[oe] [RFC]: Toolchain build sequence alteration.

[Khem Raj]

On Wednesday 16 July 2008 13:43:26 Leon Woestenberg wrote:
> Khem,
> 
> On Wed, Jul 16, 2008 at 10:10 AM, Khem Raj <raj.khem at gmail.com> wrote:
> >> I have not tested it on older compiler/library combinations but I think
> >> it will work there too as I have build various combinations in the past
> >> with same sequence outside OE.
> >>
> >> Comments and feedback is welcome.
> >>
> 
> I will try some combinations over the next few days.
> 
> anyone: This would be a perfect candidate for some good testing
> outside of .dev, what would be the best approach for that currently
> besides the posted patch?
> 
> Excellent work by the way!
> 
> Regards,

I have done some more testing using various toolchain combinations. I could build all gcc 4.x + glibc 2.4+ combos and eglibc as well as uclibc 0.9.28 and uclibc 0.9.29 and uclibc_svn. In this gcc 3.4.4 glibc 2.3.6 was the oldest combination I could  build on x86 with some surgeries.

This is yet another version of the patch with more stuff in it. ( including the fix to build sdk with gcc 4.3.1) 

Unless you folks approve of this patch is going to get bigger and bigger :)

slugOS maintainers confirmed that they do not need the old combo's so I was more concentrating on gcc 4.x and glibc 2.4+ combos.

I would be happy if someone can test it in his scenario and see if the patch has problems.

I have tested arm and x86. I would appreciate other arch builds if possible.

Thanks

-Khem
-------------- next part --------------
A non-text attachment was scrubbed...
Name: oe-new-toolchain-build-sequence.patch
Type: text/x-patch
Size: 325886 bytes
Desc: not available
URL: <http://lists.openembedded.org/pipermail/openembedded-devel/attachments/20080721/5485a0a8/attachment-0002.bin>