public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "ro at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/112959] New: install.tex needs updates on FreeBSD
Date: Mon, 11 Dec 2023 10:47:31 +0000	[thread overview]
Message-ID: <bug-112959-4@http.gcc.gnu.org/bugzilla/> (raw)

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112959

            Bug ID: 112959
           Summary: install.tex needs updates on FreeBSD
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ro at gcc dot gnu.org
                CC: andreast at gcc dot gnu.org, gerald at pfeifer dot com
  Target Milestone: ---
              Host: x86_64-unknown-freebsd14.0
            Target: x86_64-unknown-freebsd14.0
             Build: x86_64-unknown-freebsd14.0

I've recently tried a trunk bootstrap on FreeBSD 14.0/amd64, which was quite an
unpleasant experience:

* With the default, /usr/bin/ld, libgcc_s.so.1 doesn't link:

  PR target/112745      stage 1 libgcc_s.so fails to link on FreeBSD

* Ada bootstrap proved to be a hellhole: the packaged gnat12 didn't work to
  compile GCC 11.4.0:

error:
("/usr/local/gnat12/lib/gcc/x86_64-portbld-freebsd14.0/12.2.0/adalib/s-aotase.ali"
is obsolete and read-only)
error:
("/usr/local/gnat12/lib/gcc/x86_64-portbld-freebsd14.0/12.2.0/adalib/s-atoope.ali"
is obsolete and read-only)
error:
("/usr/local/gnat12/lib/gcc/x86_64-portbld-freebsd14.0/12.2.0/adalib/s-atopri.ali"
is obsolete and read-only)
error:
("/usr/local/gnat12/lib/gcc/x86_64-portbld-freebsd14.0/12.2.0/adalib/i-c.ali"
is obsolete and read-only)

  I cannot tell if this is due to trying to build with a newer version of GNAT
  or just a packaging error.

  I then tried to build lang/gcc6-aux from the ports tree, but that also failed
  compiling various instances of obstack.c.  Those could be hacked around, but
  when I later gave up on this.

  Fortunately I could successfully use GNAT 6.4.1 from
/usr/ports/distfiles/ada-bootstrap.x86_64.freebsd.120.tar.bz2.

  Not really a pleasant experience, especially if you have to figure this out
  on your own.

* libgo doesn't compile, as has been known for years.  Again, a totally waste
  of time having to rediscover this all on your own.

* Neither libada nor libphobos build with the default multilib configuration:

  PR d/112957   [13/14 regression] src/std/math/exponential.d doesn't compile
on 32-bit FreeBSD/x86
  PR ada/112958 [14 regression] s-exnllf.ads etc. don't compile on 32-bit
FreeBSD/x86

In the end, the only way to have a successful build with --disable-multilib
and Go disabled.

It would be extremely helpful if install.texi would document some of the
common quirks of the target, so regular developers won't have to go down
this route all by themselves.

             reply	other threads:[~2023-12-11 10:47 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-11 10:47 ro at gcc dot gnu.org [this message]
2023-12-11 10:47 ` [Bug target/112959] " ro at gcc dot gnu.org
2023-12-11 11:39 ` rguenth at gcc dot gnu.org
2024-04-28 13:03 ` cvs-commit at gcc dot gnu.org
2024-05-01 14:24 ` cvs-commit at gcc dot gnu.org
2024-05-01 22:48 ` cvs-commit at gcc dot gnu.org
2024-05-01 22:54 ` gerald at pfeifer dot com
2024-05-06  9:27 ` gerald at pfeifer dot com
2024-05-06 11:37 ` ro at CeBiTec dot Uni-Bielefeld.DE
2024-05-07  7:43 ` rguenth at gcc dot gnu.org
2024-05-12 13:28 ` cvs-commit at gcc dot gnu.org
2024-05-12 13:30 ` cvs-commit at gcc dot gnu.org
2024-05-12 13:32 ` cvs-commit at gcc dot gnu.org
2024-05-12 14:41 ` gerald at pfeifer dot com
2024-05-12 14:53 ` thierry at FreeBSD dot org
2024-05-13  8:57 ` ro at CeBiTec dot Uni-Bielefeld.DE
2024-05-13 10:30 ` gerald at pfeifer dot com

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=bug-112959-4@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.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).