Please read this before reporting a bug:
http://wiki.archlinux.org/index.php/Reporting_Bug_Guidelines

Do NOT report bugs when a package is just outdated, or it is in Unsupported. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!
Tasklist

FS#17804 - [xorg] X uses up 100% of cpu - seems to be evdev 2.3.2

Attached to Project: Arch Linux
Opened by Michel Brabants (MichelB) - Tuesday, 12 January 2010, 12:42 GMT-4
Last edited by Gerardo Exequiel Pozzi (djgera) - Tuesday, 12 January 2010, 18:26 GMT-4
Task Type Bug Report
Category Upstream Bugs
Status Assigned
Assigned To Jan de Groot (JGC)
Architecture x86_64
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 0%
Votes 0
Private No

Details

Description: X-prococess uses 100% of cpu.

Hello, I don't know why yet, but xorg starts using up 100% of the cpu regulary (every time I used the computer). It is enough to annoy me and this unstable situation doesn't allow me to trust the current platform-stability.

Anyway, it always happens when using firefox, but I'm not sure if it is related. I also have an xorg-log-file, which contains the error and traceback. An extract:

[mi] EQ overflowing. The server is probably stuck in an infinite loop.

Backtrace:
0: /usr/bin/X (xorg_backtrace+0x28) [0x4b04e8]
1: /usr/bin/X (mieqEnqueue+0x1f4) [0x4a9874]
2: /usr/bin/X (xf86PostMotionEventP+0xce) [0x48790e]
3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7f01aca09000+0x503f) [0x7f01aca0e03f]
4: /usr/bin/X (0x400000+0x72027) [0x472027]
5: /usr/bin/X (0x400000+0x11de63) [0x51de63]
6: /lib/libpthread.so.0 (0x7f01c46cd000+0xee80) [0x7f01c46dbe80]
7: /usr/bin/X (0x400000+0x61a80) [0x461a80]
8: /lib/libpthread.so.0 (0x7f01c46cd000+0xee80) [0x7f01c46dbe80]
9: /lib/libc.so.6 (ioctl+0x7) [0x7f01c3a21a57]
10: /usr/lib/libdrm.so.2 (drmDMA+0x7d) [0x7f01c25ff26d]
11: /usr/lib/xorg/modules/drivers/radeon_drv.so (0x7f01c1f58000+0x12922) [0x7f01c1f6a922]
12: /usr/lib/xorg/modules/drivers/radeon_drv.so (0x7f01c1f58000+0x12c6f) [0x7f01c1f6ac6f]
13: /usr/lib/xorg/modules/drivers/radeon_drv.so (0x7f01c1f58000+0x682e5) [0x7f01c1fc02e5]
14: /usr/lib/xorg/modules/libexa.so (0x7f01c16fd000+0x548f) [0x7f01c170248f]
15: /usr/lib/xorg/modules/libexa.so (0x7f01c16fd000+0x5861) [0x7f01c1702861]
16: /usr/lib/xorg/modules/libexa.so (0x7f01c16fd000+0x61e7) [0x7f01c17031e7]
17: /usr/lib/xorg/modules/libexa.so (0x7f01c16fd000+0xcf07) [0x7f01c1709f07]
18: /usr/lib/xorg/modules/libexa.so (0x7f01c16fd000+0xdee2) [0x7f01c170aee2]
19: /usr/bin/X (0x400000+0xe0680) [0x4e0680]
20: /usr/lib/xorg/modules/libexa.so (0x7f01c16fd000+0xcaf8) [0x7f01c1709af8]
21: /usr/bin/X (0x400000+0xd8487) [0x4d8487]
22: /usr/bin/X (0x400000+0x41754) [0x441754]
23: /usr/bin/X (0x400000+0x219ec) [0x4219ec]
24: /lib/libc.so.6 (__libc_start_main+0xfd) [0x7f01c3978b6d]
25: /usr/bin/X (0x400000+0x21599) [0x421599]

So, it seems related to the evdev-driver, which is version 2.3.2-1

Additional info:
* package version(s):extra/xf86-input-evdev 2.3.2-1 (xorg-input-drivers)
* config and/or log files etc.


Steps to reproduce:
This task depends upon

Comment by Michel Brabants (MichelB) - Wednesday, 13 January 2010, 06:28 GMT-4
Hello,

just for reference. Yesterday I updated my ati-driver (as I no longer believe it is evdev) to the latest from git. Up till now ... there have been no issues, but the test-period is too short to confirm anything. Anyway, I don't like using something from the development-branch, but if it is stable ..., then I'm happy.

Kind regards,

Michel
Comment by Kai Samelin (Kai) - Wednesday, 13 January 2010, 13:15 GMT-4
Hmm I have the same problem but normally no errors in the Xorg.log;

May you test if the site http://studip.uni-passau.de lags, please?
I get a hang over 15s; I'll try the new drivers next.
Comment by Kai Samelin (Kai) - Wednesday, 13 January 2010, 14:36 GMT-4
Ok, the new drivers do NOT fix the problem for me.

Any other solutions?

@Michel does this work for you smootlhy in ff?

http://studip.uni-passau.de ?

I have the bad feeling that embedded iframes causes the problem to uncover.

Kate and okular scrolling is slow as well (X @ 90-100%)
Comment by Michel Brabants (MichelB) - Wednesday, 13 January 2010, 17:49 GMT-4
Hey Kai,

I'm not at home anymore, so I can't test it at the moment. I can test it during the weekend. The new drivers seem to fix my 100% cpu problem however. Not one lockup today ...

So, I'll add a comment in the weekend.

Michel
Comment by Kai Samelin (Kai) - Wednesday, 13 January 2010, 18:02 GMT-4
Hi Michael,

ok, sounds fine.

I discovered something; if DRI is disabled everything but 3D works fine.
Comment by Michel Brabants (MichelB) - Saturday, 16 January 2010, 10:07 GMT-4
Hey,

I have no problem with the site you mentionned.

Kind regards,

Michel
Comment by Warren Huard (Warren) - Wednesday, 20 January 2010, 03:48 GMT-4
I also seem to have this problem of X using 100% of the CPU, but only when running KDE. It works fine in Openbox.
Comment by Kai Samelin (Kai) - Wednesday, 20 January 2010, 06:19 GMT-4
@Warren
What kind of graphic card are you using?
Comment by Warren Huard (Warren) - Wednesday, 20 January 2010, 12:07 GMT-4
I'm using an ATI card with the xf86-video-ati 6.12.4-3 & ati-dri 7.7-1 drivers on x86_64.
Comment by Kai Samelin (Kai) - Thursday, 21 January 2010, 11:37 GMT-4
@Warren
What card exactly?

I guess there are still many problems related to ATI Cards >= RV600 esp. with AGP connector.
Comment by Warren Huard (Warren) - Thursday, 21 January 2010, 17:13 GMT-4
@Kai

From lspci:
01:00.0 VGA compatible controller: ATI Technologies Inc Mobility Radeon HD 3400 Series

Like I said though, I only have this problem in KDE, specifically KDEmod. (It works fine in Openbox.) Is this also the case for you?
Comment by Kai Samelin (Kai) - Thursday, 21 January 2010, 17:38 GMT-4
Do you have KMS or 3D enabled?
Comment by Warren Huard (Warren) - Thursday, 21 January 2010, 22:22 GMT-4
... I don't have KMS enabled, and I don't think it is enabled by default for relatively recent cards such as mine. DRI is enabled and indeed seemed to work nicely out-of-the-box; I didn't need to set-up a xorg.conf file to get 3D and the like working.
Comment by Kai Samelin (Kai) - Friday, 22 January 2010, 06:55 GMT-4
@Warren

same problem here; may you try the git driver and enable KMS late start?

This solved the problem for me but I dont have 3d then.
Comment by Kai Samelin (Kai) - Friday, 22 January 2010, 08:18 GMT-4
@Warren

same problem here; may you try the git driver and enable KMS late start?

This solved the problem for me but I dont have 3d then.
Comment by Warren Huard (Warren) - Saturday, 23 January 2010, 16:03 GMT-4
@Kai

I might try that later to see what happens, but now isn't a good time.
Comment by Warren Huard (Warren) - Wednesday, 10 February 2010, 20:04 GMT-4
@Kai

I switched to KDE 4.4 in the main Arch repositories today and now it all works fine.
Comment by Paul Mattal (paul) - Friday, 05 March 2010, 08:23 GMT-4
Michel, are you still seeing the same problem? If so, does it go away if you downgrade to xf86-input-evdev 2.3.1?

Loading...