Dean Herington reports: > '-dy -G' (note: lower-case 'd') did work! Great!
Was that combined with the -Wl, prefix to push it through gcc? Could I ask you to do a few more experiments to see what change we need in the configure script? Try tweaking hugs98/configure.ac by moving the call to HUGS_TRY_DYNLINK(-G) up about 6 lines to just before the -r flag (adding -dy and/or -Wl, if required) then delete any cache, rebuild ./configure and rerun. Hopefully it will pick the right flags this time. [But why was the config test deciding that -r was good enough? If it doesn't work in Hugs at runtime, it shouldn't work at config time. I wonder if the test needs to be strengthened to invoke a function from some other shared library?] Hmmmm, I'm having trouble ssh-ing into cvs.haskell.org so I can't update the repository directly but if you report back to hugs-bugs, I'm sure some kind soul will add it in. -- Alastair Reid _______________________________________________ Hugs-Bugs mailing list [EMAIL PROTECTED] http://www.haskell.org/mailman/listinfo/hugs-bugs
