Package Details: htop-git 1:

Git Clone URL: (read-only, click to copy)
Package Base: htop-git
Description: Interactive text-mode process viewer
Upstream URL:
Licenses: GPL
Conflicts: htop
Provides: htop
Submitter: yar
Maintainer: yar (zhimsel)
Last Packager: yar
Votes: 34
Popularity: 0.55
First Submitted: 2014-02-13 21:17 (UTC)
Last Updated: 2021-01-13 23:32 (UTC)

Dependencies (6)

Required by (5)

Sources (1)

Latest Comments

OlegKrikun commented on 2021-09-24 18:14 (UTC) (edited on 2021-09-24 18:18 (UTC) by OlegKrikun)

Have error today:

htop-git/PKGBUILD: line 31: ./ No such file or directory

UPD: But looks like problem is

warning: remote HEAD refers to nonexistent ref, unable to checkout.

haawda commented on 2020-12-28 23:45 (UTC)

This is no longer a confifure-option.

configure: WARNING: unrecognized options: --enable-cgroup

yar commented on 2020-05-23 22:09 (UTC)

@Hekuran Upstream doesn't support gcc10. The pull request to fix this hasn't been merged, even after 4 months:

I just set this PKGBUILD to fetch from the PR's fork for now. If hishamhm (or someone else) ever resumes activity we'll have to change it again.

Hekuran commented on 2020-05-21 08:57 (UTC)

Im unable to compile, either htop from abs, htop-git, nor htop-vim-git. Here are my logs:

htop ABS:


htop VIM GIT:

linkmauve commented on 2020-01-18 12:11 (UTC)

Hi, you can replace python2 with python in makedepends, just like in extra:

yar commented on 2018-02-06 06:48 (UTC)

@HalJordan have you tried building with devtools?

HalJordan commented on 2018-02-06 05:59 (UTC)

It seems to require a module that I can't find, either in the official repo, aur, or in cpan. I've already tried reinstalling Perl.

Can't locate IO/ /usr/lib/perl5/5.26/site_perl/IO/ Permission denied at /usr/share/autoconf/Autom4te/ line 41. BEGIN failed--compilation aborted at /usr/share/autoconf/Autom4te/ line 41. Compilation failed in require at /usr/bin/autoreconf line 42. BEGIN failed--compilation aborted at /usr/bin/autoreconf line 42.

yar commented on 2015-04-11 17:15 (UTC)

FYI the latest builds are failing. The last good build I could find is e3d1ca, so do this for now: source=("git+${url}.git#commit=e3d1ca0d2074ec89155f9412c6caa621265a8824")

Barthalion commented on 2015-04-11 17:02 (UTC)

1. If bumping the epoch isn't worth "it", I wonder how you manage to change $pkgver at all. 2. Probably nobody cares about epochs in AUR, especially for VCS packages.

yar commented on 2014-10-29 22:20 (UTC)

@eworm, I'm afraid that would require bumping the epoch. Not sure that's worth it.

eworm commented on 2014-10-27 13:05 (UTC)

Any chance to get the last tag into the version string? Something like git describe --tags | sed 's/-/./g' would result in "".

yar commented on 2014-05-04 17:20 (UTC)

@Brottweiler Thanks for reporting that! Looks like an upstream bug. I wrote a patch to fix this PKGBUILD, and sent them a pull request:

Brottweiler commented on 2014-05-04 14:08 (UTC)

Fails to build: Makefile:890: recipe for target 'htop-ProcessList.o' failed make[1]: *** [htop-ProcessList.o] Error 1 make[1]: Leaving directory '/home/brott/build/htop-git/src/htop' Makefile:402: recipe for target 'all' failed make: *** [all] Error 2 ==> ERROR: A failure occurred in build(). Aborting...

wlhlm commented on 2014-04-24 19:27 (UTC)

The patch can be dropped as it has been fixed:

commented on 2013-05-31 09:53 (UTC)

arch=('i686' 'x86_64' 'armv7h') Compile ok in ^_^

dot commented on 2012-06-17 16:49 (UTC)

Please add Python to the makedepends, otherwise make fails with the following message: ... scripts/ Affinity.c make: scripts/ Command not found make: *** [Affinity.h] Error 127

misc commented on 2012-04-28 16:28 (UTC)

If you want "Memory[Bar]" to display the value of used memory as (used+buffers+cache) instead of merely (used), apply this minimal patch that I've written (against r288): Use the tarball and add below line 33 ("cd $srcdir/htop/") in the PKGBUILD eg "patch -Np0 -i ${srcdir}/<name you gave it>". Its checksum needs to be added, too. The patch simply sets the value before (buffers+cache) gets subtracted, ie no 'actual' code change. I've submitted a feature request to make this a proper option: