https://sourceware.org/bugzilla/show_bug.cgi?id=16544
Cary Coutant <ccoutant at google dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
Assignee|ian at airs dot com |ccoutant at google dot
com
--- Comment #2 from Cary Coutant <ccoutant at google dot com> ---
The problem is that gold hasn't yet seen any real (i.e., non-plugin) .o files
to establish what the target architecture is. Before plugins, this really was a
"can't happen" path; hence the internal error. It shouldn't be diagnosed as an
internal error in this case, but we really need to see a real .o before
anything else. Normally, the compiler passes in various startup files.
There's no simple way of relaxing this requirement, short of changing the
plugin API to allow the plugin to specify the target architecture if necessary.
I can fix it to print a real diagnostic, but I don't plan to support the case
where all input files are claimed by the plugin.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils
