[PATCH v2 0/1] builddeb: add arm64 in the supported architectures

[PATCH v2 0/1] builddeb: add arm64 in the supported architectures

[Fathi Boudra]

Changes from v1:
 - per Ben Hutching's suggestion, move arm64 case above the arm* case
   since arm64 is a separate architecture in terms of kernel source.

Fathi Boudra (1):
  builddeb: add arm64 in the supported architectures

 scripts/package/builddeb | 2 ++
 1 file changed, 2 insertions(+)

-- 
1.9.1

[PATCH v2 1/1] builddeb: add arm64 in the supported architectures

[Fathi Boudra]

Signed-off-by: Fathi Boudra <fathi.boudra@linaro.org>
---
 scripts/package/builddeb | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/scripts/package/builddeb b/scripts/package/builddeb
index f46e4dd..0bf29a6 100644
--- a/scripts/package/builddeb
+++ b/scripts/package/builddeb
@@ -42,6 +42,8 @@ create_package() {
 		debarch=hppa ;;
 	mips*)
 		debarch=mips$(grep -q CPU_LITTLE_ENDIAN=y $KCONFIG_CONFIG && echo el || true) ;;
+	arm64)
+		debarch=arm64 ;;
 	arm*)
 		debarch=arm$(grep -q CONFIG_AEABI=y $KCONFIG_CONFIG && echo el || true) ;;
 	*)
-- 
1.9.1

Re: [PATCH v2 1/1] builddeb: add arm64 in the supported architectures

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 796 bytes --]

On Sat, 2014-04-12 at 15:53 +0300, Fathi Boudra wrote:
> Signed-off-by: Fathi Boudra <fathi.boudra@linaro.org>

Reviewed-by: Ben Hutchings <ben@decadent.org.uk>

> ---
>  scripts/package/builddeb | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/scripts/package/builddeb b/scripts/package/builddeb
> index f46e4dd..0bf29a6 100644
> --- a/scripts/package/builddeb
> +++ b/scripts/package/builddeb
> @@ -42,6 +42,8 @@ create_package() {
>  		debarch=hppa ;;
>  	mips*)
>  		debarch=mips$(grep -q CPU_LITTLE_ENDIAN=y $KCONFIG_CONFIG && echo el || true) ;;
> +	arm64)
> +		debarch=arm64 ;;
>  	arm*)
>  		debarch=arm$(grep -q CONFIG_AEABI=y $KCONFIG_CONFIG && echo el || true) ;;
>  	*)

-- 
Ben Hutchings
I haven't lost my mind; it's backed up on tape somewhere.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 811 bytes --]

Re: [PATCH v2 1/1] builddeb: add arm64 in the supported architectures

[Michal Marek]

On 2014-04-13 15:26, Ben Hutchings wrote:
> On Sat, 2014-04-12 at 15:53 +0300, Fathi Boudra wrote:
>> Signed-off-by: Fathi Boudra <fathi.boudra@linaro.org>
> 
> Reviewed-by: Ben Hutchings <ben@decadent.org.uk>

Applied to kbuild.git#misc, thanks.

Michal

Re: [PATCH v2 1/1] builddeb: add arm64 in the supported architectures

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 1080 bytes --]

On Sun, 2014-04-13 at 14:26 +0100, Ben Hutchings wrote:
> On Sat, 2014-04-12 at 15:53 +0300, Fathi Boudra wrote:
> > Signed-off-by: Fathi Boudra <fathi.boudra@linaro.org>
> 
> Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
> 
> > ---
> >  scripts/package/builddeb | 2 ++
> >  1 file changed, 2 insertions(+)
> > 
> > diff --git a/scripts/package/builddeb b/scripts/package/builddeb
> > index f46e4dd..0bf29a6 100644
> > --- a/scripts/package/builddeb
> > +++ b/scripts/package/builddeb
> > @@ -42,6 +42,8 @@ create_package() {
> >  		debarch=hppa ;;
> >  	mips*)
> >  		debarch=mips$(grep -q CPU_LITTLE_ENDIAN=y $KCONFIG_CONFIG && echo el || true) ;;
> > +	arm64)

Now I'm not so sure.  As we are comparing with the 'machine' name
($UTS_MACHINE, not $ARCH or $SRCARCH), shouldn't this actually check for
aarch64?

Ben.

> > +		debarch=arm64 ;;
> >  	arm*)
> >  		debarch=arm$(grep -q CONFIG_AEABI=y $KCONFIG_CONFIG && echo el || true) ;;
> >  	*)
> 

-- 
Ben Hutchings
One of the nice things about standards is that there are so many of them.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: [PATCH v2 1/1] builddeb: add arm64 in the supported architectures

[Fathi Boudra]

On 9 June 2014 03:15, Ben Hutchings <ben@decadent.org.uk> wrote:
> On Sun, 2014-04-13 at 14:26 +0100, Ben Hutchings wrote:
>> On Sat, 2014-04-12 at 15:53 +0300, Fathi Boudra wrote:
>> > Signed-off-by: Fathi Boudra <fathi.boudra@linaro.org>
>>
>> Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
>>
>> > ---
>> >  scripts/package/builddeb | 2 ++
>> >  1 file changed, 2 insertions(+)
>> >
>> > diff --git a/scripts/package/builddeb b/scripts/package/builddeb
>> > index f46e4dd..0bf29a6 100644
>> > --- a/scripts/package/builddeb
>> > +++ b/scripts/package/builddeb
>> > @@ -42,6 +42,8 @@ create_package() {
>> >             debarch=hppa ;;
>> >     mips*)
>> >             debarch=mips$(grep -q CPU_LITTLE_ENDIAN=y $KCONFIG_CONFIG && echo el || true) ;;
>> > +   arm64)
>
> Now I'm not so sure.  As we are comparing with the 'machine' name
> ($UTS_MACHINE, not $ARCH or $SRCARCH), shouldn't this actually check for
> aarch64?

not sure. I've seen Ian comment (added to the cc list). He's proposing:

[...]
--- a/scripts/package/builddeb
+++ b/scripts/package/builddeb
@@ -28,7 +28,7 @@ create_package() {
        # Attempt to find the correct Debian architecture
        local forcearch="" debarch=""
        case "$UTS_MACHINE" in
-       i386|ia64|alpha)
+       i386|ia64|alpha|arm64)
                debarch="$UTS_MACHINE" ;;
        x86_64)
                debarch=amd64 ;;
[...]

I'm taking a closer look now.

> Ben.
>
>> > +           debarch=arm64 ;;
>> >     arm*)
>> >             debarch=arm$(grep -q CONFIG_AEABI=y $KCONFIG_CONFIG && echo el || true) ;;
>> >     *)
>>
>
> --
> Ben Hutchings
> One of the nice things about standards is that there are so many of them.

Re: [PATCH v2 1/1] builddeb: add arm64 in the supported architectures

[Ian Campbell]

On Tue, 2014-06-10 at 11:15 +0300, Fathi Boudra wrote:
> On 9 June 2014 03:15, Ben Hutchings <ben@decadent.org.uk> wrote:
> > On Sun, 2014-04-13 at 14:26 +0100, Ben Hutchings wrote:
> >> On Sat, 2014-04-12 at 15:53 +0300, Fathi Boudra wrote:
> >> > Signed-off-by: Fathi Boudra <fathi.boudra@linaro.org>
> >>
> >> Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
> >>
> >> > ---
> >> >  scripts/package/builddeb | 2 ++
> >> >  1 file changed, 2 insertions(+)
> >> >
> >> > diff --git a/scripts/package/builddeb b/scripts/package/builddeb
> >> > index f46e4dd..0bf29a6 100644
> >> > --- a/scripts/package/builddeb
> >> > +++ b/scripts/package/builddeb
> >> > @@ -42,6 +42,8 @@ create_package() {
> >> >             debarch=hppa ;;
> >> >     mips*)
> >> >             debarch=mips$(grep -q CPU_LITTLE_ENDIAN=y $KCONFIG_CONFIG && echo el || true) ;;
> >> > +   arm64)
> >
> > Now I'm not so sure.  As we are comparing with the 'machine' name
> > ($UTS_MACHINE, not $ARCH or $SRCARCH), shouldn't this actually check for
> > aarch64?
> 
> not sure. I've seen Ian comment (added to the cc list).

The patch above produced the right thing when crossbuilt with
ARCH=arm64. e.g. linux-image-3.14.5+_3.14.5+-2_arm64.deb

With just Ben's original patch it produced linux-image-3.14.5+_3.14.5
+-2_arm.deb instead.

I think UTS_MACHINE is correct because:

Makefile:UTS_MACHINE    := $(ARCH)

and there is no override in arch/arm*/Makefile.

So the kernel arch is the correct thing to use here.

Ian.

Re: [PATCH v2 1/1] builddeb: add arm64 in the supported architectures

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 1928 bytes --]

On Tue, 2014-06-10 at 09:31 +0100, Ian Campbell wrote:
> On Tue, 2014-06-10 at 11:15 +0300, Fathi Boudra wrote:
> > On 9 June 2014 03:15, Ben Hutchings <ben@decadent.org.uk> wrote:
> > > On Sun, 2014-04-13 at 14:26 +0100, Ben Hutchings wrote:
> > >> On Sat, 2014-04-12 at 15:53 +0300, Fathi Boudra wrote:
> > >> > Signed-off-by: Fathi Boudra <fathi.boudra@linaro.org>
> > >>
> > >> Reviewed-by: Ben Hutchings <ben@decadent.org.uk>
> > >>
> > >> > ---
> > >> >  scripts/package/builddeb | 2 ++
> > >> >  1 file changed, 2 insertions(+)
> > >> >
> > >> > diff --git a/scripts/package/builddeb b/scripts/package/builddeb
> > >> > index f46e4dd..0bf29a6 100644
> > >> > --- a/scripts/package/builddeb
> > >> > +++ b/scripts/package/builddeb
> > >> > @@ -42,6 +42,8 @@ create_package() {
> > >> >             debarch=hppa ;;
> > >> >     mips*)
> > >> >             debarch=mips$(grep -q CPU_LITTLE_ENDIAN=y $KCONFIG_CONFIG && echo el || true) ;;
> > >> > +   arm64)
> > >
> > > Now I'm not so sure.  As we are comparing with the 'machine' name
> > > ($UTS_MACHINE, not $ARCH or $SRCARCH), shouldn't this actually check for
> > > aarch64?
> > 
> > not sure. I've seen Ian comment (added to the cc list).
> 
> The patch above produced the right thing when crossbuilt with
> ARCH=arm64. e.g. linux-image-3.14.5+_3.14.5+-2_arm64.deb
> 
> With just Ben's original patch it produced linux-image-3.14.5+_3.14.5
> +-2_arm.deb instead.
> 
> I think UTS_MACHINE is correct because:
> 
> Makefile:UTS_MACHINE    := $(ARCH)
> 
> and there is no override in arch/arm*/Makefile.
> 
> So the kernel arch is the correct thing to use here.

OK, so $UTS_MACHINE is a misnomer for this and some other architectures.
I wonder whether we should be using it at all?

Ben.

-- 
Ben Hutchings
DNRC Motto:  I can please only one person per day.
Today is not your day.  Tomorrow isn't looking good either.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]