I second this, would be nice to have P7zip already included, and even configured with desktop context, because I am lazy and too dumb (mostly the latter :p) to get P7zip to work correctly and have context menus on my WMs.

network-manager-applet-1.8.12:

http://ftp.gnome.org/pub/gnome/sourc...et-1.8.12.news
http://ftp.gnome.org/pub/gnome/sourc...-1.8.12.tar.xz

emacs-26.1:

http://www.gnu.org/software/emacs/index.html#Releases
ftp://ftp.gnu.org/gnu/emacs/emacs-26.1.tar.xz

I just built emacs 26.1 here. Seems ok. I don't know about anyone else but I prefer it built against the lucid toolkit: it avoids some gtk warning messages and a longstanding bug with the emacs server when built against gtk.


BTW, there are some new configure options.

--with-gameuser=":games" is now required to keep using the current shared games scores files in /var/games/emacs. Without it games scores are per user and stored under ~/.emacs/games. If you choose not to add this option then you might as well remove /var/games/emacs/ entirely.

--disable-libsystemd without this the emacsclient/emacs --daemon server process connection tries to use systemd socket activation, it still appears to work, but throws out warnings. update: actually it looks like it still throws out the warnings even with this option, so perhaps it's just something that happens now when you run emacsclient -a "" and have the daemon process implicitly started. Apologies for the noise on this one.


--disable-build-details This is a new config option, it's not essential, but I added it here as I don't have any use for build details.

Other than those, my personal preference is for:
--with-x-toolkit=lucid
--without-toolkit-scroll-bars
... but that's just me. :)

I have noticed that a few graphics aren't showing in its webbrowser mode though. I need to investigate whether that's down to my choices of toolkit and whether it was happening with 25.3. I'll report back. update: ok, the missing image is also happening with the stock 25.3 package from current, so that's not new to 26.1. update2: Mystery solved. The idiotic GNU webpage designer created a black image with transparent background... which is fine unless you're using a theme that has a black background. LOL. Nothing to see here, move along! ;)

Can't be done with amdgpu-pro. It repackages a lot of extras like proprietary builds of a lot of xorg packages, llvm, mesa rebuilds, libdrm... it's just a bloody f---ing mess. If you need the extra functionality, just grab libtxc_dxtn and driconf from SlackBuilds. Much less a headache to deal with.

fftw-3.3.8:

http://www.fftw.org/release-notes.html
ftp://ftp.fftw.org/pub/fftw/fftw-3.3.8.tar.gz

stunnel-5.46:

https://www.stunnel.org/sdf_ChangeLog.html
https://www.stunnel.org/downloads/stunnel-5.46.tar.gz

Just a heads up that with the mesa git master you no longer will need libtxc_dxtn. This might also apply to the mesa version in current?

Did the embargo on s3tc and dxtc get lifted finally Orbea?

Yes.

https://phoronix.com/scan.php?page=n...ires-Next-Week

That was an itch I had to scratch. Adrien's repository was something I ran into by accident some years ago, perhaps it was ponce even then who told me. It looks like Adrien simply commits the whole of ./source directory to his git repository, using a .gitignore file to avoid the source tarballs getting committed as well.

But I thought that's too simple, and wrote a script to create a (more generally useful IMO) repository tracking the whole slackware64-current directory tree, additionally unzipping all the compressed files like the patches, doinst.sh scripts etc.
More readability in the cgit interface and better diffing.

See https://git.slackware.nl/current/ - there's only one commit (the initial one) but I scheduled the script in cron to update the repository on every public update of the ChangeLog.txt in slackware64-current.
Hope it's useful to someone :-)

git-2.17.1:

(CVE fix)
https://raw.githubusercontent.com/gi...tes/2.17.1.txt
https://mirrors.edge.kernel.org/pub/...-2.17.1.tar.xz

rust-1.26.1:

https://blog.rust-lang.org/2018/05/29/Rust-1.26.1.html
https://static.rust-lang.org/dist/ru...6.1-src.tar.gz

ImageMagick-6.9.9.48:

https://www.imagemagick.org/download....9.9-48.tar.lz

FVWM 2.6.8
 

Forwarding an email from Thomas Adam to the main mailing list:

After upgrading some machines to the latest slackware-current (from kernel 4.9.67 to 4.14.44) I get following warning in syslog after each reboot
This disappeared after updating from mcelog-153 to mcelog-158:
https://git.kernel.org/pub/scm/utils...ce/mcelog.git/

Another new imagemagick-6.9.9-49:

https://www.imagemagick.org/download....9.9-49.tar.lz

The mesa driver, for the most part, works just as good, if not better, than amdgpu-pro with a lot less headache. But, AFAIK, the amdgpu-pro driver doesn't support Xorg 1.20, so you can't get a working amdgpu-pro driver on -current unless you downgrade Xorg. I do have SlackBuilds available for several different versions of the driver, but I don't think any will work on -current (but 17.10 and below should function on 14.2).

https://github.com/bassmadrigal/slac...gpu-pro-driver

How many times do I have to tell you this? amdgpu-pro *can* be repackaged into a Slackware package that doesn't affect the base packages (and uninstalls cleanly). I've done it with past versions, but I haven't kept it up to date for three reasons:

1. I don't have the hardware to check it.
2. There doesn't seem to be much of a demand for it (especially with the performance of the mesa driver).
3. amdgpu-pro 17.30 and newer require Xorg 1.19, which, at the time, was only available in -current (which I don't normally run), but current has now been upgraded to 1.20, so there isn't even a version that will work with -current.

tput error message fixed using tty -s
 

When running ssh with upgradepkg I get an annoying error message from tput:

tput: No value for $TERM and no -T specified

Both installpkg and upgradepkg have the bug and both can be fixed by using "tty -s"
There is an interesting discussion of the tput error message here:
https://community.hpe.com/t5/System-...d/td-p/5255040

The tty command is part of coreutils and should be safe to use.

Instead of using "tty -s" another way is to use "test -t 0" or "test -t 1".

pygobject3-3.28.3

http://ftp.gnome.org/pub/gnome/sourc...-3.28.3.tar.xz

Some xorg drivers updates

xf86-input-evdev-2.10.6
https://www.x.org/releases/individua...2.10.6.tar.bz2

xf86-input-synaptics-1.9.1
https://www.x.org/releases/individua...-1.9.1.tar.bz2

xf86-video-fbdev-0.5.0
https://www.x.org/releases/individua...-0.5.0.tar.bz2

cmake-3.11.3:

https://blog.kitware.com/cmake-3-11-...-for-download/
https://cmake.org/files/v3.11/cmake-3.11.3.tar.gz

Please enable OpenMP in ImageMagick.

Many applications and scripts depends on ImageMagick, being capped at one thread is an major issue.

mesa 18.1.1


https://lists.freedesktop.org/archiv...ne/196523.html

https://mesa.freedesktop.org/archive/mesa-18.1.1.tar.xz

sysstat:

http://sebastien.godard.pagesperso-orange.fr/
http://perso.orange.fr/sebastien.god...-11.6.4.tar.xz

librsvg-2.42.5:

http://ftp.gnome.org/pub/gnome/sourc...vg-2.42.5.news
http://ftp.gnome.org/pub/gnome/sourc...-2.42.5.tar.xz

nano-2.9.8:

https://www.nano-editor.org/news.php
http://ftp.gnu.org/gnu/nano/nano-2.9.8.tar.xz

p11-kit-0.23.12

https://github.com/p11-glue/p11-kit/...0.23.12.tar.gz

It will solve building gnutls-3.6.2 on regression test.

Thank you for the recent changes in slackpkg :D

dhcpcd-7.0.5:

https://roy.marples.name/git/dhcpcd.git/
ftp://roy.marples.name/pub/dhcpcd/dhcpcd-7.0.5.tar.xz

ModemManager-1.8.0:

https://cgit.freedesktop.org/ModemMa.../log/?h=mm-1-8
https://www.freedesktop.org/software...r-1.8.0.tar.xz

I have a small suggestion for the installer.

The hibernation mechanism in the -current kernel seems to be working flawlessly.

When the installer is formatting swap and adding it to the system, why not add a 'resume=/dev/$nameoftheswappartition' to elilo.conf?

This shouldn't hurt, and would make hibernation work automatically.

Midnight Commander-4.8.21

What's new," https://midnight-commander.org/wiki/NEWS-4.8.21

The tarball, http://www.midnight-commander.org/do...4.8.21.tar.bz2

Bluez

+++ b/ChangeLog
+ver 5.50:
+ Fix issue with GATT and reading long values.
+ Fix issue with GATT and reading multiple includes.
+ Fix issue with GATT and service changes when offline.
+ Fix issue with handling secondary service discovery.
+ Fix issue with handling persistency of CCC values.
+ Fix issue with handling Mesh session on disconnection.
+ Fix issue with handling Mesh proxy PDU SAR message length.
+ Fix issue with handling Mesh default heartbeat TTL value.
+ Add support for Mesh node-reset operation handling.
+ Add support for GATT authorization request handling.
+ Add support for GATT minimum key size requirements.
+ Add support for GATT server and included services.
+ Add support for handling additional advertising data.
+ Add support for handling separate discoverable state.
+ Add support for enabling HFP version 1.7 features.
+ Add support for dedicated Bluetooth logging daemon.
+
Source: http://www.kernel.org/pub/linux/blue...ez-5.50.tar.xz

I finally got around to testing this:
git send-email failed.

After installing these perl modules on '14.2+1528036609' git-send-email works:
IO-Socket-SSL
Net-SSLeay

I just used the SBo build scripts and installed the resulting packages.

To test from a git repo:
git send-email --smtp-debug=1 0001-foo.patch

Thank you for implementing this Pat!

yes, those are needed as dependencies of libnet: I forgot to check that all was ok after asking for libnet inclusion, sorry.

gnuplot-5.2.4:

https://sourceforge.net/projects/gnu...gnuplot/5.2.4/
https://freefr.dl.sourceforge.net/pr...t-5.2.4.tar.gz

Hi!

I ran 'slackpkg update' and 'slackpkg upgrade-all' (current) today, slackpkg was updated, so I ran those commands again, and ended up with this:
Any ideas what happened and what to do?

UPDATE: Sorry, I missed this post: https://www.linuxquestions.org/quest...te-4175630917/

This new version is very unstable on my current - it usually breaks after wakeup, and sometimes when changing network.

UPDATE: this is solved in Thu May 31 04:55:33 UTC 2018 update.

re-install of current (installer, gparted oddness)
 

Not sure if this is the right thread to use for this, but it's close...

I was occasionally testing slackware64-current (plus some of my "required" extras from SBo and Alien) on one of my laptops (Thinkpad T420) when I was initimidated by the massive Changelog of April 19 ... and so I let things slide for a while - (wasn't all that "current" even then as I was running a 4.14.17 kernel on that box). I ultimately decided I should get back to trying it out again.

Given how much I'd have to catch up, I decided the best way to proceed was to backup the installation so I could retrieve all of my usual customizations, etc.. and re-install from scratch and start anew. I used Alien's fine "mirror-slackware-current.sh" script on my 14.2 daily driver to create and burn an install dvd image from yesterday's (June 3, 2018) ChangeLog... things did not go as effortlessly as I expected, but I expect that by tomorrow or the day after, I'll be back to where I was. Here is a summary of unexpected behavior so far. Really, just 2 things.
-----
Installer - because I wanted to make sure nothing was left of previous install, I made sure to ask it to format the partitions I was going to be using (just a swap and a single ext4 partition). I haven't made the leap to using UEFI yet, so I asked lilo to be installed to MBR (as it was before) - I also asked for a basic not an expert install, thinking I would boot to huge.s and convert to generic plus initrd in a separate step (haven't done that yet). I'm used to the installer finding my Windows OS partitions and creating a stanza for them - but it did not. Not an issue (as the previous lilo.conf is one of the things that I made sure to have a backup of), but I did expect it.
-----
gparted - One of the customizations that I had done was to change to using UUIDs to identify all my partitions - so once I booted and got my network configured, I thought I'd use gparted to verify the UUIDs I had on file. but gparted did not run successfully; instead it produced a dialog pop-up with the following text [ "Libparted Bug - Assertion (metadata_length > 0 at dos.c:2305 in function add_logical_part_metadata() failed" ]. I used cfdisk to verify the UUIDs - and they all appeared ok), but I thought this should be reported, too.
-----
Really looking forward to the next release. Thx

bash-4.4 (023):

http://ftp.gnu.org/gnu/bash/bash-4.4-patches/

elf-utils-0.171:

https://sourceware.org/git/?p=elfutils.git;a=summary
ftp://sourceware.org/pub/elfutils/0....-0.171.tar.bz2

readline-7.0 (005):

http://git.savannah.gnu.org/cgit/readline.git/log/
ftp://ftp.gnu.org/pub/gnu/readline/readline-7.0-patches

libinput-1.11.0:

https://cgit.freedesktop.org/wayland/libinput/
https://www.freedesktop.org/software...-1.11.0.tar.xz

OpenLDAP server (2.4.46)

I've added the full support of the OpenLDAP server in Slackware (this is the SlackBuilds: https://slackbuilds.org/repository/1...enldap-server/) and I hope for the official inclusion in Slackware.

cups-2.2.8:

https://github.com/apple/cups/releases/tag/v2.2.8
https://github.com/apple/cups/releas...-source.tar.gz


sqlite-3.24.0:

http://www.sqlite.org/releaselog/3_24_0.html
https://github.com/apple/cups/releas...-source.tar.gz

rust-1.26.2

announce note --> https://blog.rust-lang.org/2018/06/05/Rust-1.26.2.html

harfbuzz (1.7.7 or 1.8.0)

https://cgit.freedesktop.org/harfbuzz/
https://www.freedesktop.org/software...-1.7.7.tar.bz2
https://www.freedesktop.org/software...-1.8.0.tar.bz2

Thanks Pat,
'git send-email' with TLS authentication is working for me now.

Thunar-1.8

Thunar has been ported to GTK+3.
(not sure if "ported" is the corrent term).

The announcement, https://andreldm.com/2018/06/06/thun...0-release.html
:)

Thunar 1.8 requires libxfce4ui 4.13 and or libxfce4util 4.13 so perhaps no ready for prime time yet.

libxfce4ui-4.13.4
https://github.com/xfce-mirror/libxf...-4.13.4.tar.gz

libxfce4util-4.13.1
https://github.com/xfce-mirror/libxf...-4.13.1.tar.gz

gnupg-2.2.8 is released with security fixes.

https://gnupg.org/ftp/gcrypt/gnupg/g....8.tar.bz2.sig
https://gnupg.org/ftp/gcrypt/gnupg/gnupg-2.2.8.tar.bz2

1. Can we have things like crond and atd started from separate rc.$service scripts instead of being directly called in rc.M? They are services just like syslog, acpid or all the other things rc.M starts rc.<service> for.
2. Please make crond be called with something like $CROND_OPTS after sourcing /etc/default/crond, so those who want to log what is cron doing can use CROND_OPTS="-l info" instead of editing rc.M each time sysvinit-scripts is updated

PS. when creating rc.crond, please use "pkill --ns $$ ..." in stop() so that cron daemons in other namespaces (containers) are not getting hit. That's something I mentioned a while ago here but sadly it didn't get much traction: most of rc scripts suffer from this issue - worst case being rc.httpd, which is always overwritten by doinst.sh when upgrading httpd (why?)

Thanks in advance,

xkbcomp-1.4.2:

https://cgit.freedesktop.org/xorg/app/xkbcomp/
ftp://ftp.x.org/pub/individual/app/xkbcomp-1.4.2.tar.gz

iproute2-4.17.0:

https://git.kernel.org/pub/scm/netwo...2/iproute2.git
https://mirrors.edge.kernel.org/pub/...-4.17.0.tar.gz