[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Bug binutils/20553] New: Bug in objdump when disassembling fmul,fmla an
From: |
weixiangyu at huawei dot com |
Subject: |
[Bug binutils/20553] New: Bug in objdump when disassembling fmul,fmla and fmls |
Date: |
Mon, 05 Sep 2016 06:56:28 +0000 |
https://sourceware.org/bugzilla/show_bug.cgi?id=20553
Bug ID: 20553
Summary: Bug in objdump when disassembling fmul,fmla and fmls
Product: binutils
Version: 2.26
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: binutils
Assignee: unassigned at sourceware dot org
Reporter: weixiangyu at huawei dot com
Target Milestone: ---
1. Overview:
I have recently found that 3 half-precision instructions( e.g. FMUL <Vd>.<T>,
<Vn>.<T>, <Vm>.<Ts>[<index>] ) have some problems when doing disassemble: the
insn can't be recognized when its index is larger than 1.
2. Steps to reproduce:
address@hidden:~/work> cat test.s
fmla v0.8h, v1.8h, v2.h[2]
fmls v0.8h, v1.8h, v2.h[2]
fmul v0.8h, v1.8h, v2.h[2]
address@hidden:~/work> aarch64_be-linux-gnu-gcc -c test.s
address@hidden:~/work> aarch64_be-linux-gnu-objdump -d test.o
3. Actual results
test.o: file format elf64-bigaarch64
Disassembly of section .text:
0000000000000000 <.text>:
0: 4f221020 .inst 0x4f221020 ; undefined
4: 4f225020 .inst 0x4f225020 ; undefined
8: 4f229020 .inst 0x4f229020 ; undefined
My toolchain was built with binutils 2.26 and gcc-4.9.3, and as far as I
know,this problem also exists in binutils 2.27.
--
You are receiving this mail because:
You are on the CC list for the bug.
- [Bug binutils/20553] New: Bug in objdump when disassembling fmul,fmla and fmls,
weixiangyu at huawei dot com <=