Bug 1353 - [Intel/i830] Error in I830WaitLpRing(). Server lockup
Summary: [Intel/i830] Error in I830WaitLpRing(). Server lockup
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: 6.9.0
Hardware: x86 (IA32) Linux (All)
: high critical
Assignee: Alan Hourihane
QA Contact:
URL:
Whiteboard:
Keywords:
: 1966 4466 (view as bug list)
Depends on:
Blocks:
 
Reported: 2004-09-13 05:58 UTC by Scott Webber
Modified: 2007-01-05 12:54 UTC (History)
9 users (show)

See Also:
i915 platform:
i915 features:


Attachments
Log file containing Error in I830WaitLpRing() (54.52 KB, text/plain)
2005-02-08 10:05 UTC, Kevin J. Butler
no flags Details
X Server log for 6.8.2 containing same or similar bug (57.11 KB, text/plain)
2005-03-13 17:48 UTC, peter
no flags Details
The logs for GDM describing the similar error. (2.99 KB, text/plain)
2006-01-11 17:14 UTC, Abhas Kumar
no flags Details

Description Scott Webber 2004-09-13 05:58:46 UTC
When I leave my work machine run overnight, the screen is black and everything
is locked up.  The keyboard is unresponsive and the log has this:

Error in I830WaitLpRing(), now is 3544, start is 1543
pgetbl_ctl: 0x17f60001 pgetbl_err: 0x49
ipeir: 0 iphdr: 7f1400cf
LP ring tail: 30 head: 0 len: 1f001 start 0
eir: 0 esr: 10 emr: ff7b
instdone: 68c1 instpm: 0
memmode: 0 instps: 53
hwstam: ffff ier: 0 imr: ffff iir: 0
space: 131016 wanted 131064
                                                                               
                                                                               
                    
Fatal server error:
lockup
                                                                               
                                                                               
                    
                                                                               
                                                                               
                    
Please consult the The X.Org Foundation support
         at http://wiki.X.Org
 for help.
Please also check the log file at "/var/log/Xorg.0.log" for additional information.
                                                                               
                                                                               
                    
Error in I830WaitLpRing(), now is 6547, start is 4546
pgetbl_ctl: 0x17f60001 pgetbl_err: 0x49
ipeir: 0 iphdr: 7f1400cf
LP ring tail: 38 head: 0 len: 1f001 start 0
eir: 0 esr: 10 emr: ff7b
instdone: 68c1 instpm: 0
memmode: 0 instps: 53
hwstam: ffff ier: 0 imr: ffff iir: 0
space: 131008 wanted 131064
                                                                               
                                                                               
                    
FatalError re-entered, aborting
lockup


I found a thread that has a possible fix for this issue in an old Xfree86
discussion:

http://www.mail-archive.com/xfree86@xfree86.org/msg10075.html

However, I didn't find any similar code on my Fedora2 machine yet.
Comment 1 Ed Koster 2004-10-18 10:58:27 UTC
The same issue is happening on my box, which also runs FC2.  During X's startup
my log additionally sports the following error, which may neither be related nor
important:

  ...
  drmOpenByBusid: drmGetBusid reports PCI:0:2:0
  (II) I810(0): [drm] drmSetBusid failed (9, PCI:0:2:0), Device or resource busy
  (EE) I810(0): [dri] DRIScreenInit failed. Disabling DRI.
  (II) I810(0): Updated framebuffer allocation size from 10240 to 16384 kByte
  ...

However, saying that "everything is locked up" is a little harsh in my opinion.
 It's just a matter of the X display having crashed.  I can still perform other
functions remotely, including VNCing to another display.  I can even attempt to
restart X, which sends some video output to the monitor but ultimately fails and
goes "unresponsive" again.
Comment 2 Donnie Berkholz 2004-11-15 09:59:42 UTC
Duplicate of #1379 and #1771.
Comment 3 Matthias Hopf 2004-11-25 07:06:43 UTC
Please post your xorg.conf and Xorg.0.log. Which kernel version?
Comment 4 Alan Hourihane 2005-02-07 06:58:32 UTC
Scott, is this still a problem. If so, can you post a log ?
Comment 5 Scott Webber 2005-02-07 07:25:31 UTC
I haven't experienced this anymore, but I think I worked around the problem. 
The problem seemed to be caused when my machine changed video mode or resolution
when the screensaver kicked in - so I just changed my screensaver to blank so
that the res wouldn't change anymore.

Ed Koster was right, only X is locked.  You can login remotely and kill X, but
if I recall, I don't think it restarts properly.

I'm sorry I don't have any more information.  Unfortunately this is my work
machine so I can't really spare time to try to find out anything else.
Comment 6 Kevin J. Butler 2005-02-08 10:05:13 UTC
Created attachment 1864 [details]
Log file containing Error in I830WaitLpRing()

This is currently repeatable on my machine. For more info, contact
freedesktop-kbutler AT sabaydi.com
Comment 7 Alan Hourihane 2005-02-18 10:00:43 UTC
Kevin, unfortunately, your log is of XFree86 4.3.0 which is pretty old now and
that should be reported to the XFree86 bugzilla database and not X.Org.

If you can retry with X.Org 6.8.2 and see if it's still a problem that would be
good.
Comment 8 Egbert Eich 2005-03-03 06:56:11 UTC
*** Bug 1966 has been marked as a duplicate of this bug. ***
Comment 9 Alan Hourihane 2005-03-04 07:49:05 UTC
Closing, no furthur log data produced. Re-open if there's trouble again.
Comment 10 peter 2005-03-09 13:09:42 UTC
It is very likely that I have the very same problem (some numbers are different)
and it is easily reproducable.

I have a Fedora Core 3 Installation running linux 2.6.10 (fedora errata kernel)
using the FC provided errata Xorg X11 server (called 6.8.1-12.FC3.21). 

Here is the message I get:

Error in I830WaitLpRing(), now is 10611, start is 8610
pgetbl_ctl: 0xffe0001 pgetbl_err: 0x0
ipeir: 0 iphdr: 7dac80
LP ring tail: 30 head: 0 len: 1f001 start 7fdf000
eir: 0 esr: 1 emr: ffff
instdone: ffc1 instpm: 0
memmode: 108 instps: 20
hwstam: ffff ier: 0 imr: ffff iir: 0
space: 131016 wanted 131064

Fatal server error:
lockup



It happens when I start acrobat reader (5.0.10) or with one particular latex dvi
file containing some xfig-generated EPS stuff when ghostscript has to
(re)display it.  I also saw this when setting up the screensaver.

This is extremly annoying, as I cannot use this machine (my laptop) for any
serious work beyond web-surfing.

I did not see this problem using XFree86 under 2.4.2x.

I would prefer to not have to compile X11 for myself, so if you have RPMs of
6.8.2 I might give them a try.
Comment 11 Richard Liska 2005-03-10 04:22:45 UTC
(In reply to comment #10)
Try to turn off dri in your /etc/X11/xorg.conf:

Section "Device"
	Identifier  "Videocard0"
	Driver      "i810"
	Option     "DRI" "False"
	VendorName  "Videocard vendor"
EndSection

It might avoid X server crash, however show down graphics. It did helped me 
in FC2.


> It is very likely that I have the very same problem (some numbers are different)
> and it is easily reproducable.
> 
> I have a Fedora Core 3 Installation running linux 2.6.10 (fedora errata kernel)
> using the FC provided errata Xorg X11 server (called 6.8.1-12.FC3.21). 
> 
> Here is the message I get:
> 
> Error in I830WaitLpRing(), now is 10611, start is 8610
....
> 
> 
> It happens when I start acrobat reader (5.0.10) or with one particular latex dvi
> file containing some xfig-generated EPS stuff when ghostscript has to
> (re)display it.  I also saw this when setting up the screensaver.
> 
> This is extremly annoying, as I cannot use this machine (my laptop) for any
> serious work beyond web-surfing.
> 
> I did not see this problem using XFree86 under 2.4.2x.
> 
> I would prefer to not have to compile X11 for myself, so if you have RPMs of
> 6.8.2 I might give them a try.
Comment 12 peter 2005-03-10 16:45:32 UTC
(In reply to comment #11)
> (In reply to comment #10)
> Try to turn off dri in your /etc/X11/xorg.conf:
> 
> Section "Device"
> 	Identifier  "Videocard0"
> 	Driver      "i810"
> 	Option     "DRI" "False"
> 	VendorName  "Videocard vendor"
> EndSection
> 
> It might avoid X server crash, however show down graphics. It did helped me 
> in FC2.
> 
> 
Thanks for that tip: it seems to have solved one of the problems, but not all:
Acrobat Reader now at least starts (and it can show at least some PDFs), but the
latex file still crashes the server with the same error (again, different numbers).
Comment 13 peter 2005-03-13 17:43:29 UTC
I gave 6.8.2 a try: same crash with unchanged xorg.conf (starting acroread
caused the problem).

This is my current Device section in xorg.conf:

Section "Device"
        Identifier  "Videocard0"
        Driver      "i810"
        VendorName  "Videocard vendor"
        BoardName   "Intel 830"
        Option "SWCursor" "true"
        Option "Dri" "False"
        Option "XaaNoSolidFillRect" "false"
EndSection

The SWCursor and XaaNoSolidFillRect lines fix other problems I have seen. The
Dri line was added in response to a suggestion in this bug. I will attach the
full Xorg.0.log file to this bug.

I now reverted back to 6.8.1 for gnome reasons...
Comment 14 peter 2005-03-13 17:48:45 UTC
Created attachment 2096 [details]
X Server log for 6.8.2 containing same or similar bug
Comment 15 Alan Hourihane 2005-04-08 09:12:11 UTC
Ok I'll take a look over here to see if I can replicate it with acrobat reader.
Comment 16 Alan Hourihane 2005-04-08 09:19:38 UTC
O.k. tried acrobat reader 5.0.9 here and it works fine.

You can try downloading my latest i810 driver from

http://www.fairlite.demon.co.uk/intel.html

and copying it to /usr/X11R6/lib/modules/drivers, but make sure you back up the
old one.

If that doesn't cure it, then can you give me a pointer to a PDF file that
causes the problem ?
Comment 17 Alan Hourihane 2005-04-13 09:32:21 UTC
Peter, Have you had chance to try the new driver at all ?
Comment 18 Alan Hourihane 2005-04-20 11:00:13 UTC
Closing again. Re-open if there's still a problem with the driver from my homepage.
Comment 19 j^ 2005-06-10 08:42:36 UTC
i had the same issue using pitivi (http://pitivi.org)
using your the driver from http://www.fairlite.demon.co.uk/intel.html
i am able to start pitivi without X crashing, but it 
X caches now quite often playing back video in 
mplayer/totem/xine (using xv)

this is with DRI "on" and using xorg from ubuntu/hoary on
an ibm x30 / Intel 830m
Comment 20 Alan Hourihane 2005-06-10 08:52:52 UTC
Please attach a log from /var/log/Xorg.0.log 
Comment 21 j^ 2005-06-11 00:54:27 UTC
Created attachment 2874 [details]
xorg.conf.optiplex
Comment 22 j^ 2005-06-11 00:56:30 UTC
so i was able to narrow the crash down a bit, 
it always happens after playing video with xv
suspending/resuming with apm and playing video again.
Comment 23 Alan Hourihane 2005-06-11 01:28:48 UTC
You claimed you are using the driver from
http://www.fairlite.demon.co.uk/intel.html whereas the log shows that you are not.

Can you retry this and post a log if it crashes with the new driver.
Comment 24 j^ 2005-06-11 01:57:30 UTC
ok, downloaded the driver again, and now it seems to work. 
the driver from your website causes a green strip at the top
of the screen while switching vts/resuming, other than that,
all looks good; i could not crash the system right now. 
will let you know if i run into problems again... 

thanks for your work,
Comment 25 j^ 2005-06-11 02:17:25 UTC
Created attachment 2875 [details] [review]
New Iranian keyboard layout, based on proposed new standard

md5sum /usr/X11R6/lib/modules/drivers/i810_drv.o
f05681d9154e392d59a3c9f57d2e8940  /usr/X11R6/lib/modules/drivers/i810_drv.o

xvinfo
X-Video Extension version 2.2
screen #0
 no adaptors present

suspend / resume works, but i do not get XV at all.
Comment 26 Alan Hourihane 2005-06-11 05:13:18 UTC
Sorry, I'd disabled Xv in .100, I've uploaded .101 to re-enable it. Try that.
Comment 27 j^ 2005-06-11 07:20:59 UTC
Created attachment 2880 [details]
dmesg.optiplex

with 1.5.101 i get an X lockup again after suspend using xv
Comment 28 Alan Hourihane 2005-06-14 07:33:21 UTC
I've uploaded 1.5.105 - can you try that and report back.
Comment 29 j^ 2005-06-14 08:30:37 UTC
with 1.5.105 xv worked without a hangup for several apm suspend/resume runs.
without any problem so far. 

great work and thank you for fixing this.
Comment 30 Alan Hourihane 2005-06-15 08:32:59 UTC
Closing.
Comment 31 Mike 2005-09-15 18:13:50 UTC
*** Bug 4466 has been marked as a duplicate of this bug. ***
Comment 32 Abhas Kumar 2006-01-11 17:14:18 UTC
Created attachment 4327 [details]
The logs for GDM describing the similar error.

I think this is the similar bug. I am using 

X Window System Version 6.8.2 (Debian 6.8.2.dfsg.1-11 20051129054125 David
Nusinow <dnusinow@debian.org>)

The error is there when I switch to a console terminal and then try to come
back to the Graphical console (screen 7). The X tries to restart, but in the
end computer become un-responsive.
Comment 33 Filipus Klutiero 2006-03-11 10:14:39 UTC
Note that I have experienced such a crash *once*, using xorg-x11 6.9.0.dfsg.1-4.
This bug should probably be reopened, but as I'm unable to reproduce, I'm not
going to be the owner.
I have also used X.org with i810 on the Z71A laptop where this just happened
during over 7 months without experiencing this problem.
Comment 34 lolo1984 2006-03-21 00:33:41 UTC
I still have the bug which appears when using some KDE screensavers.
Using tuxsaver (http://www.kde-look.org/content/show.php?content=4510), I click
on test screensaver , the 2/3 bottom stays white. Then trying to configure the
screensaver leads to this bug.

X Window System Version 6.9.0 (Debian 6.9.0.dfsg.1-4 20060114230205 David
Nusinow <dnusinow@debian.org>)
Release Date: 21 December 2005
X Protocol Version 11, Revision 0, Release 6.9
Build Operating System: Linux 2.6.15-1-686 i686 [ELF]
Current Operating System: Linux debian 2.6.15-stable #1 PREEMPT Sat Mar 18
21:26:12 CET 2006 i686
Build Date: 14 January 2006
        Before reporting problems, check http://wiki.X.Org
        to make sure that you have the latest version.
Module Loader present
OS Kernel: Linux version 2.6.15-stable (2.6.15-486-10.00.Custom) (root@debian)
(version gcc 4.0.3 (Debian 4.0.3-1)) #1 PREEMPT Sat Mar 18 21:26:12 CET 2006
[..]
(WW) I810(0): Bad V_BIOS checksum
[..]
(WW) I810(0): Detected stolen memory (8000 kB) doesn't match what the BIOS
reports (12288 kB)
[..]
(II) I810(0): [drm] DRM interface version 1.2
(II) I810(0): [drm] created "i915" driver at busid "pci:0000:00:02.0"
(II) I810(0): [drm] added 8192 byte SAREA at 0xe039f000
(II) I810(0): [drm] mapped SAREA 0xe039f000 to 0xb7a3c000
(II) I810(0): [drm] framebuffer handle = 0x98020000
(II) I810(0): [drm] added 1 reserved context for kernel
[..]
(II) I810(0): vgaHWGetIOBase: hwp->IOBase is 0x03d0, hwp->PIOOffset is 0x0000
(II) I810(0): xf86BindGARTMemory: bind key 7 at 0x007df000 (pgoffset 2015)
(II) I810(0): xf86BindGARTMemory: bind key 0 at 0x07fff000 (pgoffset 32767)
(II) I810(0): xf86BindGARTMemory: bind key 1 at 0x07ffb000 (pgoffset 32763)
(II) I810(0): xf86BindGARTMemory: bind key 3 at 0x07fea000 (pgoffset 32746)
(II) I810(0): xf86BindGARTMemory: bind key 2 at 0x07ffa000 (pgoffset 32762)
(II) I810(0): xf86BindGARTMemory: bind key 4 at 0x07c00000 (pgoffset 31744)
(II) I810(0): xf86BindGARTMemory: bind key 5 at 0x07a00000 (pgoffset 31232)
(II) I810(0): xf86BindGARTMemory: bind key 6 at 0x079f8000 (pgoffset 31224) 
(WW) I810(0): PRB0_CTL (0x0001f001) indicates ring buffer enabled
(WW) I810(0): PRB0_HEAD (0x00000000) and PRB0_TAIL (0x00000040) indicate ring
buffer not flushed
[..]
Error in I830WaitLpRing(), now is 3944, start is 1943
pgetbl_ctl: 0x1ffe0001 pgetbl_err: 0x0
ipeir: 0 iphdr: 7f0002e7
LP ring tail: 38 head: 0 len: 1f001 start 0
eir: 0 esr: 0 emr: ffff
instdone: ffc1 instpm: 0
memmode: 108 instps: 53
hwstam: ffff ier: 2 imr: 9 iir: a0
space: 131008 wanted 131064
(II) I810(0): [drm] removed 1 reserved context for kernel
(II) I810(0): [drm] unmapping 8192 bytes of SAREA 0xe039f000 at 0xb7a3c000

Fatal server error:
lockup
Please consult the The X.Org Foundation support
         at http://wiki.X.Org
 for help. 
Please also check the log file at "/var/log/Xorg.0.log" for additional information.

Error in I830WaitLpRing(), now is 5959, start is 3958
pgetbl_ctl: 0x1ffe0001 pgetbl_err: 0x0
ipeir: 0 iphdr: 7f0002e7
LP ring tail: 40 head: 0 len: 1f001 start 0
eir: 0 esr: 0 emr: ffff
instdone: ffc1 instpm: 0
memmode: 108 instps: 53
hwstam: ffff ier: 0 imr: ffff iir: a0
space: 131000 wanted 131064

FatalError re-entered, aborting
lockup

I'm using a 855GM on HP/Compaq NX5000.

The bug doesn't appear with LoadModule "dri" commented but my tux is too slow...

I can add or test things if it's needed.


Comment 35 Pascal Terjan 2006-03-25 22:39:02 UTC
I got this crash again (while using totem) today on my i915 (I get it every few
months usually when using mplayer).
I'm using 6.9.0 on Mandriva Cooker.
Comment 36 Alan Hourihane 2006-03-28 04:28:50 UTC
Closing this bug as it's related to Xorg 6.7.0.

Please open a new bug report if it's related to a different version of Xorg.
Comment 37 Andreas 2006-12-16 01:45:11 UTC
Hello!

This bug happens too with Xorg 6.9. I'm running here SuSE 10.1 

rpm -qa | grep xorg
xorg-x11-Xvnc-6.9.0-48
xorg-x11-6.9.0-50.17
xorg-x11-fonts-75dpi-6.9.0-48
xorg-x11-server-glx-6.9.0-50.17
xorg-x11-server-6.9.0-50.24
xorg-x11-driver-video-6.9.0-46.20
xorg-x11-fonts-scalable-6.9.0-48
xorg-x11-driver-video-nvidia-6.9.0-46.15
xorg-x11-devel-6.9.0-48
xorg-x11-fonts-100dpi-6.9.0-48
xorg-x11-libs-6.9.0-50.17

and I'm getting this bug most time after resuming and some clicks onto play or
stop or pause icon in kaffeine (kaffeine-0.8.1-17.pm.0).

Error in I830WaitLpRing(), now is 3598, start is 1597
pgetbl_ctl: 0x1ffc0001 pgetbl_err: 0x0
ipeir: 0 iphdr: 1810000
LP ring tail: 30 head: 0 len: 1f801 start 0
eir: 0 esr: 0 emr: ffff
instdone: ffc0 instpm: 0
memmode: 306 instps: f0000
hwstam: ffff ier: 0 imr: ffff iir: 0
space: 131016 wanted 131064

Fatal server error:
lockup
Comment 38 Alan Hourihane 2006-12-17 14:03:11 UTC
Use X.org 7.1 and the 1.7.3 version of the driver.

Closing.
Comment 39 Matthieu Baechler 2007-01-05 12:34:14 UTC
(In reply to comment #38)
> Use X.org 7.1 and the 1.7.3 version of the driver.
> 
> Closing.

I'm afraid these versions don't fix the issue on my laptop.
I got a macbook 2, and I can't manage having suspend/resume to work.

I use debian X.org 7.1, and tried with the shipped 1.7.2 intel driver (this
driver seems to be patched for xv lock bugs) and the 1.7.3 driver taken from
xorg.freedesktop.org (without any patch).

Using vesa driver make suspend/resume to work.

However, after a suspend/resume cycle with vesa, I can't restart the X server
with i810 driver. So the problem is really on the resume side.

Error logs are the same as for others comments of this bug.

My system is running linux 2.6.19.1 SMP from kernel.org compiled for EMT64 (aka
AMD64).

Comment 40 Alan Hourihane 2007-01-05 12:54:52 UTC
This bug was marked against 6.7.0, and then 6.9.0.

Please open a new bug with logs from 7.1 and the 1.7.2/1.7.3 driver.

So we don't clutter this bug.

Thanks.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.