[Bug ld/22920] Segfault in ld with RISC-V binary target

[geoffrey at allott dot email]

https://sourceware.org/bugzilla/show_bug.cgi?id=22920

--- Comment #2 from Geoffrey Allott <geoffrey at allott dot email> ---
Hi Jim, thanks for responding. Objcopy was indeed the way I ended up doing it -
since it's a relatively simple workaround it makes sense that this is low
priority, though an error message would have been useful here if it's known to
be unimplemented.

Sent from my Samsung Galaxy smartphone.
-------- Original message --------From: "wilson at gcc dot gnu.org"
<address@hidden> Date: 06/03/2018  1:11 am  (GMT+00:00) To:
address@hidden Subject: [Bug ld/22920] Segfault in ld with RISC-V binary
target 
https://sourceware.org/bugzilla/show_bug.cgi?id=22920

Jim Wilson <wilson at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2018-03-06
                 CC|                            |wilson at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #1 from Jim Wilson <wilson at gcc dot gnu.org> ---
Your linker script sets the output format to binary.  Unfortunately, the
current RISC-V linker only works for ELF output format.  If you want non-elf
output, you need to generate ELF first, and then use objcopy to convert it to
another format.

I've seen a few similar problems.  There are srec testcases in the linker
testsuite that fail because we can't produce srec directly.  I think there is
also a related bug report for this in the github riscv-binutils project.

This should be fixed, but we have no time table for when it will be fixed, as
we have many higher priority issues to look at first.

-- 
You are receiving this mail because:
You are on the CC list for the bug.