uClibc patches

Robert Connolly robert at linuxfromscratch.org
Sat Sep 9 15:41:47 PDT 2006


I should have mentioned it before, and I don't know if it will make a 
difference. The embryo toolchain shouldn't get any uclibc patches, assuming 
they're linked to your host system's glibc. The butterfly toolchain should 
get the uclibc patches with --host/build/target=uclibc. I have been planning 
to add a /tools/share/config.site in the adjusting page to set 
these --build/host/target values for all packages. A couple months ago I was 
able to get to chapter 6 with uclibc and gcc-4.1.1 with ssp disabled, but 
this was before fortified_source was added, I think I did it with both 0.9.28 
and uclibc-snapshot.

I'd like to make the hardened-specs header modular before trying uclibc again, 
so that ssp, and/or fortify_source, can be disabled for uclibc if its causing 
the problem. There's also the fastmath issue I haven't looked into.

robert

On Saturday 09 September 2006 16:00, Christophe CACHAT wrote:
> Thanks Robert,
>
> I managed to get a few steps further with the patches I took from
> uclibc.org, but I now end up with some weird behaviour in my
> "manually-automated" process (derived from jhalfs) to rebuild my HLFS.
> I think I am missing something either in the initial configuration of
> the uClibc or in the cocoon toolchain. I think I am going to give you
> the time to upload the patches to linuxfromscratch.org and/or to the
> NEO archive... BTW do you plan to do that any time soon?
>
> Anyway thanks for your help, I keep on trying to figure out my
> problems (and try not to embarrass myself with silly questions to the
> list).
>
> Best regards,
>
> Christophe.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.linuxfromscratch.org/pipermail/hlfs-dev/attachments/20060909/3c9fd9e7/attachment.sig>


More information about the hlfs-dev mailing list