https://sourceware.org/bugzilla/show_bug.cgi?id=32722
Bug ID: 32722
Summary: [objdump] vmovdqu considered to have 24-bit
displacement
Product: binutils
Version: unspecified
Status: NEW
Severity: normal
Priority: P2
Component: binutils
Assignee: unassigned at sourceware dot org
Reporter: vries at gcc dot gnu.org
Target Milestone: ---
Consider the following test-case:
...
$ cat test.s
.file "test.c"
.text
.globl main
.type main, @function
main:
.byte 0xc5
.byte 0xfe
.byte 0x6f
.byte 0x0d
.byte 0xff
.byte 0xff
.byte 0xff
.byte 0xff
.size main, .-main
.ident "GCC: (SUSE Linux) 7.5.0"
.section .note.GNU-stack,"",@progbits
...
Compiled with:
...
$ gcc test.s -c
...
Objdump thinks the instruction is 7 bytes long:
...
$ objdump -d test.o
...
0000000000000000 <main>:
0: c5 fe 6f 0d ff ff ff vmovdqu -0x1(%rip),%ymm1
7: ff
...
However, llvm-objdump thinks it's 8 bytes long:
...
$ llvm-objdump -d test.o
...
0000000000000000 <main>:
0: c5 fe 6f 0d ff ff ff ff vmovdqu -0x1(%rip), %ymm1
...
GDB likewise:
...
$ gdb -q -batch test.o -ex "disassemble /r main"
Dump of assembler code for function main:
0x0000000000000000 <+0>: c5 fe 6f 0d ff ff ff ff vmovdqu
-0x1(%rip),%ymm1
End of assembler dump.
...
The instruction is in fact 8 bytes long.
It consists of:
- c5 fe: vex2 prefix
- 6f : opcode
- 0d : modrm
- ff ff ff ff: 32-bit displacement
--
You are receiving this mail because:
You are on the CC list for the bug.
