https://sourceware.org/bugzilla/show_bug.cgi?id=16504
Cary Coutant <ccoutant at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution|--- |FIXED
--- Comment #6 from Cary Coutant <ccoutant at gmail dot com> ---
Fixed on master branch.
The provided test case leaves me with a nervous feeling that
neither IFUNCs nor .symver directives are a good idea when
using LTO. For .symver at least, the compiler doesn't see the
real symbol name (it's just an asm pass-thru), so it could be
missing some important linkage information. For example, GCC
doesn't even tell the linker about the reference to
sd_get_seats@LIBSYSTEMD_209 when reading the IR, which could
affect the correctness of its optimizations.
It could theoretically figure out how IFUNCs work by
looking at the resolve function, but I'm not sure it does.
--
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
