Greg Schafer wrote:
> Ryan Oliver wrote:
>
>> LFS isn't affected by the "-specs handling bug" as we do not pass
>> -specs=/some/specfile on the gcc command line
>>
>
> ??? Not affected? LFS doesn't have the clean split between the 2 phases
> like DIY does. I can simply wipe the chroot phase files and start clean
> with a pristine temptools system. This is a HUGE efficiency gain for
> testing purposes.
>
LFS not affected in regards to the fact we can set any of
md_startfile_prefix{,_1} or startfile_prefix_spec in the specs file and
have it work because we DO use a standard specs file in the appropriate
place.
As you've noticed the -specs command line switch (for user specs) comes
in far too late to adjust any of these, for which you've added your
current gcc hack and in the past proliferated the use -B/foo/lib (and
some unfair fud about the use of startfile_prefix_spec)
In your case wanting to keep using -specs you are probably better off
setting
STANDARD_STARTFILE_PREFIX_1=/tools/lib
STANDARD_STARTFILE_PREFIX_2=""
for your temptools phase, and later when you need to override the default
search path just set either
LIBRARY_PATH=/usr/lib
or
LPATH=/usr/lib
on the environment (along with passing your -specs=foo) and just dispense with
the gcc hack and -B/foo/lib guff.
LIBRARY_PATH / LPATH are added to startfile_prefixes and processed by the
multilib spec.
Best Regards
[R]
--
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
