linux-next - multi-omap image fails to boot on omap3/4

linux-next - multi-omap image fails to boot on omap3/4

[Anand Gadiyar]

Hi all,

linux-next, as of 20101019, built with the omap2plus_defconfig fails to 
boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or 
CONFIG_SWP_EMULATE to get the image to build). Building with only 
ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
an image built with only ARCH_OMAP4 boots up on OMAP4 boards.

earlyprintk does not provide any additional prints after
"Uncompressing Linux... done, booting the kernel."

Any ideas where to look?


- Anand

Re: linux-next - multi-omap image fails to boot on omap3/4

[Tony Lindgren]

* Anand Gadiyar <gadiyar@ti.com> [101019 07:41]:
> Hi all,
> 
> linux-next, as of 20101019, built with the omap2plus_defconfig fails
> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or
> CONFIG_SWP_EMULATE to get the image to build). Building with only
> ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
> 
> earlyprintk does not provide any additional prints after
> "Uncompressing Linux... done, booting the kernel."
> 
> Any ideas where to look?

Hmm I did a quick test merge of linux-omap master and rmk/devel
branches and that boots just fine.

So it's probably something that already got fixed in rmk/devel
but is not yet in next, or something that came from elsewhere,
or something that we have in omap-testing branch that's not in
for-next for some reason.

Regards,

Tony

Re: linux-next - multi-omap image fails to boot on omap3/4

[Gadiyar, Anand]

On Tue, Oct 19, 2010 at 11:51 PM, Tony Lindgren <tony@atomide.com> wrote:
> * Anand Gadiyar <gadiyar@ti.com> [101019 07:41]:
>> Hi all,
>>
>> linux-next, as of 20101019, built with the omap2plus_defconfig fails
>> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or
>> CONFIG_SWP_EMULATE to get the image to build). Building with only
>> ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
>> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
>>
>> earlyprintk does not provide any additional prints after
>> "Uncompressing Linux... done, booting the kernel."
>>
>> Any ideas where to look?
>
> Hmm I did a quick test merge of linux-omap master and rmk/devel
> branches and that boots just fine.
>
> So it's probably something that already got fixed in rmk/devel
> but is not yet in next, or something that came from elsewhere,
> or something that we have in omap-testing branch that's not in
> for-next for some reason.

I tried bisecting linux-next between v2.6.36-rc8 and HEAD, but
there's a commit in between that breaks the build. I'll take another
stab at this in a while.

- Anand

Re: linux-next - multi-omap image fails to boot on omap3/4

[Tony Lindgren]

* Gadiyar, Anand <gadiyar@ti.com> [101019 11:26]:
> On Tue, Oct 19, 2010 at 11:51 PM, Tony Lindgren <tony@atomide.com> wrote:
> > * Anand Gadiyar <gadiyar@ti.com> [101019 07:41]:
> >> Hi all,
> >>
> >> linux-next, as of 20101019, built with the omap2plus_defconfig fails
> >> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or
> >> CONFIG_SWP_EMULATE to get the image to build). Building with only
> >> ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
> >> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
> >>
> >> earlyprintk does not provide any additional prints after
> >> "Uncompressing Linux... done, booting the kernel."
> >>
> >> Any ideas where to look?
> >
> > Hmm I did a quick test merge of linux-omap master and rmk/devel
> > branches and that boots just fine.
> >
> > So it's probably something that already got fixed in rmk/devel
> > but is not yet in next, or something that came from elsewhere,
> > or something that we have in omap-testing branch that's not in
> > for-next for some reason.
> 
> I tried bisecting linux-next between v2.6.36-rc8 and HEAD, but
> there's a commit in between that breaks the build. I'll take another
> stab at this in a while.

Looks like current next at 80f8f1f8b33750d954beb386c0c8142d0c01c25c
boots again except on 2430sdp it produces a NULL pointer at ubi_io_write.

Tony

Re: linux-next - multi-omap image fails to boot on omap3/4

[Tony Lindgren]

* Tony Lindgren <tony@atomide.com> [101019 15:48]:
> * Gadiyar, Anand <gadiyar@ti.com> [101019 11:26]:
> > On Tue, Oct 19, 2010 at 11:51 PM, Tony Lindgren <tony@atomide.com> wrote:
> > > * Anand Gadiyar <gadiyar@ti.com> [101019 07:41]:
> > >> Hi all,
> > >>
> > >> linux-next, as of 20101019, built with the omap2plus_defconfig fails
> > >> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or
> > >> CONFIG_SWP_EMULATE to get the image to build). Building with only
> > >> ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
> > >> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
> > >>
> > >> earlyprintk does not provide any additional prints after
> > >> "Uncompressing Linux... done, booting the kernel."
> > >>
> > >> Any ideas where to look?
> > >
> > > Hmm I did a quick test merge of linux-omap master and rmk/devel
> > > branches and that boots just fine.
> > >
> > > So it's probably something that already got fixed in rmk/devel
> > > but is not yet in next, or something that came from elsewhere,
> > > or something that we have in omap-testing branch that's not in
> > > for-next for some reason.
> > 
> > I tried bisecting linux-next between v2.6.36-rc8 and HEAD, but
> > there's a commit in between that breaks the build. I'll take another
> > stab at this in a while.
> 
> Looks like current next at 80f8f1f8b33750d954beb386c0c8142d0c01c25c
> boots again except on 2430sdp it produces a NULL pointer at ubi_io_write.

Oops, sorry it's still broken, wrong tree.

Tony

Re: linux-next - multi-omap image fails to boot on omap3/4

[Gadiyar, Anand]

On Wed, Oct 20, 2010 at 5:02 AM, Tony Lindgren <tony@atomide.com> wrote:
> * Tony Lindgren <tony@atomide.com> [101019 15:48]:
>> * Gadiyar, Anand <gadiyar@ti.com> [101019 11:26]:
>> > On Tue, Oct 19, 2010 at 11:51 PM, Tony Lindgren <tony@atomide.com> wrote:
>> > > * Anand Gadiyar <gadiyar@ti.com> [101019 07:41]:
>> > >> Hi all,
>> > >>
>> > >> linux-next, as of 20101019, built with the omap2plus_defconfig fails
>> > >> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or
>> > >> CONFIG_SWP_EMULATE to get the image to build). Building with only
>> > >> ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
>> > >> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
>> > >>
>> > >> earlyprintk does not provide any additional prints after
>> > >> "Uncompressing Linux... done, booting the kernel."
>> > >>
>> > >> Any ideas where to look?
>> > >
>> > > Hmm I did a quick test merge of linux-omap master and rmk/devel
>> > > branches and that boots just fine.
>> > >
>> > > So it's probably something that already got fixed in rmk/devel
>> > > but is not yet in next, or something that came from elsewhere,
>> > > or something that we have in omap-testing branch that's not in
>> > > for-next for some reason.
>> >
>> > I tried bisecting linux-next between v2.6.36-rc8 and HEAD, but
>> > there's a commit in between that breaks the build. I'll take another
>> > stab at this in a while.
>>
>> Looks like current next at 80f8f1f8b33750d954beb386c0c8142d0c01c25c
>> boots again except on 2430sdp it produces a NULL pointer at ubi_io_write.
>
> Oops, sorry it's still broken, wrong tree.
>

I've started bisecting again - I needed to pick commit 5bac0926121e
(driver core: platform_bus: allow runtime override of dev_pm_ops) to
solve the build break.

Will report back if I find something.

- Anand

Re: linux-next - multi-omap image fails to boot on omap3/4

[Gadiyar, Anand]

On Wed, Oct 20, 2010 at 5:16 AM, Gadiyar, Anand <gadiyar@ti.com> wrote:
> On Wed, Oct 20, 2010 at 5:02 AM, Tony Lindgren <tony@atomide.com> wrote:
>> * Tony Lindgren <tony@atomide.com> [101019 15:48]:
>>> * Gadiyar, Anand <gadiyar@ti.com> [101019 11:26]:
>>> > On Tue, Oct 19, 2010 at 11:51 PM, Tony Lindgren <tony@atomide.com> wrote:
>>> > > * Anand Gadiyar <gadiyar@ti.com> [101019 07:41]:
>>> > >> Hi all,
>>> > >>
>>> > >> linux-next, as of 20101019, built with the omap2plus_defconfig fails
>>> > >> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or
>>> > >> CONFIG_SWP_EMULATE to get the image to build). Building with only
>>> > >> ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
>>> > >> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
>>> > >>
>>> > >> earlyprintk does not provide any additional prints after
>>> > >> "Uncompressing Linux... done, booting the kernel."
>>> > >>
>>> > >> Any ideas where to look?
>>> > >
>>> > > Hmm I did a quick test merge of linux-omap master and rmk/devel
>>> > > branches and that boots just fine.
>>> > >
>>> > > So it's probably something that already got fixed in rmk/devel
>>> > > but is not yet in next, or something that came from elsewhere,
>>> > > or something that we have in omap-testing branch that's not in
>>> > > for-next for some reason.
>>> >
>>> > I tried bisecting linux-next between v2.6.36-rc8 and HEAD, but
>>> > there's a commit in between that breaks the build. I'll take another
>>> > stab at this in a while.
>>>
>>> Looks like current next at 80f8f1f8b33750d954beb386c0c8142d0c01c25c
>>> boots again except on 2430sdp it produces a NULL pointer at ubi_io_write.
>>
>> Oops, sorry it's still broken, wrong tree.
>>
>
> I've started bisecting again - I needed to pick commit 5bac0926121e
> (driver core: platform_bus: allow runtime override of dev_pm_ops) to
> solve the build break.
>
> Will report back if I find something.
>

I tried disabling CONFIG_SMP and CONFIG_SMP_ON_UP, and the resulting
image booted up just fine.

- Anand

Re: linux-next - multi-omap image fails to boot on omap3/4

[Tony Lindgren]

* Gadiyar, Anand <gadiyar@ti.com> [101019 23:21]:
> On Wed, Oct 20, 2010 at 5:16 AM, Gadiyar, Anand <gadiyar@ti.com> wrote:
> > On Wed, Oct 20, 2010 at 5:02 AM, Tony Lindgren <tony@atomide.com> wrote:
> >> * Tony Lindgren <tony@atomide.com> [101019 15:48]:
> >>> * Gadiyar, Anand <gadiyar@ti.com> [101019 11:26]:
> >>> > On Tue, Oct 19, 2010 at 11:51 PM, Tony Lindgren <tony@atomide.com> wrote:
> >>> > > * Anand Gadiyar <gadiyar@ti.com> [101019 07:41]:
> >>> > >> Hi all,
> >>> > >>
> >>> > >> linux-next, as of 20101019, built with the omap2plus_defconfig fails
> >>> > >> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2 or
> >>> > >> CONFIG_SWP_EMULATE to get the image to build). Building with only
> >>> > >> ARCH_OMAP3 allows the resultant image to boot up on OMAP3. Likewise,
> >>> > >> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
> >>> > >>
> >>> > >> earlyprintk does not provide any additional prints after
> >>> > >> "Uncompressing Linux... done, booting the kernel."
> >>> > >>
> >>> > >> Any ideas where to look?
> >>> > >
> >>> > > Hmm I did a quick test merge of linux-omap master and rmk/devel
> >>> > > branches and that boots just fine.
> >>> > >
> >>> > > So it's probably something that already got fixed in rmk/devel
> >>> > > but is not yet in next, or something that came from elsewhere,
> >>> > > or something that we have in omap-testing branch that's not in
> >>> > > for-next for some reason.
> >>> >
> >>> > I tried bisecting linux-next between v2.6.36-rc8 and HEAD, but
> >>> > there's a commit in between that breaks the build. I'll take another
> >>> > stab at this in a while.
> >>>
> >>> Looks like current next at 80f8f1f8b33750d954beb386c0c8142d0c01c25c
> >>> boots again except on 2430sdp it produces a NULL pointer at ubi_io_write.
> >>
> >> Oops, sorry it's still broken, wrong tree.
> >>
> >
> > I've started bisecting again - I needed to pick commit 5bac0926121e
> > (driver core: platform_bus: allow runtime override of dev_pm_ops) to
> > solve the build break.
> >
> > Will report back if I find something.
> >
> 
> I tried disabling CONFIG_SMP and CONFIG_SMP_ON_UP, and the resulting
> image booted up just fine.

Looks like it's caused by this patch:

"ARM: hotplug cpu: Keep processor information, startup code & __lookup_processor_type"

It's probably trashing registers it shouldn't. Will take a look and post
something to linux-arm-kernel.

Tony

RE: linux-next - multi-omap image fails to boot on omap3/4

[Anand Gadiyar]

Tony Lindgren wrote:
> * Gadiyar, Anand <gadiyar@ti.com> [101019 23:21]:

...

> > >>> > >> linux-next, as of 20101019, built with the
omap2plus_defconfig fails
> > >>> > >> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2
or
> > >>> > >> CONFIG_SWP_EMULATE to get the image to build). Building with
only
> > >>> > >> ARCH_OMAP3 allows the resultant image to boot up on OMAP3.
Likewise,
> > >>> > >> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
> > >>> > >>
> > >>> > >> earlyprintk does not provide any additional prints after
> > >>> > >> "Uncompressing Linux... done, booting the kernel."
> > >>> > >>
> > >>> > >> Any ideas where to look?
> > >>> > >

...

> >
> > I tried disabling CONFIG_SMP and CONFIG_SMP_ON_UP, and the resulting
> > image booted up just fine.
>
> Looks like it's caused by this patch:
>
> "ARM: hotplug cpu: Keep processor information, startup code &
> __lookup_processor_type"
>
> It's probably trashing registers it shouldn't. Will take a look and post
> something to linux-arm-kernel.
>

I'm not sure what's going on, but looks like linux-next as of
20101022 boots up fine on panda. I can't explain it, as I couldn't
find anything related to the patch you mentioned.

- Anand

Re: linux-next - multi-omap image fails to boot on omap3/4

[Tony Lindgren]

* Anand Gadiyar <gadiyar@ti.com> [101022 06:59]:
> Tony Lindgren wrote:
> > * Gadiyar, Anand <gadiyar@ti.com> [101019 23:21]:
> 
> ...
> 
> > > >>> > >> linux-next, as of 20101019, built with the
> omap2plus_defconfig fails
> > > >>> > >> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2
> or
> > > >>> > >> CONFIG_SWP_EMULATE to get the image to build). Building with
> only
> > > >>> > >> ARCH_OMAP3 allows the resultant image to boot up on OMAP3.
> Likewise,
> > > >>> > >> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
> > > >>> > >>
> > > >>> > >> earlyprintk does not provide any additional prints after
> > > >>> > >> "Uncompressing Linux... done, booting the kernel."
> > > >>> > >>
> > > >>> > >> Any ideas where to look?
> > > >>> > >
> 
> ...
> 
> > >
> > > I tried disabling CONFIG_SMP and CONFIG_SMP_ON_UP, and the resulting
> > > image booted up just fine.
> >
> > Looks like it's caused by this patch:
> >
> > "ARM: hotplug cpu: Keep processor information, startup code &
> > __lookup_processor_type"
> >
> > It's probably trashing registers it shouldn't. Will take a look and post
> > something to linux-arm-kernel.
> >
> 
> I'm not sure what's going on, but looks like linux-next as of
> 20101022 boots up fine on panda. I can't explain it, as I couldn't
> find anything related to the patch you mentioned.

Yeah, it's an alignment issue that's why it sometimes works depending
on the Kconfig options selected..

For rererence, posted a fix for that at:

https://patchwork.kernel.org/patch/276161/

Tony

Re: linux-next - multi-omap image fails to boot on omap3/4

[Anand Gadiyar]

On 10/22/2010 2:52 PM, Tony Lindgren wrote:
> * Anand Gadiyar<gadiyar@ti.com>  [101022 06:59]:
>> Tony Lindgren wrote:
>>> * Gadiyar, Anand<gadiyar@ti.com>  [101019 23:21]:
>>
>> ...
>>
>>>>>>>>>> linux-next, as of 20101019, built with the
>> omap2plus_defconfig fails
>>>>>>>>>> to boot on omap3 and omap4. (I've disabled CONFIG_ARCH_OMAP2
>> or
>>>>>>>>>> CONFIG_SWP_EMULATE to get the image to build). Building with
>> only
>>>>>>>>>> ARCH_OMAP3 allows the resultant image to boot up on OMAP3.
>> Likewise,
>>>>>>>>>> an image built with only ARCH_OMAP4 boots up on OMAP4 boards.
>>>>>>>>>>
>>>>>>>>>> earlyprintk does not provide any additional prints after
>>>>>>>>>> "Uncompressing Linux... done, booting the kernel."
>>>>>>>>>>
>>>>>>>>>> Any ideas where to look?
>>>>>>>>>
>>
>> ...
>>
>>>>
>>>> I tried disabling CONFIG_SMP and CONFIG_SMP_ON_UP, and the resulting
>>>> image booted up just fine.
>>>
>>> Looks like it's caused by this patch:
>>>
>>> "ARM: hotplug cpu: Keep processor information, startup code&
>>> __lookup_processor_type"
>>>
>>> It's probably trashing registers it shouldn't. Will take a look and post
>>> something to linux-arm-kernel.
>>>
>>
>> I'm not sure what's going on, but looks like linux-next as of
>> 20101022 boots up fine on panda. I can't explain it, as I couldn't
>> find anything related to the patch you mentioned.
>
> Yeah, it's an alignment issue that's why it sometimes works depending
> on the Kconfig options selected..
>
> For rererence, posted a fix for that at:
>
> https://patchwork.kernel.org/patch/276161/
>
> Tony

The patch seems to work - thanks! Just curious, how do you go about 
debugging an issue like this?

- Anand

Re: linux-next - multi-omap image fails to boot on omap3/4

[Tony Lindgren]

* Anand Gadiyar <gadiyar@ti.com> [101022 13:07]:
> On 10/22/2010 2:52 PM, Tony Lindgren wrote:
> >
> >Yeah, it's an alignment issue that's why it sometimes works depending
> >on the Kconfig options selected..
> >
> >For rererence, posted a fix for that at:
> >
> >https://patchwork.kernel.org/patch/276161/
> 
> The patch seems to work - thanks! Just curious, how do you go about
> debugging an issue like this?

Well it was a bit tricky to find out as random changes to .config
would make it work. Found the breaking patch with git bisect, but
could not see anything wrong with it. Then thought it somehow
misbehaves with CONFIG_SMP_ON_UP, but some printhex8 debugging
showed that it happens even earlier. Then went back few lines in
the boot to narrow it down, but finally had to step through it
with jtag to see that the address loaded into r5 was not aligned.

Tony