Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Looking to move on from udev - static /dev vs mdevd?
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, 4  Next  
Reply to topic    Gentoo Forums Forum Index Unsupported Software
View previous topic :: View next topic  
Author Message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Wed Mar 01, 2023 1:29 pm    Post subject: Reply with quote

NeddySeagoon wrote:
stefan11111,

I just use alsa. Its not problem free yet. For good sound output, with no stuttering, I need to keep the load average over 40. :)
That's nothing to do with static dev though.
Until alsa works on its own, nothing else can as everything uses alsa at the bottom, well except OSS.

I have dbus installed but not started. VirtualBox wants it.
I should really give up VirtualBox but its easier than QEMU for a quick and dirty throw away Virtual Machine. For long term KVMs its all QEMU.

Have you disabled dinamic minors in the kernel config?
https://github.com/alsa-project/alsa-lib/issues/296
Also, how do you see the load average?
And how do I mark a post as solved on this forum? I want to mark the librewolf topic from this forum as solved.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
pietinger
Moderator
Moderator


Joined: 17 Oct 2006
Posts: 4160
Location: Bavaria

PostPosted: Wed Mar 01, 2023 2:06 pm    Post subject: Reply with quote

stefan11111 wrote:
And how do I mark a post as solved on this forum? I want to mark the librewolf topic from this forum as solved.

Just edit your first post in this thread and add a [Solved] into the title. Beware it must not be too long.
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Wed Mar 01, 2023 2:18 pm    Post subject: Reply with quote

pietinger wrote:
stefan11111 wrote:
And how do I mark a post as solved on this forum? I want to mark the librewolf topic from this forum as solved.

Just edit your first post in this thread and add a [Solved] into the title. Beware it must not be too long.

Thanks, just marked that post.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
colo-des
Tux's lil' helper
Tux's lil' helper


Joined: 20 May 2011
Posts: 97

PostPosted: Wed Mar 01, 2023 2:34 pm    Post subject: Reply with quote

stefan11111 wrote:
Has anyone gotten rid of this too?

net-print/cups-filters need gdbus-codegen at compile time, then don't.
In any case, it can be seen that at the time I have masked it for all its versions.
Code:
my-user@pc-gentoo ~ $ eix dev-util/gdbus-codegen
* dev-util/gdbus-codegen
     Available versions:  [m]2.70.4 [m](~)2.72.0 [m](~)2.72.1 {PYTHON_SINGLE_TARGET="python3_8 python3_9 python3_10"}
     Homepage:            https://www.gtk.org/
     Description:         GDBus code and documentation generator

Maybe it's the DNSSD, CUPS options in the src_configure() section, we should try... pending task.
Code:
--with-browseremoteprotocols=DNSSD,CUPS

https://manpages.debian.org/jessie/cups-browsed/cups-browsed.conf.5
gtk+ is not the culprit because in BLFS it didn't ask me for anything from gdbus-codegen to create the package.

Regards.
Back to top
View user's profile Send private message
Hu
Moderator
Moderator


Joined: 06 Mar 2007
Posts: 21635

PostPosted: Wed Mar 01, 2023 3:47 pm    Post subject: Reply with quote

colo-des wrote:
making system packages exclusively depend on pam and having no options to make them work without
pam...that should never happen.
Which Gentoo packages force PAM, but can be built without PAM without patching the upstream source? You can say nothing should require PAM, but if upstream designed their software to require PAM (or udev, or systemd, ...), then Gentoo's choices are to hard-depend on PAM or to not ship the package at all.
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Wed Mar 01, 2023 3:54 pm    Post subject: Reply with quote

Hu wrote:
colo-des wrote:
making system packages exclusively depend on pam and having no options to make them work without
pam...that should never happen.
Which Gentoo packages force PAM, but can be built without PAM without patching the upstream source? You can say nothing should require PAM, but if upstream designed their software to require PAM (or udev, or systemd, ...), then Gentoo's choices are to hard-depend on PAM or to not ship the package at all.

What packages have a hard dependency on pam?
The only one I found was su. What else does not work without pam?
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
ukky
Tux's lil' helper
Tux's lil' helper


Joined: 26 Feb 2023
Posts: 109
Location: Montreal, Canada

PostPosted: Wed Mar 01, 2023 4:17 pm    Post subject: Reply with quote

stefan11111 wrote:
Have firefox/librewolf's maintainers been notified of this?

This issue has nothing to do with Firefox/Librewolf.
Firefox build system supports building Firefox without D-Bus installed, also it supports building without ATK (Accessibility Toolkit).
Note, that building Firefox without ATK is not supported by Gentoo ebuild, but ebuild can be easily modified to build Firefox without ATK. Only ebuild has to be modified to build Firefox without ATK, no patches to Firefox sources are needed.
Thus, it is a Gentoo system issue that Firefox cannot be built without D-Bus if ATK is installed. And any Gentoo system with GTK:2 or GTK:3 installed requires ATK, which requires D-Bus. Thus, we have a soft dependency on D-Bus in Firefox (via ATK).
Back to top
View user's profile Send private message
Hu
Moderator
Moderator


Joined: 06 Mar 2007
Posts: 21635

PostPosted: Wed Mar 01, 2023 4:40 pm    Post subject: Reply with quote

stefan11111 wrote:
What packages have a hard dependency on pam?
The only one I found was su. What else does not work without pam?
There is no package su in ::gentoo. I see two packages in ::gentoo which can provide /bin/su. shadow optionally uses PAM for su. util-linux requires PAM for su, but util-linux can be built without su if you wish, which appears to remove the dependency on PAM. If you override the ebuild, can util-linux build a working /bin/su that does not use PAM? If yes, then the ebuild is unnecessarily strict. If no, then your issue is with upstream for their hard dependency on PAM.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54250
Location: 56N 3W

PostPosted: Wed Mar 01, 2023 5:47 pm    Post subject: Reply with quote

stefan11111,

Code:
$ grep CONFIG_SND_DYNAMIC_MINORS /usr/src/linux/.config
CONFIG_SND_DYNAMIC_MINORS=y


I live with that as
Code:
$ cat /proc/asound/devices
  2: [ 0- 3]: digital audio playback
  3: [ 0- 7]: digital audio playback
  4: [ 0- 8]: digital audio playback
  5: [ 0- 9]: digital audio playback
  6: [ 0-10]: digital audio playback
  7: [ 0-11]: digital audio playback
  8: [ 0- 0]: hardware dependent
  9: [ 0]   : control
 10: [ 1- 0]: digital audio playback
 11: [ 1- 0]: digital audio capture
 12: [ 1- 1]: digital audio playback
 13: [ 1- 2]: digital audio capture
 14: [ 1- 0]: hardware dependent
 15: [ 1]   : control
 33:        : timer
tells me the minor device numbers to use in /dev/snd. They are fixed, even if they are dynamically allocated.

I had a read of your issue that you linked.

Quote:
for convenience, I added this to my .xinitrc:

Scripts to run at startup go into /etc/local.d/ with a name ending in .start. e.g. foo.start

Adding scripts to .xinitrc runs the scripts when Xorg is started. It may not matter,
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
colo-des
Tux's lil' helper
Tux's lil' helper


Joined: 20 May 2011
Posts: 97

PostPosted: Wed Mar 01, 2023 6:03 pm    Post subject: Reply with quote

@Hu

I understand it perfectly, maybe I misinterpreted what I meant, I was saying that system packages should not be reached
they depended on pam compulsorily, saying it as an ideal. If the creators of the code do that, they will find beautiful forks or divine patches
which will be separate packages with other names. The creators of ebuilds are the owners to do what they want and the users to use them or not
It's just a matter of choice and taste.

Regards.
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Wed Mar 01, 2023 8:53 pm    Post subject: Reply with quote

NeddySeagoon wrote:
stefan11111,

Code:
$ grep CONFIG_SND_DYNAMIC_MINORS /usr/src/linux/.config
CONFIG_SND_DYNAMIC_MINORS=y


I live with that as
Code:
$ cat /proc/asound/devices
  2: [ 0- 3]: digital audio playback
  3: [ 0- 7]: digital audio playback
  4: [ 0- 8]: digital audio playback
  5: [ 0- 9]: digital audio playback
  6: [ 0-10]: digital audio playback
  7: [ 0-11]: digital audio playback
  8: [ 0- 0]: hardware dependent
  9: [ 0]   : control
 10: [ 1- 0]: digital audio playback
 11: [ 1- 0]: digital audio capture
 12: [ 1- 1]: digital audio playback
 13: [ 1- 2]: digital audio capture
 14: [ 1- 0]: hardware dependent
 15: [ 1]   : control
 33:        : timer
tells me the minor device numbers to use in /dev/snd. They are fixed, even if they are dynamically allocated.

I had a read of your issue that you linked.

Quote:
for convenience, I added this to my .xinitrc:

Scripts to run at startup go into /etc/local.d/ with a name ending in .start. e.g. foo.start

Adding scripts to .xinitrc runs the scripts when Xorg is started. It may not matter,

Thanks for the info.
I no longer need those scripts to make /dev nodes, fix permissions and load modules as I no longer have an initramfs that mounts /dev and I build everything into the kernel(=y not =m).
Thanks for helping me with that.
Regarding /etc/local.d, is it enough to just add the script in that folder and it will run at boot with no further configuration?
I have a line in my .xinitrc to mount all my network drives.
Code:
rdo mount -a &

rdo is just an app that escalates privileges. Treat in like sudo/doas.
I have to do this because during startup, I get a mount error saying that the network is not accessible.
I have everything set up in my fstab.
Code:
//10.0.15.220/hdd1      /home/stefan/HDD1       cifs    username=admin,password=raspberry,uid=stefan,_netdev    0 0
//10.0.15.220/hdd2      /home/stefan/HDD2       cifs    username=admin,password=raspberry,uid=stefan,_netdev    0 0
//10.0.15.220/MyBook    /home/stefan/MyBook     cifs    username=admin,password=raspberry,uid=stefan,_netdev    0 0
//10.0.15.115/DuneHDD_349216fe_819a_4f56_b771_d977ce430eb3/     /home/stefan/DuneHDD    cifs    username=root,password=,vers=1.0,uid=stefan,_netdev     0 0

Any idea how to fix that?
Also, did you steal a pc from a record company?
What is going on with all those sound cards?
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 54250
Location: 56N 3W

PostPosted: Wed Mar 01, 2023 10:27 pm    Post subject: Reply with quote

stefan11111,

The *.start scripts in /etc/local.d are run as root by the local service starting.
Its the last service to be started before the login prompt.

Likewise the *.stop scripts are run when the local service is stopped.

local should be in the default runlevel.

I have two sound cards. Card0 is my Video Card. It claims that it has 6 digital outputs but it has a lot less connectors than that.
Card 1 is the on board analogue sound card.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Wed Mar 01, 2023 10:32 pm    Post subject: Reply with quote

NeddySeagoon wrote:
stefan11111,

The *.start scripts in /etc/local.d are run as root by the local service starting.
Its the last service to be started before the login prompt.

Likewise the *.stop scripts are run when the local service is stopped.

local should be in the default runlevel.

I have two sound cards. Card0 is my Video Card. It claims that it has 6 digital outputs but it has a lot less connectors than that.
Card 1 is the on board analogue sound card.

What video card do you have?
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Thu Mar 02, 2023 12:50 am    Post subject: Reply with quote

colo-des wrote:
stefan11111 wrote:
Has anyone gotten rid of this too?

net-print/cups-filters need gdbus-codegen at compile time, then don't.
In any case, it can be seen that at the time I have masked it for all its versions.
Code:
my-user@pc-gentoo ~ $ eix dev-util/gdbus-codegen
* dev-util/gdbus-codegen
     Available versions:  [m]2.70.4 [m](~)2.72.0 [m](~)2.72.1 {PYTHON_SINGLE_TARGET="python3_8 python3_9 python3_10"}
     Homepage:            https://www.gtk.org/
     Description:         GDBus code and documentation generator

Maybe it's the DNSSD, CUPS options in the src_configure() section, we should try... pending task.
Code:
--with-browseremoteprotocols=DNSSD,CUPS

https://manpages.debian.org/jessie/cups-browsed/cups-browsed.conf.5
gtk+ is not the culprit because in BLFS it didn't ask me for anything from gdbus-codegen to create the package.

Regards.

Code:
$ doas emerge -cav dev-util/gdbus-codegen

Calculating dependencies... done!
  dev-util/gdbus-codegen-2.74.6 pulled in by:
    x11-libs/gtk+-3.24.36 requires >=dev-util/gdbus-codegen-2.48

>>> No packages selected for removal by depclean

_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
colo-des
Tux's lil' helper
Tux's lil' helper


Joined: 20 May 2011
Posts: 97

PostPosted: Thu Mar 02, 2023 2:05 am    Post subject: Reply with quote

@stefan11111

I was looking and the version I have installed is the 3.24.22 that does not have BDEPEND and is EAPI=6, the 3.24.33 has BDEPEND and is EAPI=7
and there if it is in the block:
Code:
BDEPEND="
   >=dev-util/gdbus-codegen-2.48

In BLFS when compiling 3.24.34 it is not a dependency.
I no longer remember what I did back then with 3.24.22, if I installed gdbus-codegen and emerged the 3.24.22 and then uninstalled and masked
gdbus-codegen or if it was that it didn't even ask me, I know that cups-filters asks for it when configuring, but then it run without it.
Removing it from the BDEPEND= block would suffice.

These are the options that I used in BLFS with 3.24.34, and the most striking thing may be having used --disable-cups

Code:
# Configure:
   ./configure --prefix=/usr   \
   --sysconfdir=/etc           \
   --disable-cloudproviders    \
   --disable-papi              \
   --enable-profiler=no        \
   --disable-broadway-backend  \
   --disable-wayland-backend   \
   --disable-cups              \
   --enable-colord=no          \
   --disable-glibtest          \
   --disable-gtk-doc           \
   --disable-gtk-doc-html      \
   --enable-xkb                \
   --enable-xinerama           \
   --enable-xrandr             \
   --enable-xfixes             \
   --enable-xcomposite         \
   --enable-xdamage            \
   --enable-x11-backend        \
   --without-atk-bridge        \
   --enable-introspection=yes  \
   --disable-installed-tests   \
   --with-x

Regards.
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Thu Mar 02, 2023 7:04 pm    Post subject: Reply with quote

colo-des wrote:
@stefan11111

I was looking and the version I have installed is the 3.24.22 that does not have BDEPEND and is EAPI=6, the 3.24.33 has BDEPEND and is EAPI=7
and there if it is in the block:
Code:
BDEPEND="
   >=dev-util/gdbus-codegen-2.48

In BLFS when compiling 3.24.34 it is not a dependency.
I no longer remember what I did back then with 3.24.22, if I installed gdbus-codegen and emerged the 3.24.22 and then uninstalled and masked
gdbus-codegen or if it was that it didn't even ask me, I know that cups-filters asks for it when configuring, but then it run without it.
Removing it from the BDEPEND= block would suffice.

These are the options that I used in BLFS with 3.24.34, and the most striking thing may be having used --disable-cups

Code:
# Configure:
   ./configure --prefix=/usr   \
   --sysconfdir=/etc           \
   --disable-cloudproviders    \
   --disable-papi              \
   --enable-profiler=no        \
   --disable-broadway-backend  \
   --disable-wayland-backend   \
   --disable-cups              \
   --enable-colord=no          \
   --disable-glibtest          \
   --disable-gtk-doc           \
   --disable-gtk-doc-html      \
   --enable-xkb                \
   --enable-xinerama           \
   --enable-xrandr             \
   --enable-xfixes             \
   --enable-xcomposite         \
   --enable-xdamage            \
   --enable-x11-backend        \
   --without-atk-bridge        \
   --enable-introspection=yes  \
   --disable-installed-tests   \
   --with-x

Regards.

Is there any way to pass these options to portage?
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
ukky
Tux's lil' helper
Tux's lil' helper


Joined: 26 Feb 2023
Posts: 109
Location: Montreal, Canada

PostPosted: Thu Mar 02, 2023 7:45 pm    Post subject: Reply with quote

stefan11111 wrote:
I have a line in my .xinitrc to mount all my network drives.
... edited ...
I have to do this because during startup, I get a mount error saying that the network is not accessible.
I have everything set up in my fstab.
... edited ...
Any idea how to fix that?

To avoid having errors at boot, you need to add noauto option to all your network drives in /etc/fstab.
Depending on your boot and network configuration, even in .xinitrc it might be too early for network to be available.
If you are using net-misc/netifrc, then you can mount your network drives from postup() function in /etc/conf.d/net.
See /usr/share/doc/netifrc-0.7.3/net.example.bz2 on how to add postup() function to your /etc/conf.d/net.
Similarly, you can umount network drives from predown() function.
Make sure network interface for your network drives matches "${IFACE}" parameter when postup() or predown() is called.
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Thu Mar 02, 2023 8:06 pm    Post subject: Reply with quote

ukky wrote:
stefan11111 wrote:
I have a line in my .xinitrc to mount all my network drives.
... edited ...
I have to do this because during startup, I get a mount error saying that the network is not accessible.
I have everything set up in my fstab.
... edited ...
Any idea how to fix that?

To avoid having errors at boot, you need to add noauto option to all your network drives in /etc/fstab.
Depending on your boot and network configuration, even in .xinitrc it might be too early for network to be available.
If you are using net-misc/netifrc, then you can mount your network drives from postup() function in /etc/conf.d/net.
See /usr/share/doc/netifrc-0.7.3/net.example.bz2 on how to add postup() function to your /etc/conf.d/net.
Similarly, you can umount network drives from predown() function.
Make sure network interface for your network drives matches "${IFACE}" parameter when postup() or predown() is called.

Thanks.
I do not use netifrc. I only use dhcpcd.
That seemed to do the trick.
I also get this message when booting
Code:
/lib/dhcpcd/dhcpcd-run-hooks: line 132: /etc/resolv.conf: Operation not permitted
chmod: changing permissions of '/etc/resolv.conf': Operation not permitted

_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
ukky
Tux's lil' helper
Tux's lil' helper


Joined: 26 Feb 2023
Posts: 109
Location: Montreal, Canada

PostPosted: Thu Mar 02, 2023 8:20 pm    Post subject: Reply with quote

stefan11111 wrote:
I also get this message when booting
/lib/dhcpcd/dhcpcd-run-hooks: line 132: /etc/resolv.conf: Operation not permitted
chmod: changing permissions of '/etc/resolv.conf': Operation not permitted

net-dns/openresolv might help you to fix this error.
Please also see Resolv.conf being blank written
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Thu Mar 02, 2023 8:40 pm    Post subject: Reply with quote

ukky wrote:
stefan11111 wrote:
I also get this message when booting
/lib/dhcpcd/dhcpcd-run-hooks: line 132: /etc/resolv.conf: Operation not permitted
chmod: changing permissions of '/etc/resolv.conf': Operation not permitted

net-dns/openresolv might help you to fix this error.
Please also see Resolv.conf being blank written

No need for that package. It appears that editing /etc/dhcpcd.conf to stop touching resolv.conf made the error go away.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Thu Mar 02, 2023 8:54 pm    Post subject: Reply with quote

NeddySeagoon wrote:
stefan11111,

The *.start scripts in /etc/local.d are run as root by the local service starting.
Its the last service to be started before the login prompt.

Likewise the *.stop scripts are run when the local service is stopped.

local should be in the default runlevel.

I have two sound cards. Card0 is my Video Card. It claims that it has 6 digital outputs but it has a lot less connectors than that.
Card 1 is the on board analogue sound card.

I have separated my init scripts according to this. Hope nothing breaks.
Code:
$ cat /etc/local.d/startup.start
#! /bin/sh

qbittorrent-nox &
curl https://raw.githubusercontent.com/StevenBlack/hosts/master/hosts > /etc/hosts

Yes, I see the irony with my signature. :)
Code:
$ cat .xinitrc
#!/bin/sh

slstatus &
xcompmgr &
/usr/bin/feh --bg-fill '/home/stefan/images/wallpapers/2.jpg' &
xset -dpms &
xset s off &
exec dwm


Now my .xinitrc has only what needs to be loaded with X11.

Regarding the sound devices, here is my /proc/asound/devices
Code:
$ cat /proc/asound/devices
  0: [ 0]   : control
  1:        : sequencer
  4: [ 0- 0]: hardware dependent
 16: [ 0- 0]: digital audio playback
 17: [ 0- 1]: digital audio playback
 24: [ 0- 0]: digital audio capture
 26: [ 0- 2]: digital audio capture
 32: [ 1]   : control
 33:        : timer
 36: [ 1- 0]: hardware dependent
 51: [ 1- 3]: digital audio playback
 64: [ 2]   : control
 80: [ 2- 0]: digital audio playback
 88: [ 2- 0]: digital audio capture

As you can see, there are more than 8. This seems to list all the sound outputs.
Code:
$ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: PCH [HDA Intel PCH], device 0: ALC892 Analog [ALC892 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 1: ALC892 Digital [ALC892 Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: NVidia [HDA NVidia], device 3: Generic Digital [Generic Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 2: Ultimate [Razer Kraken Ultimate], device 0: USB Audio [USB Audio]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

And this seems to list the sound cards.
Code:
$ grep -i dynamic /usr/src/linux/.config
# CONFIG_PREEMPT_DYNAMIC is not set
CONFIG_DYNAMIC_MEMORY_LAYOUT=y
CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT=y
CONFIG_HAVE_PREEMPT_DYNAMIC=y
CONFIG_HAVE_PREEMPT_DYNAMIC_CALL=y
CONFIG_DYNAMIC_SIGFRAME=y
# CONFIG_SND_DYNAMIC_MINORS is not set
# CONFIG_USB_DYNAMIC_MINORS is not set
# CONFIG_DYNAMIC_DEBUG is not set
# CONFIG_DYNAMIC_DEBUG_CORE is not set
CONFIG_HAVE_DYNAMIC_FTRACE=y
CONFIG_HAVE_DYNAMIC_FTRACE_WITH_REGS=y
CONFIG_HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS=y
CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS=y
CONFIG_HAVE_DYNAMIC_FTRACE_NO_PATCHABLE=y

Everything works fine without dynamic minors enabled.
This is the only thing I see about sound devices:
Code:
$ dmesg | grep freq
[  729.842482] usb 1-12: 2:1: cannot get freq at ep 0x1
[  736.969264] usb 1-12: 2:1: cannot get freq at ep 0x1
[  757.884458] usb 1-12: 2:1: cannot get freq at ep 0x1

This appears in my dmesg only three times at these specific timestamps(or very close to them).
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Thu Mar 02, 2023 9:29 pm    Post subject: Reply with quote

stefan11111 wrote:
ukky wrote:
stefan11111 wrote:
I have a line in my .xinitrc to mount all my network drives.
... edited ...
I have to do this because during startup, I get a mount error saying that the network is not accessible.
I have everything set up in my fstab.
... edited ...
Any idea how to fix that?

To avoid having errors at boot, you need to add noauto option to all your network drives in /etc/fstab.
Depending on your boot and network configuration, even in .xinitrc it might be too early for network to be available.
If you are using net-misc/netifrc, then you can mount your network drives from postup() function in /etc/conf.d/net.
See /usr/share/doc/netifrc-0.7.3/net.example.bz2 on how to add postup() function to your /etc/conf.d/net.
Similarly, you can umount network drives from predown() function.
Make sure network interface for your network drives matches "${IFACE}" parameter when postup() or predown() is called.

Thanks.
I do not use netifrc. I only use dhcpcd.
That seemed to do the trick.
I also get this message when booting
Code:
/lib/dhcpcd/dhcpcd-run-hooks: line 132: /etc/resolv.conf: Operation not permitted
chmod: changing permissions of '/etc/resolv.conf': Operation not permitted

While adding noauto gets rid of the error, it also prevents mounting the file systems.
For now, I am back to mounting in .xinitrc
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
ukky
Tux's lil' helper
Tux's lil' helper


Joined: 26 Feb 2023
Posts: 109
Location: Montreal, Canada

PostPosted: Thu Mar 02, 2023 9:51 pm    Post subject: Reply with quote

stefan11111 wrote:
While adding noauto gets rid of the error, it also prevents mounting the file systems.

That is the purpose of noauto option.
When your system is ready, i.e. network interface is up, you can do, for example:
Code:
mount /home/stefan/HDD1

stefan11111 wrote:
For now, I am back to mounting in .xinitrc

Using dhcpcd hooks might be another option:
Code:
man 8 dhcpcd-run-hooks

Mounting from .xinitrc will not work if your router (or DHCP server) is not available when you start your X session.
Back to top
View user's profile Send private message
stefan11111
l33t
l33t


Joined: 29 Jan 2023
Posts: 922
Location: Romania

PostPosted: Thu Mar 02, 2023 10:09 pm    Post subject: Reply with quote

ukky wrote:
stefan11111 wrote:
While adding noauto gets rid of the error, it also prevents mounting the file systems.

That is the purpose of noauto option.
When your system is ready, i.e. network interface is up, you can do, for example:
Code:
mount /home/stefan/HDD1

stefan11111 wrote:
For now, I am back to mounting in .xinitrc

Using dhcpcd hooks might be another option:
Code:
man 8 dhcpcd-run-hooks

Mounting from .xinitrc will not work if your router (or DHCP server) is not available when you start your X session.

The network is available after I log into a tty. I can mount my network filesystems from a tty. Mounting in .xinitrc also works.
_________________
My overlay: https://github.com/stefan11111/stefan_overlay
INSTALL_MASK="/etc/systemd /lib/systemd /usr/lib/systemd /usr/lib/modules-load.d *udev* /usr/lib/tmpfiles.d *tmpfiles* /var/lib/dbus /usr/bin/gdbus /lib/udev"
Back to top
View user's profile Send private message
colo-des
Tux's lil' helper
Tux's lil' helper


Joined: 20 May 2011
Posts: 97

PostPosted: Fri Mar 03, 2023 12:44 am    Post subject: Reply with quote

@stefan11111
You must create a local repository if you don't already have one and copy the port you want to modify.
https://wiki.gentoo.org/wiki/Basic_guide_to_write_Gentoo_Ebuilds
By having the same version of the port in both repositories, emerge will give priority to the local repository when emerging.
Code:
# mkdir -pv /var/db/repos/local_repo/x11-libs
# cp -rpv /usr/portage/x11-libs/gtk+ /var/db/repos/local_repo/x11-libs
# cd /var/db/repos/local_repo/x11-libs/gtk+
# nano gtk+-3.24.22.ebuild

In the multilib_src_configure() function you can see that a use with an alias is assigned to each real option in autotools ./configure
The uses are nothing more than the usual ./configure options.
Code:
multilib_src_configure() {
        local myconf=(
                $(use_enable cups cups auto)

In the case of version 3.4.22, it has the use cups which can be enabled or disabled as always through package.use
If this use is not declared through IUSE= in the ebuild then the option can be passed directly in that section
that is, putting under the $(use_enable the options that one wants to add or remove as it is done in a normal ./configure.
Code:
  --disable-cups

At the end it is time to rebuild the Manifest file since the ebuild has changed, it will download the source if it is not already downloaded in /usr/portage/distfiles and then it will compute for all files used in the port and create the new Manifest
Code:
# ebuild gtk+-3.24.22.ebuild manifest

If you use eix you can update your database so that it appears and list it to see the port of the modified copy in the local tree:
Code:
# eix-update
# eix -I x11-libs/gtk+
# emerge -av1 x11-libs/gtk+

Regards.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Unsupported Software All times are GMT
Goto page Previous  1, 2, 3, 4  Next
Page 3 of 4

 
Jump to:  
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