bug#63009: etc/team.scm cc home #2

[Janneke Nieuwenhuizen]

Seems only one X-Debbugs-Cc header is honoured at a time, forgot them
initiially...

bug#63009: etc/team.scm cc home #3

[Janneke Nieuwenhuizen]

Seems only one X-Debbugs-Cc header is honoured at a time, forgot them
initiially...

bug#62948: etc/team.scm cc home #2

[Janneke Nieuwenhuizen]

Seems only one X-Debbugs-Cc header is honoured at a time, forgot them
initiially...

bug#62948: etc/team.scm cc home #3

[Janneke Nieuwenhuizen]

Seems only one X-Debbugs-Cc header is honoured at a time, forgot them
initiially...

bug#62890: StumpWM fails to start - Read only file system

[Guillaume Le Vaillant]

I think the issue comes from grafts.

Without grafts, the ASDF configuration is coherent; the path for
stumpwm's sources and libraries is the same everywhere (here in
stumpwm:lib and sbcl-stumpwm-cpu):

--8<---cut here---start->8---
$ cat $(guix build --no-grafts stumpwm | grep 
'lib$')/etc/common-lisp/*/50-stumpwm.conf
(("/gnu/store/v18hzbid3rjblz35s69w4c0gcsx9f9w9-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*)
 
("/gnu/store/v18hzbid3rjblz35s69w4c0gcsx9f9w9-stumpwm-22.11-lib/lib/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*))

(:tree 
"/gnu/store/v18hzbid3rjblz35s69w4c0gcsx9f9w9-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm")

$ cat $(guix build --no-grafts 
sbcl-stumpwm-cpu)/etc/common-lisp/*/50-stumpwm.conf
(("/gnu/store/v18hzbid3rjblz35s69w4c0gcsx9f9w9-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*)
 
("/gnu/store/v18hzbid3rjblz35s69w4c0gcsx9f9w9-stumpwm-22.11-lib/lib/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*))

(:tree 
"/gnu/store/v18hzbid3rjblz35s69w4c0gcsx9f9w9-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm")
--8<---cut here---end--->8---


With grafts, we get different paths, which probably explains why ASDF
fails to find the compiled Lisp files it is looking for:

--8<---cut here---start->8---
$ cat $(guix build stumpwm | grep 'lib$')/etc/common-lisp/*/50-stumpwm.conf
(("/gnu/store/wsjyjqf20ldbmgdq96h73qikmwhxv36c-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*)
 
("/gnu/store/wsjyjqf20ldbmgdq96h73qikmwhxv36c-stumpwm-22.11-lib/lib/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*))

(:tree 
"/gnu/store/wsjyjqf20ldbmgdq96h73qikmwhxv36c-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm")

$ cat $(guix build sbcl-stumpwm-cpu)/etc/common-lisp/*/50-stumpwm.conf
(("/gnu/store/h8i8mwsgjb96r407xqa2sf56clgy7r7c-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*)
 
("/gnu/store/h8i8mwsgjb96r407xqa2sf56clgy7r7c-stumpwm-22.11-lib/lib/common-lisp/sbcl/stumpwm"
  :**/
  :*.*.*))

(:tree 
"/gnu/store/h8i8mwsgjb96r407xqa2sf56clgy7r7c-stumpwm-22.11-lib/share/common-lisp/sbcl/stumpwm")
--8<---cut here---end--->8---


signature.asc
Description: PGP signature

bug#63047: Can't load glib debug symbols in gdb

[Maxim Cournoyer]

Hi Andrew,

Josselin Poiret via Bug reports for GNU Guix  writes:

> Hi Andrew,
>
> Andrew Tropin  writes:
>
>> I try to run emacs in gdb with debug symbols for some libs available, I
>> succeed with gtk+, but it doesn't work for glib and glibc.  It looks
>> strange to me, but maybe I am doing something wrong.
>>
>> Reproducer:
>>
>> guix shell gdb emacs-next-pgtk glibc:debug gtk+:debug glib:debug \
>>  --with-debug-info=glibc --with-debug-info=glib --with-debug-info=gtk+ \
>>  --no-grafts -- gdb .emacs-30.0.50-real
>
> At least for glibc, the glibc that is linked against is the one in (gnu
> packages commencement), which is hidden from the user.  The one in (gnu
> packages base), which you can refer to with "glibc" is different.  You
> can try to find the proper debug output by looking at `guix size` of
> your store path, then finding out the deriver for glibc with `guix gc
> --derivers` and finally looking at the .drv to find out what the debug
> output should be.
>
> For glib, it might be similar, make sure that you're using exactly the
> right store path for it.

Perhaps try on the core-updates branch, where glibc no longer has its
symbols stripped.  Another thing that can cause the debug symbols to not
be found is grafts, as described in #48907, so when debugging I'd
recommend using --no-grafts for now.

-- 
Thanks,
Maxim

bug#62176: Close

[Andreas Enge]

Mesa has been updated in the latest core-updates merge.

Andreas

bug#61364: Close

[Andreas Enge]

Core-updates has been merged, and hopefully addresses all your problems.
If some remain, please feel free to open new issues and submit corresponding
patches.

Thanks!

Andreas

bug#62954: Valgrind blocks R on powerpc64le

[Andreas Enge]

Closing this bug, the topic is treated through the patches in
   https://issues.guix.gnu.org/62967

Andreas

bug#63073: News entry for ‘core-updates’

[Ludovic Courtès]

Hi!

As a followup to our discussion on IRC, here’s a proposed news entry for
‘core-updates’ that we should push soonish (I can do that later today or
tomorrow but feel free to do it before!).

I probably missed important items, maybe build system changes?

Ludo’.

diff --git a/etc/news.scm b/etc/news.scm
index 7f38f8f294..6a6fbffa79 100644
--- a/etc/news.scm
+++ b/etc/news.scm
@@ -26,6 +26,28 @@
 (channel-news
  (version 0)
 
+ (entry (commit "c919bfefd98bf2e29549539b4e28e6dc2a8a6f32")
+(title
+ (en "Core packages updated"))
+(body
+ (en "Core packages have been updated, following months of hard work
+by contributors.  Noteworthy package upgrades include:
+
+@itemize
+@item glibc 2.35;
+@item Python 3.10;
+@item Perl 5.36;
+@item Mesa 22;
+@item GCC 11 is now used as the default compiler.
+@end itemize
+
+A major highlight is the introduction of the so-called @dfn{full-source
+bootstrap}: packages are all built starting from a 500-byte program called
+stage0, which is then used to build a higher-level interpreter, a basic Scheme
+interpreter and C compiler (GNU Mes), and so on, until @acronym{GCC, the GNU
+Compiler Collection} is finally built.  This is a premiere and a huge step
+forward in terms of transparency of auditability.")))
+
  (entry (commit "21564fada141bfba25d471518b293b6004244c3a")
 (title
  (en "Linux-libre LTS kernel updated to 6.1")

bug#63068: Apache mirrors are not available anymore

[Julian Flake]

- The apache mirrors defined in guix/download.scm aren't available 
 anymore.

- Compare the info at https://www.apache.org/mirrors
- (At least Some of the) packages are available at 
 https://dlcdn.apache.org/

bug#63076: [PATCH] gnu: libunistring: Update to 1.1.

[Vivien Kraus via Bug reports for GNU Guix]

* gnu/packages/libunistring.scm (libunistring): Update to 1.1.
---
 gnu/packages/libunistring.scm | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/gnu/packages/libunistring.scm b/gnu/packages/libunistring.scm
index eccd053a35..f0b071f11c 100644
--- a/gnu/packages/libunistring.scm
+++ b/gnu/packages/libunistring.scm
@@ -35,7 +35,7 @@ (define-module (gnu packages libunistring)
 (define-public libunistring
   (package
(name "libunistring")
-   (version "1.0")
+   (version "1.1")
(source (origin
 (method url-fetch)
 (uri (string-append
@@ -43,7 +43,7 @@ (define-public libunistring
   version ".tar.xz"))
 (sha256
  (base32
-  "0h97qx3c4970wvh25g51sjgz54xn37lrfy95dg97xmvmkys5basv"
+  "164vfbh3sf0qjcimjbnz4icqr9cai30dli8p2y5p6z3frfwiwz42"
(propagated-inputs (libiconv-if-needed))
(outputs '("out" "static"))
(build-system gnu-build-system)

base-commit: 74038cf66e3a8cdcd9d97a81d9176c3cdefdaaf7
-- 
2.39.2

bug#63076: This is for core-updates

[Vivien Kraus via Bug reports for GNU Guix]

Hello,

This patch should go to core-updates, but I can’t find the branch.

Best regards,

Vivien

bug#63077: [PATCH] gnu: gnutls: skip examples on mingw.

[Vivien Kraus via Bug reports for GNU Guix]

* gnu/packages/tls.scm (gnutls): Skip the doc/examples subdir on mingw,
because the gnulib modules are not correct.
---
 gnu/packages/tls.scm | 9 +
 1 file changed, 9 insertions(+)

diff --git a/gnu/packages/tls.scm b/gnu/packages/tls.scm
index 6d7cff41b0..9e16b04702 100644
--- a/gnu/packages/tls.scm
+++ b/gnu/packages/tls.scm
@@ -253,6 +253,15 @@ (define-public gnutls
(substitute* "tests/fastopen.sh"
  (("^unset RETCODE")
   "exit 77\n"   ;skip
+   #$@(if (target-mingw?)
+  #~((add-after 'unpack 'skip-doc-examples
+   ;; The examples in doc do not link to correct
+   ;; gnulib modules.
+   (lambda _
+ (substitute* "doc/Makefile.in"
+   (("SUBDIRS = examples")
+"SUBDIRS =")
+  #~())
#$@(if (target-ppc32?)
   ;; https://gitlab.com/gnutls/gnutls/-/issues/1354
   ;; Extend the test timeout from the default of 20 * 1000

base-commit: e287a0e402ee15eefa80c2804e487025507fe3b1
-- 
2.39.2

bug#63050: "guix pull" requires graphical libraries

[Ludovic Courtès]

Hi,

Andreas Enge  skribis:

> While trying out a "guix pull" on an aarch64 machine, for which many
> packages are currently not available as substitutes, I notice an extra-
> ordinary amount of dependencies, see below (and since I interrupted and
> restarted it, there are even more dependencies in reality; I remember
> X11 libraries such as libxi and libxt). Many of them are related to
> graphical environments, which should not happen for a command line
> program. Chances are they are pulled in for building documentation
> (not necessarily of Guix, but of packages that are needed for pulling);
> but this is still undesirable and should be sorted out.

This is apparently coming from Graphviz:

--8<---cut here---start->8---
$ guix graph --path guix libx11
guix@1.4.0-5.286cdf0
graphviz@2.49.0
libx11@1.7.3.1
$ guix graph --path guix libxt
guix@1.4.0-5.286cdf0
graphviz@2.49.0
libxaw@1.0.14
libxt@1.2.1
--8<---cut here---end--->8---

Surprising to me, but apparently it’s been this way from the start,
commit b1b07d72c755ea314fb0c8333cd88293ee504ce4 (2013!).

Maybe these are optional dependencies?

Ludo’.

bug#31769: mpd / pulse control issue

[Maxim Cournoyer]

Hi,

iyzs...@member.fsf.org (宋文武) writes:

> Bradley Haggerty  writes:
>
>> I have mpd enabled as a herd service. When I start my computer, mpd
>> can play my music fine, but pavucontrol can't connect to pulse and
>> show my volume controls. If I
>> kill pulse and open pavucontrol, all my volume controls are visible,
>> but then mpd can't play music. If I kill pulse again and this time
>> play a song in mpd, mpd works again,
>> but pavucontrol can't connect. Basically, if mpd starts pulse, mpd
>> works, but other things can't connect, and if pavucontrol starts
>> pulse, mpd can't connect to pulse. Since
>> I have mpd enabled as a service, it's the one to start pulse on a
>> fresh boot. I also tried starting pulse as my user by killing it and
>> opening pavucontrol, then restarting the
>> mpd service in case it could find my pulse service this way. No
>> success there. While mpd has control of pulse, other applications
>> also lack sounds, such as mpv (video
>> players) and icecat.
>
> According to the wiki of ArchLinux, users of PulseAudio with a
> system-wide MPD configuration have to implement a workaground:
>
> 

Could this happen to have been fixed by
bc30a9ee889fb1b81c43a7f94ea4c2b95a15db75 ("services: mpd: Set
PulseAudio-related variables.") ?  I'm not sure I understand what the
issue is exactly, but that commit did do some pulseaudio-related configuration.

-- 
Thanks,
Maxim

bug#63082: mpd defaul configuration does not work ('No database' error)

[Maxim Cournoyer]

Hello,

Our MPD service does not work out of the box.  Running under its default
'mpd' user, clients would get the following error attempting to update
its database, e.g. by running 'mpc update':

--8<---cut here---start->8---
MPD error: database
--8<---cut here---end--->8---

This particular problem was solved by adding the following field to my
mpd-configuration record (thanks to Bruno):

--8<---cut here---start->8---
(database (mpd-plugin
   (plugin "simple")
   (extra-options
'((path . "/var/lib/mpd/database")
  (cache-directory . "/var/cache/mpd")
--8<---cut here---end--->8---

and running

--8<---cut here---start->8---
mkdir /var/cache/mpd && chown mpd:users /var/cache/mpd
--8<---cut here---end--->8---

Now the above error is gone, but I do not hear any audio coming out, so
there seems to be another problem.

I'm opening this ticket so that we can document and resolve all these
issues in the configuration, so that hopefully our mpd system service
can work out of the box.

-- 
Thanks,
Maxim

bug#63082: mpd defaul configuration does not work ('No database' error)

[Maxim Cournoyer]

Hello,

Maxim Cournoyer  writes:

> Hello,
>
> Our MPD service does not work out of the box.  Running under its default
> 'mpd' user, clients would get the following error attempting to update
> its database, e.g. by running 'mpc update':
>
> MPD error: database
>
>
> This particular problem was solved by adding the following field to my
> mpd-configuration record (thanks to Bruno):
>
> (database (mpd-plugin
>(plugin "simple")
>(extra-options
> '((path . "/var/lib/mpd/database")
>   (cache-directory . "/var/cache/mpd")
>
>
> and running
>
> mkdir /var/cache/mpd && chown mpd:users /var/cache/mpd

The cache-directory configuration doesn't seem to be useful so far;
nothing was being populated under /var/cache/mpd.

in /var/log/messages, pulseaudio throws a couple errors:

--8<---cut here---start->8---
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] 
module-jackdbus-detect.c: Unable to contact D-Bus session bus: 
org.freedesktop.DBus.Error.NotSupported: Unable to autolaunch a dbus-daemon 
without a $DISPLAY for X11
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] module.c: Failed to 
load module "module-jackdbus-detect" (argument: "channels=2"): initialization 
failed.
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] cli-command.c: 
stat('/gnu/store/5ahapvp7rnd2ymakyjv1pwwdav7w9wdc-pulseaudio-16.1/etc/pulse/default.pa.d'):
 No such file or directory
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] main.c: No card found 
by this name or index.
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] main.c: Source 
alsa_input.pci-_01_01.0.analog-mono does not exist.
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] main.c: Sink 
alsa_output.pci-_01_01.0.analog-surround-40 does not exist.
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] server-lookup.c: 
Unable to contact D-Bus: org.freedesktop.DBus.Error.NotSupported: Unable to 
autolaunch a dbus-daemon without a $DISPLAY for X11
Apr 25 23:06:52 localhost pulseaudio[10356]: [pulseaudio] main.c: Unable to 
contact D-Bus: org.freedesktop.DBus.Error.NotSupported: Unable to autolaunch a 
dbus-daemon without a $DISPLAY for X11
--8<---cut here---end--->8---

-- 
Thanks,
Maxim