for me.  I am
neither a kernel developer nor hacker, my coding skills applying
mostly to userspace applications.  If there is anything you need from
me to help diagnose this problem please let me know.

I am not subscribed so please CC me on replies.

Thanks all for your time,

-- 
Chris Howie
http://www.chrishowie.com
http://en.wikipedia.org/wiki/User:Crazycomputers

Re: Boot problems with 2.6.29.1 - cannot smoothly boot on battery

[Rafael J. Wysocki]

On Monday 18 May 2009, Chris Howie wrote:
> Folks:
> 
> I've been running 2.6.29.1 for some time and reported an issue to this
> mailing list a while ago about the boot process.  I believe I have now
> nailed down the one factor that determines whether I see this issue.
> 
> The problem arises only when booting or halting my laptop when there
> is no power plugged in, e.g. the system is running on battery.
> Sometime after init begins starting services I must hold down a key on
> the keyboard to convince it to continue booting, otherwise it just
> hangs.  If the system is running on AC then it boots fine.  This same
> problem repeats itself during the halt process if running on battery.
> 
> This also happens when resuming a suspend-to-disk image, after loading
> the data from the hard drive, but before control is returned to
> userspace.  The keyboard has no effect on this hang, presumably
> because the kernel is not listening to the keyboard at this point.
> When this happens the only solution is to hold the power button and
> boot normally instead.
> 
> An issue like this is a serious usability problem for me.  I am
> neither a kernel developer nor hacker, my coding skills applying
> mostly to userspace applications.  If there is anything you need from
> me to help diagnose this problem please let me know.
> 
> I am not subscribed so please CC me on replies.

Let's try to make your report a bit more visible (some CCs added).

Best,
Rafael

Re: Boot problems with 2.6.29.1 - cannot smoothly boot on battery

[H. Peter Anvin]

Rafael J. Wysocki wrote:
> On Monday 18 May 2009, Chris Howie wrote:
>> Folks:
>>
>> I've been running 2.6.29.1 for some time and reported an issue to this
>> mailing list a while ago about the boot process.  I believe I have now
>> nailed down the one factor that determines whether I see this issue.
>>
>> The problem arises only when booting or halting my laptop when there
>> is no power plugged in, e.g. the system is running on battery.
>> Sometime after init begins starting services I must hold down a key on
>> the keyboard to convince it to continue booting, otherwise it just
>> hangs.  If the system is running on AC then it boots fine.  This same
>> problem repeats itself during the halt process if running on battery.
>>
>> This also happens when resuming a suspend-to-disk image, after loading
>> the data from the hard drive, but before control is returned to
>> userspace.  The keyboard has no effect on this hang, presumably
>> because the kernel is not listening to the keyboard at this point.
>> When this happens the only solution is to hold the power button and
>> boot normally instead.
>>
>> An issue like this is a serious usability problem for me.  I am
>> neither a kernel developer nor hacker, my coding skills applying
>> mostly to userspace applications.  If there is anything you need from
>> me to help diagnose this problem please let me know.
>>
>> I am not subscribed so please CC me on replies.
> 
> Let's try to make your report a bit more visible (some CCs added).
> 

I presume it boots with "acpi=off"?

	-hpa

-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.

Re: Boot problems with 2.6.29.1 - cannot smoothly boot on battery

[yakui_zhao]

On Mon, 2009-05-25 at 05:20 +0800, H. Peter Anvin wrote:
> Rafael J. Wysocki wrote:
> > On Monday 18 May 2009, Chris Howie wrote:
> >> Folks:
> >>
> >> I've been running 2.6.29.1 for some time and reported an issue to this
> >> mailing list a while ago about the boot process.  I believe I have now
> >> nailed down the one factor that determines whether I see this issue.
> >>
> >> The problem arises only when booting or halting my laptop when there
> >> is no power plugged in, e.g. the system is running on battery.
> >> Sometime after init begins starting services I must hold down a key on
> >> the keyboard to convince it to continue booting, otherwise it just
> >> hangs.  If the system is running on AC then it boots fine.  This same
> >> problem repeats itself during the halt process if running on battery.
> >>
> >> This also happens when resuming a suspend-to-disk image, after loading
> >> the data from the hard drive, but before control is returned to
> >> userspace.  The keyboard has no effect on this hang, presumably
> >> because the kernel is not listening to the keyboard at this point.
> >> When this happens the only solution is to hold the power button and
> >> boot normally instead.
> >>
> >> An issue like this is a serious usability problem for me.  I am
> >> neither a kernel developer nor hacker, my coding skills applying
> >> mostly to userspace applications.  If there is anything you need from
> >> me to help diagnose this problem please let me know.
> >>
> >> I am not subscribed so please CC me on replies.
> > 
> > Let's try to make your report a bit more visible (some CCs added).
> > 
> 
> I presume it boots with "acpi=off"?
How about using the following boot option?
   a. idle=poll
   b. nolapic_timer
   c. processor.max_cstate=1

If the box can be booted with the above option, please open a new bug at
http://bugzilla.kernel.org/enter_bug.cgi?product=ACPI and attach the
output of acpidump, lspci -xxx.

Thanks.
> 
> 	-hpa
> 

Re: Boot problems with 2.6.29.1 - cannot smoothly boot on battery

[Shaohua Li]

On Mon, May 25, 2009 at 05:20:09AM +0800, H. Peter Anvin wrote:
> Rafael J. Wysocki wrote:
> > On Monday 18 May 2009, Chris Howie wrote:
> >> Folks:
> >>
> >> I've been running 2.6.29.1 for some time and reported an issue to this
> >> mailing list a while ago about the boot process.  I believe I have now
> >> nailed down the one factor that determines whether I see this issue.
> >>
> >> The problem arises only when booting or halting my laptop when there
> >> is no power plugged in, e.g. the system is running on battery.
> >> Sometime after init begins starting services I must hold down a key on
> >> the keyboard to convince it to continue booting, otherwise it just
> >> hangs.  If the system is running on AC then it boots fine.  This same
> >> problem repeats itself during the halt process if running on battery.
> >>
> >> This also happens when resuming a suspend-to-disk image, after loading
> >> the data from the hard drive, but before control is returned to
> >> userspace.  The keyboard has no effect on this hang, presumably
> >> because the kernel is not listening to the keyboard at this point.
> >> When this happens the only solution is to hold the power button and
> >> boot normally instead.
> >>
> >> An issue like this is a serious usability problem for me.  I am
> >> neither a kernel developer nor hacker, my coding skills applying
> >> mostly to userspace applications.  If there is anything you need from
> >> me to help diagnose this problem please let me know.
> >>
> >> I am not subscribed so please CC me on replies.
> > 
> > Let's try to make your report a bit more visible (some CCs added).
> > 
> 
> I presume it boots with "acpi=off"?
if this is a AMD CPU, please check http://bugzilla.kernel.org/show_bug.cgi?id=13233

Re: Boot problems with 2.6.29.1 - cannot smoothly boot on battery

[Chris Howie]

On Sun, May 24, 2009 at 9:18 PM, Shaohua Li <shaohua.li@intel.com> wrote:
> if this is a AMD CPU, please check http://bugzilla.kernel.org/show_bug.cgi?id=13233

Ahh, that sounds like it.  I first noticed the problem in 2.6.28 after
switching from 2.6.26 (but never tried 2.6.27).  So the information
about versions is indeed consistent.  I'm on an amd64 cpu.

Thanks everyone for the suggestions, I will try to take a look within
the next couple of days and report my findings.

-- 
Chris Howie
http://www.chrishowie.com
http://en.wikipedia.org/wiki/User:Crazycomputers

Re: Boot problems with 2.6.29.1 - cannot smoothly boot on battery

[Chris Howie]

On Sun, May 24, 2009 at 9:53 PM, Chris Howie <cdhowie@gmail.com> wrote:
> On Sun, May 24, 2009 at 9:18 PM, Shaohua Li <shaohua.li@intel.com> wrote:
>> if this is a AMD CPU, please check http://bugzilla.kernel.org/show_bug.cgi?id=13233
>
> Ahh, that sounds like it.  I first noticed the problem in 2.6.28 after
> switching from 2.6.26 (but never tried 2.6.27).  So the information
> about versions is indeed consistent.  I'm on an amd64 cpu.
>
> Thanks everyone for the suggestions, I will try to take a look within
> the next couple of days and report my findings.

The bug linked was in fact my problem, and after forward-porting the
patch to 2.6.29.4 (which required manual patching for one hunk --
luckily this was straightforward) my system boots perfectly on battery
and AC.

Many thanks all, this was the only issue I've had with the latest
kernels.  I'm very happy that it's out of the way!

Cheers,
-- 
Chris Howie
http://www.chrishowie.com
http://en.wikipedia.org/wiki/User:Crazycomputers