public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
From: "H.J. Lu" <hjl.tools@gmail.com>
To: Fangrui Song <i@maskray.me>
Cc: Binutils <binutils@sourceware.org>,
	GNU C Library <libc-alpha@sourceware.org>
Subject: Re: [PATCH 05/10] ld: Initial DT_RELR support
Date: Sat, 8 Jan 2022 05:19:55 -0800	[thread overview]
Message-ID: <CAMe9rOp7AEYnP8-i0jQVcXA93c4Kw6zZ-ids2fp2rUpiD5CizQ@mail.gmail.com> (raw)
In-Reply-To: <20220108083044.mkom7nyt72kedoxz@gmail.com>

On Sat, Jan 8, 2022 at 12:30 AM Fangrui Song <i@maskray.me> wrote:
>
> On 2022-01-07, H.J. Lu via Binutils wrote:
> >Add a -z pack-relative-relocs option to enable DT_RELR and create a
> >relr.dyn section for DT_RELR.
> >
> >-z pack-relative-relocs implies -z combreloc and --relax.  --no-relax
> >and -z nocombreloc imply -z nopack-relative-relocs.
> >
> >-z pack-relative-relocs is chosen over the similar option in lld,
> >--pack-dyn-relocs=relr, to implement a glibc binary lockout mechanism
> >with a special glibc version symbol, to avoid random crashes of DT_RELR
> >binaries with the existing glibc binaries.
>
> I think -z pack-relative-relocs is orthogonal to -z combreloc.
> -z combreloc affects how .rela.dyn is sorted and whether RELACOUNT is
> generated. It has nothing to do with .relr.dyn, which is a different section.

With -z nocombreloc, ld can generate:

$ readelf -r tmpdir/pr25754-3ba

Relocation section '.rela.data.rel.ro' at offset 0x540 contains 1 entry:
  Offset          Info           Type           Sym. Value    Sym. Name + Addend
000000003de8  000000000008 R_X86_64_RELATIVE                    3de8

Relocation section '.rela.got' at offset 0x558 contains 5 entries:
  Offset          Info           Type           Sym. Value    Sym. Name + Addend
000000003fd0  000100000006 R_X86_64_GLOB_DAT 0000000000000000
__libc_start_main@GLIBC_2.34 + 0
000000003fd8  000200000006 R_X86_64_GLOB_DAT 0000000000000000
_ITM_deregisterTM[...] + 0
000000003fe8  000400000006 R_X86_64_GLOB_DAT 0000000000000000 __gmon_start__ + 0
000000003ff0  000500000006 R_X86_64_GLOB_DAT 0000000000000000
_ITM_registerTMCl[...] + 0
000000003ff8  000600000006 R_X86_64_GLOB_DAT 0000000000000000
__cxa_finalize@GLIBC_2.2.5 + 0

Relocation section '.rela.fini_array' at offset 0x5d0 contains 1 entry:
  Offset          Info           Type           Sym. Value    Sym. Name + Addend
000000003de0  000000000008 R_X86_64_RELATIVE                    10f0

Relocation section '.rela.init_array' at offset 0x5e8 contains 1 entry:
  Offset          Info           Type           Sym. Value    Sym. Name + Addend
000000003dd8  000000000008 R_X86_64_RELATIVE                    1130

Relocation section '.rela.plt' at offset 0x600 contains 1 entry:
  Offset          Info           Type           Sym. Value    Sym. Name + Addend
000000004018  000300000007 R_X86_64_JUMP_SLO 0000000000000000
puts@GLIBC_2.2.5 + 0

3 R_X86_64_RELATIVEs are 3 different .rela sections.
If they are put in a single section, they are combined.

-- 
H.J.

  reply	other threads:[~2022-01-08 13:20 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-07 19:06 [PATCH 0/10] ld: Implement DT_RELR for x86 H.J. Lu
2022-01-07 19:06 ` [PATCH 01/10] ld: Extract _bfd_elf_link_iterate_on_relocs H.J. Lu
2022-01-07 19:06 ` [PATCH 02/10] elf: Add .relr.dyn to special_sections_r H.J. Lu
2022-01-07 19:06 ` [PATCH 03/10] elf: Extract _bfd_elf_process_reverse_copy H.J. Lu
2022-01-07 19:06 ` [PATCH 04/10] elf: Pass need_layout to _bfd_elf_map_sections_to_segments H.J. Lu
2022-01-07 19:06 ` [PATCH 05/10] ld: Initial DT_RELR support H.J. Lu
2022-01-08  2:10   ` Fangrui Song
2022-01-08  2:42     ` H.J. Lu
2022-01-08 18:32       ` H.J. Lu
2022-01-08  8:30   ` Fangrui Song
2022-01-08 13:19     ` H.J. Lu [this message]
2022-01-07 19:06 ` [PATCH 06/10] elf: Add size_relative_relocs and finish_relative_relocs H.J. Lu
2022-01-07 19:06 ` [PATCH 07/10] elf: Support DT_RELR in linker tests H.J. Lu
2022-01-08  2:42   ` Fangrui Song
2022-01-08 13:46     ` H.J. Lu
2022-01-07 19:06 ` [PATCH 08/10] x86: Add DT_RELR support H.J. Lu
2022-01-07 19:06 ` [PATCH 09/10] ld: Add simple DT_RELR tests H.J. Lu
2022-01-08  2:39   ` Fangrui Song
2022-01-08  2:44     ` H.J. Lu
2022-01-08  8:37   ` Fangrui Song
2022-01-08 18:33     ` H.J. Lu
2022-01-07 19:06 ` [PATCH 10/10] ld: Add glibc dependency for DT_RELR H.J. Lu
2022-01-08  4:43   ` Fangrui Song
2022-01-08  4:51     ` H.J. Lu
2022-01-08  8:19       ` Fangrui Song
2022-01-08 18:37         ` H.J. Lu
2022-01-07 23:45 ` [PATCH 0/10] ld: Implement DT_RELR for x86 Fangrui Song
2022-01-07 23:56   ` H.J. Lu
2022-01-08  1:31     ` Fangrui Song
2022-01-08  1:55       ` H.J. Lu
2022-01-08 18:30         ` H.J. Lu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAMe9rOp7AEYnP8-i0jQVcXA93c4Kw6zZ-ids2fp2rUpiD5CizQ@mail.gmail.com \
    --to=hjl.tools@gmail.com \
    --cc=binutils@sourceware.org \
    --cc=i@maskray.me \
    --cc=libc-alpha@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).