https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
--- Comment #8 from Alexander von Gluck ---
Looks good to me. Now also matches other architectures.
Thanks! (viva riscv)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
Jim Wilson changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
--- Comment #6 from Jim Wilson ---
Author: wilson
Date: Thu Jun 6 23:18:48 2019
New Revision: 272021
URL: https://gcc.gnu.org/viewcvs?rev=272021=gcc=rev
Log:
RISC-V: Move STARTFILE_PREFIX_SPEC into target OS files.
gcc/
PR
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
--- Comment #5 from Jim Wilson ---
OK, sounds like we need to move STARTFILE_PREFIX_SPEC into various OS header
files then. That will require some testing. I caught a virus last week and am
behind on everything, so I haven't had a chance to
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
--- Comment #4 from Alexander von Gluck ---
oh.. also some context I missed.
Our libraries aren't at /lib,/usr/lib,etc. (which is also reflected in our os
config headers)
Am I wrong to think making that assumption in the architecture level
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
--- Comment #3 from Alexander von Gluck ---
The issue here is RISC-V is the only one that does this. We do override the
STARTFILE_PREFIX_SPEC for our OS, however the architecture headers come later
and undo our undef.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
Jim Wilson changed:
What|Removed |Added
CC||wilson at gcc dot gnu.org
--- Comment #2
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89955
--- Comment #1 from Andreas Schwab ---
In which way is STARTFILE_PREFIX_SPEC interfering with sysroot? The sysroot is
expected to use the standard layout, and the directories in
STARTFILE_PREFIX_SPEC are prefixed with the sysroot before being
