kisslinux / repo Goto Github PK
View Code? Open in Web Editor NEWKISS Linux - Official Repositories
Home Page: https://kisslinux.github.io
License: MIT License
KISS Linux - Official Repositories
Home Page: https://kisslinux.github.io
License: MIT License
If anyone has any recommendations on what to enable/disable, please reply here.
libelf
is required to build the kernel and it is tiny enough to include in the chroot
without causing a noticeable size increase.
โ ./kiss-size libelf
4.0K /var/db/kiss/installed/libelf/version
4.0K /var/db/kiss/installed/libelf/sources
8.0K /var/db/kiss/installed/libelf/patches/preprocessor-warnings.patch
4.0K /var/db/kiss/installed/libelf/manifest
4.0K /var/db/kiss/installed/libelf/checksums
4.0K /var/db/kiss/installed/libelf/build
4.0K /usr/lib/pkgconfig/libelf.pc
92.0K /usr/lib/libelf.so.0.8.13
0 /usr/lib/libelf.so.0
0 /usr/lib/libelf.so
176.0K /usr/lib/libelf.a
4.0K /usr/include/nlist.h
4.0K /usr/include/libelf/sys_elf.h
4.0K /usr/include/libelf/nlist.h
12.0K /usr/include/libelf/libelf.h
8.0K /usr/include/libelf/gelf.h
28.0K /usr/include/libelf/elf_repl.h
4.0K /usr/include/libelf.h
4.0K /usr/include/gelf.h
368.0K total
The llvm build script executes make -j "$(nproc || echo 2)"
which sets the number of jobs to the number of cores (or 2). This ignores and overrides any values present in MAKEFLAGS.
Solutions:
Packages (so far):
Why can you uninstall busybox, kiss and / or musl? That makes no sense. There should be packages that can not be uninstalled.
=> Mounting /dev, /proc and /sys from host.
=> Copying /etc/resolv.conf from host.
=> You may need to run 'update-ca-certificates' for SSL.
=> You may need to run 'kiss update'.
=> Entering chroot.
/ # kiss remove musl
=> Removing packages...
/usr/bin/kiss: line 624: rm: not found
=> [musl]: Failed to remove '/usr/lib/libc.a'.
/usr/bin/kiss: line 624: rm: not found
=> [musl]: Failed to remove '/usr/lib/ld-musl-x86_64.so.1'.
/usr/bin/kiss: line 624: rm: not found
=> [musl]: Failed to remove '/usr/lib/crtn.o'.
/usr/bin/kiss: line 624: rm: not found
=> [musl]: Failed to remove '/usr/lib/crti.o'.
/usr/bin/kiss: line 624: rm: not found
=> [musl]: Failed to remove '/usr/lib/crt1.o'.
/usr/bin/kiss: line 624: rm: not found
=> [musl]: Failed to remove '/usr/lib/Scrt1.o'.
/usr/bin/kiss: line 624: rm: not found
=> [musl]: Failed to remove '/usr/include/wordexp.h'.
getting this:
/var/db/kiss/repo/extra/cmake/build: line 10: can't open /var/db/kiss/repo/extra/cmake/build/version: no such file
!> [cmake]: Build failed.
I'm guessing it should be /var/db/kiss/repo/extra/cmake/version
instead of /var/db/kiss/repo/extra/cmake/build/version
below is the relevant line in the build script, but I don't know bash enough to know what I am looking at.
IFS=. read -r ver1 ver2 _ < "${0%*/}/version"
It seems like sdhcp would be better aligned with the philosophy of KISS Linux so im posting this to see why dhcpcd was choosen.
grub
to 2.02
.
busybox.conf
to fix SUID binaries.info
files.zip
man directory.info
files.polkit
files.gettext
files.man
is broken(?).fakeroot
doesn't actually fake root(?).I'm attempting to install xorg-server
and python-setuptools
fails installation with the above error. Is there a way to determine what the conflict is and how to resolve it?
In the spirit of keeping things simple, I'd like to install some minimal bootloaders instead of grub which has way too many features compared to what I really need.
If these packages are out of the scope of the project, please feel free to close the issue.
I will attempt to build systemd-boot or syslinux once I manage to compile the kernel. If everything goes as expected I'll open a PR.
EDIT: Removed efilinux as it seems just an overhead since both syslinux and systemd-boot handle UEFI booting already.
While the package count in the main repository is small, I want to explore the idea of splitting the repository into categories. This is as simple as adding an additional sub-directory in repo/
for each category.
The user would then be able to only add the categories they need to KISS_PATH
. The code is already in place for something like this. Categories could be split into shells
, wm
, coreutils
, etc. This would make discoverability of alternative packages easy.
An issue arises when the user wants to enable all categories hence this issue. I'd like to discuss this and see what others think.
Xorg-Server-1.20.5 (must be built with glamor enabled)
Source: http://www.linuxfromscratch.org/blfs/view/svn/x/x7driver.html#xorg-amdgpu-driver
For grub
to find the kernel it needs to be named vmlinuz-SOMETHING
. We need to fix the linux
package to install it as vmlinuz-VERSION
.
=> [ca-certificates] Running post-install script.
Clearing symlinks in /etc/ssl/certs...
done.
Updating certificates in /etc/ssl/certs...
W: /usr/share/ca-certificates/mozilla/AC_Raรญz_Certicรกmara_S.A..crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/Certinomis_-_Root_CA.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/Certplus_Class_2_Primary_CA.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/Certplus_Root_CA_G1.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/Certplus_Root_CA_G2.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/Deutsche_Telekom_Root_CA_2.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/OpenTrust_Root_CA_G1.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/OpenTrust_Root_CA_G2.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/OpenTrust_Root_CA_G3.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/Swisscom_Root_CA_2.crt not found, but listed in /etc/ca-certificates.conf.
W: /usr/share/ca-certificates/mozilla/UTN_USERFirst_Email_Root_CA.crt not found, but listed in /etc/ca-certificates.conf.
137 added, 0 removed; done.
=> [ca-certificates] Installed successfully.
fd
is a sane alternative to find
. It tries to provide good defaults for 80% of cases. It also has some "smart" features in that it ignores dot folders and adheres to .gitignore . It's also very fast. It makes a good development tool.
Very simple in our case:
Line 15 in 23640d4
Is this intentional? Am I reading the script wrong? Is there a good reason to do this?
Easy once again.
building glib fails. The below is what looks to be the relevant portion of the build output. Let me know if you need more info.
rm: '/var/cache/kiss/pkg-6459/glib/usr/share/gettext' is a directory
!> [glib]: Build failed.
diff --git a/.travis.yml b/.travis.yml
old mode 100644
new mode 100755
diff --git a/LICENSE.md b/LICENSE.md
old mode 100644
new mode 100755
diff --git a/README.md b/README.md
old mode 100644
new mode 100755
diff --git a/core/autoconf/checksums b/core/autoconf/checksums
old mode 100644
new mode 100755
diff --git a/core/autoconf/depends b/core/autoconf/depends
old mode 100644
new mode 100755
diff --git a/core/autoconf/sources b/core/autoconf/sources
old mode 100644
new mode 100755
diff --git a/core/autoconf/version b/core/autoconf/version
old mode 100644
new mode 100755
diff --git a/core/automake/checksums b/core/automake/checksums
old mode 100644
new mode 100755
diff --git a/core/automake/depends b/core/automake/depends
old mode 100644
new mode 100755
diff --git a/core/automake/sources b/core/automake/sources
and so on
Current list (Tue Aug 13 18:42:47 GMT 2019)
Delayed packages.
rust
.Public (Time given to maintainers)
Packages with unknown upstream versions.
cairo
libpng
dependency if possible.aria2
chroot:
bootable system:
Not sure if this is the right place to bring up this issue. I am getting errors trying to build python-setuptools. I am in a kiss-chroot.
/ # kiss b python-setuptools
=> Resolving dependencies...
=> Building: python-setuptools.
=> Checking to see if any dependencies have already been built...
=> Installing any pre-built dependencies...
=> [python-setuptools]: Checking repository files...
=> [python-setuptools]: Downloading sources...
=> [python-setuptools]: Found cached source 'v41.1.0.tar.gz'.
=> [python-setuptools]: Extracting sources...
adding minimal entry_points
Regenerating egg_info
Traceback (most recent call last):
File "setup.py", line 9, in <module>
import setuptools
File "/var/cache/kiss/build-11734/python-setuptools/setuptools/__init__.py", line 20, in <module>
from setuptools.dist import Distribution, Feature
File "/var/cache/kiss/build-11734/python-setuptools/setuptools/dist.py", line 35, in <module>
from setuptools import windows_support
File "/var/cache/kiss/build-11734/python-setuptools/setuptools/windows_support.py", line 2, in <module>
import ctypes
File "/usr/lib/python3.7/ctypes/__init__.py", line 7, in <module>
from _ctypes import Union, Structure, Array
ModuleNotFoundError: No module named '_ctypes'
Traceback (most recent call last):
File "bootstrap.py", line 64, in <module>
__name__ == '__main__' and main()
File "bootstrap.py", line 61, in main
run_egg_info()
File "bootstrap.py", line 54, in run_egg_info
subprocess.check_call(cmd)
File "/usr/lib/python3.7/subprocess.py", line 347, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', 'setup.py', 'egg_info']' returned non-zero exit status 1.
!> [python-setuptools]: Build failed.
Lots of Linux users love games; I think it is inevitable then that we will include game packages.
My question: do we put these packages in public, or do we create a games repository?
If you guys decide to create a games repo, then I believe it would be best to create the directory for the first games-related PR.
Hi,
I tried to rebuild GTK+2 (after the update of mesa), but I get the following error:
Couldn't find include 'Pango-1.0.gir' (search path: '['../gdk', '/usr/share', 'gir-1.0', '/usr/share/gir-1.0', '/usr/share/gir-1.0']')
make[4]: *** [/usr/share/gobject-introspection-1.0/Makefile.introspection:156: Gdk-2.0.gir] Error 1
make[4]: Leaving directory '/home/macquat/.cache/kiss/build-16468/gtk+2/gdk'
make[3]: *** [Makefile:1381: all-recursive] Error 1
make[3]: Leaving directory '/home/macquat/.cache/kiss/build-16468/gtk+2/gdk'
make[2]: *** [Makefile:1161: all] Error 2
make[2]: Leaving directory '/home/macquat/.cache/kiss/build-16468/gtk+2/gdk'
make[1]: *** [Makefile:733: all-recursive] Error 1
make[1]: Leaving directory '/home/macquat/.cache/kiss/build-16468/gtk+2'
make: *** [Makefile:625: all] Error 2
!> gtk+2 Build failed
The problem is the following compilation option in the build of pango:
-Dgir=false \
Do you also have this problem ?
ripgrep
is a rust written alternative to grep
. It would supplement the capabilities of normal grep for a few cases. I specifically use it a lot for when browsing around a code repo.
Wayland is a much, much simpler design. The idea that X11 is simpler or better than Wayland is not based in fact, but cargo culting.
llvm 9.0.0
rust 1.38.0
.clang 9.0.0
llvm 9.0.0
.mesa 19.2.0
firefox 70.0
.Checksum mismatch error when trying to build python.
version file has 3.7.4
checksums file has 3.6.8
No one knows for sure how big the x86-64 instruction set is. It takes hundreds of instructions and esoteric hacks to boot up to long mode. EFI exists. Supporting cross-compiling is important to ensure the simpler architectures of the future are accounted for; a port to RISC-V for example should be relatively easy.
mandoc.bsd.lv is down (DNS doesn't resolve). This might be a temporary issue, but as long as the site is down, it's impossible to build mandoc.
https
release tarballs exist.I really love this project and would like to get involved by contributing. I love shell scripting (especially POSIX sh) and use your pure bash bible regularly. Further I am package maintainer for some packages of voidlinux and packaged for archlinux in the past. So I am not a beginner I guess.
Is there any way to get involved?
Seems like initramfs required to boot from encrypted /boot
https://cryptsetup-team.pages.debian.net/cryptsetup/encrypted-boot.html
https://wiki.gentoo.org/wiki/Custom_Initramfs
https://wiki.voidlinux.org/Full_Disk_Encryption_w/Encrypted_Boot
Seems like grub-mkimage can build initramfs archives https://manpages.debian.org/buster/grub-common/grub-mkimage.1.en.html
Using BLFS as reference, the strict dependencies are:
Of course, this does not include makedepends
.
it seems that the included GCC compiler does not properly work. I tried compiling linux(skipping the bootstrapping step at first) and got this error:
After that, I tried bootstrapping the packages first, but it gave me an error right away when automake checks for a c99 compatible compiler(for which gcc returns false).
The last thing I tried was recompiling gcc itself via kiss build gcc
but I also had no luck there. It returned the same internal compiler error: Illegal instruction
. Do you guys run into the same problem or have any idea on how to fix this?
udhcpc
needs a script to function.
udhcpc
pkill
unsupported arguments on shutdown.cryptsetup
currently cannot be compiled in a standard KISS chroot.puke
to something "nicer".$PATH
.contrib
repository for smaller example scripts to interface with the repository.=> Mounting /dev, /proc and /sys from host.
=> Copying /etc/resolv.conf from host.
=> You may need to run 'update-ca-certificates' for SSL.
=> You may need to run 'kiss update'.
=> Entering chroot.
/ # update-ca-certificates --fresh
Clearing symlinks in /etc/ssl/certs...
done.
Updating certificates in /etc/ssl/certs...
openssl:Error: 'rehash' is an invalid command.
Standard commands
asn1parse ca certhash ciphers
crl crl2pkcs7 dgst dh
dhparam dsa dsaparam ec
ecparam enc errstr gendh
gendsa genpkey genrsa nseq
ocsp passwd pkcs12 pkcs7
pkcs8 pkey pkeyparam pkeyutl
prime rand req rsa
rsautl s_client s_server s_time
sess_id smime speed spkac
ts verify version x509
Message Digest commands (see the `dgst' command for more details)
gost-mac md4 md5 md_gost94
ripemd160 sha1 sha224 sha256
sha384 sha512 sm3 sm3WithRSAEncryption
streebog256 streebog512 whirlpool
Cipher commands (see the `enc' command for more details)
aes-128-cbc aes-128-ecb aes-192-cbc aes-192-ecb
aes-256-cbc aes-256-ecb base64 bf
bf-cbc bf-cfb bf-ecb bf-ofb
camellia-128-cbc camellia-128-ecb camellia-192-cbc camellia-192-ecb
camellia-256-cbc camellia-256-ecb cast cast-cbc
cast5-cbc cast5-cfb cast5-ecb cast5-ofb
chacha des des-cbc des-cfb
des-ecb des-ede des-ede-cbc des-ede-cfb
des-ede-ofb des-ede3 des-ede3-cbc des-ede3-cfb
des-ede3-ofb des-ofb des3 desx
idea idea-cbc idea-cfb idea-ecb
idea-ofb rc2 rc2-40-cbc rc2-64-cbc
rc2-cbc rc2-cfb rc2-ecb rc2-ofb
rc4 rc4-40 sm4 sm4-cbc
sm4-cfb sm4-ecb sm4-ofb
148 added, 0 removed; done.
vdev
is a minimal udev
replacement which also includes its own version of libudev
(for compatability with libinput
and other software which depends on udev
). Sadly development stopped a few years ago and it is now too out of date with libudev
to be of any use.
I propose we attempt to revive development through a direct fork of the project. From what I understand, the libudev
compatibility library is just udev
's with changes to work with any device manager available!
This isn't a set in stone idea. I'd like to first figure out just how much work is required and what exactly needs fixing. It's an interesting idea nonetheless.
NOTE: Splitting vdev
into libudev
/vdev
allows for libudev
to be used with mdev
, vdev
, smdev
etc etc while still allowing for a working libinput
etc.
More information:
Vdev comes with a reference configuration and set of helper programs that make it usable out-of-the-box, with no subsequent tweaking required. The Linux port is a drop-in replacement for udev: it generates an equivalent /dev tree, as well as an equivalent /run/udev tree. It also ships with a "libudev-compat" library that is ABI-compatible with libudev 219. Libudev-linked software such as the X.org X server can use vdev and libudev-compat without recompilation. In addition, the Linux port interoperates with but does not depend on the devtmpfs filesystem.
For example, if I installed nodejs, and PYTHON
was set to python3
before, will it then be set to python2
?
Line 3 in 23640d4
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.