[Bug ld/3132] New: Incorrect 16-bit relocations in ELF

[markus+sw at gyger dot org]

ELF on ld 2.17 incorrectly calculates 8- and 16-bit relocations that worked with
COFF on ld 2.16.2.

  h8300-coff-as -oreloc.o reloc.s
  h8300-coff-ld -mh8300s --oformat binary reloc.o
  od -tx1 a.out
correctly yields:
  0000000 6a 08 e0 02 00 02

while

  h8300-elf-as -oreloc.o reloc.s
  h8300-elf-ld -mh8300self --oformat binary reloc.o
yields:
  reloc.o:(.text+0x2): relocation truncated to fit: R_H8_DIR16A8 against `addr'

and

  h8300-elf-as -oreloc.o reloc.s
  h8300-elf-ld -mh8300self reloc.o
  h8300-elf-objdump -d a.out
  h8300-elf-objcopy -Obinary a.out
  od -tx1 a.out
yields:
  0000000 6a 08 e0 02 01 06


Content of reloc.s:
.h8300s
_start = 0
.global _start

mov @addr + 2 :16, r0l
addr = 0xffe000

.word label - .
label:

-- 
           Summary: Incorrect 16-bit relocations in ELF
           Product: binutils
           Version: 2.17
            Status: NEW
          Severity: normal
          Priority: P2
         Component: ld
        AssignedTo: unassigned at sources dot redhat dot com
        ReportedBy: markus+sw at gyger dot org
                CC: bug-binutils at gnu dot org
 GCC build triplet: i386-pc-solaris2.10
  GCC host triplet: i386-pc-solaris2.10
GCC target triplet: h8300-elf


http://sourceware.org/bugzilla/show_bug.cgi?id=3132

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.