Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > alt.comp.software.seamonkey > #8410 > unrolled thread

[FreeBSD 15.0] Successfull build

Started byJKB <JKB@hilbert.invalid>
First post2026-01-15 10:23 +0000
Last post2026-03-14 21:55 +0900
Articles 10 on this page of 30 — 11 participants

Back to article view | Back to alt.comp.software.seamonkey


Contents

  [FreeBSD 15.0] Successfull build JKB <JKB@hilbert.invalid> - 2026-01-15 10:23 +0000
    Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-03-14 08:38 +0000
      Re: [FreeBSD 15.0] Successfull build Daniel70 <daniel47@nomail.afraid.org> - 2026-03-14 20:25 +1100
        Re: [FreeBSD 15.0] Successfull build Nuno Silva <nunojsilva@invalid.invalid> - 2026-03-14 11:34 +0000
          Re: [FreeBSD 15.0] Successfull build Daniel70 <daniel47@nomail.afraid.org> - 2026-03-15 19:58 +1100
        Re: [FreeBSD 15.0] Successfull build Gerhard Strangar <g.s@arcor.de> - 2026-03-15 07:27 +0100
          Re: [FreeBSD 15.0] Successfull build Geoff Welsh <GeeDubya@some.rr.com> - 2026-03-16 09:43 -1000
            Re: [FreeBSD 15.0] Successfull build Gerhard Strangar <g.s@arcor.de> - 2026-03-17 17:26 +0100
              Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-24 15:16 +0000
                Re: [FreeBSD 15.0] Successfull build Gerhard Strangar <g.s@arcor.de> - 2026-06-24 17:09 +0000
                  Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-24 21:56 +0000
                    Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-24 22:14 +0000
                      Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-25 04:17 +0000
                        Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-25 16:43 +0000
                          Re: [FreeBSD 15.0] Successfull build Gerhard Strangar <g.s@arcor.de> - 2026-06-26 18:45 +0000
                            Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-27 05:16 +0000
                              Re: [FreeBSD 15.0] Successfull build Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2026-06-27 18:22 +0000
                                Re: [FreeBSD 15.0] Successfull build Hank Rogers <Hank@nospam.invalid> - 2026-06-27 19:50 -0500
                                Re: [FreeBSD 15.0] Successfull build Nuno Silva <nunojsilva@invalid.invalid> - 2026-06-28 07:40 +0100
                                  Re: [FreeBSD 15.0] Successfull build groenveld@acm.org (John D Groenveld) - 2026-06-28 13:51 +0000
                                Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-28 19:41 +0000
                                  Re: [FreeBSD 15.0] Successfull build Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2026-06-29 16:54 +0000
                          Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-27 05:29 +0000
                            Re: [FreeBSD 15.0] Successfull build Dave Yeo <dave.r.yeo@gmail.com> - 2026-06-27 21:38 -0700
                              Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-28 19:38 +0000
                                Re: [FreeBSD 15.0] Successfull build Louis Epstein <le@lekno.ws> - 2026-06-28 19:45 +0000
                                  Re: [FreeBSD 15.0] Successfull build Dave Yeo <dave.r.yeo@gmail.com> - 2026-06-28 22:20 -0700
                                    Re: [FreeBSD 15.0] Successfull build Nuno Silva <nunojsilva@invalid.invalid> - 2026-06-29 07:32 +0100
                                Re: [FreeBSD 15.0] Successfull build Dave Yeo <dave.r.yeo@gmail.com> - 2026-06-28 22:25 -0700
      Re: [FreeBSD 15.0] Successfull build Yuchan Seo <hexagonwin@disroot.org> - 2026-03-14 21:55 +0900

Page 2 of 2 — ← Prev page 1 [2]


#8619

FromLouis Epstein <le@lekno.ws>
Date2026-06-28 19:41 +0000
Message-ID<111rtdv$1va$1@reader1.panix.com>
In reply to#8613
In comp.unix.bsd.freebsd.misc Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
> On 2026-06-27, Louis Epstein <le@lekno.ws> wrote:
> 
>> Gerhard Strangar <g.s@arcor.de> wrote:
>>
>>> Louis Epstein wrote:
>>> 
>>>> I have downloaded the source.tar.xz file into a seamonkey253 directory
>>>> but what is the exact tar -x suboption?
>>> 
>>> tar xf source.tar.xz
>>> But if that's not related to Seamonkey.
>>
>> Just trying to get the exact sequence down.
>>
>> The guy who usually does tech stuff for me was making 
>> installing Seamonkey sound impossible as a long term
>> excuse for not doing it.
>>  
>> I miss its being a port/package.
> 
> Why make things so complicated?  I don't need no steenkin'
> installer - I just download the release, unzip it into
> /opt/seamonkey, and the update is complete.  I have an
> icon that points to it, but that doesn't change after
> an update.  Easy-peasy.

Is that directory a global or in your home directory?
You're taking the binary straight off the seamonkey-project site?

>>>> Created a sub on my home directory though I'd like the program
>>>> to be in available path for all accounts.
>>> 
>>> Ask root to create one somwhere else and put it in your .mozconfig, for
>>> example:
>>> ac_add_options --prefix=/usr/local/seamonkey-2.53.23
>>> 
>>> But if that's what's stopping you, you might want to reconsider using
>>> thunderbird + firefox, they come as packages in FreeBSD.
>>
>> I use Firefox as it is but I want my Seamonkey back.
> 
> Seamonkey is my default browsesr; unfortunately I have to
> fall back to Firefox more and more, as much as I dislike
> its user interface.
> 

-=-=-
The World Trade Center towers MUST rise again,
at least as tall as before...or terror has triumphed.

[toc] | [prev] | [next] | [standalone]


#8624

FromCharlie Gibbs <cgibbs@kltpzyxm.invalid>
Date2026-06-29 16:54 +0000
Message-ID<frx0S.121477$81a2.104320@fx17.iad>
In reply to#8619
On 2026-06-28, Louis Epstein <le@lekno.ws> wrote:

> In comp.unix.bsd.freebsd.misc Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
>
>> Why make things so complicated?  I don't need no steenkin'
>> installer - I just download the release, unzip it into
>> /opt/seamonkey, and the update is complete.  I have an
>> icon that points to it, but that doesn't change after
>> an update.  Easy-peasy.
>
> Is that directory a global or in your home directory?

It's an absolute path, i.e. off the root.  Normaly I run
seamonkey by clicking on a panel icon I've set up, which
contains the path.  To run it from the command line I'd
have to type "/opt/seamonkey/seamonkey".

> You're taking the binary straight off the seamonkey-project site?

Yes.

-- 
/~\  Charlie Gibbs                  |  No artificial
\ /  <cgibbs@kltpzyxm.invalid>      |  intelligence was
 X   I'm really at ac.dekanfrus     |  used in the creation
/ \  if you read it the right way.  |  of this post.

[toc] | [prev] | [next] | [standalone]


#8612

FromLouis Epstein <le@lekno.ws>
Date2026-06-27 05:29 +0000
Message-ID<111nn4i$7hn$1@reader1.panix.com>
In reply to#8609
Louis Epstein <le@lekno.ws> wrote:
> Louis Epstein <le@lekno.ws> wrote:
>> Louis Epstein <le@lekno.ws> wrote:
>>> Louis Epstein <le@lekno.ws> wrote:
>>>> Gerhard Strangar <g.s@arcor.de> wrote:
>>>>> Louis Epstein wrote:
>>>>> 
>>>>>> I have now moved from FBSD 13.5 to 14.4.
>>>>>> 
>>>>>> I have had great problems building Rust lately.
>>>>> 
>>>>> I used the one from packages. The latest one, which is why I disabled
>>>>> the rust SIMD.
>>>>> 
>>>>>> What is the exact install process you used once you downloaded?
>>>>> 
>>>>> - untar
> 
> I have downloaded the source.tar.xz file into a seamonkey253 directory
> but what is the exact tar -x suboption?

Have now done this.
 
>>>>> - create .mozconfig

A simple text edit of the previously cited?

>>>>> - ./mach build install
>>>> 
>>>> mach not make?
>>>> "mach" is something formed from the untar?

Yes,I see that it is.

>>>> Is there a particular directory to download into?
> 
> Created a sub on my home directory though I'd like the program
> to be in available path for all accounts.
> 
>>>>  
>>>>> And that's it.
>>> 
>>> Just looking at old .mozconfig files on my machine.

No analysis of which of these options are workable?

>>> One is
>>> 
>>> ==
>>> ac_add_options --enable-application=suite
>>> mk_add_options MOZ_OBJDIR=/path/to/comm-central/obj-sm-release
>>> mk_add_options MOZ_MAKE_FLAGS="-j4"
>>> ==
>>> (assigned to 2.49.5 which I think is the last I had working)
>>> 
>>> Another is
>>> ==      
>>> ac_add_options --enable-application=suite
>>> ac_add_options --enable-ldap
>>> ac_add_options --enable-calendar
>>> ac_add_options --prefix="/usr/local"
>>> ac_add_options --enable-jemalloc=4
>>> ac_add_options --enable-system-cairo
>>> ac_add_options --with-system-libevent
>>> ac_add_options --enable-system-ffi
>>> ac_add_options --with-system-graphite2
>>> ac_add_options --with-system-harfbuzz
>>> ac_add_options --enable-system-hunspell
>>> ac_add_options --with-system-icu
>>> ac_add_options --with-intl-api
>>> ac_add_options --with-system-jpeg=/usr/local
>>> ac_add_options --with-system-nspr
>>> ac_add_options --with-system-nss
>>> ac_add_options --with-system-png=/usr/local
>>> ac_add_options --enable-system-pixman
>>> ac_add_options --enable-system-sqlite
>>> ac_add_options --with-system-libvpx
>>> ac_add_options --with-system-vorbis
>>> ac_add_options --with-system-ogg
>>> ac_add_options --enable-chrome-format=omni
>>> ac_add_options --enable-default-toolkit=cairo-gtk3
>>> ac_add_options --enable-update-channel=release
>>> ac_add_options --disable-updater
>>> ac_add_options --enable-pie
>>> ac_add_options --with-system-zlib
>>> ac_add_options --with-system-bz2
>>> ac_add_options --enable-optimize
>>> ac_add_options --enable-startup-notification
>>> ac_add_options --enable-gconf
>>> ac_add_options --disable-libproxy
>>> ac_add_options --enable-alsa
>>> ac_add_options --enable-jack
>>> ac_add_options --enable-pulseaudio
>>> ac_add_options --disable-rust
>>> ac_add_options --disable-debug
>>> ac_add_options --disable-debug-symbols
>>> ac_add_options --enable-release
>>> ac_add_options --enable-dtrace
>>> ac_add_options --disable-gold
>>> ac_add_options --enable-profiling
>>> ac_add_options --disable-tests
>>> ac_add_options --disable-strip
>>> ac_add_options --disable-install-strip
>>> mk_add_options MOZ_OBJDIR="/usr/ports/www/seamonkey/work/.build"
>>> export MAKE=gmake
>>> export ac_cv_path_PERL=/usr/local/bin/perl
>>> export ac_cv_path_PERL_PATH=/usr/local/bin/perl
>>> export PERL_USE_UNSAFE_INC=1
>>> export PKG_CONFIG=pkgconf
>>> export PYTHON="/usr/local/bin/python2.7"
>>> export XDG_DATA_HOME=/usr/ports/www/seamonkey/work
>>> export XDG_CONFIG_HOME=/usr/ports/www/seamonkey/work
>>> export HOME=/usr/ports/www/seamonkey/work
>>> export PATH=/usr/ports/www/seamonkey/work/.bin:/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/root/bin
>>> export SHELL=/bin/sh
>>> export CONFIG_SHELL=/bin/sh
>>> export RUSTFLAGS=""
>>> export PERL="/usr/local/bin/perl"
>>> export MOZ_JEMALLOC4=1
>>> export MOZ_GOOGLE_API_KEY=AIzaSyBsp9n41JLW8jCokwn7vhoaMejDFRd1mp8
>>> export MOZ_OPTIMIZE_FLAGS="-O2 -O3"
>>> ==
>>> (that was in 2.49.4 which definitely worked here)
>>> 
>>> Also present is
>>> 
>>> ==
>>> export CC=clang
>>> export CXX=clang++
>>> ac_add_options --enable-application=comm/suite
>>> ac_add_options --enable-calendar
>>> ac_add_options --enable-irc
>>> ac_add_options --enable-dominspector
>>> mk_add_options MOZ_OBJDIR=/home/ribo/mozilla-253/objdir
>>> ac_add_options --enable-optimize
>>> ac_add_options --enable-js-shell
>>> #RAF ac_add_options --enable-elf-hack
>>> ac_add_options --disable-debug-symbols
>>> ac_add_options --disable-tests
>>> 
>>> # Disable checking that add-ons are signed by the trusted root
>>> MOZ_ADDON_SIGNING=0
>>> # Disable enforcing that add-ons are signed by the trusted root
>>> MOZ_REQUIRE_SIGNING=0
>>> 
>>> # Package js shell
>>> export MOZ_PACKAGE_JSSHELL=1
>>> ==
>>> 
>>> ...which is from an attempt to install an early 2.53 
>>> by someone who gave up on doing it for me and tells me
>>> to forget Seamonkey which I refuse to do,since I like the
>>> linked browser and email clients. 
>> 
>> AND I have had this one recommended to me for 2.53.12
>> which I never did get installed:
>> 
>> 
>> ==
>> export CC=clang80
>> export CXX=clang++80
>> ac_add_options --enable-application=comm/suite
>> ac_add_options --enable-calendar
>> ac_add_options --enable-irc
>> ac_add_options --enable-dominspector
>> ac_add_options --enable-optimize
>> ac_add_options --disable-debug
>> 
>> ==
>> 
>>  
>>>> -=-=-
>>>> The World Trade Center towers MUST rise again,
>>>> at least as tall as before...or terror has triumphed.

[toc] | [prev] | [next] | [standalone]


#8615

FromDave Yeo <dave.r.yeo@gmail.com>
Date2026-06-27 21:38 -0700
Message-ID<Wy10S.78$%a4.69@fx38.iad>
In reply to#8612
Louis Epstein wrote:
>>>> Just looking at old .mozconfig files on my machine.
> No analysis of which of these options are workable?
>

Start out with a minimal .mozconfig, this one that you posted might be a 
good start
----
export CC=clang
export CXX=clang++
ac_add_options --enable-application=comm/suite
ac_add_options --enable-calendar
ac_add_options --enable-irc
ac_add_options --enable-dominspector
mk_add_options MOZ_OBJDIR=/home/ribo/mozilla-253/objdir
ac_add_options --enable-optimize
ac_add_options --enable-js-shell
#RAF ac_add_options --enable-elf-hack
ac_add_options --disable-debug-symbols
ac_add_options --disable-tests

# Disable checking that add-ons are signed by the trusted root
MOZ_ADDON_SIGNING=0
# Disable enforcing that add-ons are signed by the trusted root
MOZ_REQUIRE_SIGNING=0

# Package js shell
export MOZ_PACKAGE_JSSHELL=1
----

Which to me who doesn't use any BSD, seems a good start. IIRC the BSD's 
use clang.
Then if it turns out that the build breaks building one of various libs 
that SM includes, switch to using the system one, adding for example if 
cairo is broken (unlikely but who knows) adding the right ac_add_options 
such as
ac_add_options --enable-system-cairo
and retry.
I assume you have lots of memory so using the in tree libraries as 
recommended shouldn't be a problem as long as they work on your FreeBSD
I don't know about Rust and you may need something in .mozconfig to 
handle differences between what you have and what is expected. I think 
one example disabled Rust simd to work around an issue. Slight 
performance decrease I assume but likely not really noticeable.
You should perhaps add a Prefix unless you don't mind installing into 
/usr/local. The install phase is important as it will build omni.ja but 
you can test by running the new binary from dist/bin. Actually at the 
end of the build, mach should suggest a test drive where it will create 
a scratch profile and use the right options to not interfere with your 
installed SM, if you have one.
Dave
ps run mach --help to see some of the capabilities of mach if you are 
curious.

[toc] | [prev] | [next] | [standalone]


#8618

FromLouis Epstein <le@lekno.ws>
Date2026-06-28 19:38 +0000
Message-ID<111rt7j$4tn$1@reader1.panix.com>
In reply to#8615
Dave Yeo <dave.r.yeo@gmail.com> wrote:
> Louis Epstein wrote:
>>>>> Just looking at old .mozconfig files on my machine.
>> No analysis of which of these options are workable?
>>
> 
> Start out with a minimal .mozconfig, this one that you posted might be a 
> good start
> ----
> export CC=clang
> export CXX=clang++
> ac_add_options --enable-application=comm/suite
> ac_add_options --enable-calendar
> ac_add_options --enable-irc
> ac_add_options --enable-dominspector
> mk_add_options MOZ_OBJDIR=/home/ribo/mozilla-253/objdir

(This is actually the home directory of my tech friend who
gave up on Seamonkey as uninstallable and is currently off
to Maine for the summer)

> ac_add_options --enable-optimize
> ac_add_options --enable-js-shell
> #RAF ac_add_options --enable-elf-hack
> ac_add_options --disable-debug-symbols
> ac_add_options --disable-tests
> 
> # Disable checking that add-ons are signed by the trusted root
> MOZ_ADDON_SIGNING=0
> # Disable enforcing that add-ons are signed by the trusted root
> MOZ_REQUIRE_SIGNING=0
> 
> # Package js shell
> export MOZ_PACKAGE_JSSHELL=1
> ----
> 
> Which to me who doesn't use any BSD, seems a good start. IIRC the BSD's 
> use clang.

In the FreeBSD ports/packages Clang comes as part of LLVM,I believe.

> Then if it turns out that the build breaks building one of various libs 
> that SM includes, switch to using the system one, adding for example if 
> cairo is broken (unlikely but who knows) adding the right ac_add_options 
> such as
> ac_add_options --enable-system-cairo
> and retry.
> I assume you have lots of memory so using the in tree libraries as 
> recommended shouldn't be a problem as long as they work on your FreeBSD
> I don't know about Rust and you may need something in .mozconfig to 
> handle differences between what you have and what is expected. I think 
> one example disabled Rust simd to work around an issue. Slight 
> performance decrease I assume but likely not really noticeable.

I have had enormous problems building Rust.

> You should perhaps add a Prefix unless you don't mind installing into 
> /usr/local. The install phase is important as it will build omni.ja but 
> you can test by running the new binary from dist/bin. Actually at the 
> end of the build, mach should suggest a test drive where it will create 
> a scratch profile and use the right options to not interfere with your 
> installed SM, if you have one.
> Dave
> ps run mach --help to see some of the capabilities of mach if you are 
> curious.

-=-=-
The World Trade Center towers MUST rise again,
at least as tall as before...or terror has triumphed.

[toc] | [prev] | [next] | [standalone]


#8620

FromLouis Epstein <le@lekno.ws>
Date2026-06-28 19:45 +0000
Message-ID<111rtk7$d6u$1@reader1.panix.com>
In reply to#8618
Louis Epstein <le@lekno.ws> wrote:
> Dave Yeo <dave.r.yeo@gmail.com> wrote:
>> Louis Epstein wrote:
>>>>>> Just looking at old .mozconfig files on my machine.
>>> No analysis of which of these options are workable?
>>>
>> 
>> Start out with a minimal .mozconfig, this one that you posted might be a 
>> good start
>> ----
>> export CC=clang
>> export CXX=clang++
>> ac_add_options --enable-application=comm/suite
>> ac_add_options --enable-calendar
>> ac_add_options --enable-irc
>> ac_add_options --enable-dominspector
>> mk_add_options MOZ_OBJDIR=/home/ribo/mozilla-253/objdir
> 
> (This is actually the home directory of my tech friend who
> gave up on Seamonkey as uninstallable and is currently off
> to Maine for the summer)
> 
>> ac_add_options --enable-optimize
>> ac_add_options --enable-js-shell
>> #RAF ac_add_options --enable-elf-hack
>> ac_add_options --disable-debug-symbols
>> ac_add_options --disable-tests
>> 
>> # Disable checking that add-ons are signed by the trusted root
>> MOZ_ADDON_SIGNING=0
>> # Disable enforcing that add-ons are signed by the trusted root
>> MOZ_REQUIRE_SIGNING=0
>> 
>> # Package js shell
>> export MOZ_PACKAGE_JSSHELL=1
>> ----
>> 
>> Which to me who doesn't use any BSD, seems a good start. IIRC the BSD's 
>> use clang.
> 
> In the FreeBSD ports/packages Clang comes as part of LLVM,I believe.
> 
>> Then if it turns out that the build breaks building one of various libs 
>> that SM includes, switch to using the system one, adding for example if 
>> cairo is broken (unlikely but who knows) adding the right ac_add_options 
>> such as
>> ac_add_options --enable-system-cairo
>> and retry.
>> I assume you have lots of memory so using the in tree libraries as 
>> recommended shouldn't be a problem as long as they work on your FreeBSD
>> I don't know about Rust and you may need something in .mozconfig to 
>> handle differences between what you have and what is expected. I think 
>> one example disabled Rust simd to work around an issue. Slight 
>> performance decrease I assume but likely not really noticeable.
> 
> I have had enormous problems building Rust.
> 
>> You should perhaps add a Prefix unless you don't mind installing into 
>> /usr/local. The install phase is important as it will build omni.ja but 
>> you can test by running the new binary from dist/bin. Actually at the 
>> end of the build, mach should suggest a test drive where it will create 
>> a scratch profile and use the right options to not interfere with your 
>> installed SM, if you have one.
>> Dave
>> ps run mach --help to see some of the capabilities of mach if you are 
>> curious.

While "mach" was part of what unzipped with the Seamonkey source tar,
mach is still itself a command not found and has no man page.
 
> -=-=-
> The World Trade Center towers MUST rise again,
> at least as tall as before...or terror has triumphed.

[toc] | [prev] | [next] | [standalone]


#8621

FromDave Yeo <dave.r.yeo@gmail.com>
Date2026-06-28 22:20 -0700
Message-ID<ngn0S.120842$zNZ2.30830@fx16.iad>
In reply to#8620
Louis Epstein wrote:
>>> ps run mach --help to see some of the capabilities of mach if you are
>>> curious.
> While "mach" was part of what unzipped with the Seamonkey source tar,
> mach is still itself a command not found and has no man page.
>

You can chmod it to be executable, or here on OS/2, simply run it as "sh 
mach --help" It is a Python script but also an sh script at the 
beginning. I use sh as the older mach I have is python2 and the system 
Python is Python3 and running it with sh, it figures out which Python to 
use.
Don't know if current mach is still Python2.
I should really boot into Linux and try building the latest, or if 
really adventurous try FreeBSD.
Dave

[toc] | [prev] | [next] | [standalone]


#8623

FromNuno Silva <nunojsilva@invalid.invalid>
Date2026-06-29 07:32 +0100
Message-ID<111t3i9$4tqc$1@dont-email.me>
In reply to#8621
On 2026-06-29, Dave Yeo wrote:

> Louis Epstein wrote:
>>>> ps run mach --help to see some of the capabilities of mach if you are
>>>> curious.
>> While "mach" was part of what unzipped with the Seamonkey source tar,
>> mach is still itself a command not found and has no man page.
>>
>
> You can chmod it to be executable, or here on OS/2, simply run it as
> "sh mach --help" It is a Python script but also an sh script at the
> beginning. I use sh as the older mach I have is python2 and the system
> Python is Python3 and running it with sh, it figures out which Python
> to use.
> Don't know if current mach is still Python2.
> I should really boot into Linux and try building the latest, or if
> really adventurous try FreeBSD.

One issue is that at least some systems and distributions don't put the
CWD in $PATH, so even if +x, you still have to invoke it as a path,
relative or absolute.

-- 
Nuno Silva

[toc] | [prev] | [next] | [standalone]


#8622

FromDave Yeo <dave.r.yeo@gmail.com>
Date2026-06-28 22:25 -0700
Message-ID<Hln0S.52084$fs29.44135@fx01.iad>
In reply to#8618
Louis Epstein wrote:
> Dave Yeo <dave.r.yeo@gmail.com> wrote:
>> Louis Epstein wrote:
>>>>>> Just looking at old .mozconfig files on my machine.
>>> No analysis of which of these options are workable?
>>>
>>
>> Start out with a minimal .mozconfig, this one that you posted might be a
>> good start
>> ----
>> export CC=clang
>> export CXX=clang++
>> ac_add_options --enable-application=comm/suite
>> ac_add_options --enable-calendar
>> ac_add_options --enable-irc
>> ac_add_options --enable-dominspector
>> mk_add_options MOZ_OBJDIR=/home/ribo/mozilla-253/objdir
>
> (This is actually the home directory of my tech friend who
> gave up on Seamonkey as uninstallable and is currently off
> to Maine for the summer)

Well, obviously edit to your build location in $HOME. Used to be a bug 
where the MOZ_OBJDIR had to be absolute, don't know if it ever got 
fixed, safer to use the full path.

...

>> Which to me who doesn't use any BSD, seems a good start. IIRC the BSD's
>> use clang.
>
> In the FreeBSD ports/packages Clang comes as part of LLVM,I believe.
>
>> Then if it turns out that the build breaks building one of various libs
>> that SM includes, switch to using the system one, adding for example if
>> cairo is broken (unlikely but who knows) adding the right ac_add_options
>> such as
>> ac_add_options --enable-system-cairo
>> and retry.
>> I assume you have lots of memory so using the in tree libraries as
>> recommended shouldn't be a problem as long as they work on your FreeBSD
>> I don't know about Rust and you may need something in .mozconfig to
>> handle differences between what you have and what is expected. I think
>> one example disabled Rust simd to work around an issue. Slight
>> performance decrease I assume but likely not really noticeable.
>
> I have had enormous problems building Rust.

Yes, everything I've seen is that it is a bitch to build with at first 
needing to build each version then use that to build the next. Haven't 
followed development so don't know how it is currently.
Rust was what killed the Mozilla apps on OS/2, a 32 bit system.
...
Dave

[toc] | [prev] | [next] | [standalone]


#8494

FromYuchan Seo <hexagonwin@disroot.org>
Date2026-03-14 21:55 +0900
Message-ID<10p3lq1$8mov$1@dont-email.me>
In reply to#8491
Louis Epstein wrote:
> JKB <JKB@hilbert.invalid> wrote:
>>          Hello,
>>
>>          Just a message to say that SM 2.53.23 can be built on FreeBSD 15.0
>>          with following .mozconfig file:
>>
>> export CC=clang15
>> export CXX=clang++15
>> ac_add_options --enable-application=comm/suite
>> ac_add_options --enable-calendar
>> ac_add_options --enable-irc
>> ac_add_options --enable-dominspector
>> ac_add_options --enable-optimize
>> ac_add_options --disable-debug
>> ac_add_options --disable-rust-simd
>>
>>          mach has to use system python (mach bootstrap returns python
>>          errors).
>>
>>          With same .mozconfig, I was unable to build SM 2.53.22 on FreeBSD
>>          14.x or 15.0.
> 
> How about 13.5-RELEASE?
> 
> I know it's approaching EOL but I've been without a working Seamonkey
> for far too long now and want to get a functional install somehow.
>   
> 
> -=-=-
> The World Trade Center towers MUST rise again,
> at least as tall as before...or terror has triumphed.
> 
fyi the linux binary on seamonkey homepage works fine through linuxulator

[toc] | [prev] | [standalone]


Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | alt.comp.software.seamonkey


csiph-web