FS#12304 - xf86-video-intel behaves strangely

Attached to Project: Arch Linux
Opened by Matěj Týč (bubla) - Monday, 01 December 2008, 13:16 GMT
Last edited by Jan de Groot (JGC) - Wednesday, 13 May 2009, 20:14 GMT
Task Type Bug Report
Category Packages: Extra
Status Closed
Assigned To Jan de Groot (JGC)
Architecture x86_64
Severity High
Priority Normal
Reported Version None
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 5
Private No

Details

Description:
I have upgraded to the newest intel drivers and I have unsatisfactory results.
Sometimes the screen goes to black for a second and then restores and sometimes things go completely wrong and I have to restart the computer.
I have Asus P5Q-VM with G45 chipset and Intel GMA X4500HD integrated graphics card (I had to use VESA before)

Additional info:
* package version(s)
xf86-video-intel 2.4.3-1

* config and/or log files etc.
What should I look at?

Steps to reproduce:
Just work with your computer and you will notice sooner or later :-)
This task depends upon

Closed by  Jan de Groot (JGC)
Wednesday, 13 May 2009, 20:14 GMT
Reason for closing:  Fixed
Comment by Jonathan Steel (jsteel) - Monday, 01 December 2008, 22:55 GMT
I have a similar problem. After updating to the latest Intel driver, and after starting X, the display flickers for a while and then freezes. The image each time is black with a few scrabled paterns at the top (while flickering and when frozen). I can use "ctrl, alt and F1" to switch to a console (though I cannot see that I'm doing this, the frozen image stays), log in and remove xorg.conf. This then resorts to the Vesa driver luckily!
Comment by Rabi Poudyal (rabspd) - Tuesday, 02 December 2008, 06:07 GMT
I am also facing the same problem. I just updated video driver from xf86-video-i810 to xf86-video-intel, and after that GUI stops to load. Log file is attached here.
Comment by Jonathan Steel (jsteel) - Tuesday, 02 December 2008, 07:34 GMT
I used to use the i810 driver but when I go back to it, I get the same error in my Xorg log as above (no screens found) after running an Xorg -configure and testing the configuration. (The Intel driver does not work for me, see my message above). Both the i810 and Intel drivers worked before; though the i810 driver displayed fonts better I found.
Comment by Jan de Groot (JGC) - Tuesday, 02 December 2008, 11:16 GMT
Please keep your i810 bugs outside this bugreport, they're totally unrelated. People upgrading from i810 drivers should fix their config to reflect the new driver. I won't accept bugs due to wrong xorg.conf files.
Comment by Jonathan Steel (jsteel) - Tuesday, 02 December 2008, 11:28 GMT
JGC,
Sorry maybe I was not clear. An X -configure, X -config /root/xorg.conf.new produces these problems. I mentioned the i810 as it also does not work after doing this; so it may not be unreleated (even though it produces differnt problems). Basically neither work.
Comment by Joshua Cole (purpledragon77) - Wednesday, 03 December 2008, 03:44 GMT
I am having a similar problem, I was doing a brand new arch install (after using another distro for awhile with this intel driver and it freezes after I configure X. It installed without flaw in the past.
Comment by Glenn Matthys (RedShift) - Friday, 05 December 2008, 10:52 GMT
Please post your xorg.conf.
Comment by Matěj Týč (bubla) - Monday, 08 December 2008, 11:07 GMT
Here it is (it was initially generated by hwd)
   xorg.conf (11.9 KiB)
Comment by Jan de Groot (JGC) - Monday, 08 December 2008, 11:10 GMT
Could you attach your logfile also?
Comment by Jonathan Steel (jsteel) - Monday, 08 December 2008, 22:22 GMT
I have resolved my issue. After an Xorg -configure it produced two screens in my xorg.conf. After removing the second screen it works perfectly, sorry for not looking at this before reporting the issue. Thanks.
Comment by Jan de Groot (JGC) - Tuesday, 09 December 2008, 07:36 GMT
Reopened: because the bug hijacker fixed his problem, it doesn't mean the original reporter doesn't have problems with his X4500HD anymore.
Comment by Matěj Týč (bubla) - Tuesday, 09 December 2008, 12:42 GMT
Here is /var/log/Xorg.0.log
And yes, I still experience the strange behaviour described in the report
Comment by Jan de Groot (JGC) - Tuesday, 09 December 2008, 12:57 GMT
What screen do you have? HP? Googling around for the disconnected device in the last lines of your Xorg.0.log gives me some results where exactly the same problems appear on Radeon chips.
Comment by Matěj Týč (bubla) - Tuesday, 09 December 2008, 13:03 GMT
I have HP LP2065. It is connected via DVI, nothing is connected to VGA
Comment by Rabi Poudyal (rabspd) - Wednesday, 10 December 2008, 07:44 GMT
I generated another xorg.conf by using the command xorgcfg. That xorg.conf worked for me. The xorg.conf is attached here. Previously all xorg.conf generated via hwd or xorgconfig was not working.
Comment by Jan de Groot (JGC) - Wednesday, 10 December 2008, 07:57 GMT
Please don't hijack this bugreport for other problems. This bug is about flickering issues, not about configuration errors.
Comment by Matěj Týč (bubla) - Thursday, 11 December 2008, 12:54 GMT
Yes, I still experience unpredictable (but not frequent) display blackouts.
I can see that the status of this bugreport is "waiting for response", what data would you like?
Comment by Anonymous Submitter - Wednesday, 17 December 2008, 04:45 GMT
I could be suffering from the similar problem of display blackouts. I run my work PC 24x7, with X running. While the display blackout is somewhat unpredictable, it does seem to take aruond a week to a week and a half. PC is still working, and last time it happened I was able to switch to a different console VC, and attempt to restart X. That didn't work. I was only able to get back into X after I rebooted the PC. That seems to me to be a bug with the driver that is leaving the hardware in a non-operational state.

Here's the error I got in /var/log/Xorg.0.log last time it happened:

--
Error in I830WaitLpRing(), timeout for 2 seconds
pgetbl_ctl: 0x00000001 getbl_err: 0x00000000
ipeir: 0x00000000 iphdr: 0x02000011
LP ring tail: 0x00018e10 head: 0x00018e1c len: 0x0001f001 start 0x00000000
eir: 0x0000 esr: 0x0001 emr: 0xffff
instdone: 0xfa41 instpm: 0x0000
memmode: 0x00000306 instps: 0x80007826
hwstam: 0xfffe ier: 0x0002 imr: 0x0000 iir: 0x0040
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring at virtual 0x7f30b6f11000 head 0x18e1c tail 0x18e10 count 32765
Ring end
space: 4 wanted 32
(II) intel(0): [drm] removed 1 reserved context for kernel
(II) intel(0): [drm] unmapping 8192 bytes of SAREA 0x1efff000 at 0x7f30cb063000
(II) intel(0): [drm] Closed DRM master.

Fatal server error:
lockup

(II) AIGLX: Suspending AIGLX clients for VT switch
(II) intel(0): xf86UnbindGARTMemory: unbind key 0
(II) intel(0): xf86UnbindGARTMemory: unbind key 1
(II) intel(0): xf86UnbindGARTMemory: unbind key 2
(II) intel(0): xf86UnbindGARTMemory: unbind key 3
(II) intel(0): xf86UnbindGARTMemory: unbind key 4

--

Device section from Xorg is nothing but driver defaults:

--
Section "Device"
### Available Driver options are:-
### Values: <i>: integer, <f>: float, <bool>: "True"/"False",
### <string>: "String", <freq>: "<f> Hz/kHz/MHz"
### [arg]: arg optional
#Option "NoAccel" # [<bool>]
#Option "SWcursor" # [<bool>]
#Option "ColorKey" # <i>
#Option "CacheLines" # <i>
#Option "Dac6Bit" # [<bool>]
#Option "DRI" # [<bool>]
#Option "NoDDC" # [<bool>]
#Option "ShowCache" # [<bool>]
#Option "XvMCSurfaces" # <i>
#Option "PageFlip" # [<bool>]
Identifier "Card0"
Driver "intel"
VendorName "Intel Corporation"
BoardName "Integrated Graphics Controller"
BusID "PCI:0:2:0"
EndSection
--
Comment by Andrew Mellor (quantumphaze) - Saturday, 20 December 2008, 13:34 GMT
Try adding this to the Device section:
Option "FramebufferCompression" "off"

It fixed the flickering (I would describe it like the screen was jumping to the left) I experience on the external VGA on my Acer Aspire 3680. The screen was stable on the LVDS.
Comment by Matěj Týč (bubla) - Monday, 05 January 2009, 11:37 GMT
I have added that option to my xorg.conf and it still goes black from time to time.
Comment by T (zoryfl) - Monday, 19 January 2009, 19:21 GMT
I am experiencing a similar problem with my Dell Latitude E4300.

Using xf86-video-intel version 2.4.3.
My problem is that my screen uses to freeze 'randomly', in idle-mode as well as when I am working at it.
I am able to move my mouse after the screen is frozen, but nothing else works.

xorg.conf and xorg.0.log attached.
Comment by Matěj Týč (bubla) - Wednesday, 25 February 2009, 10:43 GMT
What about updating the driver? The current version is now 2.6.0
Comment by Jan de Groot (JGC) - Wednesday, 25 February 2009, 10:48 GMT
The current version is 2.6.2, not 2.6.0 ;)

I'll update xorg-server to the latest snapshot from the server-1.6 branch in testing this week. This will require a full update and/or rebuild of all drivers, including intel.
Comment by Paride Legovini (torn) - Monday, 02 March 2009, 19:01 GMT
I'm having some strange problems with the intel driver too. It's *slow*, I mean, the whole xorg is slow. Dragging/resizing windows takes time every time, while I have no such problems with the vesa driver. I tried the xf86-video-intel-newest (2.6.2) from AUR, but it didn't help. I use the same xorg.conf and the same kernel in debian (xorg 7.4, xserver 1.5.99.902, see http://packages.debian.org/experimental/xserver-xorg), and it works fine. I hope the new release will help, I'm waiting for it impatiently :)
Comment by Ray Rashif (schivmeister) - Friday, 06 March 2009, 22:46 GMT
I believe the slow drawing is inherent and unavoidable without repercussions.
Comment by Paride Legovini (torn) - Friday, 06 March 2009, 23:49 GMT
Well, in Debian it works, so it isn't so unavoidable. Anyway, the packages in testing seem to have fixed that.
Comment by Jan de Groot (JGC) - Friday, 10 April 2009, 21:10 GMT
Is this still an issue with X from testing?
Comment by Matěj Týč (bubla) - Tuesday, 14 April 2009, 09:15 GMT
It is an issue with the
xf86-video-intel 2.6.3-3
package.
This is quite bad, I will file a bug upstream later this week...
Comment by Matěj Týč (bubla) - Tuesday, 14 April 2009, 09:55 GMT
Not to mention that glxgears crash when using UXA and with EXA I get FPS of 22 when on fullscreen... Is this normal?
Comment by Matěj Týč (bubla) - Wednesday, 15 April 2009, 12:04 GMT
OK, I have found out that this "flickering" bug has been reported upstream quite long ago.
It is now considered fixed. Could you take a look here
http://bugs.freedesktop.org/show_bug.cgi?id=17805
and perhaps update the driver?
Comment by bowo (potamota) - Thursday, 16 April 2009, 15:07 GMT
random crash, probably intel drm fault

kernel.log:
kernel: X[4834]: segfault at 10 ip b7a05579 sp bfc08960 error 4 in libdrm_intel.so.1.0.0[b7a04000+8000]
kernel: [drm:i915_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 1

Xorg.0.log:
Backtrace:
0: /usr/bin/X(xorg_backtrace+0x3b) [0x813278b]
1: /usr/bin/X(xf86SigHandler+0x51) [0x80dda61]
2: [0xb7fec400]
3: /usr/lib/xorg/modules/drivers//intel_drv.so [0xb7a5989b]
4: /usr/lib/xorg/modules/drivers//intel_drv.so(i830_crtc_dpms_video+0x4b) [0xb7a5991b]
5: /usr/lib/xorg/modules/drivers//intel_drv.so [0xb7a424de]
6: /usr/bin/X(xf86DPMSSet+0x80) [0x80eadc0]
7: /usr/bin/X(xf86SaveScreen+0x3b) [0x80eb64b]
8: /usr/bin/X(dixSaveScreens+0x23f) [0x807b3bf]
9: /usr/bin/X [0x812fa65]
10: /usr/bin/X [0x812fe87]
11: /usr/bin/X(WaitForSomething+0x52f) [0x81304bf]
12: /usr/bin/X(Dispatch+0x7e) [0x808ca7e]
13: /usr/bin/X(main+0x3bd) [0x8071f3d]
14: /lib/libc.so.6(__libc_start_main+0xe5) [0xb7bc36c5]
15: /usr/bin/X [0x8071401]
Fatal server error:
Caught signal 11. Server aborting
Comment by Matěj Týč (bubla) - Tuesday, 12 May 2009, 12:14 GMT
The original bug (so-called flickering) is fixed with the new drivers (xf86-video-intel 2.7.0-3).
However, the graphics stuff is still very very slow, but I guess that the only solution is to wait for better drivers.

Loading...