> Op 28 okt. 2014, om 23:22 heeft Richard Purdie 
> <richard.pur...@linuxfoundation.org> het volgende geschreven:
> 
> On Tue, 2014-10-28 at 10:36 +0100, Koen Kooi wrote:
>>> Op 27 okt. 2014, om 14:15 heeft Burton, Ross <ross.bur...@intel.com> het 
>>> volgende geschreven:
>>> 
>>> 
>>> On 27 October 2014 13:06, Koen Kooi <k...@dominion.thruhere.net> wrote:
>>> Ah, two different problems :/
>>> 
>>> The problem that lzip-native isn't in oe-core is only a problem when 
>>> there's a .lz SRC_URI in oe-core, but the false-positives are a problem now.
>> 
>> Apart from this dependency problem, bitbake has a problem as well. Does 
>> anyone know why bitbake does:
>> 
>>      compressiontool --stdout file.tar.compressed | tar x
>> 
>> instead of:
>> 
>>      tar xf file.tar.compressed 
> 
> The unpack code is rather old, its age being around a decade. A decade
> ago, tar without inbuilt decompression support was not uncommon. The
> code has remained unchanged...
> 
> zlib was added, I know that most tar has bzip2 now. I'm not 100% sure
> which formats we could rely on being in tar though.

What hapenneded over here was:

tar xf foo.tar.lz
<error>
install lzip
tar xf foo.tar.lz
<success>

So I'm not sure if tar has the knowledge already or not, but installing lzip 
made things work without rebuilding tar.

> It would also be
> interesting to know which performs better on a multicore system.

Indeed.

regards,

Koen
-- 
_______________________________________________
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Reply via email to