| View previous topic :: View next topic |
| Author |
Message |
miroR
l33t
Joined: 05 Mar 2008
Posts: 826
|
 Posted: Thu Dec 27, 2012 11:51 am Post subject: |
 |
|
| miroR wrote: | It just might not boot. All is different now. One detail. On a reboot about a week and a day ago, I think, there was a message to the effect of:
"No /sbin/udevd" and no devices were created...
|
That was because udev-init-scripts wasn't installed, I remember (I was in a rush, for lack of time).
I just now backed up my Gentoo, which is perfectly booting, as 99% of the time.
EDIT start
Confirmed all working fine by cloning the backed up system, as per somewhere in this thread
https://forums.gentoo.org/viewtopic-t-940916.html
or pointers therefrom I explained.
I now have to go back to:
https://forums.gentoo.org/viewtopic-p-7207178.html
because I promised, and mentioned it in a post in this thread a few days ago.
Only one more remark.
It doesn't matter not having succeeded yet. Devs who tried to fix things with this fork are appreciated for having tried the rught thing! Thanks!
EDIT end |
|
| Back to top |
|
 |
grey_dot
Tux's lil' helper
Joined: 15 Jul 2012
Posts: 142
|
| Posted: Thu Dec 27, 2012 1:38 pm Post subject: |
|
|
| cach0rr0 wrote: | | grey_dot wrote: |
Wait, what? A web browser, which is by definition pure user application, requires an exact implementation of one of the core system components? Huh...
|
try hacking the ebuild to remove the udev dependency. observe smoke and fire if udev is not installed (presumably eudev or your fork would fulfill the requirements) :)
|
I think I just believe you. Sucks anyway. |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6214
Location: Dallas area
|
| Posted: Fri Dec 28, 2012 10:40 am Post subject: |
|
|
| cach0rr0 wrote: | | grey_dot wrote: |
Wait, what? A web browser, which is by definition pure user application, requires an exact implementation of one of the core system components? Huh...
|
try hacking the ebuild to remove the udev dependency. observe smoke and fire if udev is not installed (presumably eudev or your fork would fulfill the requirements)  |
Looking at the chromium ebuild it technically requires virtual/udev (under RDEPEND)
which should be able to be built in a local directory and
changed by adding mdev or whatever other replacement is used.
From virtual/udev
| Code: | RDEPEND="|| ( ~sys-fs/udev-171[gudev?,hwdb?,introspection?,keymap?,selinux?]
~sys-fs/eudev-0[gudev?,hwdb?,introspection?,keymap?,selinux?] )" |
Note: I haven't chased it down completely but it might even be possible to put in static-dev as a RDEPEND of virtual/udev.
But thanks for the heads up, I don't have either kde or gnome dependencies on my system
and simply find a replacement if something is being adamant about trying to force me
to add such silliness when I don't want it
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
cach0rr0
Bodhisattva
Joined: 13 Nov 2008
Posts: 4123
Location: Houston, Republic of Texas
|
| Posted: Sat Dec 29, 2012 10:58 pm Post subject: |
|
|
| Anon-E-moose wrote: |
Looking at the chromium ebuild it technically requires virtual/udev (under RDEPEND)
which should be able to be built in a local directory and
changed by adding mdev or whatever other replacement is used.
From virtual/udev
| Code: | RDEPEND="|| ( ~sys-fs/udev-171[gudev?,hwdb?,introspection?,keymap?,selinux?]
~sys-fs/eudev-0[gudev?,hwdb?,introspection?,keymap?,selinux?] )" |
Note: I haven't chased it down completely but it might even be possible to put in static-dev as a RDEPEND of virtual/udev.
But thanks for the heads up, I don't have either kde or gnome dependencies on my system
and simply find a replacement if something is being adamant about trying to force me
to add such silliness when I don't want it |
not a runtime failure
it's a build failure, as the chromium build expects libudev
do this:
-unmerge udev (completely - presumably, any package, including this fork and eudev, that provide libudev, will get you past this error)
-hack the chromium ebuild to remove the RDEPEND for virtual/udev
-digest the ebuild
-emerge
-observe the following:
| Code: |
>>> Configuring source in /var/tmp/portage/www-client/chromium-24.0.1312.35/work/chromium-24.0.1312.35 ...
build/gyp_chromium --depth=. -Ddisable_sse2=1 -Dlinux_use_tcmalloc=0 -Ddisable_nacl=1 -Dflapper_version_h_file=/var/tmp/portage/www-client/chromium-24.0.1312.35/temp/flapper_version.h -Duse_system_bzip2=1 -Duse_system_flac=1 -Duse_system_icu=1 -Duse_system_libevent=1 -Duse_system_libjpeg=1 -Duse_system_libpng=1 -Duse_system_libusb=1 -Duse_system_libvpx=1 -Duse_system_libwebp=1 -Duse_system_libxml=1 -Duse_system_minizip=1 -Duse_system_opus=1 -Duse_system_speex=1 -Duse_system_v8=1 -Duse_system_xdg_utils=1 -Duse_system_yasm=1 -Duse_system_zlib=1 -Duse_cups=1 -Duse_gconf=0 -Duse_gnome_keyring=0 -Dlinux_link_gnome_keyring=0 -Duse_kerberos=0 -Duse_pulseaudio=0 -Dselinux=0 -Dlinux_link_gsettings=1 -Dlinux_sandbox_path=/usr/lib64/chromium-browser/chrome_sandbox -Dlinux_sandbox_chrome_path=/usr/lib64/chromium-browser/chrome -Dlinux_use_gold_binary=0 -Dlinux_use_gold_flags=0 -Dproprietary_codecs=1 -Dffmpeg_branding=Chrome -Dtarget_arch=x64 -Dwerror=
Updating projects from gyp files...
Package libudev was not found in the pkg-config search path.
Perhaps you should add the directory containing `libudev.pc'
to the PKG_CONFIG_PATH environment variable
No package 'libudev' found
gyp: Call to 'pkg-config --cflags libudev' returned exit status 1. while loading dependencies of /var/tmp/portage/www-client/chromium-24.0.1312.35/work/chromium-24.0.1312.35/base/base.gyp while loading dependencies of /var/tmp/portage/www-client/chromium-24.0.1312.35/work/chromium-24.0.1312.35/build/all.gyp while trying to load /var/tmp/portage/www-client/chromium-24.0.1312.35/work/chromium-24.0.1312.35/build/all.gyp
* ERROR: www-client/chromium-24.0.1312.35 failed (configure phase):
* (no error message)
*
* Call stack:
* ebuild.sh, line 89: Called src_configure
* environment, line 6483: Called die
* The specific snippet of code:
* egyp_chromium ${myconf} || die
*
* If you need support, post the output of `emerge --info '=www-client/chromium-24.0.1312.35'`,
* the complete build log and the output of `emerge -pqv '=www-client/chromium-24.0.1312.35'`.
* The complete build log is located at '/var/tmp/portage/www-client/chromium-24.0.1312.35/temp/build.log'.
* The ebuild environment file is located at '/var/tmp/portage/www-client/chromium-24.0.1312.35/temp/environment'.
* Working directory: '/var/tmp/portage/www-client/chromium-24.0.1312.35/work/chromium-24.0.1312.35'
* S: '/var/tmp/portage/www-client/chromium-24.0.1312.35/work/chromium-24.0.1312.35'
>>> Failed to emerge www-client/chromium-24.0.1312.35, Log file:
|
you can, of course, merge udev just for the chromium build, and then unmerge it after, since it doesnt cause any issues at runtime
but that's a royal pain in the ass.
I found an issue in the chromium bug tracker (not the gentoo one, as in, the one from google) where, long story short, there is zero interest in ever removing this as a build dependency. So, you will not have an easy time ever unudev'ing your system if you are a chromium user.
_________________
Lost configuring your system?
dump lspci -n here | see Pappy's guide | Link Stash |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6214
Location: Dallas area
|
| Posted: Sun Dec 30, 2012 10:42 am Post subject: |
|
|
Interesting.
1st off I don't use chromium, and with the (to me) insane need for any part of udev available to compile I probably never will.
It seems to be looking for "libudev.pc" which makes me wonder if a dummy udev package would work.
I have something similar for sys-kernel/sources, as I roll my own kernels.
I may look into it simply out of curiousity.
Note: I'm not trying to be argumentative, but if this is cropping up in chromium
it may show up in other apps and a work around to installing *udev* would be nice.
Edit to add: The latest chromium also appears to need "dbus" to compile.
It does appear to link to libudev.so, so not sure if it would run without udev.
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
cach0rr0
Bodhisattva
Joined: 13 Nov 2008
Posts: 4123
Location: Houston, Republic of Texas
|
| Posted: Mon Dec 31, 2012 2:38 am Post subject: |
|
|
| Anon-E-moose wrote: |
Note: I'm not trying to be argumentative, but if this is cropping up in chromium
it may show up in other apps and a work around to installing *udev* would be nice.
|
didnt take it as argument. methinks we're seeing eye to eye on this one.
if a f**king web browser is requiring udev, what sort of other crap is it going to pollute?
how long until you are having to apply crude hacks for every application you use?
i wouldnt be so worried had chromium upstream said "go kick rocks, we're not removing the udev dependency or making it configurable"
at this rate it won't be long before openssl depends on pulseaudio 
_________________
Lost configuring your system?
dump lspci -n here | see Pappy's guide | Link Stash |
|
| Back to top |
|
|
PaulBredbury
Watchman
Joined: 14 Jul 2005
Posts: 7310
|
| Posted: Mon Dec 31, 2012 8:47 am Post subject: |
|
|
| cach0rr0 wrote: | | a f**king web browser is requiring udev |
Reason. |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6214
Location: Dallas area
|
| Posted: Mon Dec 31, 2012 10:39 am Post subject: |
|
|
| PaulBredbury wrote: | | cach0rr0 wrote: | | a f**king web browser is requiring udev |
Reason. |
From the link.
| Quote: | you need to think bigger. Chromium supports joystick inputs (which come and
go) for playing games in the browser, so udev makes sense. it also supports
looking up your location, detecting when the network comes up/down, and
kde/gnome password stores, all of which require dbus. and with nacl, you're
loading ELF objects, so libelf makes sense.
really though, if you don't like Chrome, then don't use it. many people do. |
Seriously, how many people use joysticks for running their browser 
Firefox supports looking up location without dbus.
I don't run either kde or gnome, so that is just bloat.
Etc.
I think they do it just because they can not because there is some overriding reason
and when questioned out come the ignorant excuses.
There should be flags to shut out udev, dbus, gconf and other crap that isn't necessary
when one isn't running gnome or kde.
The real problem is that chrome/chromium/google wants to be in the OS business and
they've tried for years (so did firefox but they stopped a while back).
They also think they have to right to all my personal data. They are wrong
really though, I don't like Chrome and I don't use it and never will. (response to mikes ridiculous close in the quote)
I don't give a flying flip what many people do.
This is linux after all, choice is king.
Edit to add: in red
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland
Last edited by Anon-E-moose on Mon Dec 31, 2012 10:46 am; edited 1 time in total |
|
| Back to top |
|
|
Hypnos
Advocate
Joined: 18 Jul 2002
Posts: 2889
Location: Omnipresent
|
| Posted: Mon Dec 31, 2012 10:41 am Post subject: |
|
|
| PaulBredbury wrote: | | cach0rr0 wrote: | | a f**king web browser is requiring udev |
Reason. |
I'm not sure that's a good reason -- are there not apps which use a joystick that *don't* depend on udev?
_________________
Personal overlay | Simple backup scheme |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6214
Location: Dallas area
|
| Posted: Mon Dec 31, 2012 10:50 am Post subject: |
|
|
| Hypnos wrote: | | PaulBredbury wrote: | | cach0rr0 wrote: | | a f**king web browser is requiring udev |
Reason. |
I'm not sure that's a good reason -- are there not apps which use a joystick that *don't* depend on udev? |
Seriously what's next, requiring systemd, a certain version of compiler, signed over rights to my first born?
The quote was an excuse for all the cr*p, not a reason.
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
PaulBredbury
Watchman
Joined: 14 Jul 2005
Posts: 7310
|
| Posted: Mon Dec 31, 2012 11:20 am Post subject: |
|
|
For God's sake, the majority of this thread is just moan, moan, moan, despite there being other threads (yes plural) for such moaning.
Write patches (Edit: added link) to make the objectionable stuff optional. It's their app, not yours, you have no rights. They don't even need to provide reasons or excuses, so enough with the bitchin' already 
Oh, and happy new year 
| Quote: | | are there not apps which use a joystick that *don't* depend on udev? |
Yeah. So if more effort were put into patch-writing than hot-air-expulsion, this thread would be much nicer.
Last edited by PaulBredbury on Mon Dec 31, 2012 9:54 pm; edited 1 time in total |
|
| Back to top |
|
|
cach0rr0
Bodhisattva
Joined: 13 Nov 2008
Posts: 4123
Location: Houston, Republic of Texas
|
| Posted: Mon Dec 31, 2012 5:45 pm Post subject: |
|
|
| PaulBredbury wrote: | | cach0rr0 wrote: | | a f**king web browser is requiring udev |
Reason. |
sure looks like a great 'reason' for udev support being *optional'
not a particularly great reason to make it a mandatory build dependency.
| PaulBredbury wrote: | For God's sake, the majority of this thread is just moan, moan, moan, despite there being other threads (yes plural) for such moaning.
Write patches to make the objectionable stuff optional. It's their app, not yours, you have no rights. They don't even need to provide reasons or excuses, so enough with the bitchin' already
Oh, and happy new year
| Quote: | | are there not apps which use a joystick that *don't* depend on udev? |
Yeah. So if more effort were put into patch-writing than hot-air-expulsion, this thread would be much nicer. |
Not sure if you've quite noticed, bit you're bitching about people bitching in a thread about a now-abandoned udev fork.
Of note: the next time you see Linus bitching about the udev team being "braindead" as he put it once, please do make sure to inform him he has no rights, it's their app not his, and to stop his whinging. Because bitching about someone else's work is a faux pas, especially in the linux world you see...
_________________
Lost configuring your system?
dump lspci -n here | see Pappy's guide | Link Stash |
|
| Back to top |
|
|
PaulBredbury
Watchman
Joined: 14 Jul 2005
Posts: 7310
|
| Posted: Mon Dec 31, 2012 6:08 pm Post subject: |
|
|
Well, if a moderator says this thread's about bitchin', then that's what it's about
There was silly me, hoping that this would be a thread to facilitate maintaining the code.
Linus in a unique situation, that's why people listen to him. I doubt Google are monitoring this thread with trepidation.
Thank you consus and grey_dot (and anyone else involved), for a fork which I thought was a fantastic idea, and am still using. I'll switch to Gentoo's fork when a problem appears. |
|
| Back to top |
|
|
cach0rr0
Bodhisattva
Joined: 13 Nov 2008
Posts: 4123
Location: Houston, Republic of Texas
|
| Posted: Tue Jan 01, 2013 2:48 am Post subject: |
|
|
| PaulBredbury wrote: | Well, if a moderator says this thread's about bitchin', then that's what it's about
There was silly me, hoping that this would be a thread to facilitate maintaining the code.
|
pffttt. I'm a human spam filter (and a poor one at that)
If this were still alive I would have bit my tongue and not gone off on a tangent.
But as of this post I'm under the impression the project is pretty much dead and abandoned (understandably so)
_________________
Lost configuring your system?
dump lspci -n here | see Pappy's guide | Link Stash |
|
| Back to top |
|
|
grey_dot
Tux's lil' helper
Joined: 15 Jul 2012
Posts: 142
|
| Posted: Tue Jan 08, 2013 8:36 pm Post subject: |
|
|
| PaulBredbury wrote: |
Write patches (Edit: added link) to make the objectionable stuff optional. It's their app, not yours, you have no rights. They don't even need to provide reasons or excuses, so enough with the bitchin' already :evil:
Oh, and happy new year :wink:
| Quote: | | are there not apps which use a joystick that *don't* depend on udev? |
Yeah. So if more effort were put into patch-writing than hot-air-expulsion, this thread would be much nicer. |
Patch this, patch that, patch everything again and again... You know, I'm getting really tired of _HAVING_TO_ write patches to many programs I use on a daily basis. Yeah, I know that those are open source softwares, and developers made them open so that anybody could participate in the development process. But there is a huge difference between participating voluntarily like when adding new features and making cool things, and participating because you have to do it or your system will drown in megatons of bloated crap. I'd really like to help developers, not to fight with them because they have their brains somewhere inside their asses, but most of the time things don't go along with my expectations. So I got mdev instead of udev now, openbsd on most of my servers instead of linux, and a real headache about using a bluetooth headset in linux without having to struggle with dbus/python/gobject/other-bloated-piece-of-crap.
But you're right, it's developers' right to produce unusable unmaintainable crap. I'll just try to avoid it. |
|
| Back to top |
|
|
dmpogo
Advocate
Joined: 02 Sep 2004
Posts: 3468
Location: Canada
|
| Posted: Tue Jan 15, 2013 6:43 am Post subject: |
|
|
| grey_dot wrote: |
Patch this, patch that, patch everything again and again... You know, I'm getting really tired of _HAVING_TO_ write patches to many programs I use on a daily basis. Yeah, I know that those are open source softwares, and developers made them open so that anybody could participate in the development process. |
Part of it, is because we are here not treating our computers utilitarinly, but play with them too much continously updating them.
On my machines I would have been better off if I left then in as of a year ago state. After this spring-summer updates thay are all running worse or a bit flaky.
Just had to use my CPU time on Compute Canada high performance facilities - has noted that vast majority of the machines it links from coast to coast still use 2.6.18 kernel. And the rest - 2.6.32. |
|
| Back to top |
|
|
Dominique_71
Veteran
Joined: 17 Aug 2005
Posts: 1918
Location: Switzerland (Romandie)
|
| Posted: Sun Jun 16, 2013 1:41 pm Post subject: |
|
|
I have a simple question. In a system with udev and openrc, with an initrd that just load a boot splash theme, to install eudev, is it as simple than to mask sys-apps/systemd and sys-fs/udev, uninstall udev and update world (which just want to install eudev) ?
_________________
"Confirm You are a robot." - the singularity |
|
| Back to top |
|
|
lost+found
Guru
Joined: 15 Nov 2004
Posts: 509
Location: North~Sea~Coa~s~~t~~~
|
| Posted: Sun Jun 16, 2013 7:24 pm Post subject: |
|
|
| Dominique_71 wrote: | | I have a simple question. In a system with udev and openrc, with an initrd that just load a boot splash theme, to install eudev, is it as simple than to mask sys-apps/systemd and sys-fs/udev, uninstall udev and update world (which just want to install eudev) ? |
I only did "emerge --oneshot sys-fs/eudev" and sys-fs/udev was unmerged automatically. I didn't mask udev/systemd either, and no changes to bootsplash/initrd. It just worked with eudev. I didn't notice anything unusual, eudev works fine on my desktop (it's running *kitless KDE+pmount+bluetooth...etc.).
| Code: | 1364734201: Started emerge on: Mar 31, 2013 14:50:01
1364734201: *** emerge --oneshot sys-fs/eudev
1364734215: >>> emerge (1 of 1) sys-fs/eudev-1_beta2-r2 to /
1364734215: === (1 of 1) Cleaning (sys-fs/eudev-1_beta2-r2::/usr/portage/sys-fs/eudev/eudev-1_beta2-r2.ebuild)
1364734215: === (1 of 1) Compiling/Merging (sys-fs/eudev-1_beta2-r2::/usr/portage/sys-fs/eudev/eudev-1_beta2-r2.ebuild)
1364734265: === (1 of 1) Merging (sys-fs/eudev-1_beta2-r2::/usr/portage/sys-fs/eudev/eudev-1_beta2-r2.ebuild)
1364734268: >>> AUTOCLEAN: sys-fs/eudev:0
1364734271: === (1 of 1) Post-Build Cleaning (sys-fs/eudev-1_beta2-r2::/usr/portage/sys-fs/eudev/eudev-1_beta2-r2.ebuild)
1364734271: ::: completed emerge (1 of 1) sys-fs/eudev-1_beta2-r2 to /
1364734272: === Unmerging... (sys-fs/udev-197-r8)
1364734274: >>> unmerge success: sys-fs/udev-197-r8
1364734274: *** Finished. Cleaning up...
1364734277: *** exiting successfully.
1364734287: *** terminating. |
Elog output gives important steps to add udev-postmount to the default runlevel, and to restart udev!!!
Good luck! |
|
| Back to top |
|
|
Dominique_71
Veteran
Joined: 17 Aug 2005
Posts: 1918
Location: Switzerland (Romandie)
|
| Posted: Sat Jun 22, 2013 5:17 pm Post subject: |
|
|
Just to say it worked fine. 
_________________
"Confirm You are a robot." - the singularity |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Kernel & Hardware
