Prev: : Fix checkstack warning in bunzip2 library
Next: [PATCH] kernel/power/swap.c: fix typos in comments
From: Jaswinder Singh Rajput on 27 May 2010 18:10
Hello Sam,

On Fri, May 28, 2010 at 12:16 AM, Sam Ravnborg <sam(a)ravnborg.org> wrote:
> On Thu, May 27, 2010 at 11:49:16AM -0400, Christoph Hellwig wrote:
>> On Thu, May 27, 2010 at 05:37:36AM +0530, Jaswinder Singh Rajput wrote:
>> > Hello,
>> >
>> > I am trying to build 32 bit kernel image on 64-bit machine but after
>> > 'make menuconfig' it automatically switches 32-bit to 64-bit.
>> >
>> > Is this a BUG or did intentionally.
>>
>> I think it's intentional, but I'm really annoyed by this behaviour.
>
> The request when we did the merge of 32 and 64 bit was that we
> should continue to build 64bit kernels on 64 bit machines.
> And likewise for 32bit.
> Because this was what the users where used to.
>
> We could simplify metters by defining new rules.
> As you suggest that uname on an intel box always resulted in ARCH=x86.
> But then people would complain why a 32 bit kernel is default when
> the box is 64 bit.
>
> Something like this:
> [The escape chars align nicely in the patched Makefile]
>
> � � � �Sam
>
> diff --git a/Makefile b/Makefile
> index ebc8225..e7f97a5 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -161,10 +161,11 @@ export srctree objtree VPATH
> �# then ARCH is assigned, getting whatever value it gets normally, and
> �# SUBARCH is subsequently ignored.
>
> -SUBARCH := $(shell uname -m | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \
> - � � � � � � � � � � � � � � � � -e s/arm.*/arm/ -e s/sa110/arm/ \
> +SUBARCH := $(shell uname -m | sed -e s/i.86/x86/ -e s/x86_64/x86/ � � � �\
> + � � � � � � � � � � � � � � � � -e s/sun4u/sparc64/ � � � � � � � � � �\
> + � � � � � � � � � � � � � � � � -e s/arm.*/arm/ -e s/sa110/arm/ � � � �\
> � � � � � � � � � � � � � � � � �-e s/s390x/s390/ -e s/parisc64/parisc/ \
> - � � � � � � � � � � � � � � � � -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \
> + � � � � � � � � � � � � � � � � -e s/ppc.*/powerpc/ -e s/mips.*/mips/ �\
> � � � � � � � � � � � � � � � � �-e s/sh[234].*/sh/ )
>
> �# Cross compiling and selecting different set of gcc/bin-utils
>
>

Thanks, this patch fixes the issue. Now 'make menuconfig' is following .config.

Thanks,
--
Jaswinder Singh.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Sam Ravnborg on 29 May 2010 07:30
On Fri, May 28, 2010 at 03:37:00AM +0530, Jaswinder Singh Rajput wrote:
> Hello Sam,
>
> On Fri, May 28, 2010 at 12:16 AM, Sam Ravnborg <sam(a)ravnborg.org> wrote:
> > On Thu, May 27, 2010 at 11:49:16AM -0400, Christoph Hellwig wrote:
> >> On Thu, May 27, 2010 at 05:37:36AM +0530, Jaswinder Singh Rajput wrote:
> >> > Hello,
> >> >
> >> > I am trying to build 32 bit kernel image on 64-bit machine but after
> >> > 'make menuconfig' it automatically switches 32-bit to 64-bit.
> >> >
> >> > Is this a BUG or did intentionally.
> >>
> >> I think it's intentional, but I'm really annoyed by this behaviour.
> >
> > The request when we did the merge of 32 and 64 bit was that we
> > should continue to build 64bit kernels on 64 bit machines.
> > And likewise for 32bit.
> > Because this was what the users where used to.
> >
> > We could simplify metters by defining new rules.
> > As you suggest that uname on an intel box always resulted in ARCH=x86.
> > But then people would complain why a 32 bit kernel is default when
> > the box is 64 bit.
> >
> > Something like this:
> > [The escape chars align nicely in the patched Makefile]
> >
> > � � � �Sam
> >
> > diff --git a/Makefile b/Makefile
> > index ebc8225..e7f97a5 100644
> > --- a/Makefile
> > +++ b/Makefile
> > @@ -161,10 +161,11 @@ export srctree objtree VPATH
> > �# then ARCH is assigned, getting whatever value it gets normally, and
> > �# SUBARCH is subsequently ignored.
> >
> > -SUBARCH := $(shell uname -m | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \
> > - � � � � � � � � � � � � � � � � -e s/arm.*/arm/ -e s/sa110/arm/ \
> > +SUBARCH := $(shell uname -m | sed -e s/i.86/x86/ -e s/x86_64/x86/ � � � �\
> > + � � � � � � � � � � � � � � � � -e s/sun4u/sparc64/ � � � � � � � � � �\
> > + � � � � � � � � � � � � � � � � -e s/arm.*/arm/ -e s/sa110/arm/ � � � �\
> > � � � � � � � � � � � � � � � � �-e s/s390x/s390/ -e s/parisc64/parisc/ \
> > - � � � � � � � � � � � � � � � � -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \
> > + � � � � � � � � � � � � � � � � -e s/ppc.*/powerpc/ -e s/mips.*/mips/ �\
> > � � � � � � � � � � � � � � � � �-e s/sh[234].*/sh/ )
> >
> > �# Cross compiling and selecting different set of gcc/bin-utils
> >
> >
>
> Thanks, this patch fixes the issue. Now 'make menuconfig' is following .config.

If you think this is the right way to do things you can take it up
with the x86 maintainers.

Although the patch touches only a generic file this is x86 specific.

Sam
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Jaswinder Singh Rajput on 29 May 2010 13:20
Hello Sam and Am�rico,

On Sat, May 29, 2010 at 4:49 PM, Sam Ravnborg <sam(a)ravnborg.org> wrote:
> On Fri, May 28, 2010 at 03:37:00AM +0530, Jaswinder Singh Rajput wrote:
>> Hello Sam,
>>
>> On Fri, May 28, 2010 at 12:16 AM, Sam Ravnborg <sam(a)ravnborg.org> wrote:
>> > On Thu, May 27, 2010 at 11:49:16AM -0400, Christoph Hellwig wrote:
>> >> On Thu, May 27, 2010 at 05:37:36AM +0530, Jaswinder Singh Rajput wrote:
>> >> > Hello,
>> >> >
>> >> > I am trying to build 32 bit kernel image on 64-bit machine but after
>> >> > 'make menuconfig' it automatically switches 32-bit to 64-bit.
>> >> >
>> >> > Is this a BUG or did intentionally.
>> >>
>> >> I think it's intentional, but I'm really annoyed by this behaviour.
>> >
>> > The request when we did the merge of 32 and 64 bit was that we
>> > should continue to build 64bit kernels on 64 bit machines.
>> > And likewise for 32bit.
>> > Because this was what the users where used to.
>> >
>> > We could simplify metters by defining new rules.
>> > As you suggest that uname on an intel box always resulted in ARCH=x86.
>> > But then people would complain why a 32 bit kernel is default when
>> > the box is 64 bit.
>> >
>> > Something like this:
>> > [The escape chars align nicely in the patched Makefile]
>> >
>> > � � � �Sam
>> >
>> > diff --git a/Makefile b/Makefile
>> > index ebc8225..e7f97a5 100644
>> > --- a/Makefile
>> > +++ b/Makefile
>> > @@ -161,10 +161,11 @@ export srctree objtree VPATH
>> > �# then ARCH is assigned, getting whatever value it gets normally, and
>> > �# SUBARCH is subsequently ignored.
>> >
>> > -SUBARCH := $(shell uname -m | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \
>> > - � � � � � � � � � � � � � � � � -e s/arm.*/arm/ -e s/sa110/arm/ \
>> > +SUBARCH := $(shell uname -m | sed -e s/i.86/x86/ -e s/x86_64/x86/ � � � �\
>> > + � � � � � � � � � � � � � � � � -e s/sun4u/sparc64/ � � � � � � � � � �\
>> > + � � � � � � � � � � � � � � � � -e s/arm.*/arm/ -e s/sa110/arm/ � � � �\
>> > � � � � � � � � � � � � � � � � �-e s/s390x/s390/ -e s/parisc64/parisc/ \
>> > - � � � � � � � � � � � � � � � � -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \
>> > + � � � � � � � � � � � � � � � � -e s/ppc.*/powerpc/ -e s/mips.*/mips/ �\
>> > � � � � � � � � � � � � � � � � �-e s/sh[234].*/sh/ )
>> >
>> > �# Cross compiling and selecting different set of gcc/bin-utils
>> >
>> >
>>
>> Thanks, this patch fixes the issue. Now 'make menuconfig' is following .config.
>
> If you think this is the right way to do things you can take it up
> with the x86 maintainers.
>
> Although the patch touches only a generic file this is x86 specific.
>

Am�rico Wang, Do you mind handle this kconfig patch.

Thanks,
--
Jaswinder Singh.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Américo Wang on 2 Jun 2010 04:20
On Sat, May 29, 2010 at 10:49:25PM +0530, Jaswinder Singh Rajput wrote:
>Hello Sam and Américo,
>
>On Sat, May 29, 2010 at 4:49 PM, Sam Ravnborg <sam(a)ravnborg.org> wrote:
>> On Fri, May 28, 2010 at 03:37:00AM +0530, Jaswinder Singh Rajput wrote:
>>> Hello Sam,
>>>
>>> On Fri, May 28, 2010 at 12:16 AM, Sam Ravnborg <sam(a)ravnborg.org> wrote:
>>> > On Thu, May 27, 2010 at 11:49:16AM -0400, Christoph Hellwig wrote:
>>> >> On Thu, May 27, 2010 at 05:37:36AM +0530, Jaswinder Singh Rajput wrote:
>>> >> > Hello,
>>> >> >
>>> >> > I am trying to build 32 bit kernel image on 64-bit machine but after
>>> >> > 'make menuconfig' it automatically switches 32-bit to 64-bit.
>>> >> >
>>> >> > Is this a BUG or did intentionally.
>>> >>
>>> >> I think it's intentional, but I'm really annoyed by this behaviour.
>>> >
>>> > The request when we did the merge of 32 and 64 bit was that we
>>> > should continue to build 64bit kernels on 64 bit machines.
>>> > And likewise for 32bit.
>>> > Because this was what the users where used to.
>>> >
>>> > We could simplify metters by defining new rules.
>>> > As you suggest that uname on an intel box always resulted in ARCH=x86.
>>> > But then people would complain why a 32 bit kernel is default when
>>> > the box is 64 bit.
>>> >
>>> > Something like this:
>>> > [The escape chars align nicely in the patched Makefile]
>>> >
>>> >        Sam
>>> >
>>> > diff --git a/Makefile b/Makefile
>>> > index ebc8225..e7f97a5 100644
>>> > --- a/Makefile
>>> > +++ b/Makefile
>>> > @@ -161,10 +161,11 @@ export srctree objtree VPATH
>>> >  # then ARCH is assigned, getting whatever value it gets normally, and
>>> >  # SUBARCH is subsequently ignored.
>>> >
>>> > -SUBARCH := $(shell uname -m | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \
>>> > -                                 -e s/arm.*/arm/ -e s/sa110/arm/ \
>>> > +SUBARCH := $(shell uname -m | sed -e s/i.86/x86/ -e s/x86_64/x86/        \
>>> > +                                 -e s/sun4u/sparc64/                    \
>>> > +                                 -e s/arm.*/arm/ -e s/sa110/arm/        \
>>> >                                  -e s/s390x/s390/ -e s/parisc64/parisc/ \
>>> > -                                 -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \
>>> > +                                 -e s/ppc.*/powerpc/ -e s/mips.*/mips/  \
>>> >                                  -e s/sh[234].*/sh/ )
>>> >
>>> >  # Cross compiling and selecting different set of gcc/bin-utils
>>> >
>>> >
>>>
>>> Thanks, this patch fixes the issue. Now 'make menuconfig' is following .config.
>>
>> If you think this is the right way to do things you can take it up
>> with the x86 maintainers.
>>
>> Although the patch touches only a generic file this is x86 specific.
>>
>
>Américo Wang, Do you mind handle this kconfig patch.
>


Sorry for the delay.

Usually Michal should take this patch. Cc'ing him.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Sam Ravnborg on 2 Jun 2010 05:00
> >>
> >> Although the patch touches only a generic file this is x86 specific.
> >>
> >
> >Am�rico Wang, Do you mind handle this kconfig patch.
> >
>
>
> Sorry for the delay.
>
> Usually Michal should take this patch. Cc'ing him.

This is not generic functionality. The patch changes default
behaviour when building x86 kernes only.
So IMO it requires ack from the x86 maintainers.

Sam
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5
Prev: : Fix checkstack warning in bunzip2 library
Next: [PATCH] kernel/power/swap.c: fix typos in comments