Le 27/01/2012 19:46, Bruce Dubbs a écrit : > > Pierre Labastie wrote: >> Hi, >> >> I think I spotted something by doing ICA (not all investigated yet). >> ld.so.cache differs at the end of pass 1 and at the end of pass 2. >> It can be printed with ldconfig -p, and then diffed, which gives: > > >> Now, one differing binary file is grub-mkimage. Going back to the logs, >> I find >> >> --- build_1/107-grub 2012-01-26 23:01:10.000000000 +0100 >> +++ build_2/107-grub 2012-01-26 23:32:59.000000000 +0100 >> [snip] >> -checking for lzma_code in -llzma... no >> +checking for lzma_code in -llzma... yes >> >> It has to be investigated whether this is related to lzma being absent >> from ld.so.cache. >> >> In any event, it looks like ldconfig should be run now and then during >> the build. > > Possibly, what happens if you do ldconfig at the end of Chapter 5 in, > say, the stripping chapter. It might be useful in 'Section 6.64 - > Cleaning Up' also. > > Just add those lines to your xml and try again. > > -- Bruce I realize that some important lines for understanding my preceding post were in French. Sorry about that! The beginning of the diff between the two ld.so.cache said that there were 93 libs in ld.so.cache at the end of pass 1, and 117 at the end of pass 2. And then I only listed the 24 new libs. Of course, ldconfig _is_ run at the beginning of pass 2 (after glibc), and it finds all the installed libs of pass 1. So I guess that the 24 extra libs enter ld.so.cache early in pass 2.
Well, the reason why grub does not find liblzma is much simpler anyway: grub is built before xz! Not a big deal, unless you want to create compressed boot images (the only possible compression is xz). But shows up on ICA. Regards, Pierre -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
