bug-binutils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug ld/29998] ld terminated with signal 11 [Segmentation fault] under m


From: nickc at redhat dot com
Subject: [Bug ld/29998] ld terminated with signal 11 [Segmentation fault] under mingw with LTO
Date: Mon, 23 Jan 2023 15:38:31 +0000

https://sourceware.org/bugzilla/show_bug.cgi?id=29998

--- Comment #14 from Nick Clifton <nickc at redhat dot com> ---
(In reply to Jan Janssen from comment #12)
> Now it crashes here (sec->output_section is NULL):
> 
> #0  _bfd_coff_generic_relocate_section (output_bfd=0x55555581c000,

*sigh*.  Yes I suspected that that might happen.

(In reply to Jan Janssen from comment #13)
> Dunno why it it not occur to me to do a bisection…
> 514b4e191d5f46de8e142fe216e677a35fa9c4bb is the first bad commit

The patch itself is not necessarily bad, but it does give me an idea.  One of
the changes it makes is to change the linker's default behaviour so that unless
asked otherwise it will generate a .reloc output section.  Since this is the
section that is giving us problems, maybe that is the cause.  As a possible
workaround, please could you try adding "-Wl,--disable-reloc-section" to your
final gcc command line.

My guess is that now .reloc sections are being generated by default, there are
circumstances where there is insufficient information to generate them.  But
without a test case I can use locally, I am not going to be able to investigate
further.

-- 
You are receiving this mail because:
You are on the CC list for the bug.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]