Xorg - High CPU - Desktop hung

[JD]

  On 08/10/2010 03:47 PM, Rick Stevens wrote:
> On 08/10/2010 12:25 PM, JD wrote:
>>     On 08/10/2010 11:43 AM, Alan J. Gagne wrote:
>>> Anybody seeing an issue with Desktop hangs with Xorg using 90% CPU.
>>> This started for me with the latest kernel.
>>>
>>> Aug 03 09:54:35 Installed: kernel-2.6.33.6-147.2.4.fc13.i686
>>>
>>> Runs fine using previous kernel. (2.6.33.6-147.fc13.i686)
>>>
>>> Snip from Xorg log.
>>>
>>>> [104476.705] [mi] EQ overflowing. The server is probably stuck in an infinite loop.
>>>> [104476.705]
>>>> Backtrace:
>>>> [104476.705] 0: /usr/bin/Xorg (xorg_backtrace+0x3c) [0x80ddfac]
>>>> [104476.706] 1: /usr/bin/Xorg (mieqEnqueue+0x1b7) [0x80d91b7]
>>>> [104476.706] 2: /usr/bin/Xorg (xf86PostButtonEventP+0xcf) [0x80b535f]
>>>> [104476.706] 3: /usr/bin/Xorg (xf86PostButtonEvent+0x6c) [0x80b540c]
>>>> [104476.706] 4: /usr/lib/xorg/modules/input/evdev_drv.so (0x34a000+0x341b) [0x34d41b]
>>>> [104476.706] 5: /usr/bin/Xorg (0x8048000+0x739e0) [0x80bb9e0]
>>>> [104476.706] 6: /usr/bin/Xorg (0x8048000+0x10fd04) [0x8157d04]
>>>> [104476.706] 7: (vdso) (__kernel_sigreturn+0x0) [0x292400]
>>>> [104476.706] 8: (vdso) (__kernel_vsyscall+0x2) [0x292416]
>>>> [104476.706] 9: /lib/libc.so.6 (ioctl+0x19) [0x8bd0b9]
>>>> [104476.706] 10: /usr/lib/libdrm.so.2 (drmIoctl+0x2e) [0x4b55a7e]
>>>> [104476.707] 11: /usr/lib/libdrm.so.2 (drmCommandWrite+0x3c) [0x4b55e0c]
>>>> [104476.707] 12: /usr/lib/libdrm_nouveau.so.1 (0x181000+0x2a9a) [0x183a9a]
>>>> [104476.707] 13: /usr/lib/libdrm_nouveau.so.1 (nouveau_bo_map_range+0xf1) [0x183c91]
>>>> [104476.707] 14: /usr/lib/libdrm_nouveau.so.1 (nouveau_bo_map+0x34) [0x183d64]
>>>> [104476.707] 15: /usr/lib/xorg/modules/drivers/nouveau_drv.so (0x1b5000+0x3a36) [0x1b8a36]
>>>> [104476.707] 16: /usr/lib/xorg/modules/libexa.so (0x3cc000+0x3521) [0x3cf521]
>>>> [104476.707] 17: /usr/lib/xorg/modules/libexa.so (0x3cc000+0x66ad) [0x3d26ad]
>>>> [104476.707] 18: /usr/lib/xorg/modules/libexa.so (0x3cc000+0xf7f5) [0x3db7f5]
>>>> [104476.707] 19: /usr/lib/xorg/modules/libexa.so (0x3cc000+0xd37e) [0x3d937e]
>>>> [104476.707] 20: /usr/bin/Xorg (0x8048000+0xc8d77) [0x8110d77]
>>>> [104476.707] 21: /usr/bin/Xorg (CompositePicture+0x290) [0x81030e0]
>>>> [104476.708] 22: /usr/bin/Xorg (0x8048000+0xc1fa5) [0x8109fa5]
>>>> [104476.708] 23: /usr/bin/Xorg (0x8048000+0xbeb34) [0x8106b34]
>>>> [104476.708] 24: /usr/bin/Xorg (0x8048000+0x50ab7) [0x8098ab7]
>>>> [104476.708] 25: /usr/bin/Xorg (0x8048000+0x1a595) [0x8062595]
>>>> [104476.708] 26: /lib/libc.so.6 (__libc_start_main+0xe6) [0x7fecc6]
>>>> [104476.708] 27: /usr/bin/Xorg (0x8048000+0x1a181) [0x8062181]
>> Since no one else is reporting this (at least not that I have seen),
>> it is probably caused by the graphics chip driver used by Xorg.
>> What is your graphics chip brand and model?
> This can also be caused by the infernal nVidia MCP78S [GeForce 8200] USB
> bug present on many Asus mobos.  It has bitten me quite hard at the
> most inopportune times (my mobo is an Asus M3N78-VM).
>
> The manifestation of it is the mouse or keyboard freezes up and X
> starts eating CPU like mad.  I think the USB ports stop supplying
> data or interrupts and X sits there polling on them.  The only way out
> I've found is a restart of X.  I have been looking for a fix quite a
> while but have not been successful.  It has bitten many people using
> that "OS from the Northwest" so it's not a Linux-specific thing.  It
> truly appears to be a hardware bug.
> ----------------------------------------------------------------------
> - Rick Stevens, Systems Engineer, C2 Hosting          ricks at nerd.com -
> - AIM/Skype: therps2        ICQ: 22643734            Yahoo: origrps2 -
> -                                                                    -
> - Real Time, adj.: Here and now, as opposed to fake time, which only -
> -                        occurs there and then                       -
> ----------------------------------------------------------------------
I saw the same exact problem as you describe as far back as
2003 or 2004. The system would freeze up after logging in as soon
as I touched the KB or mouse.
The cause of the problem, I was told, was the
irq conflict between the USB devices and the NVidia card.
I recall I emailed Nvidia and complained about their driver
and they could not fix the problem.
Seems after all these years they are still not fixing the problem.