https://bugs.llvm.org/show_bug.cgi?id=42277
Bug ID: 42277
Summary: llvm-objcopy does not rip off .bss section
Product: tools
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: release blocker
Priority: P
Component: llvm-objcopy/strip
Assignee: unassignedb...@nondot.org
Reporter: masahi...@kernel.org
CC: alexander.v.shaposhni...@gmail.com,
jake.h.ehrl...@gmail.com,
jh7370.2...@my.bristol.ac.uk,
llvm-bugs@lists.llvm.org, ruppre...@google.com
objcopy is often used to convert an ELF file into a raw binary.
In my workflow, I build my firmware, covert it into raw binary by the following
command, then burn it into my embedded board.
$ llvm-objcopy -O binary -R .comment --strip-all my-input.elf my-output.bin
This worked fine until LLVM 8,
and in the same way as GNU toolchain objcopy.
If I use the llvm-objcopy from the latest trunk,
it does not work any more.
The first bad commit is:
commit 2b68379b89263178917e3bd37eaf9154078c2965 (HEAD)
Author: Peter Collingbourne <pe...@pcc.me.uk>
Date: Fri May 24 00:21:46 2019 +0000
llvm-objcopy: Change sectionWithinSegment() to use virtual addresses
instead of file offsets for SHT_NOBITS sections.
Without this, sectionWithinSegment() will return the wrong answer for bss
sections. This doesn't seem to matter now (for non-broken ELF files), but
it will matter with a change that I'm working on.
Differential Revision: https://reviews.llvm.org/D58426
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361578
91177308-0d34-0410-b5e6-96231b3b80d8
Since this commit, llvm-objcopy does not drop .bss section.
So, the produced raw binary is bigger than it should be
since it contains the .bss section.
I patched my build script by adding '-R .bss' explcitly.
$ llvm-objcopy -O binary -R .comment --strip-all -R .bss my-input.elf
my-output.bin
However, I do not think this is the right thing to do.
There is no need to load .bss section, nor should it be included in the image.
Moreover, now llvm-objcopy works differently from GNU toolchain objcopy.
Please take a look.
Thanks.
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
