[SOLVED] DPMS still blanking screen

S3NTYN3L · 2016-10-20 04:13:38

I'm reading through this in the wiki: Display Power Management Signaling

I've got the following installed.

xorg-server
xorg-server-utils
xorg-xinit
i3 (started via startx and .xinitrc file. I've no DM/DE installed)

I can't, for the life of me, get DPMS to disable.

While in the shell, without X running, the screen goes blank after the usual 10 minutes.

I've run (as root)

setterm --blank 0 --powerdown 0 --powersave off >> /etc/issue

The results in the /etc/issue file are:

^[[9;0]^[[14;0]

I then reboot and log back in, wait 10 minutes, and my screen still blanks

While in i3, I can't get DPMS disabled either.
I have the following in /etc/X11/xorg.conf.d/10-monitor.conf

        Section "Monitor"
            Identifier "LVDS0"
            Option "DPMS" "false"
        EndSection

        Section "ServerLayout"
            Identifier "ServerLayout0"
            Option "StandbyTime" "0"
            Option "SuspendTime" "0"
            Option "OffTime"     "0"
            Option "BlankTime"   "0"
        EndSection

Same results here.
I create that file, reboot, startx and xset -q reports that DPMS is still enabled.

The only way I'm able to disable it is via ~/.xinitrc with

xset s off &
xset -dpms &

However, that only gets the job done whilst running X.
I'd love to have DPMS disabled while in the shell as well as having the config file work so I can clean up my .xinitrc.

Sorry, I've OCD issues and even though it's only two lines in the file, well, it's things like this that bother me.
I'd rather have the monitor.conf file work as is intended so I don't have to edit user .xinitrc files, ya know?

Thanks in advance!

Last edited by S3NTYN3L (2016-10-20 22:28:50)

seth · 2016-10-20 06:53:09

Section "Extensions"
    Option      "DPMS" "Disable"
EndSection

Did you ensure setterm is effective when actually called (ie. is the problem the command or the escape sequence or etc/issue, or ... ;-)

Head_on_a_Stick · 2016-10-20 06:58:36

S3NTYN3L wrote:

I've run (as root)
setterm --blank 0 --powerdown 0 --powersave off >> /etc/issue
The results in the /etc/issue file are:
^[[9;0]^[[14;0]

Have you tried the recommendation on the ArchWiki page?

# echo -ne "\033[9;0]" >> /etc/issue

Also, you seem to have copied the example xorg configuration file, in particular:

    Identifier "LVDS0"

Is that the correct identifier for your screen?

S3NTYN3L · 2016-10-20 13:25:37

seth wrote:

Did you ensure setterm is effective when actually called (ie. is the problem the command or the escape sequence or etc/issue, or ... ;-)

I don't know. As;

Head_on_a_Stick wrote:

Have you tried the recommendation on the ArchWiki page?
# echo -ne "\033[9;0]" >> /etc/issue

Neither method seems to work.

Head_on_a_Stick wrote:

    Identifier "LVDS0"
Is that the correct identifier for your screen?

I've no idea. In my researching this, I can't seem to find anywhere that details how to actually find this information.
The closest I've found was in a Google search referring to EDID.

I found https://www.archlinux.org/packages/comm … read-edid/, but upon installation and running the "parse-edid" only seems to hang and the "get-edid" only throws errors my way, ei;

Attempting to use i2c interface
Looks like no buses have an EDID. Sorry!
Attempting to use the classical VBE interface
Illegal instruction (core dumped)

I noticed "(core dumped)", but dumped to where? Can I glean anything to said dump?

Where can I find definitions for these xorg.conf files?
My Google-fu has failed me in this regard.

Last edited by S3NTYN3L (2016-10-20 13:26:46)

seth · 2016-10-20 13:38:32

Just run setterm and wait whether the VT still blanks. If it does, there's no point in investigating on the escape sequence or etc/issue invocation.

To query your X11 outputs, run "xrandr -q", but the forementioned xorg.conf snippet simply disables dpms altogether (if this is what you want)

S3NTYN3L · 2016-10-20 15:05:12

seth wrote:

Just run setterm and wait whether the VT still blanks. If it does, there's no point in investigating on the escape sequence or etc/issue invocation.

I've done just that.

No joy.

seth wrote:

To query your X11 outputs, run "xrandr -q", but the forementioned xorg.conf snippet simply disables dpms altogether (if this is what you want)

Yes, that's exactly what I'm going for. I want DPMS disabled in BOTH X and the shell.

xrandr -q reports;

eDP1 connected
DP1 disconnected
HDMI1 disconnected
VGA1 disconnected
VIRTUAL1 disconnected

So, I suppose that it's safe to assume eDP1 is my laptop monitor.

Unfortunately, I've got to go run some errands at the moment so I can't test until I return in a few hours.

In the meantime, in order to get xrandr running, I was forced to install everything in the xorg group.
Obviously, some packages got reinstalled...

Any way, to my point; What packages do you think I can safely REMOVE and still keep X and/or DPMS working?
My ultimate goal is to have everything work with a minimum of package bloat.

Thanks for your help thus far guys!
I'll return to this in just a little bit.

seth · 2016-10-20 15:13:11

You don't need xrandr to control dpms. Just to be sure "DPMS woking" means "no power saving at all", right?

=> Disable the extension as shown in comment #2

If setterm has no impact that might be due to the framebuffer console. Drop it, see https://wiki.archlinux.org/index.php/GR … ramebuffer (there's no guarantee that this is the cause, though. just worth a shot)

S3NTYN3L · 2016-10-20 20:01:25

seth wrote:

You don't need xrandr to control dpms. Just to be sure "DPMS woking" means "no power saving at all", right?

I understand that I don't need xrandr, but what about all the other packages that the xorg group installed?

Can I go simply remove ALL that xorg installed and leave ONLY?;

xf86-video-intel        # I'm running a laptop with Intel HD graphics only
xf86-input-libinput     # for the touchpad, obviously ;)
xorg-server
xorg-server-utils       # not really sure if this one is necessary
xorg-xinit

Yes, I want DPMS disabled.

I've the snippet in comment #2, but that didn't work for some reason.

FWIW, I now have the following in /etc/X11/xorg.conf.d/10-monitor.conf.
(So yeah, I found the xorg wiki page. I must have had a brain fart prior to posting this thread, lol)
It's the only way I could get DPMS disabled. Not even defining "eDP1", as reported by xrandr, as my Identifier(s) worked.

Section "Screen"
    Identifier "Screen0"
    Device     "Device0"
    Monitor    "Monitor0"
EndSection

Section "Monitor"
    Identifier "Monitor0"
    Option     "DPMS" "0"
EndSection

Section "ServerLayout"
    Identifier "ServerLayout0"
    Option     "BlankTime"   "0"
    Option     "OffTime"     "0"
    Option     "StandbyTime" "0"
    Option     "SuspendTime" "0"
EndSection

If you have any suggestions on consolidating the above, I'd be happy to try it!

seth wrote:

If setterm has no impact that might be due to the framebuffer console. Drop it, see https://wiki.archlinux.org/index.php/GR … ramebuffer (there's no guarantee that this is the cause, though. just worth a shot)

I'm not running GRUB, I'm running systemd-boot.
I did have

fbcon=scrollback:128k

in my kernel parameters, but I removed it, updated mkinitcpio and bootctl, rebooted, loged in, waited the 10 minutes and...

Still no joy in the shell.

Last edited by S3NTYN3L (2016-10-20 20:15:25)

seth · 2016-10-20 20:23:15

You don't need the server utils to run a basic X server. They're just CLI tools to "manipulate" the display server.
You neither need xf86-video-intel, you'll run the modesetting driver instead.
xinit provides startx. If you start the server differently, you don't need that as well.

Now back on topic:
If you disable dpms, it should not show up in the extension list of xdpyinfo (something you probably just uninstalled ;-)
As long as DPMS is loaded, the timings can be changed at runtime any time.

I meanwhile tried setterm on a VGA console - doesn't seem to have any impact reg. powersaving (attempting to reduce the timeout to one minute), so apparently that's simply broken.

S3NTYN3L · 2016-10-20 20:39:19

seth wrote:

You don't need the server utils to run a basic X server. They're just CLI tools to "manipulate" the display server.
You neither need xf86-video-intel, you'll run the modesetting driver instead.
If you start the server differently, you don't need that as well.

Understood.

I noticed this in the i3 wiki page:

i3-wm includes i3.desktop as Xsession which starts the window manager.

Any clue as to how I could use that instead?
If I'm thinking correctly, I could do away with xinit as you've alluded to. That would be nice.

seth wrote:

Now back on topic:
If you disable dpms, it should not show up in the extension list of xdpyinfo (something you probably just uninstalled ;-)
As long as DPMS is loaded, the timings can be changed at runtime any time.

OK, so better to disable COMPLETELY? As in, use the method in comment #2?

seth wrote:

I meanwhile tried setterm on a VGA console - doesn't seem to have any impact reg. powersaving (attempting to reduce the timeout to one minute), so apparently that's simply broken.

Hmm, so this means something in Arch is broken and I'm not just crazy in thinking that I broke it, lol?

seth · 2016-10-20 21:30:58

You "somehow" need to start X. The first thing you usually get is some DM - or you use startx and get xinitrc executed.
You can also just "X & export DISPLAY=:0 && sleep 0.5 && i3" or similar, but that has some caveats. startx is usually a good choice if you don't want a DM.

Disabling the extension is better unless™ this leaves DPMS to the apparently broken VT control, yes.

S3NTYN3L · 2016-10-20 22:12:18

seth wrote:

You "somehow" need to start X. The first thing you usually get is some DM - or you use startx and get xinitrc executed.
You can also just "X & export DISPLAY=:0 && sleep 0.5 && i3" or similar, but that has some caveats. startx is usually a good choice if you don't want a DM.
Disabling the extension is better unless™ this leaves DPMS to the apparently broken VT control, yes.

Understood, thanks.

As for the setterm issue, was I correct to interpret your statement as I did?
I mean, it's something broken with setterm and not me doing something wrong?

seth · 2016-10-20 22:14:26

It very much seems so (while the bug is more likely in the kewrnel than in setterm ;-)

S3NTYN3L · 2016-10-20 22:27:59

Gotcha.

I suppose I'll just leave /etc/issue alone in the hopes that it gets fixed in a future update.
I'll go ahead and mark this as solved.

Thanks yet again for your help!

Arch Linux

#1 2016-10-20 04:13:38

[SOLVED] DPMS still blanking screen

#2 2016-10-20 06:53:09

Re: [SOLVED] DPMS still blanking screen

#3 2016-10-20 06:58:36

Re: [SOLVED] DPMS still blanking screen

#4 2016-10-20 13:25:37

Re: [SOLVED] DPMS still blanking screen

#5 2016-10-20 13:38:32

Re: [SOLVED] DPMS still blanking screen

#6 2016-10-20 15:05:12

Re: [SOLVED] DPMS still blanking screen

#7 2016-10-20 15:13:11

Re: [SOLVED] DPMS still blanking screen

#8 2016-10-20 20:01:25

Re: [SOLVED] DPMS still blanking screen

#9 2016-10-20 20:23:15

Re: [SOLVED] DPMS still blanking screen

#10 2016-10-20 20:39:19

Re: [SOLVED] DPMS still blanking screen

#11 2016-10-20 21:30:58

Re: [SOLVED] DPMS still blanking screen

#12 2016-10-20 22:12:18

Re: [SOLVED] DPMS still blanking screen

#13 2016-10-20 22:14:26

Re: [SOLVED] DPMS still blanking screen

#14 2016-10-20 22:27:59

Re: [SOLVED] DPMS still blanking screen

Board footer