wip-armhf branch ready for wider testing

wip-armhf branch ready for wider testing

[Mark H Weaver]

Hello Guix!

I'm pleased to announce that the 'wip-armhf' branch is now ready for
wider testing.  So far, I've successfully used this branch to:

(1) cross-compile armhf-linux bootstrap tarballs on my Libreboot X60,

(2) natively compile armhf bootstrap tarballs on my Novena, based on the
    cross-compiled bootstrap tarballs from (1), and

(3) build the 'hello' package based on the natively-compiled bootstrap
    tarballs from (2).

Most likely we'll find more build problems higher up in the stack, and
help with that would be most appreciated!

   Happy hacking,
       Mark

Re: wip-armhf branch ready for wider testing

[David Thompson]

Mark H Weaver <mhw@netris.org> writes:

> I'm pleased to announce that the 'wip-armhf' branch is now ready for
> wider testing.
> [...]
> Most likely we'll find more build problems higher up in the stack, and
> help with that would be most appreciated!

Wow, great progress!  I will start testing when my Novena arrives.

-- 
David Thompson
Web Developer - Free Software Foundation - http://fsf.org
GPG Key: 0FF1D807
Support the FSF: https://fsf.org/donate

Re: wip-armhf branch ready for wider testing

[Ludovic Courtès]

Mark H Weaver <mhw@netris.org> skribis:

> I'm pleased to announce that the 'wip-armhf' branch is now ready for
> wider testing.  So far, I've successfully used this branch to:
>
> (1) cross-compile armhf-linux bootstrap tarballs on my Libreboot X60,
>
> (2) natively compile armhf bootstrap tarballs on my Novena, based on the
>     cross-compiled bootstrap tarballs from (1), and
>
> (3) build the 'hello' package based on the natively-compiled bootstrap
>     tarballs from (2).

Congrats on all the progress made!

It would be nice to have a couple of build machines for this platform.
If anyone reading this wants to help, please let us know!

Ludo’.

Re: wip-armhf branch ready for wider testing

[Andreas Enge]

On Sat, Jan 03, 2015 at 08:09:42PM +0100, Ludovic Courtès wrote:
> Congrats on all the progress made!

These are good news indeed!

> It would be nice to have a couple of build machines for this platform.
> If anyone reading this wants to help, please let us know!

On which kind of machines does this armhf run? Is it related to the output
of "uname -m"? I have a raspberry pi that identifies as "armv6l", and a
seagate goflex that identifies as "armv5tel". I suppose both are too old
to run armhf? If it is not "uname -m", how do I know whether the port
will work?

Andreas

Re: wip-armhf branch ready for wider testing

[Mark H Weaver]

Andreas Enge <andreas@enge.fr> writes:

> On Sat, Jan 03, 2015 at 08:09:42PM +0100, Ludovic Courtès wrote:
>> Congrats on all the progress made!
>
> These are good news indeed!
>
>> It would be nice to have a couple of build machines for this platform.
>> If anyone reading this wants to help, please let us know!
>
> On which kind of machines does this armhf run? Is it related to the output
> of "uname -m"?

Our armhf targets the same set of machines as Debian's armhf, namely the
ARMv7-A architecture with VFP3D16 coprocessor.  So 'uname -m' must
output "armv7l" or better, and I guess the processor "Features" as
reported by /proc/cpuinfo must include "vfpv3" and either "vfpd16"
or "vfpd32".

> I have a raspberry pi that identifies as "armv6l", and a seagate
> goflex that identifies as "armv5tel". I suppose both are too old to
> run armhf?

Right, they can't run armhf.  Also, a build slave should have 1 GB RAM
minimum (preferably 2) and reasonably well performing storage (i.e. SATA
not SD).  So, I think Beaglebone Black and other systems in that class
will not be sufficient.

I think we need ARM board(s) with performance comparable to the Novena.
The Novena itself would be a fine option, but I'm not willing to give
mine up for that task :)

      Mark

Re: wip-armhf branch ready for wider testing

[Ludovic Courtès]

Mark H Weaver <mhw@netris.org> skribis:

> Our armhf targets the same set of machines as Debian's armhf, namely the
> ARMv7-A architecture with VFP3D16 coprocessor.  So 'uname -m' must
> output "armv7l" or better, and I guess the processor "Features" as
> reported by /proc/cpuinfo must include "vfpv3" and either "vfpd16"
> or "vfpd32".

Since it’s becoming a FAQ, could you add something along these lines to
the manual, right where armhf-linux is introduced?

Thanks,
Ludo’.

Re: wip-armhf branch ready for wider testing

[Mark H Weaver]

ludo@gnu.org (Ludovic Courtès) writes:

> Mark H Weaver <mhw@netris.org> skribis:
>
>> Our armhf targets the same set of machines as Debian's armhf, namely the
>> ARMv7-A architecture with VFP3D16 coprocessor.  So 'uname -m' must
>> output "armv7l" or better, and I guess the processor "Features" as
>> reported by /proc/cpuinfo must include "vfpv3" and either "vfpd16"
>> or "vfpd32".
>
> Since it’s becoming a FAQ, could you add something along these lines to
> the manual, right where armhf-linux is introduced?

I already put this in the "GNU Distribution" node (chapter 6):

--8<---------------cut here---------------start------------->8---
   The GNU distribution is currently available on the following
platforms:

[...]

‘armhf-linux’
     ARMv7-A architecture with hard float, Thumb-2 and VFP3D16
     coprocessor, using the EABI hard-float ABI, and Linux-Libre kernel.
--8<---------------cut here---------------end--------------->8---

Do you think more is needed?

      Mark

Re: wip-armhf branch ready for wider testing

[Daniel Pimentel]

On 2015-01-20 19:46, Mark H Weaver wrote:
> ludo@gnu.org (Ludovic Courtès) writes:
> 
>> Mark H Weaver <mhw@netris.org> skribis:
>> 
>>> Our armhf targets the same set of machines as Debian's armhf, namely 
>>> the
>>> ARMv7-A architecture with VFP3D16 coprocessor.  So 'uname -m' must
>>> output "armv7l" or better, and I guess the processor "Features" as
>>> reported by /proc/cpuinfo must include "vfpv3" and either "vfpd16"
>>> or "vfpd32".
>> 
>> Since it’s becoming a FAQ, could you add something along these lines 
>> to
>> the manual, right where armhf-linux is introduced?
> 
> I already put this in the "GNU Distribution" node (chapter 6):
> 
> --8<---------------cut here---------------start------------->8---
>    The GNU distribution is currently available on the following
> platforms:
> 
> [...]
> 
> ‘armhf-linux’
>      ARMv7-A architecture with hard float, Thumb-2 and VFP3D16
>      coprocessor, using the EABI hard-float ABI, and Linux-Libre 
> kernel.
> --8<---------------cut here---------------end--------------->8---
> 
> Do you think more is needed?
> 
>       Mark
Great work! thanks.

Re: wip-armhf branch ready for wider testing

[Ludovic Courtès]

Mark H Weaver <mhw@netris.org> skribis:

> ludo@gnu.org (Ludovic Courtès) writes:
>
>> Mark H Weaver <mhw@netris.org> skribis:
>>
>>> Our armhf targets the same set of machines as Debian's armhf, namely the
>>> ARMv7-A architecture with VFP3D16 coprocessor.  So 'uname -m' must
>>> output "armv7l" or better, and I guess the processor "Features" as
>>> reported by /proc/cpuinfo must include "vfpv3" and either "vfpd16"
>>> or "vfpd32".
>>
>> Since it’s becoming a FAQ, could you add something along these lines to
>> the manual, right where armhf-linux is introduced?
>
> I already put this in the "GNU Distribution" node (chapter 6):
>
>    The GNU distribution is currently available on the following
> platforms:
>
> [...]
>
> ‘armhf-linux’
>      ARMv7-A architecture with hard float, Thumb-2 and VFP3D16
>      coprocessor, using the EABI hard-float ABI, and Linux-Libre kernel.
>
> Do you think more is needed?

Perhaps in parentheses or in a footnote indicate what to look for in
“uname -m” and /proc/cpuinfo?  WDYT?

Ludo’.