Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Black screen after startx [WORKAROUND]
View unanswered posts
View posts from last 24 hours

 
Reply to topic    Gentoo Forums Forum Index Desktop Environments
View previous topic :: View next topic  
Author Message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Fri Nov 23, 2012 1:36 am    Post subject: Black screen after startx [WORKAROUND] Reply with quote

Hello Everyone,

Let me start by saying thankyou to all the forum contributors. The information gleaned has been invaluable.

In a nutshell my problem is: I boot up to a login prompt fine. I can login as root or a user and everything works great. However, as soon as I type "startx" to go to a graphical environment my screen goes black. If I enter as root I am entering twm with xterm and I can type in the blind "reboot" or "halt" and the command executes. If I enter as a user I'm trying to enter and OpenBox environment and I can only kill the computer with switch.

I first encountered this problem trying to load Funtoo. I completely reinstalled about three times (once with genkernel) and never succeeded in getting an xorg environment. So then I tried Gentoo. Initially I had everthing working although after typing "startx" and hitting enter I would have about a ten second delay before everthing turned on. Then recently after a major emerge I am facing a black screen again. I have tried downgrading the xorg-server and drivers. Then I tried the genkernel. ...so far nothing!

I have looked through the forums and found quite a few similar problems but so far nothing that seems to apply to my particular situation.

I have a dual boot with arch linux. It works fine so I infer that my hardware is fine. I have some pastebins to help complete the picture.

My emerge --info
http://pastebin.com/xgAwNDjE

My .config file (before the genkernel)
http://pastebin.com/DLUkSUrs

My Xorg.0.log file (with genkernel)
http://pastebin.com/Abqxy1f1

My lspci -n
http://pastebin.com/X1vFZDLv

My lspci -k
http://pastebin.com/UXFex63x

If I need to post any additional info please let me know. This is my first post so if I didn't quite do things kosher please make allowance. In the meantime I'm pulling my hair out and I have very little left.

Any help would be appreciated.


Last edited by Myrlynn on Tue Dec 11, 2012 12:28 am; edited 1 time in total
Back to top
View user's profile Send private message
Hypnos
Advocate
Advocate


Joined: 18 Jul 2002
Posts: 2867
Location: Omnipresent

PostPosted: Fri Nov 23, 2012 5:49 am    Post subject: Reply with quote

You should be able to kill X with the ctrl-alt-backspace key combo, not have to reboot/halt. Then, you should be able to examine dmesg to see if the kernel is complaining about something.

If the key combo doesn't work, read this.

PS: You first help request is far better than most :wink:
_________________
Personal overlay | Simple backup scheme
Back to top
View user's profile Send private message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Fri Nov 23, 2012 6:47 pm    Post subject: Reply with quote

Thanks for the response Hypnos.

I attempted to follow the directions you gave me. Perhaps I didn't do it right. First i tried adding "setxkbmap -option terminate:ctrl_alt_bksp" to the .xinitrc file both for root and user. No effect. Then I created directory /etc/X11/xorg.conf.d and copied 10-evdev.conf from my Arch distro and added a section to it for termination. Again no effect.

I have however chrooted into my gentoo system and copied the dmesg. I don't recognize anything but perhaps you or someone else might. It is located at:

http://pastebin.com/jVHhq0u8

Thanks for taking the time.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 31711
Location: 56N 3W

PostPosted: Fri Nov 23, 2012 7:23 pm    Post subject: Reply with quote

Myrlynn,

When you chroot in after a restart, you get a whole new dmesg - its not saved across reboots.

Your Xorg.0. log appears to show a normal Xorg start. Thats supported by root getting twm to work.

How do you attempt to start OpenBox?
With entries in ~/.xinitrc ?

Move the users ~/.xinitrc to say ~/.xinitrc_saved and try startx. Your user should get twm too. If that works, post your ~/.xinitrc file.
_________________
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
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Fri Nov 23, 2012 10:35 pm    Post subject: Reply with quote

NeddySeagoon,

I saved my ~/.xinitrc as ~/.xinitrc_saved at your suggestion. When I entered startx I ended up with a black screen. It did not respond to ctl-alt-bkspace or reboot this time.

I had entered "vesa" into my make.conf file in addition to "intel" this time.

I have pastebined my latest Xorg.0.log:

http://pastebin.com/AKvBcVYb

The contents of my ~/.xinitrc (now ~/.xinitrc_saved) are as follows:

Code:
numlockx &
udiskie &
setxkbmap -option terminate:ctrl_alt_bksp &

exec openbox-session


Can you tell me how to go about getting a copy of my dmesg since I seem to end up either typing reboot or doing a hard kill with the switch?

More suggestions?
Back to top
View user's profile Send private message
Jaglover
Advocate
Advocate


Joined: 29 May 2005
Posts: 4611
Location: Saint Amant, Acadiana

PostPosted: Fri Nov 23, 2012 11:06 pm    Post subject: Reply with quote

Try simple things first, comment everything out but exec openbox-session. Second, if your keyboard is responding switch to another console to get the logs.
_________________
Please learn how to denote units correctly!
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 31711
Location: 56N 3W

PostPosted: Sat Nov 24, 2012 1:28 am    Post subject: Reply with quote

Myrlynn,

You log shows Xorg starting normally, running for 40 seconds then terminating normally.

Try running
Code:
startx; sleep 20 && killall X

That starts Xorg, waits 20 seconds, then kills Xorg which should give you control again.
From your log, thats about what happened anyway.

If you have another system, you can ssh in before you startx.
_________________
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
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Sat Nov 24, 2012 5:13 am    Post subject: Reply with quote

Jaglover,

When I do exec openbox-session I have no keyboard control at all. When I do startx I usually can do either reboot or halt. Ctl_Alt_f2 gives me nothing. Ctl_Alt_bkspace gives me nothing.

However, your idea was deffinitely worth a try.
Back to top
View user's profile Send private message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Sat Nov 24, 2012 5:18 am    Post subject: Reply with quote

NeddySeagoon,

startx; sleep 20 && killall X didn't appear to do anything.

You also mentioned ssh before startx. At the present time that is beyound my learning curve.

In the meantime, I've got to get some sleep. This problem will keep until tomorrow.

Thanks everyone for your help and support.
Back to top
View user's profile Send private message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Thu Nov 29, 2012 2:07 am    Post subject: Reply with quote

Hello Everyone,

An update on my black screen situation. In a previous entry NeddySeagoon suggested
that I ssh into my system before startx. So far I have been unsucessful in this
regard. I'm using wicd and evidently need static ip address. The distro I'm trying
to use is Arch Linux. This still may be my best bet but I've been a little slow figuring it out.

In the meantime I have reverted back to my manual kernel since the genkernel was
not solving anything and I thought the manual kernel would have less clutter to
walk through.

I went back through my Xorg.0.log and came up with an interesting tidbit:

Code:
[181.441] (--) intel(0): found backlight control interface acpi_video0 (type 'firmware')


I'm dual booted with Arch Linux and when I compare a similar entry the interface
is acpi_video1. I don't know if that is relevant or not.

I ran accross some site that indicated that I could change the brightness of my
screen by modifying entries in the acpi_video file.

Code:
#cat /sys/class/backlight/acpi_video0/brightness
0


I nano in as root and changed the value to 10 which is the value showing in my
Arch Linux file.

This time when I startx, it kicked back to a prompt after the following message.

Code:
xinit: connection to X server lost

waiting for X server to shut down Server terminated successfully (0). Closing
log file.

Since I was at a prompt I did a dmesg. I'm sorry but I did not know how to save
it to be able to show everyone. I did locate a couple entries which I copied with
pen and paper that might be relevant.

Code:
[0.331100] [Firmware Bug]: ACPI: BIOS_OS(Linux) query ignored

[0.968555] [Firmware Bug]: ACPI: No _BQC method, cannot determine initial brightness


If I reboot and check brightness, I find that the value has reverted back to 0
and if I startx at that point I end up with a black screen that the only thing I have
found that works is either reboot or halt.

Anyone want to take a crack at it?
Back to top
View user's profile Send private message
Hypnos
Advocate
Advocate


Joined: 18 Jul 2002
Posts: 2867
Location: Omnipresent

PostPosted: Thu Nov 29, 2012 2:13 am    Post subject: Reply with quote

Googled your bug, found this.
_________________
Personal overlay | Simple backup scheme
Back to top
View user's profile Send private message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Fri Nov 30, 2012 3:15 am    Post subject: Reply with quote

Hello again,

I followed the previous reference and ended up at:

http://cannibalcandy.wordpress.com/2011/02/18/dsdt-editing-put-an-end-to-your-acpi-woes/

Reading through the site about DSDT editing: Put an end to your ACPI woes. I thought, I can handle this. The example
only had 5 errors to deal with. When I tried it though:

Code:
(chroot) Myrlynn# iasl -tc dsdt.dsl

Intel ACPI Component Architecture
ASL Optimizing Compiler version 20090123 [Nov 14 2012]
Copyright (C) 2000 - 2009 Intel Corporation
Supports ACPI Specification Revision 3.0a


Maximum error count (200) exceeded
dsdt.dsl  2475:                 If (And (GPL0, 0x0800))
Error    4063 -       Object does not exist ^  (GPL0)

dsdt.dsl  2507:                     If (^^PCI0.LPCB.EC0.BAL0)
Error    4064 -                                            ^ Object not found or not accessible from scope (^^PCI0.LPCB.EC0.BAL0)

                    ############ note: more stuff ##########

Maximum error count (200) exceeded
ASL Input:  dsdt.dsl - 18038 lines, 646005 bytes, 4940 keywords
Compilation complete. 201 Errors, 0 Warnings, 0 Remarks, 2 Optimizations

Maximum error count (200) exceeded
Segmentation fault


Which leads me to the question: WHY?

Surely I'm not expected to make 200+ corrections in a DSDT file to eliminate my problem. Not only that, but evidently it will have to be done everytime I upgrade my kernel. I'm dual booted with Arch Linux and am not having any of these problems. What is being handled differently and why to create this black screen problem?

I like a challenge but if this truly is my answer then I'll probably say adios. I'm hoping that there is someone out there that is a whole lot smarter than I am that is willing to help and has a simplified answer.

In the meantime let me say thanks again to those of you who have tried to help me.
Back to top
View user's profile Send private message
Hypnos
Advocate
Advocate


Joined: 18 Jul 2002
Posts: 2867
Location: Omnipresent

PostPosted: Fri Nov 30, 2012 3:27 am    Post subject: Reply with quote

I'm sure there's an easy answer, since I doubt Arch has a special DSDT for your machine :) Possibly some kernel or Xorg patch Gentoo does not apply, so may you can look at their repository and see what gets applied.

What is your machine BTW? Have you searched around by model #?
_________________
Personal overlay | Simple backup scheme
Back to top
View user's profile Send private message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Fri Nov 30, 2012 3:58 am    Post subject: Reply with quote

Hypnos,

My machine is a laptop HP pavillion DV4i. Intel core i3 cpu M330 @ 2.13 GHz. The BIOS is vendor: Insyde version: F.0F release date: 01/07/2011.

In my kernel I have been building as: Processor family (Core 2/newer Xeon) --->

And No I haven't tried searching around model # but now that you mention it, I'll try.

Thanks.
Back to top
View user's profile Send private message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Sun Dec 09, 2012 12:31 am    Post subject: Reply with quote

Hello Everyone,

UPDATE!

I was getting pretty ticked so I trashed the gentoo install and tried funtoo again just for kicks. Along the way I stumbled across changing the brightness value and being able to startx. So, I unloaded funtoo and I'm back to gentoo because I really want to solve this problem.

Here is where I am: I can boot up to runlevel 3 where I have a login prompt. At that point everything looks good. If I log in as root, I can

Code:
echo 20 > /sys/class/backlight/acpi_video0/brightness


I can then startx and arrive at twm with xterm or I can login as user and in my case startx will open openbox.

That's all fine and great but when I reboot or restart I find that the brightness value has been reset to 0. That is why in the past when I tried something like ctl-alt-bkspace or ctl-alt-f2 all I was seeing was a black screen. It seems that at login my screen is fine even though the brightness value is 0 but as soon as I startx the brightness value kicks in and it's done for unless I'm able as root to type in the blind the echo value to the brightness file to turn it back on.

So, yes I've got my system usable but I consider it half-assed. Surely there is something I'm missing that would make this right. Why is the brightness value defaulting back to zero? Is anyone else having this problem?
Back to top
View user's profile Send private message
krinn
Advocate
Advocate


Joined: 02 May 2003
Posts: 4150

PostPosted: Sun Dec 09, 2012 12:50 am    Post subject: Reply with quote

try make sure acpi daemon is install and running maybe.
Code:
emerge -pv sys-power/acpid
/etc/init.d/acpid status


install, run, add to default runlevel:
Code:
emerge -a sys-power/acpid
/etc/init.d/acpid start
rc-update add acpid default


You could still force the value on each load with local, or hack X to set the brigthness on each run...
But for now, let's just try the acpid way
Back to top
View user's profile Send private message
Hypnos
Advocate
Advocate


Joined: 18 Jul 2002
Posts: 2867
Location: Omnipresent

PostPosted: Sun Dec 09, 2012 12:54 am    Post subject: Reply with quote

You have an ATI Radeon GPU, right? Maybe this bug is not being handled for your particular chipset/motherboard combination.

You'll have to get upstream to fix it or find why it does not occur in Arch.

In the meantime, you can automate your workaround by putting it in an script in /etc/local.d (make sure to make the file executable!) . This will get executed on startup, so you don't have to type it in manually every time.

If the phenomena occurs with sleep/suspend, that has a different workaround ...
_________________
Personal overlay | Simple backup scheme
Back to top
View user's profile Send private message
Myrlynn
n00b
n00b


Joined: 22 Nov 2012
Posts: 10

PostPosted: Sun Dec 09, 2012 4:57 pm    Post subject: Reply with quote

Hello again,

I tried the acpid daemon to no effect. But thanks for the input krinn.

I'm running with intel however I did check out the "bug" you posted Hypnos. It at least gives me food for thought. In the meantime I did run a script in /etc/local.d. Everything seems to be working now even though that seems like a cheesy solution.

I probably won't try hacking the X in the near future. In fact I may take the approach of waiting to see if upgrades will handle the problem.

So, I assume this thread should be marked as solved. I'm new enough at this that I don't know the protocol or what my responsibilities are. If I'm supposed to be doing something, please instruct me.

Again, thanks to all those who looked and tried to help me. Hopefully this thread has been of assistance to others with a similar problem.
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 31711
Location: 56N 3W

PostPosted: Sun Dec 09, 2012 5:27 pm    Post subject: Reply with quote

Myrlynn,

You may edit the title of the first post in the thread to indicate the problem is solved, if you wish.
There is no real protocol for this. It will help other readers to know that you reached a solution.
_________________
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
Hypnos
Advocate
Advocate


Joined: 18 Jul 2002
Posts: 2867
Location: Omnipresent

PostPosted: Sun Dec 09, 2012 10:29 pm    Post subject: Reply with quote

Instead of adding "[SOLVED]" to the thread title, you can add "[WORKAROUND]".
_________________
Personal overlay | Simple backup scheme
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Desktop Environments All times are GMT
Page 1 of 1

 
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