From Newsgroup: muc.lists.freebsd.ports

The binutils in our ports tree was the last binutils to support the Gold linker. A previous exp-run for 2.45.1 showed us that a number of ports failures were due to the missing linker.

I've decided to import binutils 2.47 into ports. I think the path forward
is to address ports failures due do the removed feature.

Do we want to rename the current binutils to binutils244 and import 2.47 as binutils? Or would going "cold turkey" be advisable?
--
Cheers,
Cy Schubert <Cy.Schubert@cschubert.com>
FreeBSD UNIX: <cy@FreeBSD.org> Web: https://FreeBSD.org
NTP: <cy@nwtime.org> Web: https://nwtime.org

e**(i*pi)+1=0




--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.ports

On 8/26/26 12:11, Cy Schubert wrote:

The binutils in our ports tree was the last binutils to support the Gold linker. A previous exp-run for 2.45.1 showed us that a number of ports failures were due to the missing linker.

I've decided to import binutils 2.47 into ports. I think the path forward
is to address ports failures due do the removed feature.

Do we want to rename the current binutils to binutils244 and import 2.47 as binutils? Or would going "cold turkey" be advisable?

Is there a public link for the exp run so we can see how many + what
things break? How linkers work is a particular weakpoint for me so any examples of identifying and fixing some of the issues already?
It sounds like the difference between 'cold turkey' and creating binutils244 will both break stuff unless either a default to use 244 for
now is set or broken ports are patched to use it instead of the main
binutils. Any problems with some ports using 2.47 and others using 2.44?
Thank you for this work


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.ports

--000000000000de18f7065a02573e
Content-Type: text/plain; charset="UTF-8"

Thank-you for the heads-up.

I think the renaming of the last gold supported version should have a
separate version for those that need the functionality, and that 2.47
becomes the new binutils, as you suggest. I know I needed the gold plugin
in the past to build lto executables, but isn't the *need* for gold
obviated by llvm's linker (with clang's lto object files)?

--000000000000de18f7065a02573e
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><div>Thank-you for the heads-up.</div><di= v><br></div><div>I think the renaming of the last gold supported version sh= ould have a separate version for those that need the functionality, and tha=
t 2.47 becomes the new binutils, as you suggest.=C2=A0 I know I needed the = gold plugin in the past to build lto executables, but isn&#39;t the *need* = for gold obviated by llvm&#39;s linker (with clang&#39;s lto object files)?= </div></div><br></div>

--000000000000de18f7065a02573e--


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.ports


--rvbp2a45ysht736l
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Aug 26, 2026 at 12:11:21PM -0700, Cy Schubert wrote:

The binutils in our ports tree was the last binutils to support the Gold=

=20

linker. A previous exp-run for 2.45.1 showed us that a number of ports=20 failures were due to the missing linker.
=20
I've decided to import binutils 2.47 into ports. I think the path forward=

=20

is to address ports failures due do the removed feature.
=20
Do we want to rename the current binutils to binutils244 and import 2.47 =

as=20

binutils? Or would going "cold turkey" be advisable?

Well, how many ports break ? If it's like 10 obscure leaf ports that are
dead upstream, going cold turkey can work, if it's like 100 or more with
more ports depending on them that won't be available any more, please
ask for an exp-run and fix them before updating.

--=20
Mathieu Arnold

--rvbp2a45ysht736l
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQITBAABCgB9FiEE9XJBpJetWizkEBUef2IOCp6dQb4FAmqQcZFfFIAAAAAALgAo aXNzdWVyLWZwckBub3RhdGlvbnMub3BlbnBncC5maWZ0aGhvcnNlbWFuLm5ldEY1 NzI0MUE0OTdBRDVBMkNFNDEwMTUxRTdGNjIwRTBBOUU5RDQxQkUACgkQf2IOCp6d Qb53ywwAknWRTIHOrGYC80PqFPz7esqYDSsNlJ85ZBK8KZ3Z044Ie23/lF9DbdcP yt1lnqsYeWAIOGlANfGH1a3Q1Lkwrozk5m7Gol7mzGe+dVEy0QapSQt4IWWDbX9e VPtud4dGYdH+ahZWD5NoBlFyYsQpmb3QmwIFFDs5PdxuTpa68mFogsEJcjtIXDID 88Ep7b8nuK1fDBZxfEmsh9NGXEXsVMhc1WNeHvrYCt45K4pPf9iDlgFv/RN1BAPa 0Z3wC5LsYJ8tljTgwC9F6rbtC5d9+JmyUyZ3JVvZ6PwhvcdByctPINrbwhHrqG/X kp9lvXoKQgKUH1x0s8XyIXt6Z2x2jjX4Fq81Ud5J2e2p6PP8dsH9A+QEA/WwYBj4 bXpZUnMZOjAGIMno7U0sKj8PFm+IGg8YVo0y++wvOMldF/+w+ygnwXITVdLPxD8H IMotb2vgqQ/9AOZ0/1u3BDdatNV2ljbq0GrQohyGle7Jaombq/I53AUKqVv+jAAU
4WbOaJ2N
=+baF
-----END PGP SIGNATURE-----

--rvbp2a45ysht736l--


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.22a-Linux NewsLink 1.2

From Newsgroup: muc.lists.freebsd.ports

In message <BL4PR11MB8824749350088988CDF5F81DE6AE2@BL4PR11MB8824.namprd11.pr od.
outlook.com>, "Edward Sanford Sutton, III" writes:

On 8/26/26 12:11, Cy Schubert wrote:

The binutils in our ports tree was the last binutils to support the Gold linker. A previous exp-run for 2.45.1 showed us that a number of ports failures were due to the missing linker.

I've decided to import binutils 2.47 into ports. I think the path forward is to address ports failures due do the removed feature.

Do we want to rename the current binutils to binutils244 and import 2.47 as binutils? Or would going "cold turkey" be advisable?

Is there a public link for the exp run so we can see how many + what things break? How linkers work is a particular weakpoint for me so any examples of identifying and fixing some of the issues already?
It sounds like the difference between 'cold turkey' and creating binutils244 will both break stuff unless either a default to use 244 for
now is set or broken ports are patched to use it instead of the main binutils. Any problems with some ports using 2.47 and others using 2.44?
Thank you for this work

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=297985
--
Cheers,
Cy Schubert <Cy.Schubert@cschubert.com>
FreeBSD UNIX: <cy@FreeBSD.org> Web: https://FreeBSD.org
NTP: <cy@nwtime.org> Web: https://nwtime.org

e**(i*pi)+1=0




--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-admin@muc.de
--- Synchronet 3.22a-Linux NewsLink 1.2