On Wed, Dec 26, 2007 at 05:19:40PM -0200, Luís Vitório Cargnini wrote: > well according to the book > http://cross-lfs.org/view/clfs-sysroot/arm/cross-tools/binutils.html: > this is the result after make: > && mv BLD-POTFILES-t BLD-POTFILES ) > cd .. \ > && CONFIG_FILES=po/Makefile.in:po/Make-in \ > CONFIG_HEADERS= /bin/sh ./config.status > config.status: creating po/Makefile.in > config.status: executing depfiles commands > config.status: executing libtool commands > config.status: executing default-1 commands > config.status: executing bfd_stdint.h commands > config.status: executing default commands > make[3]: Leaving directory > `/home/lvcargnini/puc/instramed/toolchain/build/binutils218/bfd/po' > make[3]: Entering directory > `/home/lvcargnini/puc/instramed/toolchain/build/binutils218/bfd/po' > make[3]: Nothing to be done for `info'. > make[3]: Leaving directory > `/home/lvcargnini/puc/instramed/toolchain/build/binutils218/bfd/po' > make[3]: Entering directory > `/home/lvcargnini/puc/instramed/toolchain/build/binutils218/bfd' > make[3]: Nothing to be done for `info-am'. > make[3]: Leaving directory > `/home/lvcargnini/puc/instramed/toolchain/build/binutils218/bfd' > make[2]: *** [info-recursive] Error 1 > make[2]: Leaving directory > `/home/lvcargnini/puc/instramed/toolchain/build/binutils218/bfd' > make[1]: *** [all-bfd] Error 2 > make[1]: Leaving directory > `/home/lvcargnini/puc/instramed/toolchain/build/binutils218' > make: *** [all] Error 2 > > > how to fix this ? > this was I tried the binutils 2.17 instead of 2.18 > Please note that we often get upset with people who top-post: putting a reply underneath the appropriate part of the text helps people follow what is being written.
Also, you hadn't explained that 2.18 had failed, and you haven't commented on why you were using a different --prefix for what you posted earlier. This means everybody has to guess, and we will often guess wrongly, so please help us to help you. Any other variations or problems with it that you'd like to mention ? Thanks. But, the reason why binutils-2.18 is failing ? Can't tell - all we can see from what you posted above is that a /bin/sh ./config.status command ran, apparently in bfd/po, and that it was trying to make the info pages. Perhaps it had an error that didn't result in an error message in your log (unlikely). Probably, somewhere before that (perhaps a long way) should be the original error. Trimming the output is good, perhaps leave a few lines ahead of the original 'Error' or 'error' to show any failing command or interesting messages, then snip anything afterwards, it probably isn't going to bring any useful information (but log the output to a file so that you can go back if somebody needs to ask for more details). If you can't find anything, have a look at the config.status script (probably in bfd/po or else in bfd ? ) - look at what it does, if nothing strikes you as obviously wrong (e.g. trying to run a command from an unset variable) try instrumenting it in an editor, e.g. to echo text telling you where it is, and what it is going to do, with the values of any variables, and the return value in "$?" after running commands. That is long-winded and tedious, but you need to identify where the problem is. With luck, if you have to do this you'll get lucky and spot something which looks odd. Because it is in the info area, I wonder if your host system is missing the info headers (texinfo-dev or a similarly-named package), or perhaps has a too-old version. But a quick look into the toplevel configure script suggests it will use the 'missing' program in the toplevel if makeinfo is older than 4.4. That is supposed to touch the output file so that an empty file exists it none was present. I think there used to be a bug where something got confused by versions of makeinfo with a double-digit second level (4.10, 4.11). I never saw that, so I don't know what the symptoms are, but maybe the original error message will point to that sort of thing. If so, I would be tempted to try the binutils branch update patch we are using in trunk: http://trac.cross-lfs.org/browser/trunk/patches/binutils-2.18-branch_update-3.patch If that allows 2.18 to build, I think a better solution (unless the patch arrives in the -sysroot book in the meantime) would be to identify where in the configure script(s) the error happened, then compare that test to the new version and produce a patch for just that part of the file (the sed which checks the version) and use that in your build. I say that because the branch_update patches are on the bleeding edge, and for a project with few visible users (-sysroot) you could end up with other problems on a little-used architecture. Of course, if you are happy to debug any resulting problems, testing the branch update is better for the long-term health of the project. If you are able to pin this down to a too-new version of makeinfo on the host, please raise a ticket against binutils in clfs-sysroot (unless that gets updated first). And if I was totally wide of the mark and the problem is something else entirely, please disregard the previous 4 paragrahs ;-) ĸen -- das eine Mal als Tragödie, das andere Mal als Farce _______________________________________________ Clfs-dev mailing list Clfs-dev@lists.cross-lfs.org http://lists.cross-lfs.org/listinfo.cgi/clfs-dev-cross-lfs.org
