[REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Sitsofe Wheeler]

<posted & mailed>

When using a 32 bit linux-next-20080526 the bootup process will hang at a
random point (not even sysrq helps) with no additional output on the screen
(whereas linux-next-20080523 did boot). Mysteriously, booting with
nmi_watchdog=2 allows the boot to finish (booting with nmi_watchdog=1 still
stalls). I have bisected it down to commit
[d1b946b97d71423f365fa797d1428e1847c0bec1]:

commit d1b946b97d71423f365fa797d1428e1847c0bec1
Author: Cyrill Gorcunov <gorcunov@gmail.com>
Date:   Sat May 24 19:36:34 2008 +0400

    x86: nmi_32.c - add "panic" option

    Allow to pass "panic" option in 32bit mode

    Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
    Cc: hpa@zytor.com
    Cc: mingo@redhat.com
    Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Reverting this seemed to allow the boot to proceed without issue. Here is
the bisection log:

# bad: [5ffeee1785c30303007b3d2dce85402b0bbf9b00] Add linux-next specific
files
# good: [7b6c98f0853b86c101948f1bf4e0a21d3065944f] Add linux-next specific
files
git-bisect start 'next-20080526' 'next-20080523'
# bad: [cab464e32db5aef80c51831c20009ae7c3354396] Merge
commit 'x86/auto-x86-next'
git-bisect bad cab464e32db5aef80c51831c20009ae7c3354396
# good: [497d7294d4fb1a0d2ce14bb891a1d812b2af3745] Merge
branch 'x86/mpparse' into auto-x86-next
git-bisect good 497d7294d4fb1a0d2ce14bb891a1d812b2af3745
# good: [8315d6949d94b7513f8829b62d59d7d5699872dd] USB: driver for gotemp
device
git-bisect good 8315d6949d94b7513f8829b62d59d7d5699872dd
# good: [289c79a4bd350e8a25065102563ad1a183d1b402] vlan: Use bitmask of
feature flags instead of seperate feature bits
git-bisect good 289c79a4bd350e8a25065102563ad1a183d1b402
# bad: [c791e51837f26cba073d3c37cf04803bb3a5c6c0] Merge branch 'x86/timers'
into auto-x86-next
git-bisect bad c791e51837f26cba073d3c37cf04803bb3a5c6c0
# bad: [4039cbaacf7874ab1effd0e9fb6c60bf9217f373] Merge branch 'x86/nmi'
into auto-x86-next
git-bisect bad 4039cbaacf7874ab1effd0e9fb6c60bf9217f373
# bad: [4b82b277707a39b97271439c475f186f63ec4692] x86: nmi_32.c - add
nmi_watchdog_default helper
git-bisect bad 4b82b277707a39b97271439c475f186f63ec4692
# good: [e56b3a12c45e439169f2d1f7194be397667320a6] x86: nmi - die_nmi()
output message unification
git-bisect good e56b3a12c45e439169f2d1f7194be397667320a6
# bad: [d1b946b97d71423f365fa797d1428e1847c0bec1] x86: nmi_32.c -
add "panic" option
git-bisect bad d1b946b97d71423f365fa797d1428e1847c0bec1
# good: [c6425b9f143a75bbcd0a7684b4df40e20d0b2f32] x86: move do_nmi(),
stop_nmi() and restart_nmi() to traps_64.c
git-bisect good c6425b9f143a75bbcd0a7684b4df40e20d0b2f32

-- 
Sitsofe | http://sucs.org/~sits/

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Cyrill Gorcunov]

[Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
| <posted & mailed>
| 
| When using a 32 bit linux-next-20080526 the bootup process will hang at a
| random point (not even sysrq helps) with no additional output on the screen
| (whereas linux-next-20080523 did boot). Mysteriously, booting with
| nmi_watchdog=2 allows the boot to finish (booting with nmi_watchdog=1 still
| stalls). I have bisected it down to commit
| [d1b946b97d71423f365fa797d1428e1847c0bec1]:
| 
| commit d1b946b97d71423f365fa797d1428e1847c0bec1
| Author: Cyrill Gorcunov <gorcunov@gmail.com>
| Date:   Sat May 24 19:36:34 2008 +0400
| 
|     x86: nmi_32.c - add "panic" option
| 
|     Allow to pass "panic" option in 32bit mode
| 
|     Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
|     Cc: hpa@zytor.com
|     Cc: mingo@redhat.com
|     Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| 
| Reverting this seemed to allow the boot to proceed without issue. Here is
| the bisection log:
| 

Hi, so it helps by reverting _only_ that commit? I mean all further commits
are still appiled?

		- Cyrill -

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Cyrill Gorcunov]

[Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
| <posted & mailed>
| 
| When using a 32 bit linux-next-20080526 the bootup process will hang at a
| random point (not even sysrq helps) with no additional output on the screen
| (whereas linux-next-20080523 did boot). Mysteriously, booting with
| nmi_watchdog=2 allows the boot to finish (booting with nmi_watchdog=1 still
| stalls). I have bisected it down to commit
| [d1b946b97d71423f365fa797d1428e1847c0bec1]:
| 
| commit d1b946b97d71423f365fa797d1428e1847c0bec1
| Author: Cyrill Gorcunov <gorcunov@gmail.com>
| Date:   Sat May 24 19:36:34 2008 +0400
| 
|     x86: nmi_32.c - add "panic" option
| 
|     Allow to pass "panic" option in 32bit mode
| 
|     Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
|     Cc: hpa@zytor.com
|     Cc: mingo@redhat.com
|     Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
| 
| Reverting this seemed to allow the boot to proceed without issue. Here is
| the bisection log:
| 
| # bad: [5ffeee1785c30303007b3d2dce85402b0bbf9b00] Add linux-next specific
| files
| # good: [7b6c98f0853b86c101948f1bf4e0a21d3065944f] Add linux-next specific
| files
| git-bisect start 'next-20080526' 'next-20080523'
| # bad: [cab464e32db5aef80c51831c20009ae7c3354396] Merge
| commit 'x86/auto-x86-next'
| git-bisect bad cab464e32db5aef80c51831c20009ae7c3354396
| # good: [497d7294d4fb1a0d2ce14bb891a1d812b2af3745] Merge
| branch 'x86/mpparse' into auto-x86-next
| git-bisect good 497d7294d4fb1a0d2ce14bb891a1d812b2af3745
| # good: [8315d6949d94b7513f8829b62d59d7d5699872dd] USB: driver for gotemp
| device
| git-bisect good 8315d6949d94b7513f8829b62d59d7d5699872dd
| # good: [289c79a4bd350e8a25065102563ad1a183d1b402] vlan: Use bitmask of
| feature flags instead of seperate feature bits
| git-bisect good 289c79a4bd350e8a25065102563ad1a183d1b402
| # bad: [c791e51837f26cba073d3c37cf04803bb3a5c6c0] Merge branch 'x86/timers'
| into auto-x86-next
| git-bisect bad c791e51837f26cba073d3c37cf04803bb3a5c6c0
| # bad: [4039cbaacf7874ab1effd0e9fb6c60bf9217f373] Merge branch 'x86/nmi'
| into auto-x86-next
| git-bisect bad 4039cbaacf7874ab1effd0e9fb6c60bf9217f373
| # bad: [4b82b277707a39b97271439c475f186f63ec4692] x86: nmi_32.c - add
| nmi_watchdog_default helper
| git-bisect bad 4b82b277707a39b97271439c475f186f63ec4692
| # good: [e56b3a12c45e439169f2d1f7194be397667320a6] x86: nmi - die_nmi()
| output message unification
| git-bisect good e56b3a12c45e439169f2d1f7194be397667320a6
| # bad: [d1b946b97d71423f365fa797d1428e1847c0bec1] x86: nmi_32.c -
| add "panic" option
| git-bisect bad d1b946b97d71423f365fa797d1428e1847c0bec1
| # good: [c6425b9f143a75bbcd0a7684b4df40e20d0b2f32] x86: move do_nmi(),
| stop_nmi() and restart_nmi() to traps_64.c
| git-bisect good c6425b9f143a75bbcd0a7684b4df40e20d0b2f32
| 
| -- 
| Sitsofe | http://sucs.org/~sits/
| 
| 

and, btw, could you post your config, please?

		- Cyrill -

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Sitsofe Wheeler]

<posted & mailed>

Cyrill Gorcunov wrote:

> [Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
> | When using a 32 bit linux-next-20080526 the bootup process will hang at
> | a random point (not even sysrq helps) with no additional output on the
> | screen (whereas linux-next-20080523 did boot). Mysteriously, booting
> | with nmi_watchdog=2 allows the boot to finish (booting with
> | nmi_watchdog=1 still stalls). I have bisected it down to commit
> | [d1b946b97d71423f365fa797d1428e1847c0bec1]:
>
> Hi, so it helps by reverting only that commit? I mean all further commits
> are still appiled?

Ah that I hadn't tested. I believe I might need to revert
4b82b277707a39b97271439c475f186f63ec4692 too if later commits are applied
(but I'm still testing)

> and, btw, could you post your config, please?

http://sucs.org/~sits/test/config-20080526.txt

-- 
Sitsofe | http://sucs.org/~sits/

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Sitsofe Wheeler]

<posted & mailed>

Sitsofe Wheeler wrote:

> Cyrill Gorcunov wrote:
> 
>> [Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
>> | When using a 32 bit linux-next-20080526 the bootup process will hang at
>> | a random point (not even sysrq helps) with no additional output on the
>> | screen (whereas linux-next-20080523 did boot). Mysteriously, booting
>> | with nmi_watchdog=2 allows the boot to finish (booting with
>> | nmi_watchdog=1 still stalls). I have bisected it down to commit
>> | [d1b946b97d71423f365fa797d1428e1847c0bec1]:
>>
>> Hi, so it helps by reverting only that commit? I mean all further commits
>> are still appiled?
> 
> Ah that I hadn't tested. I believe I might need to revert
> 4b82b277707a39b97271439c475f186f63ec4692 too if later commits are applied
> (but I'm still testing)
> 
>> and, btw, could you post your config, please?
> 
> http://sucs.org/~sits/test/config-20080526.txt

OK applying the following patch (which is more or less a revert of 
[4b82b277707a39b97271439c475f186f63ec4692]) resolves the problem:

diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index d99ee8a..c55519c 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -480,8 +480,12 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
                return -EIO;
        }
 
-       /* if nmi_watchdog is not set yet, then set it */
-       nmi_watchdog_default();
+       if (nmi_watchdog == NMI_DEFAULT) {
+               if (lapic_watchdog_ok())
+                       nmi_watchdog = NMI_LOCAL_APIC;
+               else
+                       nmi_watchdog = NMI_IO_APIC;
+       }
 
        if (nmi_watchdog == NMI_LOCAL_APIC) {
                if (nmi_watchdog_enabled)
diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h
index 1e8f34d..7cd5b6a 100644
--- a/include/asm-x86/nmi.h
+++ b/include/asm-x86/nmi.h
@@ -38,9 +38,11 @@ static inline void unset_nmi_pm_callback(struct pm_dev *dev)
 
 #ifdef CONFIG_X86_64
 extern void default_do_nmi(struct pt_regs *);
+extern void nmi_watchdog_default(void);
+#else
+#define nmi_watchdog_default() do {} while (0)
 #endif
 
-extern void nmi_watchdog_default(void);
 extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
 extern int check_nmi_watchdog(void);
 extern int nmi_watchdog_enabled; 

The removal of extern void nmi_watchdog_default(void) and the inclusion
of #define nmi_watchdog_default() do {} while (0) look suspicious (why
would nmi_watchdog_default() need to be an infinite loop on 32 bit
systems?).

-- 
Sitsofe | http://sucs.org/~sits/

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Cyrill Gorcunov]

[Sitsofe Wheeler - Mon, May 26, 2008 at 08:36:54PM +0100]
| <posted & mailed>
| 
| Sitsofe Wheeler wrote:
| 
| > Cyrill Gorcunov wrote:
| > 
| >> [Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
| >> | When using a 32 bit linux-next-20080526 the bootup process will hang at
| >> | a random point (not even sysrq helps) with no additional output on the
| >> | screen (whereas linux-next-20080523 did boot). Mysteriously, booting
| >> | with nmi_watchdog=2 allows the boot to finish (booting with
| >> | nmi_watchdog=1 still stalls). I have bisected it down to commit
| >> | [d1b946b97d71423f365fa797d1428e1847c0bec1]:
| >>
| >> Hi, so it helps by reverting only that commit? I mean all further commits
| >> are still appiled?
| > 
| > Ah that I hadn't tested. I believe I might need to revert
| > 4b82b277707a39b97271439c475f186f63ec4692 too if later commits are applied
| > (but I'm still testing)
| > 
| >> and, btw, could you post your config, please?
| > 
| > http://sucs.org/~sits/test/config-20080526.txt
| 
| OK applying the following patch (which is more or less a revert of 
| [4b82b277707a39b97271439c475f186f63ec4692]) resolves the problem:
| 
| diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
| index d99ee8a..c55519c 100644
| --- a/arch/x86/kernel/nmi.c
| +++ b/arch/x86/kernel/nmi.c
| @@ -480,8 +480,12 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
|                 return -EIO;
|         }
|  
| -       /* if nmi_watchdog is not set yet, then set it */
| -       nmi_watchdog_default();
| +       if (nmi_watchdog == NMI_DEFAULT) {
| +               if (lapic_watchdog_ok())
| +                       nmi_watchdog = NMI_LOCAL_APIC;
| +               else
| +                       nmi_watchdog = NMI_IO_APIC;
| +       }
|  
|         if (nmi_watchdog == NMI_LOCAL_APIC) {
|                 if (nmi_watchdog_enabled)
| diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h
| index 1e8f34d..7cd5b6a 100644
| --- a/include/asm-x86/nmi.h
| +++ b/include/asm-x86/nmi.h
| @@ -38,9 +38,11 @@ static inline void unset_nmi_pm_callback(struct pm_dev *dev)
|  
|  #ifdef CONFIG_X86_64
|  extern void default_do_nmi(struct pt_regs *);
| +extern void nmi_watchdog_default(void);
| +#else
| +#define nmi_watchdog_default() do {} while (0)
|  #endif
|  
| -extern void nmi_watchdog_default(void);
|  extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
|  extern int check_nmi_watchdog(void);
|  extern int nmi_watchdog_enabled; 
| 
| The removal of extern void nmi_watchdog_default(void) and the inclusion
| of #define nmi_watchdog_default() do {} while (0) look suspicious (why
| would nmi_watchdog_default() need to be an infinite loop on 32 bit
| systems?).
| 
| -- 
| Sitsofe | http://sucs.org/~sits/
| 
| 

Thanks a lot! Will take a look tomorrow!

And there was NOT an infinite loop - look more closer on that string
do { } while (0) - only *one* iteration is going (well, gcc will eliminate
it at all by optimization). Anyway, it was while (0), not while (1),
so it is ok ;)

		- Cyrill -

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Cyrill Gorcunov]

On 5/26/08, Sitsofe Wheeler <sitsofe@yahoo.com> wrote:
> <posted & mailed>
>
> Sitsofe Wheeler wrote:
>
> > Cyrill Gorcunov wrote:
> >
> >> [Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
> >> | When using a 32 bit linux-next-20080526 the bootup process will hang at
> >> | a random point (not even sysrq helps) with no additional output on the
> >> | screen (whereas linux-next-20080523 did boot). Mysteriously, booting
> >> | with nmi_watchdog=2 allows the boot to finish (booting with
> >> | nmi_watchdog=1 still stalls). I have bisected it down to commit
> >> | [d1b946b97d71423f365fa797d1428e1847c0bec1]:
> >>
> >> Hi, so it helps by reverting only that commit? I mean all further commits
> >> are still appiled?
> >
> > Ah that I hadn't tested. I believe I might need to revert
> > 4b82b277707a39b97271439c475f186f63ec4692 too if later commits are applied
> > (but I'm still testing)
> >
> >> and, btw, could you post your config, please?
> >
> > http://sucs.org/~sits/test/config-20080526.txt
>
> OK applying the following patch (which is more or less a revert of
> [4b82b277707a39b97271439c475f186f63ec4692]) resolves the problem:
>
> diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
> index d99ee8a..c55519c 100644
> --- a/arch/x86/kernel/nmi.c
> +++ b/arch/x86/kernel/nmi.c
> @@ -480,8 +480,12 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
>                return -EIO;
>        }
>
> -       /* if nmi_watchdog is not set yet, then set it */
> -       nmi_watchdog_default();
> +       if (nmi_watchdog == NMI_DEFAULT) {
> +               if (lapic_watchdog_ok())
> +                       nmi_watchdog = NMI_LOCAL_APIC;
> +               else
> +                       nmi_watchdog = NMI_IO_APIC;
> +       }
>
>        if (nmi_watchdog == NMI_LOCAL_APIC) {
>                if (nmi_watchdog_enabled)
> diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h
> index 1e8f34d..7cd5b6a 100644
> --- a/include/asm-x86/nmi.h
> +++ b/include/asm-x86/nmi.h
> @@ -38,9 +38,11 @@ static inline void unset_nmi_pm_callback(struct pm_dev *dev)
>
>  #ifdef CONFIG_X86_64
>  extern void default_do_nmi(struct pt_regs *);
> +extern void nmi_watchdog_default(void);
> +#else
> +#define nmi_watchdog_default() do {} while (0)
>  #endif
>
> -extern void nmi_watchdog_default(void);
>  extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
>  extern int check_nmi_watchdog(void);
>  extern int nmi_watchdog_enabled;
>
> The removal of extern void nmi_watchdog_default(void) and the inclusion
> of #define nmi_watchdog_default() do {} while (0) look suspicious (why
> would nmi_watchdog_default() need to be an infinite loop on 32 bit
> systems?).
>
> --
> Sitsofe | http://sucs.org/~sits/
>

Hi Sitsofe, actually I can't imagine how this revert have helped you
from hang. Newly introduced nmi_watchdog_default() didn't change
the logic of the code being used before this merge-series.

Thomas, Ingo, maybe I've overlooked something (hard to believe ;)?
Only the change is the adding of new call for nmi_watchdog_default()
which was say 'inlined' before.

The thread is started on http://lkml.org/lkml/2008/5/26/146

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Sitsofe Wheeler]

<posted & mailed>

Cyrill Gorcunov wrote:

> Hi Sitsofe, actually I can't imagine how this revert have helped you
> from hang. Newly introduced nmi_watchdog_default() didn't change
> the logic of the code being used before this merge-series.

I've just retested again and I'm fairly sure it's that commit. I can do the
following:

git checkout -f 4039cbaacf7874ab1effd0e9fb6c60bf9217f373

(compiled kernel hangs during boot)

git revert 1798bc22b2790bf2a956588e6b17c36ef79ceff7
git revert fd5cea02de100197a4c26d9e103508cf09b50a82
git revert 4b82b277707a39b97271439c475f186f63ec4692

(compiled kernel boots normally)

The last thing I normally see is
ACPI: Sleep Button (CM) [SLPB
during a normal boot is followed by
ACPI: device:03 is registered as cooling_device0
input: Video Bus
as /devices/LNXSYSTM:00/device:00/PNP0A08:00/device:02/input/input3

If you can give me a patch that reverts just
4b82b277707a39b97271439c475f186f63ec4692 again the current HEAD of
linux-next then I will try that too...

-- 
Sitsofe | http://sucs.org/~sits/

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Adrian Bunk]

On Tue, May 27, 2008 at 10:35:43AM +0400, Cyrill Gorcunov wrote:
> On 5/26/08, Sitsofe Wheeler <sitsofe@yahoo.com> wrote:
> > <posted & mailed>
> >
> > Sitsofe Wheeler wrote:
> >
> > > Cyrill Gorcunov wrote:
> > >
> > >> [Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
> > >> | When using a 32 bit linux-next-20080526 the bootup process will hang at
> > >> | a random point (not even sysrq helps) with no additional output on the
> > >> | screen (whereas linux-next-20080523 did boot). Mysteriously, booting
> > >> | with nmi_watchdog=2 allows the boot to finish (booting with
> > >> | nmi_watchdog=1 still stalls). I have bisected it down to commit
> > >> | [d1b946b97d71423f365fa797d1428e1847c0bec1]:
> > >>
> > >> Hi, so it helps by reverting only that commit? I mean all further commits
> > >> are still appiled?
> > >
> > > Ah that I hadn't tested. I believe I might need to revert
> > > 4b82b277707a39b97271439c475f186f63ec4692 too if later commits are applied
> > > (but I'm still testing)
> > >
> > >> and, btw, could you post your config, please?
> > >
> > > http://sucs.org/~sits/test/config-20080526.txt
> >
> > OK applying the following patch (which is more or less a revert of
> > [4b82b277707a39b97271439c475f186f63ec4692]) resolves the problem:
>...
> 
> Hi Sitsofe, actually I can't imagine how this revert have helped you
> from hang. Newly introduced nmi_watchdog_default() didn't change
> the logic of the code being used before this merge-series.
> 
> Thomas, Ingo, maybe I've overlooked something (hard to believe ;)?
> Only the change is the adding of new call for nmi_watchdog_default()
> which was say 'inlined' before.
>...

arch/x86/kernel/smpboot.c:native_smp_prepare_cpus() also calls
the no longer empty nmi_watchdog_default() on 32bit.

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: [REGRESSION][BISECTED][X86] next-20080526 hangs on boot

[Cyrill Gorcunov]

Hi,
thanks a lot for checking this!
I'll take a look today evening.

On 5/28/08, Adrian Bunk <bunk@kernel.org> wrote:
> On Tue, May 27, 2008 at 10:35:43AM +0400, Cyrill Gorcunov wrote:
>> On 5/26/08, Sitsofe Wheeler <sitsofe@yahoo.com> wrote:
>> > <posted & mailed>
>> >
>> > Sitsofe Wheeler wrote:
>> >
>> > > Cyrill Gorcunov wrote:
>> > >
>> > >> [Sitsofe Wheeler - Mon, May 26, 2008 at 03:04:54PM +0100]
>> > >> | When using a 32 bit linux-next-20080526 the bootup process will
>> > >> hang at
>> > >> | a random point (not even sysrq helps) with no additional output on
>> > >> the
>> > >> | screen (whereas linux-next-20080523 did boot). Mysteriously,
>> > >> booting
>> > >> | with nmi_watchdog=2 allows the boot to finish (booting with
>> > >> | nmi_watchdog=1 still stalls). I have bisected it down to commit
>> > >> | [d1b946b97d71423f365fa797d1428e1847c0bec1]:
>> > >>
>> > >> Hi, so it helps by reverting only that commit? I mean all further
>> > >> commits
>> > >> are still appiled?
>> > >
>> > > Ah that I hadn't tested. I believe I might need to revert
>> > > 4b82b277707a39b97271439c475f186f63ec4692 too if later commits are
>> > > applied
>> > > (but I'm still testing)
>> > >
>> > >> and, btw, could you post your config, please?
>> > >
>> > > http://sucs.org/~sits/test/config-20080526.txt
>> >
>> > OK applying the following patch (which is more or less a revert of
>> > [4b82b277707a39b97271439c475f186f63ec4692]) resolves the problem:
>>...
>>
>> Hi Sitsofe, actually I can't imagine how this revert have helped you
>> from hang. Newly introduced nmi_watchdog_default() didn't change
>> the logic of the code being used before this merge-series.
>>
>> Thomas, Ingo, maybe I've overlooked something (hard to believe ;)?
>> Only the change is the adding of new call for nmi_watchdog_default()
>> which was say 'inlined' before.
>>...
>
> arch/x86/kernel/smpboot.c:native_smp_prepare_cpus() also calls
> the no longer empty nmi_watchdog_default() on 32bit.
>
> cu
> Adrian
>
> --
>
>        "Is there not promise of rain?" Ling Tan asked suddenly out
>         of the darkness. There had been need of rain for many days.
>        "Only a promise," Lao Er said.
>                                        Pearl S. Buck - Dragon Seed
>
>