X server in F14 horribly slow

Martín Marqués martin.marques at gmail.com
Tue Jul 5 12:14:48 UTC 2011


Since last night I'm having these messages too on my laptop:

kernel: [  388.964307] [drm:drm_mode_getfb] *ERROR* invalid framebuffer id

CPU gets really loaded and top says it's Xorg who os eating the CPU.

Any ideas? Couldn't find anything on bugzilla and no updates
available, not even on *testing

2011/3/30 JD <jd1008 at gmail.com>:
> On 03/30/2011 11:45 AM, Kevin J. Cummings wrote:
>> On 03/30/2011 01:58 PM, JD wrote:
>>> On 03/30/2011 10:39 AM, Kevin J. Cummings wrote:
>>>> On 03/30/2011 01:15 PM, JD wrote:
>>>>>> I suspect you have something mis-configured, but I have no clue as to
>>>>>> what right now.
>>>>> But I did not change any configuration between fc13 and fc14.
>>>> Nor did I, but I had *lot's* of problems initially with my working
>>>> F12/F13 configuration after I switched to F14.  Mostly due to changes in
>>>> the ati driver.
>>>>
>>>> 1)  I used to have to turn off KMS in order for suspend to work right.
>>>>
>>>>     With F14, I had to not disable it in order to prevent a whole list of
>>>> video problems, including text consoles being corrupt, X hanging
>>>> completely when it started using the ati driver (I had to use vesa to
>>>> even boot up using UMS).  I was finally convinced that by enabling KMS,
>>>> things would start working again, and they did, but there was nothing in
>>>> the upgrade that suggested doing so was the right thing to do.
>>> What is KMS and how do I enable it?
>> Kernel Mode Set.  See your "nomodeset" option on your kernel below.  Try
>> removing it.
>>
>>>> 2)  I had all sorts of problems with the xorg configuration file.  I had
>>>> a *very* complicated one because I had changed it continually since FC6.
>>>>    Once I removed most of it, things vastly improved.  Its probably a good
>>>> idea anyways since the advent of the /etc/x11/conf.d stuff.  YMMV
>>> /etc/X11/xorg.conf.d only contains  00-system-setup-keyboard.conf
>>>
>>> (/etc/X11/xorg.conf),  I do not see KMS in there at all. How would I
>>> enable it?
>> You enable it by removing the "nomodeset" option from your kernel in grub.
>>
>>>> So, it might me a good idea to list any anomalies with your
>>>> configuration, from kernel boot parameters, to xorg.conf stuff so that
>>>> we can see if you are doing anything that might now be considered
>>>> problematic.  Also, if you could re-state what your ATI hardware is,
>>>> that could be helpful as well.  Is this a desktop or laptop?
>>>>
>>>    From lspci:
>>> 01:00.0 VGA compatible controller: ATI Technologies Inc RV350 [Mobility
>>> Radeon 9600 M10]
>> Just for comparison purposes, mine is an:
>>
>>> 01:00.0 VGA compatible controller: ATI Technologies Inc M56P [Radeon Mobility X1600]
>>>    From /boot/grub/grub.conf:
>>>
>>> title Fedora (2.6.35.6-45.fc14.i686)
>>>           root (hd0,1)
>>>           kernel /boot/vmlinuz-2.6.35.6-45.fc14.i686 ro
>>> root=UUID=99c4a1b1-99c2-40e6-8d2c-44bba67ad7df nomodeset
>>                                                   ^^^^^^^^^
>>
>> That's the guy!
>>
>>> SYSFONT=latarcyrheb-sun16 LANG=en_US.UTF-8 KEYTABLE=us
>>>           initrd /boot/initramfs-2.6.35.6-45.fc14.i686.img
>>>
>>> And finally /etc/X11/xorg.conf
>>> # ls -l /etc/X11/xorg.conf
>>> -rw-r--r--. 1 skroot root 4978 Mar 27 18:07 /etc/X11/xorg.conf
>>>
>>> As you can see from the date, it was touched during the upgrade to fc14.
>> 1) rename it so X doesn't find it and see what happens upon reboot.
>>       (I suggest that you try with "nomodeset" removed from your
>>       kernels options in grub.conf)  See what X11 does to configure
>>       itself.  If you know you've configured something that your
>>       server needs, you can always add it back in its own file
>>       in /etx/x11/xorg.conf.d later.  (to me, your xorg.conf looks
>>       fairly generic, anyways, besides specifying the obvious.)
> OK,
> I removed nomodeset
> and I renamed /etc/X11/xorg.conf
> and the workspace switchover is now as crisp as before.
> With a caveat!
> dmesg shows:
>
> Mar 30 14:00:08 localhost kernel: [drm:drm_mode_getfb] *ERROR* invalid
> framebuffer id
>
> Not sure what that means. Does it have to do with the FB driver not
> being configured?
>
> here's the rest of the drm related messages:
>
> Mar 30 13:59:10 localhost kernel: [drm] Initialized drm 1.1.0 20060810
> Mar 30 13:59:10 localhost kernel: [drm] radeon defaulting to kernel
> modesetting.
> Mar 30 13:59:10 localhost kernel: [drm] radeon kernel modesetting enabled.
> Mar 30 13:59:10 localhost kernel: radeon 0000:01:00.0: PCI INT A -> GSI
> 16 (level, low) -> IRQ 16
> Mar 30 13:59:10 localhost kernel: [drm] initializing kernel modesetting
> (RV350 0x1002:0x4E50).
> Mar 30 13:59:10 localhost kernel: [drm] register mmio base: 0xFEAF0000
> Mar 30 13:59:10 localhost kernel: [drm] register mmio size: 65536
> Mar 30 13:59:10 localhost kernel: [drm:radeon_agp_init] *ERROR* Unable
> to acquire AGP: -19
> Mar 30 13:59:10 localhost kernel: [drm] Forcing AGP to PCI mode
> Mar 30 13:59:10 localhost kernel: [drm] Generation 2 PCI interface,
> using max accessible memory
> Mar 30 13:59:10 localhost kernel: radeon 0000:01:00.0: VRAM: 128M
> 0x00000000F0000000 - 0x00000000F7FFFFFF (128M used)
> Mar 30 13:59:10 localhost kernel: radeon 0000:01:00.0: GTT: 512M
> 0x00000000D0000000 - 0x00000000EFFFFFFF
> Mar 30 13:59:10 localhost kernel: [drm] Supports vblank timestamp
> caching Rev 1 (10.10.2010).
> Mar 30 13:59:10 localhost kernel: [drm] Driver supports precise vblank
> timestamp query.
> Mar 30 13:59:10 localhost kernel: [drm] radeon: irq initialized.
> Mar 30 13:59:10 localhost kernel: [drm] Detected VRAM RAM=128M, BAR=128M
> Mar 30 13:59:10 localhost kernel: [drm] RAM width 128bits DDR
> Mar 30 13:59:10 localhost kernel: [TTM] Zone  kernel: Available graphics
> memory: 442682 kiB.
> Mar 30 13:59:10 localhost kernel: [TTM] Zone highmem: Available graphics
> memory: 1036510 kiB.
> Mar 30 13:59:10 localhost kernel: [TTM] Initializing pool allocator.
> Mar 30 13:59:10 localhost kernel: [drm] radeon: 128M of VRAM memory ready
> Mar 30 13:59:10 localhost kernel: [drm] radeon: 512M of GTT memory ready.
> Mar 30 13:59:10 localhost kernel: [drm] GART: num cpu pages 131072, num
> gpu pages 131072
> Mar 30 13:59:10 localhost kernel: [drm] radeon: 1 quad pipes, 1 Z pipes
> initialized.
> Mar 30 13:59:10 localhost kernel: radeon 0000:01:00.0: WB enabled
> Mar 30 13:59:10 localhost kernel: [drm] Loading R300 Microcode
> Mar 30 13:59:10 localhost kernel: [drm] radeon: ring at 0x00000000D0001000
> Mar 30 13:59:10 localhost kernel: [drm] ring test succeeded in 2 usecs
> Mar 30 13:59:10 localhost kernel: [drm] radeon: ib pool ready.
> Mar 30 13:59:10 localhost kernel: [drm] ib test succeeded in 0 usecs
> Mar 30 13:59:10 localhost kernel: [drm] Panel ID String: Samsung
> LTN154X1 WXGA
> Mar 30 13:59:10 localhost kernel: [drm] Panel Size 1280x800
> Mar 30 13:59:10 localhost kernel: [drm] Radeon Display Connectors
> Mar 30 13:59:10 localhost kernel: [drm] Connector 0:
> Mar 30 13:59:10 localhost kernel: [drm]   VGA
> Mar 30 13:59:10 localhost kernel: [drm]   DDC: 0x60 0x60 0x60 0x60 0x60
> 0x60 0x60 0x60
> Mar 30 13:59:10 localhost kernel: [drm]   Encoders:
> Mar 30 13:59:10 localhost kernel: [drm]     CRT1: INTERNAL_DAC1
> Mar 30 13:59:10 localhost kernel: [drm] Connector 1:
> Mar 30 13:59:10 localhost kernel: [drm]   LVDS
> Mar 30 13:59:10 localhost kernel: [drm]   Encoders:
> Mar 30 13:59:10 localhost kernel: [drm]     LCD1: INTERNAL_LVDS
> Mar 30 13:59:10 localhost kernel: [drm] Connector 2:
> Mar 30 13:59:10 localhost kernel: [drm]   S-video
> Mar 30 13:59:10 localhost kernel: [drm]   Encoders:
> Mar 30 13:59:10 localhost kernel: [drm]     TV1: INTERNAL_DAC2
> Mar 30 13:59:10 localhost kernel: [drm] radeon: power management initialized
> Mar 30 13:59:10 localhost kernel: [drm] fb mappable at 0xF0040000
> Mar 30 13:59:10 localhost kernel: [drm] vram apper at 0xF0000000
> Mar 30 13:59:10 localhost kernel: [drm] size 4096000
> Mar 30 13:59:10 localhost kernel: [drm] fb depth is 24
> Mar 30 13:59:10 localhost kernel: [drm]    pitch is 5120
> Mar 30 13:59:10 localhost kernel: fb0: radeondrmfb frame buffer device
> Mar 30 13:59:10 localhost kernel: drm: registered panic notifier
> Mar 30 13:59:10 localhost kernel: [drm] Initialized radeon 2.8.0
> 20080528 for 0000:01:00.0 on minor 0
> Mar 30 14:00:08 localhost kernel: [drm:drm_mode_getfb] *ERROR* invalid
> framebuffer id
>
> --
> users mailing list
> users at lists.fedoraproject.org
> To unsubscribe or change subscription options:
> https://admin.fedoraproject.org/mailman/listinfo/users
> Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
>



-- 
Martín Marqués
select 'martin.marques' || '@' || 'gmail.com'
DBA, Programador, Administrador


More information about the users mailing list