<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW --- - cairo fails to compile with gcc 4.9" href="https://bugs.freedesktop.org/show_bug.cgi?id=77060#c6">Comment # 6</a> on <a class="bz_bug_link bz_status_NEW " title="NEW --- - cairo fails to compile with gcc 4.9" href="https://bugs.freedesktop.org/show_bug.cgi?id=77060">bug 77060</a> from <a class="email" href="mailto:psychon@znc.in" title="Uli Schlachter <psychon@znc.in>"> Uli Schlachter</a> <pre>Created <a href="attachment.cgi?id=102703" name="attach_102703" title="Self-contained Makefile to reproduce this issue">attachment 102703</a> <a href="attachment.cgi?id=102703&action=edit" title="Self-contained Makefile to reproduce this issue">[details]</a> Self-contained Makefile to reproduce this issue So I tried to reproduce this issue outside of cairo. The result is the attached Makefile. Then I found the GCC LTO FAQ which states that you have to use gcc-ar, gcc-ranlib and gcc-nm instead of the tools without those prefixes. With my Makefile I tested this and noticed (after some comparing to what cairo does) that "ranlib" breaks the .a files that cairo builds by replacing the symbol index. Using my Makefile with "make RANLIB=gcc-ranlib" is enough to make it work. Then I fought a bit against cairo's configure script and can now build cairo again, if I use ./configure ac_cv_prog_RANLIB=gcc-ranlib RANLIB="gcc-ranlib" AR="gcc-ar" (No, I do not know why just "RANLIB=gcc-ranlib" isn't enough. It's just silently ignored and libtool will happily continue to use ranlib.) We can just ignore this issue until a binutils version with plugin support is used everywhere, at which point "ranlib" and "gcc-ranlib" are equally fine. Besides that, any idea for fixing this properly?</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>