From: "Fāng-ruì Sòng" <maskray@google.com>
To: GNU C Library <libc-alpha@sourceware.org>
Subject: Re: [PATCH 2/3] elf: Replace a --defsym trick with an object file to be compatible with lld
Date: Mon, 11 Jan 2021 12:06:04 -0800 [thread overview]
Message-ID: <CAFP8O3+gnft4TVCAcjON7xmdJ_Wgn8rL7HR8hS2gWvSq0FOqFw@mail.gmail.com> (raw)
In-Reply-To: <20201228194855.510315-3-maskray@google.com>
On Mon, Dec 28, 2020 at 11:49 AM Fangrui Song <maskray@google.com> wrote:
>
> The existing code specifies -Wl,--defsym=malloc=0 and other malloc.os
> definitions before libc_pic.a so that libc_pic.a(malloc.os) is not
> fetched. This trick is used to avoid multiple definition errors which
> would happen as a chain result:
>
> dl-allobjs.os has an undefined __libc_scratch_buffer_set_array_size
> __libc_scratch_buffer_set_array_size fetches libc_pic.a(scratch_buffer_set_array_size.os)
> libc_pic.a(scratch_buffer_set_array_size.os) has an undefined free
> free fetches libc_pic.a(malloc.os)
> libc_pic.a(malloc.os) has an undefined __libc_message
> __libc_message fetches libc_pic.a(libc_fatal.os)
>
> libc_fatal.os will cause a multiple definition error (__GI___libc_fatal)
> >>> defined at dl-fxstatat64.c
> >>> /tmp/p/glibc/Release/elf/dl-allobjs.os:(__GI___libc_fatal)
> >>> defined at libc_fatal.c
> >>> libc_fatal.os:(.text+0x240) in archive /tmp/p/glibc/Release/libc_pic.a
>
> lld processes --defsym after all input files, so this trick does not
> suppress multiple definition errors with lld. Split the step into two
> and use an object file to make the intention more obvious and make lld
> work.
>
> This is conceptually more appropriate because --defsym defines a SHN_ABS
> symbol while a normal definition is relative to the image base.
>
> See https://sourceware.org/pipermail/libc-alpha/2020-March/111910.html
> for discussions about the --defsym semantics.
> ---
> elf/Makefile | 11 ++++-------
> 1 file changed, 4 insertions(+), 7 deletions(-)
Ping on this
next prev parent reply other threads:[~2021-01-11 20:06 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-28 19:48 [PATCH 0/3] Make glibc build with LLD Fangrui Song
2020-12-28 19:48 ` [PATCH 1/3] configure: Allow LD to be a linker other than GNU ld and gold Fangrui Song
2020-12-28 20:46 ` H.J. Lu
2020-12-28 21:15 ` [PATCH 1/3] configure: Allow LD to be LLD 9.0.0 or above " Fangrui Song
2020-12-28 19:48 ` [PATCH 2/3] elf: Replace a --defsym trick with an object file to be compatible with lld Fangrui Song
2021-01-11 20:06 ` Fāng-ruì Sòng [this message]
2021-01-18 22:08 ` Fāng-ruì Sòng
2021-01-19 0:03 ` H.J. Lu
2021-01-19 8:50 ` Fāng-ruì Sòng
2021-01-19 12:30 ` H.J. Lu
2021-01-20 18:51 ` Fāng-ruì Sòng
2021-01-28 1:03 ` Fāng-ruì Sòng
2021-01-29 14:38 ` Adhemerval Zanella
2021-01-29 15:29 ` H.J. Lu
2021-01-29 18:04 ` Adhemerval Zanella
2021-01-29 19:28 ` [PATCH v2] elf: Replace a --defsym trick with an object file to be compatible with LLD Fangrui Song
2021-02-01 13:19 ` Adhemerval Zanella
2021-02-01 13:45 ` H.J. Lu
2021-02-01 13:43 ` Florian Weimer
2020-12-28 19:48 ` [PATCH 3/3] install: Replace scripts/output-format.sed with objdump -f [BZ #26559] Fangrui Song
2021-01-08 19:38 ` Adhemerval Zanella
2021-01-12 11:20 ` Florian Weimer
2021-01-12 12:00 ` Adhemerval Zanella
2021-01-12 17:46 ` Fāng-ruì Sòng
2021-01-12 11:32 ` Andreas Schwab
2021-01-12 11:58 ` Florian Weimer
2020-12-28 21:11 ` [PATCH 0/3] Make glibc build with LLD H.J. Lu
2020-12-28 21:45 ` Fangrui Song
2020-12-28 21:49 ` H.J. Lu
2020-12-28 22:54 ` H.J. Lu
2021-01-05 23:03 ` Fāng-ruì Sòng
2021-01-05 23:33 ` H.J. Lu
2021-01-05 23:41 ` Fāng-ruì Sòng
2021-01-05 23:51 ` H.J. Lu
2021-01-06 0:50 ` Fāng-ruì Sòng
2021-01-06 1:43 ` H.J. Lu
2021-01-06 2:00 ` Fāng-ruì Sòng
2021-01-06 2:14 ` H.J. Lu
2021-01-05 22:57 ` Fāng-ruì Sòng
2021-01-08 5:04 ` Fāng-ruì Sòng
2021-01-08 17:09 ` H.J. Lu
2021-01-08 18:04 ` Fāng-ruì Sòng
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=CAFP8O3+gnft4TVCAcjON7xmdJ_Wgn8rL7HR8hS2gWvSq0FOqFw@mail.gmail.com \
--to=maskray@google.com \
--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).