> Date: Tue, 17 Dec 2013 16:39:03 +0000 > From: lf...@cruziero.com (akhiezer) > To: LFS Support List <lfs-support@linuxfromscratch.org> > Subject: Re: [lfs-support] GCC build first pass: mpc build looks for > libgmp.la in the wrong place > > > Date: Fri, 06 Dec 2013 21:05:51 +0000 > > From: lf...@cruziero.com (akhiezer) > > To: LFS Support List <lfs-support@linuxfromscratch.org> > > Subject: Re: [lfs-support] GCC build first pass: mpc build looks for > > libgmp.la in the wrong place > > > . > . > > > > > > lfs-6.7 and lfs-6.8 both use './configure ...' options: > > > --with-gmp-include=$(pwd)/gmp --with-gmp-lib=$(pwd)/gmp/.libs > > > ; and they don't use any mpfr/mpc stuff. > > > > > > And lfs-7.0 thru lfs-7.4 uses (I've replaced the gcc version numbers with > > > the > > > literal string '${GCC_VERNUM}'): > > > --with-mpfr-include=$(pwd)/../gcc-${GCC_VERNUM}/mpfr/src \ > > > --with-mpfr-lib=$(pwd)/mpfr/src/.libs > > > ; and they don't use any gmp/mpc stuff. > > > > > > Had a brief look at trac to see why the switchover at 6.8 -> 7.0 , and > > > why > > > wanting/needing to specify those options to './configure ...' explicitly; > > > but > > > nothing obvious in the time spent. > > > > > > > > > Will leave the above remarks here, as likely to return to it. > > > . > . > > > > > Matt, Bruce, > > > It looks like, from Hazel's sterling work to be communic to list shortly, > that there may be a strong case for adding the two '--with-gmp-*' configure > command-line flags back in for at least pass-1 of gcc in chapter-5. > > > Looking back through the changelog for gcc-pass1 for the past 5 years > (>= rev 8755), via > > http://wiki.linuxfromscratch.org/lfs/log/trunk/BOOK/chapter05/gcc-pass1.xml?rev=10400 > > , plus lfs-dev mailing list since lfs-6.5, it looks like the > '--with-gmp-*' may have been removed inadvertently, or at least without > explicit/overt explanation, between lfs 6.8 and 7.0, as follows. > > > Looks like the main relevant-here commits are, in chron order from > earliest-first: > ---- > * http://wiki.linuxfromscratch.org/lfs/changeset/9309 > Timestamp: 06/22/10 02:25:36 (3 years ago) ((Between lfs 6.6 & 6.7 )). > Added '--with-gmp-*' configure cmdline-opts, plus explanations in > sub-section 'The meaning of the configure options'. > (Sim. for gcc-pass2). > > * http://wiki.linuxfromscratch.org/lfs/changeset/9493 > Timestamp: 03/30/11 13:24:53 (3 years ago) ((Between lfs 6.8 and 7.0 )). > Removed '--with-gmp-*' from configure cmdline-opts, and their > explanations from sub-section 'The meaning of the configure options'. > (Sim. for gcc-pass2). > ((Also, "Message: Upgrade to GCC-4.6.0. Fixes #2847." likely should have > read '#2848' .)) > > * http://wiki.linuxfromscratch.org/lfs/changeset/9630 > Timestamp: 10/12/11 14:35:47 (2 years ago) > '--with-mpfr-*' added; no explanations under 'The meaning of ...' . > (Sim. for gcc-pass2). > > * (( > http://wiki.linuxfromscratch.org/lfs/changeset/9645 > Timestamp: 10/28/11 12:42:21 (2 years ago) > Adjust mpfr paths to use '$(pwd)' . (Sim. for gcc-pass2). > )) > > * http://wiki.linuxfromscratch.org/lfs/changeset/10380 > Timestamp: 11/23/13 15:35:47 (3 weeks ago) > Added explanation for '--with-mpfr-*' configure command-line opts, in > sub-section 'The meaning of ...' .
- that was added for gcc-pass1 only, and not for gcc-pass2; whereas the old(er) '--with-gmp-*' explanations were on both of gcc-pass1 & gcc-pass2 pages. > ((Also, "Message: Add description of GCC's --enable-mpfr switches." should > perhaps have read '--with-' rather than '--enable-' .)) > ---- > > > Any partic reason(s) for the removal in rev-9493 per above? > - and there doesn't seem to be any usual-format 'formal' email message to lfs-book for r9493 - instead it jumps from r9492 to r9494 - at least as shown at 'http://news.gmane.org/gmane.linux.lfs.book' (ca. pages 35-36 in the drop-down menu at top-right); ( 'http://www.mail-archive.com/lfs-book@linuxfromscratch.org/' seems to not go back that far.) > > Naturally you'll likely want to see Hazel's findings before any re-addition > of the '--with-gmp-*'; but thought I'd put this in a separate parallel note. > > > It may be argued that mpc/gmp/mpfr might just go into host-requirements: but > I'd say best to instead have both the '--with-gmp-*' and '--with-mpfr-*' - > perhaps also '--with-mpc-*' to be on safe side - stuff on configure > command-line for gcc-pass1 & perh/prob also gcc-pass2. > > > > rgds, > akh > > > -- -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page