This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
Hello As I proposed on http://gcc.gnu.org/ml/gcc/2006-11/msg00803.html and http://gcc.gnu.org/ml/gcc/2006-11/msg00804.html it could be useful to propose some options (flags at GCC runtime) which can be fully disabled at configure time. Also the Condition keyword in *.opt file seems never used anymore, but does not disable entirely an option when the preprocessor condition is false. (the generated flag variables in $GCCBUILD/gcc/options.h still exist, and the --help option mentions such a flag). Since Condition is not used, I suggest to mark it as obsolete. The motivation for such a feature is to permit some expensive processing in GCC to be enabled or disabled at configure time, and when it is enabled, to offer an additional GCC?runtime flag which triggers it. As a simplistic scenario, the --enable-check=tree configure option enable the build of $GCCBUILD/gcc/tree-browser.o (which is not even built from $GCCTOP/gcc/tree-browser.c without this --enable-check!), this tree-browser.o offers a browse_tree function, and we could in that case only propose a -fbrowse-tree flag which calls this function at several interesting points in the compiler. If configured without any --enable-check the compiler won't even accept such a -fbrowse-tree flag I intend to propose in the future another patch which make a real use of the CppCondition offered here. The attached patch to rev. 119231 of GCC trunk does only provide option handling (and does not propose its use, as illustrated above). In addition it changes the generted comment in generated options.[ch] files (which mentions an opts.sh which don't exist anymore). Bootstraped with --disable-multilib --enable-languages=c on x86_64-unknown-linux-gnu (Debian/Sid/AMD64) Regarding legalese: my copyright assignment (signed for CEA-LIST my employer to FSF) has been sent and the FSF (J.Jacobson) should have recieved it before monday 27th november 2006 evening. Jonas Jacobson have all the relevant details. Changelog follows my signature Regards -- Basile STARYNKEVITCH http://starynkevitch.net/Basile/ email: basile<at>starynkevitch<dot>net mobile: +33 6 8501 2359 8, rue de la Faïencerie, 92340 Bourg La Reine, France *** opinions {are only mines, sont seulement les miennes} *** *********************************** Changelog 2006-11-26 Basile Starynkevitch <basile@starynkevitch.net> * gcc/doc/options.texi: Condition is marked as obsolete, and added a description of CppCondition for flags which can be entirely disabled at configure time. * gcc/optc-gen.awk: generated comment tells about optc-gen.awk. Added handling of CppCondition. * gcc/opth-gen.awk:? generated comment tells about opth-gen.awk. Added handling of CppCondition.
Attachment:
gcc-trunk-rev119231-patch-cppoption.diff
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |