You are not logged in.
I use kde 4.2.2 and xorg 1.6.My notebook has nvidia Geforce 8400M GS with the latest beta driver.
After ~10min of use, the X process binds ~200-300MB of memory
I tried with new ~/.kde4 and the problem remains(with no difference if kwin effects are on/off).
Now with ~6 hours uptime the X process consumes 955MB.
Also when the "memory leak" has already begun,if I try to execute nvidia-settings -q gpucoretemp(displays gpu's temperature),
I get this message:
Maximum number of clients reached
ERROR: Cannot open display ':0.0'.
ERROR: Unable to query attribute GPUCoreTemp specified in query 'gpucoretemp' (no Display connection).
What I can do to fix it?
My xorg.conf:
Section "Files"
FontPath "/usr/share/fonts/misc"
FontPath "/usr/share/fonts/cyrilic"
FontPath "/usr/share/fonts/100dpi/:unscaled"
FontPath "/usr/share/fonts/75dpi/:unscaled"
FontPAth "/usr/share/fonts/Type1"
FontPath "/usr/share/fonts/100dpi"
FontPath "/usr/share/fonts/75dpi"
FontPath "/usr/share/fonts/TTF"
EndSection
Section "ServerLayout"
Identifier "Layout0"
Screen 0 "Screen0"
#InputDevice "Keyboard0" "CoreKeyboard"
#InputDevice "Mouse0" "CorePointer"
EndSection
Section "Server Flags"
Option "DontZap" "false"
EndSection
Section "Module"
Load "dbe"
Load "extmod"
Load "glx"
Load "freetype"
EndSection
Section "Monitor"
Identifier "Monitor0"
VendorName "Unknown"
ModelName "Unknown"
HorizSync 28.0 - 33.0
VertRefresh 43.0 - 72.0
Option "DPMS" "True"
EndSection
Section "Device"
Identifier "Device0"
Driver "nvidia"
VendorName "NVIDIA Corporation"
BoardName "GeForce 8400M GS"
Option "RenderAccel" "true"
Option "TwinViewOrientation" "Clone"
Option "CursorSdadow" "true"
Option "HWCursor" "true"
Option "TripleBuffer" "true"
EndSection
Section "Extensions"
Option "Composite" "enable"
EndSection
Section "Screen"
Identifier "Screen0"
Device "Device0"
Monitor "Monitor0"
Option "UseEvents" "true"
DefaultDepth 24
SubSection "Display"
Depth 24
Modes "1280x800"
EndSubSection
Option "AddARGBGLXVisuals" "true"
EndSection
xrestop:
xrestop - Display: localhost:0
Monitoring 208 clients. XErrors: 2
Pixmaps: 678355K total, Other: 176K total, All: 678531K total
res-base Wins GCs Fnts Pxms Misc Pxm mem Other Total PID Identifier
1a00000 71 32 0 491 661 6409K 17K 6427K 19656 Qt-subapplication
17400000 0 2 0 1 0 6000K 48B 6000K ? <unknown>
2800000 22 5 0 511 558 4848K 13K 4862K 19727 amarok
18c00000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
18800000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
18600000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
18000000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
17c00000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
17a00000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
17800000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
17600000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
17200000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
17000000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
16e00000 0 0 0 1 0 4000K 0B 4000K ? <unknown>
P.S: Sorry for my english
Last edited by tas0s (2009-04-28 17:41:00)
Offline
Can you provide the outputs of:
1. xlsclients | wc -l
2. xwininfo -root -children | wc -l
3. top
Offline
thanx for your reply!
1.xlsclients | wc -l
122
2.xwininfo -root -children | wc -l
172
3.top
top - 22:21:39 up 8:10, 3 users, load average: 0.54, 0.31, 0.22
Tasks: 120 total, 1 running, 119 sleeping, 0 stopped, 0 zombie
Cpu(s): 6.5%us, 3.6%sy, 0.0%ni, 87.2%id, 2.5%wa, 0.0%hi, 0.2%si, 0.0%st
Mem: 2058260k total, 2029300k used, 28960k free, 23460k buffers
Swap: 2096472k total, 88688k used, 2007784k free, 737324k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
19697 tas0s 20 0 614m 40m 22m S 3 2.0 3:36.20 ktorrent
19453 root 20 0 1164m 818m 8692 S 2 40.7 12:29.40 X
4880 root 20 0 88800 4016 1636 S 1 0.2 1:42.08 wicd
3773 root 15 -5 0 0 0 S 1 0.0 0:03.87 kjournald2
364 root 15 -5 0 0 0 S 0 0.0 0:04.99 ata/0
19729 tas0s 20 0 185m 19m 6588 S 0 1.0 0:04.65 wicd-client
28315 tas0s 20 0 520m 160m 23m S 0 8.0 0:08.59 firefox
28863 tas0s 20 0 10532 1172 892 R 0 0.1 0:00.02 top
1 root 20 0 3800 548 520 S 0 0.0 0:00.52 init
2 root 15 -5 0 0 0 S 0 0.0 0:00.00 kthreadd
3 root RT -5 0 0 0 S 0 0.0 0:00.04 migration/0
4 root 15 -5 0 0 0 S 0 0.0 0:00.59 ksoftirqd/0
Last edited by tas0s (2009-04-28 19:29:07)
Offline
It seems your plasma bin is leaking (well not exactly leaking, but causes X to leak).
If you give a: xwininfo -root -children > list and then you read the list file it generated, I'm pretty sure you'll see a lot of " "Qt-subapplication": ("Qt-subapplication" "Plasma")" in there. For some reason the plasma process is spawning too many X clients and your X's sockets are all (or most of them) reserved by those clients.
This perhaps explains the large memory usage of X too.
I think I saw a blog post of Aaron Seigo discussing this problem and the solution but I'm not sure.
Offline
Probably you mean this http://aseigo.blogspot.com/2009/04/plas … usage.html
Thank you again for your help!
Just one more silly question.Is anything I can do to fix it or I must wait for the next kde update?
Offline
Is anything I can do to fix it or I must wait for the next kde update?
Have you found an answer for this?
Offline
Nope.I have installed kde 4.2.3 from testing with the same results.
I also try the same setup in sidux and the problem occurs.So it is not arch specific.
Now I am thinking to switch in gnome until kde 4.3 is out.
Sad,I really like the new kde.
Offline
Now I am thinking to switch in gnome until kde 4.3 is out.
you could also try to use latest QT from git repository as this leak is from QT lib.
Offline
@newgargamel
Could you be a little more specific?
When you say the latest Qt,you mean this package :http://www.archlinux.org/packages/extra/i686/qt/,
but not from the extra repository;I must compile it from git?
(Any how-to or guide? )
thanx for your help!
Offline
He probably means this package from AUR: http://aur.archlinux.org/packages.php?ID=24093
The problem is that it doesn't compile in my machine. I tried but couldn't get a working PKGBUILD.
Offline
He probably means this package from AUR: http://aur.archlinux.org/packages.php?ID=24093
yes, this one but I'm not able to help you with building it, sorry. Here is the bug report regarding your problem: https://bugs.kde.org/show_bug.cgi?id=183191 You can also try to find a patch from the day Alexis fixed this problem(it would be the simplest way)
Offline
I tried with kde svn and qt-copy from markc repo .
Set up the same panel layout etc..
The same problem again.
Xorg process memory consuption is rising.Too fast.
Right now, after 1-2 hours of use,it consumes 500+MB
So,the only solution I can think rigt now is:
a)new .kde4,no use of plasmoids and wait to see if the problem persists
b)gnome (no offence for gnome,but I'm using kde for ~2 years)
EDIT:
I forgot to mention that I have removed the digital plasmoid
from the panel as it is reported in the bug report but with no
change in memory consumption.
Last edited by tas0s (2009-05-09 00:37:17)
Offline
I read that it is not a problem of kde or qt, but it's a xorg 1.6 bug. just try to launch for example 100 copies of konsole or terminal or anything else, then kill them and see the information provided by top. Xorg doesn't empty the unused memory I think
ArchLinux x86-64 KDE 4.3 Beta
Offline
shagrath I think that you have right.I started ~20 instances of konsole.The memory consuption is ~450MB(before the konsole copies was ~290).Even if I closed all konsoles,the memory consuption remains 450MB!
The funny thing is that when this "leak" came up in my machine,the first think I tried was to downgrade to previous xorg+nvidia, with no results.
So here we are again!I'm going to try again with the previous version of Xorg and I will inform you.
thank you very much for your help!
EDIT:
augegr said that
It seems your plasma bin is leaking (well not exactly leaking, but causes X to leak).
after the output of some commands..:/
Last edited by tas0s (2009-05-10 07:06:13)
Offline
No change with the previous xorg (1.5).
So it is kde-specific or kde+nvidia specific
The nouveau driver when will be ready (for 3D etc)?
Offline
tas0s, it's strange that previous xorg didn't help you. I was sure that it's a xorg 1.6 problem
I was trying previous nvidia drivers and even new betas - no effect. Now I don't know what to do too.
ArchLinux x86-64 KDE 4.3 Beta
Offline
So,you are facing the same problem?
The strange thing is that with new ~/.kde4.it takes about 5-6 hours or even more to see the memory leak.
But from the moment that memory leakage starts,in the next login(with restarted xorg-server) it only takes 1/2 hour!
Could be a corrupted file or something relevant (cache)?
I finally gave up/I switched to gnome
With my ram filled up,I was unable to run VirtualBox(WinXP) & NetBeans for my university assigments.
But I'm still searching for a solution
Last edited by tas0s (2009-05-10 18:41:01)
Offline
I did have the same problem about 2 or 3 weeks ago - X used up to 1.5GB of RAM.
It simply went away with one of upgrades
Thinkpad T61p : T7700 | 4GB RAM | nVidia FX 570M | Intel 4965
Arch64 @ Openbox
Offline
@vi3dr0:
Are you using any plasmoid?Could you describe your setup?
I mean nvidia driver,xorg version,panels,plasmoids etc.
Even a xorg.conf would be helpfull
Sorry for the many questions,but I dont know what is the cause of the memory leak
I see that you are using kdemod..
hmm i know that kdemod4 is almost vanilla,so are there any differences in libraries?(ex.Qt)
Offline
I just tried to switch to fluxbox. and I launched just about 100 terminals (not somehow connected to kde), xorg was occupying about 200 mb, then I killed my terminals, but xorg was still taking 200 mb...
Last edited by shagrath (2009-05-10 19:52:23)
ArchLinux x86-64 KDE 4.3 Beta
Offline
!?!?!
so its not kde-specific??
In gnome:
Xorg -> 36MB!!
Is anything common between fluxbox and kde4?
Offline
tas0s, I don't think so. but maybe my memory leak is not connected with your memory leak
ArchLinux x86-64 KDE 4.3 Beta
Offline
by the way my X in that moment takes 26 mb. that problem appears when I'm trying to launch ~100 applications and then close them. fortunately, I don't need to do that.
ArchLinux x86-64 KDE 4.3 Beta
Offline
Maybe you have right.But the "terminal test":P gave me the same results:P
So is there anything we can do to limit down the possible causes of the leak?
Offline
@vi3dr0:
Are you using any plasmoid?Could you describe your setup? [...]
Panels and plasmoids as seen on the screenshot.
Nvidia driver: 185.19 (however it could be non-beta 180.xx back then when I had this problem - don't remember )
Xorg 1.6
...and here is my xorg.conf.
Section "ServerLayout"
Identifier "Layout0"
Screen 0 "Screen0" 0 0
InputDevice "Keyboard0" "CoreKeyboard"
InputDevice "Mouse0" "CorePointer"
InputDevice "SynapticsTouchpad" "SendCoreEvents"
Option "AutoAddDevices" "false"
EndSection
Section "Files"
EndSection
Section "Module"
Load "dbe"
Load "extmod"
# Load "type1"
Load "freetype"
Load "glx"
Load "synaptics"
EndSection
Section "ServerFlags"
Option "AllowEmptyInput" "false"
Option "DontZap" "false"
EndSection
Section "InputDevice"
#Option "Device" "/dev/input/mice"
# Option "VertTwoFingerScroll" "true" # vertical scroll anywhere with two fingers
Identifier "SynapticsTouchpad"
Driver "synaptics"
Option "AlwaysCore" "true" # send events to CorePointer
Option "Device" "/dev/psaux"
Option "Protocol" "auto-dev"
Option "SHMConfig" "true" # configurable at runtime? security risk
Option "LeftEdge" "1700" # x coord left
Option "RightEdge" "5300" # x coord right
Option "TopEdge" "1700" # y coord top
Option "BottomEdge" "4200" # y coord bottom
Option "FingerLow" "25" # pressure below this level triggers release
Option "FingerHigh" "30" # pressure above this level triggers touch
Option "MaxTapTime" "180" # max time in ms for detecting tap
Option "VertEdgeScroll" "true" # enable vertical scroll zone
Option "HorizEdgeScroll" "true" # enable horizontal scroll zone
Option "VertScrollDelta" "100" # edge-to-edge scroll distance of the vertical scroll
Option "HorizScrollDelta" "100" # edge-to-edge scroll distance of the horizontal scroll
Option "MinSpeed" "0.40" # speed factor for low pointer movement
Option "MaxSpeed" "0.70" # maximum speed factor for fast pointer movement
Option "AccelFactor" "0.010" # acceleration factor for normal pointer movements
Option "HorizTwoFingerScroll" "true" # horizontal scroll anywhere with two fingers
Option "TapButton1" "1"
# Option "TapButton2" "3"
# Option "TapButton3" "2"
Option "FastTaps" "1"
EndSection
Section "InputDevice"
# generated from default
Identifier "Mouse0"
Driver "mouse"
Option "Protocol" "auto"
Option "Device" "/dev/psaux"
Option "Emulate3Buttons" "no"
Option "ZAxisMapping" "4 5"
EndSection
Section "InputDevice"
# generated from default
Identifier "Keyboard0"
Driver "kbd"
EndSection
Section "Monitor"
Identifier "Monitor0"
VendorName "Unknown"
ModelName "Unknown"
HorizSync 28.0 - 33.0
VertRefresh 43.0 - 72.0
Option "DPMS"
EndSection
Section "Device"
Identifier "Device0"
Driver "nvidia"
VendorName "NVIDIA Corporation"
EndSection
Section "Screen"
Identifier "Screen0"
Device "Device0"
Monitor "Monitor0"
DefaultDepth 24
Option "UseEDIDDpi" "False"
SubSection "Display"
Depth 24
EndSubSection
EndSection
I had this memory issue for about 4 or 5 days (using GNOME) and it simply gone away with one of upgrades.
Right now X is using ~200MB which I consider completely typical.
Thinkpad T61p : T7700 | 4GB RAM | nVidia FX 570M | Intel 4965
Arch64 @ Openbox
Offline