https://sourceware.org/bugzilla/show_bug.cgi?id=21961
Bug ID: 21961
Summary: --orphan-handling=error fails in 2.29 because of
changed .group handling
Product: binutils
Version: 2.29
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: ld
Assignee: unassigned at sourceware dot org
Reporter: franz.sirl at lauterbach dot com
Target Milestone: ---
This testcase highlights a change in 2.29 .group handling:
/* group.c
*
* Compile/link like this:
* gcc -c -o group.o group.c
* ld --orphan-handling=error -emain -Bstatic -o a.out group.o
*
* Produces output like this:
* ld: error: unplaced orphan section '.group' from 'group.o'.
*
* Problem: .group not handled in default linker script(s)
* + binutils-2.29 producing .group sections for C++ code
*
* Tested with binutils-2.29 + gcc-7.2.0 on x86_64-linux-gnu and
powerpc-unknown-eabi
*/
/* assembler source copied from binutils testsuite group-3.s */
asm( "\t.section .text.foo3,\"axG\",%progbits,foo3,comdat\n"
"\t.global foo3\n"
"foo3:\n"
"\t.word 0\n"
"\t.section .data.bar3,\"awG\",%progbits,foo3,comdat\n"
"\t.global bar3\n"
"bar3:\n"
"\t.word 0");
int main(int argc, char *argv[])
{
return 0;
}
void __eabi(void)
{
}
I'm not sure what exactly the bug is, either the .group sections shouldn't
reach the final link stage or the default linker scripts need updating.
binutils-2.28 links the same testcase without error.
The problem originally happened with an embedded C++ codebase
(PowerPC32+newlib).
--
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
