| Summary: | sci-chemistry/rasmol-2.7.5.2-r1 fails to build on ppc due to "vector" being munged during imake | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Pacho Ramos <pacho> |
| Component: | Current packages | Assignee: | Gentoo Chemistry-Related Packages <sci-chemistry> |
| Status: | RESOLVED WONTFIX | ||
| Severity: | normal | CC: | ppc, vapier, x11 |
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | PPC | ||
| OS: | Linux | ||
| See Also: |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63413 https://bugs.gentoo.org/show_bug.cgi?id=445606 |
||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: | no-vector ebuild | ||
|
Description
Pacho Ramos
2015-04-21 16:54:20 UTC
It look sot me as if it is a toolchain bug. Guys, what do you say? (In reply to Justin Lecher from comment #1) it depends on the context. sometimes "vector" is a reserved word on ppc, and sometimes it is not. maybe something like this would help in src_prepare: append-cppflags -Uvector (In reply to SpanKY from comment #2) > src_prepare: > append-cppflags -Uvector Pacho, could you please try this? Looks to fail with the same :( >>> Emerging (1 of 1) sci-chemistry/rasmol-2.7.5.2-r1::gentoo * rasmol-2.7.5.2-13May11.tar.gz SHA256 SHA512 WHIRLPOOL size ;-) ... [ ok ] * Using following Fortran compiler: * F77: powerpc-unknown-linux-gnu-gfortran * FC: powerpc-unknown-linux-gnu-gfortran >>> Unpacking source... >>> Unpacking rasmol-2.7.5.2-13May11.tar.gz to /var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r1/work >>> Source unpacked in /var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r1/work >>> Preparing source in /var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r1/work/RasMol-2.7.5.2 ... * Applying rasmol-2.7.5.2-glib.h.patch ... [ ok ] * Applying rasmol-2.7.5.2-format-security.patch ... [ ok ] * Applying rasmol-2.7.5.2-longlong.patch ... [ ok ] * Applying 2.7.5-bundled-lib.patch ... [ ok ] * Adjusting to prefix / * Imakefile ... [ ok ] imake -DGTKWIN -DUseInstalled -I/usr/lib/X11/config In file included from /usr/lib/X11/config/Imake.tmpl:2194:0, from Imakefile.c:34: /tmp/IIf.S4s1DY:292:0: error: detected recursion whilst expanding macro "vector" multiple.h multiple.c vector.h vector.c \ ^ /tmp/IIf.S4s1DY:292:0: error: detected recursion whilst expanding macro "vector" /tmp/IIf.S4s1DY:300:0: error: detected recursion whilst expanding macro "vector" multiple.o vector.o wbrotate.o langsel_unix.o maps.o $(GUIOBJ) ^ imake: Exit code 1. Stop. * ERROR: sci-chemistry/rasmol-2.7.5.2-r1::gentoo failed (prepare phase): * xmkmf failed with * * Call stack: * ebuild.sh, line 93: Called src_prepare * environment, line 2430: Called die * The specific snippet of code: * xmkmf -DGTKWIN ${myconf} || die "xmkmf failed with ${myconf}"; * * If you need support, post the output of `emerge --info '=sci-chemistry/rasmol-2.7.5.2-r1::gentoo'`, * the complete build log and the output of `emerge -pqv '=sci-chemistry/rasmol-2.7.5.2-r1::gentoo'`. * The complete build log is located at '/var/log/portage/build/sci-chemistry/rasmol-2.7.5.2-r1:20150423-191214.log'. * For convenience, a symlink to the build log is located at '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r1/temp/build.log'. * The ebuild environment file is located at '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r1/temp/environment'. * Working directory: '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r1/work/RasMol-2.7.5.2/src' * S: '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r1/work/RasMol-2.7.5.2' it's failing because imake runs the preprocessor with -traditional. the powerpc target will '#define vector ...' when it is not operating in iso mode. which is exactly what the -traditional flag does -- turn off iso mode. if imake wants to keep the -traditional flag, then it can be worked around in rasmol itself by adding to the src/Imakefile a line like: #undef vector since gcc is specifically being invoked in non-iso mode, i don't think we'll look at changing it. +*rasmol-2.7.5.2-r2 (18 May 2015) + + 18 May 2015; Justin Lecher <jlec@gentoo.org> +rasmol-2.7.5.2-r2.ebuild: + Drop -traditional from CPPFLAGS, bug #547318; enhance preprocessor definitions + Could you please try -r2? Still fails :( >>> Emerging (1 of 1) sci-chemistry/rasmol-2.7.5.2-r2::gentoo * rasmol-2.7.5.2-13May11.tar.gz SHA256 SHA512 WHIRLPOOL size ;-) ... [ ok ] * Using following Fortran compiler: * F77: powerpc-unknown-linux-gnu-gfortran * FC: powerpc-unknown-linux-gnu-gfortran >>> Unpacking source... >>> Unpacking rasmol-2.7.5.2-13May11.tar.gz to /var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r2/work >>> Source unpacked in /var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r2/work >>> Preparing source in /var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r2/work/RasMol-2.7.5.2 ... * Applying rasmol-2.7.5.2-glib.h.patch ... [ ok ] * Applying rasmol-2.7.5.2-format-security.patch ... [ ok ] * Applying rasmol-2.7.5.2-longlong.patch ... [ ok ] * Applying 2.7.5-bundled-lib.patch ... [ ok ] * Adjusting to prefix / * Imakefile ... [ ok ] imake -DGTKWIN -DUseInstalled -I/usr/lib/X11/config In file included from /usr/lib/X11/config/Imake.tmpl:2194:0, from Imakefile.c:34: /tmp/IIf.beXG8v:2:0: warning: "GTKWIN" redefined [enabled by default] #define GTKWIN ^ <command-line>:13:0: note: this is the location of the previous definition In file included from /usr/lib/X11/config/Imake.tmpl:2194:0, from Imakefile.c:34: /tmp/IIf.beXG8v:294:0: error: detected recursion whilst expanding macro "vector" multiple.h multiple.c vector.h vector.c \ ^ /tmp/IIf.beXG8v:294:0: error: detected recursion whilst expanding macro "vector" /tmp/IIf.beXG8v:302:0: error: detected recursion whilst expanding macro "vector" multiple.o vector.o wbrotate.o langsel_unix.o maps.o $(GUIOBJ) ^ imake: Exit code 1. Stop. * ERROR: sci-chemistry/rasmol-2.7.5.2-r2::gentoo failed (prepare phase): * xmkmf failed with * * Call stack: * ebuild.sh, line 93: Called src_prepare * environment, line 2140: Called die * The specific snippet of code: * xmkmf -DGTKWIN || die "xmkmf failed with ${myconf}" * * If you need support, post the output of `emerge --info '=sci-chemistry/rasmol-2.7.5.2-r2::gentoo'`, * the complete build log and the output of `emerge -pqv '=sci-chemistry/rasmol-2.7.5.2-r2::gentoo'`. * The complete build log is located at '/var/log/portage/build/sci-chemistry/rasmol-2.7.5.2-r2:20150518-195100.log'. * For convenience, a symlink to the build log is located at '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r2/temp/build.log'. * The ebuild environment file is located at '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r2/temp/environment'. * Working directory: '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r2/work/RasMol-2.7.5.2/src' * S: '/var/tmp/portage/sci-chemistry/rasmol-2.7.5.2-r2/work/RasMol-2.7.5.2' I have no idea. The -traditional seems not to be the problem. Created attachment 403606 [details]
no-vector ebuild
Could you please try this ebuild?
(In reply to Justin Lecher from comment #8) imake itself has -traditional in it: $ strings -a /usr/bin/imake | grep -e-trad -traditional as does the X11 config files that imake reads: $ grep -e-traditional /usr/lib/X11/* /usr/lib/X11/config/linux.cf:#define StandardCppOptions -traditional that's why i cc-ed the x11 team here. (In reply to Justin Lecher from comment #9) > Created attachment 403606 [details] > no-vector ebuild > > Could you please try this ebuild? Should I wait then for x11 team reply to vapier's comment #10? :) (In reply to Pacho Ramos from comment #11) > (In reply to Justin Lecher from comment #9) > > Created attachment 403606 [details] > > no-vector ebuild > > > > Could you please try this ebuild? > > Should I wait then for x11 team reply to vapier's comment #10? :) X11 must help. The ebuild works around a problem which I found in the web in where the preprocessor gets confused by vector in comments. It's just a try, but could work. @jlec, it works! \o/ :D (In reply to Pacho Ramos from comment #13) > @jlec, it works! \o/ :D So it is the bug Mike referenced to. Thanks for testing. I'm marking as WONTFIX since I'm dropping rasmol's ppc keywords. |