Patchwork [2/2] sam-ba: only propose it if the host is x86 or x86_64

login
register
mail settings
Submitter Jean-Christophe PLAGNIOL-VILLARD
Date April 17, 2012, 3:05 p.m.
Message ID <1334675140-20303-2-git-send-email-plagnioj@jcrosoft.com>
Download mbox | patch
Permalink /patch/153234/
State Accepted
Headers show

Comments

Jean-Christophe PLAGNIOL-VILLARD - April 17, 2012, 3:05 p.m.
As sam-ba is deliver as a binary for x86

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
---
 package/sam-ba/Config.in.host |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)
Jean-Christophe PLAGNIOL-VILLARD - April 23, 2012, 6:50 a.m.
On 17:05 Tue 17 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> As sam-ba is deliver as a binary for x86
> 
> Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> ---
ping

Best Regards,
J.
Jean-Christophe PLAGNIOL-VILLARD - April 27, 2012, 10:19 a.m.
On 08:50 Mon 23 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 17:05 Tue 17 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> > As sam-ba is deliver as a binary for x86
> > 
> > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> > ---
> ping
ping

Best Regards,
J.
Jean-Christophe PLAGNIOL-VILLARD - May 4, 2012, 5:01 a.m.
On 12:19 Fri 27 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 08:50 Mon 23 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 17:05 Tue 17 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > As sam-ba is deliver as a binary for x86
> > > 
> > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> > > ---
> > ping
> ping
> 
ping

Best Regards,
J.
Jean-Christophe PLAGNIOL-VILLARD - May 7, 2012, 1:45 p.m.
On 07:01 Fri 04 May     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> On 12:19 Fri 27 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 08:50 Mon 23 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > On 17:05 Tue 17 Apr     , Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > > As sam-ba is deliver as a binary for x86
> > > > 
> > > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> > > > ---
> > > ping
> > ping
> > 
> ping
it's the for ping

did I need to understand that noe one care of non x86 host?

Best Regards,
J.
Peter Korsgaard - May 7, 2012, 2:34 p.m.
>>>>> "Jean-Christophe" == Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> writes:

 Jean-Christophe> it's the for ping

 Jean-Christophe> did I need to understand that noe one care of non x86 host?

To me it is more of a question of tradeoffs. Either we do it really
right and use this kind of infrastructure for the external toolchains as
well (where it really matters), or we decide the extra complexity isn't
worth it for the few non-x86 users.

I haven't made up my mind about this yet, but am leaning towards the
2nd.
Jean-Christophe PLAGNIOL-VILLARD - May 7, 2012, 3:57 p.m.
On 16:34 Mon 07 May     , Peter Korsgaard wrote:
> >>>>> "Jean-Christophe" == Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com> writes:
> 
>  Jean-Christophe> it's the for ping
> 
>  Jean-Christophe> did I need to understand that noe one care of non x86 host?
> 
> To me it is more of a question of tradeoffs. Either we do it really
> right and use this kind of infrastructure for the external toolchains as
> well (where it really matters), or we decide the extra complexity isn't
> worth it for the few non-x86 users.
> 
> I haven't made up my mind about this yet, but am leaning towards the
> 2nd.
as a non x86 user I really care

and as the ARM A15 will come soon on the market for serer as example it will
be more and more problematic for binary package

Best Regards,
J.
> 
> -- 
> Bye, Peter Korsgaard
Arnout Vandecappelle - May 11, 2012, 7:42 p.m.
On 05/07/12 17:57, Jean-Christophe PLAGNIOL-VILLARD wrote:
>> >    Jean-Christophe>  did I need to understand that noe one care of non x86 host?
>> >
>> >  To me it is more of a question of tradeoffs. Either we do it really
>> >  right and use this kind of infrastructure for the external toolchains as
>> >  well (where it really matters), or we decide the extra complexity isn't
>> >  worth it for the few non-x86 users.
>> >
>> >  I haven't made up my mind about this yet, but am leaning towards the
>> >  2nd.
> as a non x86 user I really care
>
> and as the ARM A15 will come soon on the market for serer as example it will
> be more and more problematic for binary package

  ... except that it's not really problematic.  The only problem is that you can
configure buildroot to download something that won't actually run on your
system.  This problem is a lot worse for external toolchains than for sam-ba.

  That said, the patch is really simple and doesn't create a maintenance burden.
Having it in paves the way for disallowing other downloaded host binaries, and
for downloading host binaries for the correct platform.  Therefore:

Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

  Jean-Christophe, maybe it's worthwhile to resend and perhaps add
a patch for external toolchains as well?

  Regards,
  Arnout

Patch

diff --git a/package/sam-ba/Config.in.host b/package/sam-ba/Config.in.host
index f2ea01f..20b2b36 100644
--- a/package/sam-ba/Config.in.host
+++ b/package/sam-ba/Config.in.host
@@ -1,5 +1,6 @@ 
 config BR2_PACKAGE_HOST_SAM_BA
 	bool "host sam-ba"
+	depends on BR2_HOSTARCH = "x86_64" || BR2_HOSTARCH = "x86"
 	help
 	  Atmel SAM-BA software provides an open set of tools for
 	  programming the Atmel SAM3, SAM7 and SAM9 ARM-based