[Bug binutils/19145] New: [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 Bug ID: 19145 Summary: [regression] arm-none-eabi big-endian LTO broken Product: binutils Version: unspecified Status: NEW Severity: normal Priority: P2 Component: binutils Assignee: unassigned at sourceware dot org Reporter: jonathan at codesourcery dot com Target Milestone: --- This is a continuation of: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67871 $ touch empty.c $ arm-none-eabi-gcc -mbig-endian -flto empty.c /tmp/ccUtMyOv.o: file not recognized: File format is ambiguous /tmp/ccUtMyOv.o: matching formats: elf32-bigarm elf32-big collect2: error: ld returned 1 exit status I've git-bisected, keeping the gcc revision fixed to ToT, and following commit seems to be the culprit: commit 5ae0078cd2b6b69e6119864e20987c8724916b29 Author: H.J. Lu Date: Wed Feb 11 05:01:03 2015 -0800 Merge linker plugin handling into BFD plugin support Linker plugin_maybe_claim is the interface of linker plugin support. This patch extracts linker plugin_maybe_claim into plugin_object_p and makes it available to BFD via a new function: void register_ld_plugin_object_p (const bfd_target *(*) (bfd *)); bfd_plugin_object_p calls plugin_object_p registered by linker first. It adds an enum bfd_plugin_format field and a pointer to plugin dummy BFD so that plugin_object_p stores plugin dummy BFD to allow plugin_maybe_claim to retrieve it later. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 Jonathan Roelofs changed: What|Removed |Added CC||hjl.tools at gmail dot com -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 --- Comment #2 from Jonathan Roelofs --- Created attachment 8727 --> https://sourceware.org/bugzilla/attachment.cgi?id=8727&action=edit object file compiled from empty source file $ rm empty.c $ touch empty.c $ ./install/bin/arm-none-eabi-gcc -mbig-endian -flto empty.c -save-temps empty.o: file not recognized: File format is ambiguous empty.o: matching formats: elf32-bigarm elf32-big collect2: error: ld returned 1 exit status $ ./install/bin/arm-none-eabi-readelf -h empty.o ELF Header: Magic: 7f 45 4c 46 01 02 01 00 00 00 00 00 00 00 00 00 Class: ELF32 Data: 2's complement, big endian Version: 1 (current) OS/ABI:UNIX - System V ABI Version: 0 Type: REL (Relocatable file) Machine: ARM Version: 0x1 Entry point address: 0x0 Start of program headers: 0 (bytes into file) Start of section headers: 888 (bytes into file) Flags: 0x500, Version5 EABI Size of this header: 52 (bytes) Size of program headers: 0 (bytes) Number of program headers: 0 Size of section headers: 40 (bytes) Number of section headers: 15 Section header string table index: 12 $ ./install/bin/arm-none-eabi-objdump -f empty.o empty.o: file format elf32-bigarm architecture: armv4t, flags 0x0010: HAS_SYMS start address 0x -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 --- Comment #3 from Jonathan Roelofs --- Unfortunately, that doesn't fix it. My gcc and binutils configurations are described here: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67871#c3 I've attached the object file that I get from compiling an empty source file (in case that reduces the number of things you need to build in order to reproduce this). With that, you should see: $ ./install/bin/arm-none-eabi-ld -EB -flto empty.o -plugin ./install/libexec/gcc/arm-none-eabi/6.0.0/liblto_plugin.so empty.o: file not recognized: File format is ambiguous empty.o: matching formats: elf32-bigarm elf32-big -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145
--- Comment #5 from Jonathan Roelofs ---
(In reply to H.J. Lu from comment #4)
> There are 2 issues:
>
> 1. arm-none-eabi binutils only supports litte-endian.
> 2. arm-none-eabi GCC only supports little-endian.
>
> To support big-endian, you should configure binutils and GCC for
> armeb-none-eabi. Or you fix ARM EABI binutils and GCC to support
> both little-endian and big-endian.
I'm not convinced that's actually true. We've shipped working
arm-none-eabi-{gcc,binutils} toolchains with both the `be;@mbig-endian`
multilib and the `.;` (little-endian) multilib (among others) before.
>From the gcc testsuite results, it very much looks like the only difference in
failures between the little-endian multilib and the big-endian one is that any
testcase that uses -flto fails for the big-endian multilib.
This is definitely a regression.
--
You are receiving this mail because:
You are on the CC list for the bug.
___
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 --- Comment #7 from Jonathan Roelofs --- $ arm-none-eabi-ld -V GNU ld (GNU Binutils) 2.25.51.20150929 Supported emulations: armelf -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 --- Comment #9 from Jonathan Roelofs --- Created attachment 8735 --> https://sourceware.org/bugzilla/attachment.cgi?id=8735&action=edit fix implementation (binutils) Ah, thanks. That will work. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 --- Comment #10 from Jonathan Roelofs --- Created attachment 8736 --> https://sourceware.org/bugzilla/attachment.cgi?id=8736&action=edit fix implementation (gcc) -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 --- Comment #11 from Jonathan Roelofs --- One of my coworkers reminds me that we've done the same thing for linux before: https://sourceware.org/ml/binutils/2006-10/msg00181.html -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/19145] [regression] arm-none-eabi big-endian LTO broken
https://sourceware.org/bugzilla/show_bug.cgi?id=19145 --- Comment #12 from Jonathan Roelofs --- https://sourceware.org/ml/binutils/2015-10/msg00181.html -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
