Re: building package: gcc 7.5.0 referenced instead of latest gcc; not sure why
On Mon, Feb 22, 2021 at 04:10:16PM -0800, Andy Tai wrote:
[...]
> I am not sure why gcc 7.5 is being used; no where in the package
> definition scm file references explicitly gcc 7.5. I did not install
> gcc 7.5 explicitly either.
This is the "default" GCC package used when building packages, gcc-final
from %final-inputs in 'gnu/packages/commencement.scm':
https://git.savannah.gnu.org/cgit/guix.git/tree/gnu/packages/commencement.scm?h=v1.2.0#n3753
If you follow the chain of inheritance and references from gcc-final,
you'll see it goes like this:
gcc-final -> gcc-boot0 -> gcc -> gcc-7.5
... and in gcc-7.5, the version is defined.
You can override it by adding other packages of gcc to the native-inputs
of vtk. For example:
--
diff --git a/gnu/packages/image-processing.scm
b/gnu/packages/image-processing.scm
index c3ea8491d9..195b21c508 100644
--- a/gnu/packages/image-processing.scm
+++ b/gnu/packages/image-processing.scm
@@ -49,6 +49,7 @@
#:use-module (gnu packages documentation)
#:use-module (gnu packages flex)
#:use-module (gnu packages fontutils)
+ #:use-module (gnu packages gcc)
#:use-module (gnu packages geo)
#:use-module (gnu packages ghostscript)
#:use-module (gnu packages gl)
@@ -291,6 +292,8 @@ many popular formats.")
"-DVTK_USE_SYSTEM_TIFF:BOOL=TRUE"
"-DVTK_USE_SYSTEM_ZLIB:BOOL=TRUE")
#:tests? #f));XXX: test data not included
+(native-inputs
+ `(("gcc" ,gcc-8)))
(inputs
`(("double-conversion" ,double-conversion)
("eigen" ,eigen)
--
... and then it will be compiled using our package of GCC 8, which is
precisely:
--
$ guix show gcc-toolchain@8
name: gcc-toolchain
version: 8.4.0
outputs: out debug static
systems: x86_64-linux i686-linux
dependencies: binutils@2.34 gcc@8.4.0 glibc@2.31 ld-wrapper@0
location: gnu/packages/commencement.scm:3836:4
homepage: https://gcc.gnu.org/
license: GPL 3+
synopsis: Complete GCC tool chain for C/C++ development
description: This package provides a complete GCC tool chain for C/C++
development to be installed in user profiles. This includes GCC, as well as
libc (headers and binaries, plus debugging
+ symbols in the `debug' output), and Binutils. GCC is the GNU Compiler
Collection.
--
I hope that helps!
building package: gcc 7.5.0 referenced instead of latest gcc; not sure why
Hi, I checked out guix git following instructions in https://github.com/pjotrp/guix-notes/blob/master/INSTALL.org#fetch-repository-with-git and everything checked out and built fine. the gcc installed is version 10.2.0 ./pre-inst-env gcc --version gcc (GCC) 10.2.0 Copyright (C) 2020 Free Software Foundation, Inc. I tried to update a package definition (vtk) and then tried to build it ./pre-inst-env guix build -k -K vtk and the log file shows gcc 7.5 being used (log fragment) [ 28%] Building CXX object Common/DataModel/CMakeFiles/vtkCommonDataModel.dir/vtkQuadraticLinearWedge.cxx.o cd /tmp/guix-build-vtk-8.2.0.drv-0/build/Common/DataModel && /gnu/store/rn75fm7adgx3pw5j8pg3bczfqq1y17lk-gcc-7.5.0/bin/c++ -DVTK_IN_VTK -DvtkCommonDataModel_EXPORTS -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/DataModel -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/DataModel -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Core -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Core -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Utilities/KWIML -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Utilities/KWIML -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Utilities/KWSys -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Utilities/KWSys -I/tmp/guix-build-vtk-8.2.0.drv-0/build/ThirdParty/utf8 -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/ThirdParty/utf8 -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Math -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Math -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Misc -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Misc -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/System -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/System -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Transforms -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Transforms -O3 -DNDEBUG -fPIC -fvisibility=hidden -std=c++11 -o CMakeFiles/vtkCommonDataModel.dir/vtkQuadraticLinearWedge.cxx.o -c /tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/DataModel/vtkQuadraticLinearWedge.cxx I am not sure why gcc 7.5 is being used; no where in the package definition scm file references explicitly gcc 7.5. I did not install gcc 7.5 explicitly either. (gcc 7.5 has problem compiling the package source so I want to make sure later gcc used) Thanks for any hint on this
building package: gcc 7.5.0 referenced instead of latest gcc; not sure why
Hi, I checked out guix git following instructions in https://github.com/pjotrp/guix-notes/blob/master/INSTALL.org#fetch-repository-with-git and everything checked out and built fine. the gcc installed is version 10.2.0 ./pre-inst-env gcc --version gcc (GCC) 10.2.0 Copyright (C) 2020 Free Software Foundation, Inc. I tried to update a package definition (vtk) and then tried to build it ./pre-inst-env guix build -k -K vtk and the log file shows gcc 7.5 being used (log fragment) [ 28%] Building CXX object Common/DataModel/CMakeFiles/vtkCommonDataModel.dir/vtkQuadraticLinearWedge.cxx.o cd /tmp/guix-build-vtk-8.2.0.drv-0/build/Common/DataModel && /gnu/store/rn75fm7adgx3pw5j8pg3bczfqq1y17lk-gcc-7.5.0/bin/c++ -DVTK_IN_VTK -DvtkCommonDataModel_EXPORTS -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/DataModel -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/DataModel -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Core -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Core -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Utilities/KWIML -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Utilities/KWIML -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Utilities/KWSys -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Utilities/KWSys -I/tmp/guix-build-vtk-8.2.0.drv-0/build/ThirdParty/utf8 -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/ThirdParty/utf8 -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Math -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Math -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Misc -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Misc -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/System -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/System -I/tmp/guix-build-vtk-8.2.0.drv-0/build/Common/Transforms -I/tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/Transforms -O3 -DNDEBUG -fPIC -fvisibility=hidden -std=c++11 -o CMakeFiles/vtkCommonDataModel.dir/vtkQuadraticLinearWedge.cxx.o -c /tmp/guix-build-vtk-8.2.0.drv-0/VTK-8.2.0/Common/DataModel/vtkQuadraticLinearWedge.cxx I am not sure why gcc 7.5 is being used; no where in the package definition scm file references explicitly gcc 7.5. I did not install gcc 7.5 explicitly either. (gcc 7.5 has problem compiling the package source so I want to make sure later gcc used) Thanks for any hint on this
Re: Foreign distribution: errors when loading contrib modules of stumpwm
Roland Everaert --- Use the F.O.S.S., Luke Sent with ProtonMail Secure Email. ‐‐‐ Original Message ‐‐‐ Le lundi 22 février 2021 à 10:41, Guillaume Le Vaillant a écrit : > Roland Everaert via help-guix@gnu.org skribis: > > > I have installed StumpWM and some contrib modules, on a Fedora 33 > > distribution using the guix PM, for many month, without much problem. Since > > the begining of the week (last guix update), whenever I load a contrib > > module, StumpWM complains about a FASL file being read-only somewhere in > > /gnu/store/-stumpwm-lib. > > > > Any idea what has changed in stumpwm and its contrib modules? > > > > Below is my StumpWM configuration: > > > > Start of sources - > > > > ;; --lisp-- > > > > ;; > > > > (in-package :stumpwm) > > > > ;; * Adding modules > > > > (add-to-load-path "~/.guix-profile/share/common-lisp/sbcl/clx-xembed") > > > > (add-to-load-path > > "~/.guix-profile/share/common-lisp/sbcl/stumpwm-stumptray") > > > > (load-module "stumptray") > > > > End of sources - > > This is strange, as the last change directly related to StumpWM was > > made several months ago (around 2020-12-10). > > How do you start StumpWM? Do you use a '.xsession' file? > > What are your XDG_CONFIG_DIRS and XDG_DATA_DIRS environment variables > > set to? I start Stumpwm from gdm, and I don't use an .xsession file. I manage to configure emacs and StumpWM correctly, to be able to interact with StumpWM using SLY. You will find below the code I have executed in the REPL and the exception generated by the last instruction. The value of the Environment variable: STUMPWM> (asdf:load-system :uiop) T STUMPWM> (uiop:getenv "XDG_CONFIG_DIRS") "/home/roland/.guix-profile/etc" STUMPWM> (uiop:getenv "XDG_DATA_DIRS") "/home/roland/.guix-profile/share:/home/roland/.guix-profile/share:/home/roland/.local/share/flatpak/exports/share:/var/lib/flatpak/exports/share:/usr/local/share/:/usr/share/" The code: CL-USER> (in-package stumpwm) # STUMPWM> (set-module-dir "~/.guix-profile/share/common-lisp/sbcl") (#P"/gnu/store/yi6dypcphhjbz9gp0l8z8r3pwjr71q6v-sbcl-clx-0.7.5/share/common-lisp/sbcl/clx/" #P"/gnu/store/lskhxlvddjjkg3c4v49ppm9y022ajzi6-sbcl-stumpwm-pass-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-pass/" #P"/gnu/store/ia41a2gd9klix8xdmhgsr16j998q7xfx-sbcl-stumpwm-ttf-fonts-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-ttf-fonts/" #P"/gnu/store/dvs3f9jrw4g3p3g17v34a71l3ccjas8b-sbcl-stumpwm-net-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-net/" #P"/gnu/store/4fjpm820912fy2rmybydncg06xb3svjx-sbcl-stumpwm-swm-gaps-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-swm-gaps/" #P"/gnu/store/y3602x94y7myl25srcghgh1fw5ldaxfj-sbcl-stumpwm-stumptray-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-stumptray/" #P"/gnu/store/7mm2idvr4jrbwidp79phkg6nm9wx5ly4-sbcl-clx-xembed-0.1-1.a5c4b84/share/common-lisp/sbcl/clx-xembed/") STUMPWM> (add-to-load-path "~/.guix-profile/share/common-lisp/sbcl/clx-xembed") (add-to-load-path "~/.guix-profile/share/common-lisp/sbcl/stumpwm-stumptray") (#P"/gnu/store/yi6dypcphhjbz9gp0l8z8r3pwjr71q6v-sbcl-clx-0.7.5/share/common-lisp/sbcl/clx/" #P"/gnu/store/lskhxlvddjjkg3c4v49ppm9y022ajzi6-sbcl-stumpwm-pass-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-pass/" #P"/gnu/store/ia41a2gd9klix8xdmhgsr16j998q7xfx-sbcl-stumpwm-ttf-fonts-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-ttf-fonts/" #P"/gnu/store/dvs3f9jrw4g3p3g17v34a71l3ccjas8b-sbcl-stumpwm-net-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-net/" #P"/gnu/store/4fjpm820912fy2rmybydncg06xb3svjx-sbcl-stumpwm-swm-gaps-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-swm-gaps/" #P"/gnu/store/y3602x94y7myl25srcghgh1fw5ldaxfj-sbcl-stumpwm-stumptray-0.0.1-2.920f8fc/share/common-lisp/sbcl/stumpwm-stumptray/" #P"/gnu/store/7mm2idvr4jrbwidp79phkg6nm9wx5ly4-sbcl-clx-xembed-0.1-1.a5c4b84/share/common-lisp/sbcl/clx-xembed/") STUMPWM> (load-module "stumptray") The Backtrace: Error opening #P"/gnu/store/9f98ids4s8hg306chy36d3bhk7zmyg9a-stumpwm-20.11-lib/lib/common-lisp/sbcl/stumpwm/package-tmp5GEXGEG5.fasl": Read-only file system [Condition of type SB-INT:SIMPLE-FILE-ERROR] Restarts: 0: [CONTINUE] Retry opening. 1: [USE-VALUE] Try opening a different file. 2: [RETRY] Retry compiling #. 3: [ACCEPT] Continue, treating compiling # as having been successful. 4: [RETRY] Retry ASDF operation. 5: [CLEAR-CONFIGURATION-AND-RETRY] Retry ASDF operation after resetting the configuration. 6: [RETRY] Retry ASDF operation. 7: [CLEAR-CONFIGURATION-AND-RETRY] Retry ASDF operation after resetting the configuration. 8: [RETRY] Retry SLY mREPL evaluation request. 9: [*ABORT] Return to SLY's top level. 10: [ABORT] abort thread (#) Backtrace: 0: (SB-IMPL::FILE-PERROR #P"/gnu/store/9f98ids4s8hg306chy36d3bhk7zmyg9a-stumpwm-20.11-lib/lib/common-lisp/sbcl/stumpwm/package-tmp5GEXGEG5.fasl" 30 "Error opening ~S" #P"/gnu/store/9f98ids4s8hg306chy36d3.. 1:
Installed from guix on fedora 33, nyxt display blank buffer instead of opening urls
Hello, I already opened an issue with the Nyxt project and they think the problem is related to a conflict between guix and fedora. The issue in the Nyxt project is: https://github.com/atlas-engineer/nyxt/issues/1132 As the initial post and discussion is already rather long it seems better, to me to reference the issue instead of posting my description again. Regards, Roland Everaert --- Use the F.O.S.S., Luke Sent with [ProtonMail](https://protonmail.com/) Secure Email.
Re: error: "cannot download for method git-fetch"
On Sun, 21 Feb 2021 14:06:20 -0500 Leo Famulari wrote: > On Sun, Feb 21, 2021 at 05:54:34PM +0100, Wiktor Żelazny wrote: > > On Sun, Feb 21, 2021 at 01:27:58PM +0100, Hartmut Goebel wrote: > > > > > Updating some other package, which is using http-fetch, works. So > > > this seams to be related to git. > > > > Hi, > > > > Perhaps you’re behind a firewall that blocks connections that use > > the git protocol? > > Perhaps, but most uses of Git sources in our packages fetch over HTTP > or HTTPS, so ports 80 and 443. It's unusual for these to be blocked. > The actual git:// protocol, which is rarely used, uses port 9418. > I'm not behind a firewall and have been getting the same (or at least similar?) error. (So far I haven't saved any logs from it.) Just a guess, but maybe it's caused by some packages using string-append "v" and the updated can't figure out how to reverse that??? Purely just guessing here.
Re: glib-networking not found by Gnome ?
Le lundi 22 février 2021 à 15:59 +0100, Pierre Neidhardt a écrit :
> What's in the etc/profile of your profile?
>
> It should have something like
>
> --8<---cut here---start->8---
> export GIO_EXTRA_MODULES=\
> "${GUIX_PROFILE:-/gnu/store/2amj32ic81b9inf32a88y8gg5y7019wx-
> profile}/lib/gio/modules${GIO_EXTRA_MODULES:+:}$GIO_EXTRA_MODULES"
> --8<---cut here---end--->8---
Looks like it's here :
$ cat ~/.guix-profile/etc/profile | grep GIO
export GIO_EXTRA_MODULES="${GUIX_PROFILE:-
/gnu/store/z2d3rjsa97y2wh9amzhr577yi6snvv9d-
profile}/lib/gio/modules${GIO_EXTRA_MODULES:+:}$GIO_EXTRA_MODULES"
> Maybe try to install gsettings-desktop-schemas, see if that helps.
Still not OK, even after a reboot.
Re: glib-networking not found by Gnome ?
Hello Pierre, Thank you for your help. Despite the several package conflicts I had (meld, polari, gnucash) when installing glib, it didn't fix the issue. I still have the pop up with the error message. Do I have to install glib as root maybe ? Jérémy Le 22 février 2021 08:24:42 GMT+01:00, Pierre Neidhardt a écrit : >Hi Jérémy, > >You need to install glib as well :) >It should add a GIO_EXTRA_MODULES entry in the etc/profile. > >Cheers! > >-- >Pierre Neidhardt >https://ambrevar.xyz/ -- Envoyé de mon appareil Android avec Courriel K-9 Mail. Veuillez excuser ma brièveté.
Re: Foreign distribution: errors when loading contrib modules of stumpwm
Roland Everaert via skribis: > I have installed StumpWM and some contrib modules, on a Fedora 33 > distribution using the guix PM, for many month, without much problem. Since > the begining of the week (last guix update), whenever I load a contrib > module, StumpWM complains about a FASL file being read-only somewhere in > /gnu/store/-stumpwm-lib. > > Any idea what has changed in stumpwm and its contrib modules? > > Below is my StumpWM configuration: > > Start of sources - > ;; -*-lisp-*- > ;; > (in-package :stumpwm) > > ;; * Adding modules > (add-to-load-path "~/.guix-profile/share/common-lisp/sbcl/clx-xembed") > (add-to-load-path "~/.guix-profile/share/common-lisp/sbcl/stumpwm-stumptray") > > (load-module "stumptray") > End of sources - This is strange, as the last change directly related to StumpWM was made several months ago (around 2020-12-10). How do you start StumpWM? Do you use a '.xsession' file? What are your XDG_CONFIG_DIRS and XDG_DATA_DIRS environment variables set to? signature.asc Description: PGP signature
