You are not logged in.
So here is the issue. In 3 diffrent arch installs and my new archbang install, I have been plagued by a strange glitch that seems to persist. This glitch has a few diffrent varitations, the most prevalent being the screen wrapping from an uncentered end to the other side. Here are some pictures to clarify.
Arch
====
http://imgur.com/qOJWdR5
http://imgur.com/ECU4FpU
Archbang
=======
http://imgur.com/kGqG1Y7
The other variations of the error are as follows:
==============================
-A flickering screen after grub bootup, highly distorted.
-A black screen after grub bootup and intitial system processes.
-The same error, but with the rift closer to the middle of the screen.
-A normal bootup
Here are some system specs as well
===================
Ram- 4 gigs of corsair vengance
Cpu- AMD Phenom II x4 965
Graphics card- Raedon HD6670
The screen for my mint and archbang live usbs are just fine. The screen during the initial install is usually just fine as well. I am postulating that this error is arch-specific, as my old ubuntu and mint installs never had this problem. I am also curious because for my arch installs, even before I installed graphics drivers and mesa and whatnot, I still would get the error. Other then this strange glitch, my arch based operating systems would work just fine and dandy.
Here are some fixes I tried
==================
-Adjusting my monitor- This seemingly obvious fix did abslolutely nothing. It would move the screen over, but a black space would appear where the screen was, almost as if I tried doing the same thing with a fixed screen..
-Installing a display manager (this was in arch)- The screen was still glitched over
-Trying a new monitor- The same glitch happened, although both monitors are the exact same Dell monitor, and the only distinguishing feature is the size.
Here are some theories I have
====================
-The most likely issue is my graphics card itself, as Raedon and Linux tend to have issues as I have heard.
-A corrupted iso file in my live usb (this is highly unlikely since most of the arch install process is done via downloading files from the web). However I used the same flashdrive for both live usbs. I made the archbang one by formatting the previous arch live usb (in FAT32) and then using the dd method.
-A possible monitor size issue, but I feel as though this is highly unlikely
-A corrupted HDD (this is just me grasping at straws, therefore this possibilty is possible, but highly unlikely as other symptoms would have presented themselves).
Does anybody have any idea what to do at this point?
Offline
I have had a similar problem since the 3.14 kernel. The lts kernel did not have this issue, so I have been running it. Today after trying a lot of things I finally fixed it. I added the following kernel parameter to my boot loader, radeon.dpm=1.
From the ATI Wiki
"dpm" is only supported on R6xx and newer chips.
See http://www.x.org/wiki/RadeonFeature/#index3h2 for details.
Intel Core i7-4770
Radeon HD 8490 / R5 235X OEM
Open source ati driver
Dual DVI monitors
Last edited by Talonas (2014-06-27 23:14:34)
Offline
I have had a similar problem since the 3.14 kernel. The lts kernel did not have this issue, so I have been running it. Today after trying a lot of things I finally fixed it. I added the following kernel parameter to my boot loader, radeon.dpm=1.
From the ATI Wiki
"dpm" is only supported on R6xx and newer chips.
See http://www.x.org/wiki/RadeonFeature/#index3h2 for details.Intel Core i7-4770
Radeon HD 8490 / R5 235X OEM
Open source ati driver
Dual DVI monitors
HOLY MOTHER OF ALL THINGS LINUX! YOU HAVE FINALLY PROVIDED ME WITH SOMETHING THAT WORKS! YOU HAVE GIVEN ME THE FIX I HAVE BEEN PILLAGING THE INTERNET FOR! MAY THE OLD GODS THEMSELVES BOW IN YOUR PRESENCE!
Offline
HOLY MOTHER OF ALL THINGS LINUX! YOU HAVE FINALLY PROVIDED ME WITH SOMETHING THAT WORKS! YOU HAVE GIVEN ME THE FIX I HAVE BEEN PILLAGING THE INTERNET FOR! MAY THE OLD GODS THEMSELVES BOW IN YOUR PRESENCE!
I'm glad things are going well. Please never use all caps again. kthnxbai.
All the best,
-HG
Offline
I'm having a very similar problem. I tried installing Arch on a laptop of mine I'm not really using anymore, and every time I'm either trying to startx or startxfce4 my screen looks like this: http://imgur.com/7wZohhw.png (sorry for the blurry image).
As you can see only roughly 40% of the screen are drawn at the top of the monitor, whereas the bottom 60% are artifacts. Note that the system is perfectly responsive, the right click context menu opens up etc. It is safe to assume that this is not a hardware error since the device was running a Win7 installation flawlessly just 2 hours ago. The laptop in question is a stock ASUS X7BJN with a built-in Nvidia GTX 425M graphics card. I tried switching from proprietary Nvidia drivers to the nouveau ones, however all that does is forcing the X server to terminate on startup (no devices detected, etc.). Forcing vesa in the boot options didn't help either.
I can't even play around with the xorg.conf since this will lead me to the same error as above (no devices), I have to delete xorg.conf every time I want to get X running in the first place.
Hooking up an external monitor and trying to move the alignment did nothing for me, same as OP. Unforunately the solution proposed here won't work for me since I don't have an AMD card.
Offline
I'm having a very similar problem. I tried installing Arch on a laptop of mine I'm not really using anymore, and every time I'm either trying to startx or startxfce4 my screen looks like this: http://imgur.com/7wZohhw.png (sorry for the blurry image).
As you can see only roughly 40% of the screen are drawn at the top of the monitor, whereas the bottom 60% are artifacts. Note that the system is perfectly responsive, the right click context menu opens up etc. It is safe to assume that this is not a hardware error since the device was running a Win7 installation flawlessly just 2 hours ago. The laptop in question is a stock ASUS X7BJN with a built-in Nvidia GTX 425M graphics card. I tried switching from proprietary Nvidia drivers to the nouveau ones, however all that does is forcing the X server to terminate on startup (no devices detected, etc.). Forcing vesa in the boot options didn't help either.
I can't even play around with the xorg.conf since this will lead me to the same error as above (no devices), I have to delete xorg.conf every time I want to get X running in the first place.
Hooking up an external monitor and trying to move the alignment did nothing for me, same as OP. Unforunately the solution proposed here won't work for me since I don't have an AMD card.
While I do not know much about nvidia graphics cards, I do know that the command that I entered into the grub boot (and made persistant through other means) was a command that edited the power profile of my graphics card. It was suggested to me earlier to edit the power profile, but it didn't work because I didn't mkconfig the changes. However, it would have (in theory) worked. Perhaps you need to find a way to edit your nvidia graphcis driver power profile, and tinker with it until you find a setting that works.
Offline
The situation is worse with the 3.15 kernel with my 6850. I hadn't seen instability like this in the radeon driver since before I switched from mesa-git to normal mesa at the 10.2 -> 10.3 branching. The 3.14 kernel worked nearly perfectly for me after that. Since 3.15.1 things have gotten really bad at times. I'm currently using 3.14.6 to try to prove that things are better before I file a bug report. It's either that or I finally replace this card with my ancient nVidia card that worked better in every single way.
I am curious for any of you that are seeing this problem. Does DPM actually work for you or is it just the only way to get the card to run? DPM was doing pretty well for me for a while, but with the 3.15.x series, the card is on high power nearly always. I'm *guessing* that the graphics glitches are happening when the card is trying to change power states, but it's hard to tell when nothing is being logged anywhere anymore.
--EDIT--
User error on that last line. I forgot that I need to search for "power" and not "radeon" to get clock info out of the journal.
Last edited by skottish (2014-06-28 17:02:52)
Offline