Wise Penguin
Re: Digikam & xorg heavy cpu load
Actually i think it is a memory leak in digikam. I started it after i posted the last stats and once i work with it and quit it i still have 2903824k used. This time i have something left, but not good enough. For some reason the mem does not get released after i quit the program.
I have to try to install the newer version in hope it will work. I think it did not last time with 11.3.
+++ ATH0
. . . . . . . .
LOGOFF COMPLETE
Flux Capacitor Penguin
Re: Digikam & xorg heavy cpu load
But the 2.9 GB (total?) is mainly disk cache is it not? I don't know whether that disk cache would be reduced unless the memory is required by running more processes. So you could find out by firing up some more applications (though not digikam) and see if the cached stat decreases and the total memory stat either increases or stays roughly the same. You should be able to see if the cache mem is being released.
Once again is your system fully updated? What kernel and release are you running? You will find the information in My Computer (sysinfo). Cannot tell that from your repo list. Also you didn't mention what release of digikam you have been running.
Wise Penguin
Re: Digikam & xorg heavy cpu load
I just updated the system with digikam.
Digikam version is now 1.5, was 1.2 before.
The leak, not sure.. its still eating the mem. With what program can i determain where the ram goes? And whats in the buffer?
KDE is 4.5.2 now. Used to run 4.4.4 before.
But now, even with the high load, its smoother and not sluggish at all. At least one good thing.
Code:top - 19:55:33 up 21 min, 3 users, load average: 1.01, 0.48, 0.31 Tasks: 147 total, 1 running, 146 sleeping, 0 stopped, 0 zombie Cpu(s): 5.0%us, 1.2%sy, 0.0%ni, 53.2%id, 40.5%wa, 0.2%hi, 0.0%si, 0.0%st Mem: 4055392k total, 3495752k used, 559640k free, 94860k buffers Swap: 4193276k total, 0k used, 4193276k free, 2771452k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 5043 joerg 20 0 1189m 234m 60m S 9 5.9 6:20.98 digikam 1489 root 20 0 143m 47m 15m S 1 1.2 2:10.02 Xorg 4815 joerg 20 0 91816 34m 13m S 1 0.9 0:10.49 skype 4638 joerg 20 0 533m 71m 47m S 0 1.8 0:17.28 kwin 1 root 20 0 12408 768 628 S 0 0.0 0:00.30 init 2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd 3 root RT 0 0 0 0 S 0 0.0 0:00.00 migration/0 4 root 20 0 0 0 0 S 0 0.0 0:00.06 ksoftirqd/0 5 root RT 0 0 0 0 S 0 0.0 0:00.00 watchdog/0 6 root RT 0 0 0 0 S 0 0.0 0:00.00 migration/1 7 root 20 0 0 0 0 S 0 0.0 0:00.06 ksoftirqd/1 8 root RT 0 0 0 0 S 0 0.0 0:00.00 watchdog/1 9 root 20 0 0 0 0 S 0 0.0 0:00.05 events/0 10 root 20 0 0 0 0 S 0 0.0 0:00.05 events/1 11 root 20 0 0 0 0 S 0 0.0 0:00.00 netns 12 root 20 0 0 0 0 S 0 0.0 0:00.00 async/mgr 13 root 20 0 0 0 0 S 0 0.0 0:00.00 pm 14 root 20 0 0 0 0 S 0 0.0 0:00.00 sync_supers 15 root 20 0 0 0 0 S 0 0.0 0:00.00 bdi-default 16 root 20 0 0 0 0 S 0 0.0 0:00.00 kintegrityd/0 17 root 20 0 0 0 0 S 0 0.0 0:00.00 kintegrityd/1 18 root 20 0 0 0 0 S 0 0.0 0:00.14 kblockd/0 19 root 20 0 0 0 0 S 0 0.0 0:00.14 kblockd/1 20 root 20 0 0 0 0 S 0 0.0 0:00.00 kacpid 21 root 20 0 0 0 0 S 0 0.0 0:00.00 kacpi_notify 22 root 20 0 0 0 0 S 0 0.0 0:00.00 kacpi_hotplug 23 root 20 0 0 0 0 S 0 0.0 0:00.00 khubd 24 root 20 0 0 0 0 S 0 0.0 0:00.00 kseriod 27 root 20 0 0 0 0 S 0 0.0 0:00.00 kondemand/0 28 root 20 0 0 0 0 S 0 0.0 0:00.00 kondemand/1 29 root 20 0 0 0 0 S 0 0.0 0:00.00 khelper
+++ ATH0
. . . . . . . .
LOGOFF COMPLETE
Wise Penguin
Re: Digikam & xorg heavy cpu load
Actually see what happens. Digikam is still working, but my mem is shrinking by the seconds.
The cpu load is at a normal level, which is a big difference from before.
In some sense, i don't understand it really and i am really puzzled.
Code:top - 20:04:07 up 29 min, 3 users, load average: 1.93, 1.67, 0.99 Tasks: 150 total, 1 running, 149 sleeping, 0 stopped, 0 zombie Cpu(s): 9.2%us, 2.6%sy, 0.0%ni, 49.5%id, 38.5%wa, 0.2%hi, 0.0%si, 0.0%st Mem: 4055392k total, 4024480k used, 30912k free, 154192k buffers Swap: 4193276k total, 0k used, 4193276k free, 3169148k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 5043 joerg 20 0 1189m 234m 60m S 9 5.9 7:03.84 digikam 1489 root 20 0 150m 56m 19m S 5 1.4 2:28.18 Xorg 5048 joerg 20 0 475m 38m 21m S 3 1.0 0:02.57 konsole 4638 joerg 20 0 543m 78m 51m S 2 2.0 0:25.11 kwin 4815 joerg 20 0 91816 34m 13m S 1 0.9 0:15.10 skype 5170 joerg 20 0 486m 87m 32m S 1 2.2 0:07.48 firefox 4764 joerg 20 0 483m 37m 27m S 0 1.0 0:02.18 kopete 1 root 20 0 12408 768 628 S 0 0.0 0:00.30 init 2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd 3 root RT 0 0 0 0 S 0 0.0 0:00.00 migration/0 4 root 20 0 0 0 0 S 0 0.0 0:00.09 ksoftirqd/0 5 root RT 0 0 0 0 S 0 0.0 0:00.00 watchdog/0 6 root RT 0 0 0 0 S 0 0.0 0:00.00 migration/1 7 root 20 0 0 0 0 S 0 0.0 0:00.09 ksoftirqd/1 8 root RT 0 0 0 0 S 0 0.0 0:00.00 watchdog/1 9 root 20 0 0 0 0 S 0 0.0 0:00.07 events/0 10 root 20 0 0 0 0 S 0 0.0 0:00.08 events/1 11 root 20 0 0 0 0 S 0 0.0 0:00.00 netns 12 root 20 0 0 0 0 S 0 0.0 0:00.00 async/mgr 13 root 20 0 0 0 0 S 0 0.0 0:00.00 pm 14 root 20 0 0 0 0 S 0 0.0 0:00.00 sync_supers 15 root 20 0 0 0 0 S 0 0.0 0:00.00 bdi-default 16 root 20 0 0 0 0 S 0 0.0 0:00.00 kintegrityd/0 17 root 20 0 0 0 0 S 0 0.0 0:00.00 kintegrityd/1 18 root 20 0 0 0 0 S 0 0.0 0:00.33 kblockd/0 19 root 20 0 0 0 0 S 0 0.0 0:00.34 kblockd/1 20 root 20 0 0 0 0 S 0 0.0 0:00.00 kacpid 21 root 20 0 0 0 0 S 0 0.0 0:00.00 kacpi_notify 22 root 20 0 0 0 0 S 0 0.0 0:00.00 kacpi_hotplug 23 root 20 0 0 0 0 S 0 0.0 0:00.00 khubd 24 root 20 0 0 0 0 S 0 0.0 0:00.00 kseriod
+++ ATH0
. . . . . . . .
LOGOFF COMPLETE
Wise Penguin
Re: Digikam & xorg heavy cpu load
and here the kernel version. sorry, forgot it.
Code:Linux 2.6.34.7-0.4-desktop (linux-sf0t) 25.10.2010 _x86_64_ (2 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 16,09 0,11 2,53 24,93 0,00 56,35 Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn sda 160,97 2257,18 942,55 5640310 2355280 sdb 0,07 0,90 0,01 2254 16 sdc 21,97 1529,87 2,71 3822880 6784 scd0 0,00 0,02 0,00 48 0
+++ ATH0
. . . . . . . .
LOGOFF COMPLETE
Flux Capacitor Penguin
Re: Digikam & xorg heavy cpu load
Thanks for the software info. Although I now have no confidence in knowing what you were running for the earlier tests. I'm not surprised that upgrading KDE has improved your operation as it feels better on my test system.
How are you getting on with that test I suggested?
Load your system with more applications (not digikam - close it) step by step, and observe the changes to "used" and "cached" at each step as here (in blue), using top :But the 2.9 GB (total?) is mainly disk cache is it not? I don't know whether that disk cache would be reduced unless the memory is required by running more processes. So you could find out by firing up some more applications (though not digikam) and see if the cached stat decreases and the used memory stat either increases or stays roughly the same. You should be able to see if the cache mem is being released.
You don't need to paste the results, just tell us what you observed.Code:Mem: 4055392k total, 4024480k used, 30912k free, 154192k buffers Swap: 4193276k total, 0k used, 4193276k free, 3169148k cached
Wise Penguin
Re: Digikam & xorg heavy cpu load
Is there a way to log these stats? Also can you specifically see what program x does with the ram?
I am not really that great with commands but i am getting better.
+++ ATH0
. . . . . . . .
LOGOFF COMPLETE
Flux Capacitor Penguin
Re: Digikam & xorg heavy cpu load
A quick scan of the man page for top command suggests it can be output to file, but I haven't tried that. Take a look with Konqueror: enter man:top
Top shows you the memory being used with each task, but that's all I can think of right now. I wasn't looking for great precision here, just to see if the disk cache is lowered to release memory to the new apps that are loaded, after running digikam to take up the memory and having closed it.
Wise Penguin
Re: Digikam & xorg heavy cpu load
Well, there is a funny story to it.Originally Posted by consused
Thanks for the software info. Although I now have no confidence in knowing what you were running for the earlier tests. I'm not surprised that upgrading KDE has improved your operation as it feels better on my test system.
How are you getting on with that test I suggested?
Load your system with more applications (not digikam - close it) step by step, and observe the changes to "used" and "cached" at each step as here (in blue), using top :
You don't need to paste the results, just tell us what you observed.Code:Mem: 4055392k total, 4024480k used, 30912k free, 154192k buffers Swap: 4193276k total, 0k used, 4193276k free, 3169148k cached
The digikam version i ran was 1.2. So i looked around on the digikam website and searched for leaks and yes, past versions had memory leaks.
I suspect that the older version had it, since it seemed to fill up over time and got slower in the process.
With the new version 1.5 its smooth and it runs through without slowing down. So a big improvement.
But, to the install of 1.5 i choose one click install. Big mistake.
Why? Because it installed also (not sure why) also KDE 4.5 with it. Of course i said yes, but what i did not know was, that it only upgrade some libs and not all. So i ended up with a broken KDE.
I had to fix that, install it again and, well now i am on KDE again.
With that i updated the whole system via zypper. So the system should be up to date now.
The one thing i don't get still is the memory management in linux. It seem, and i never paid attention to this before, that it fills the cache to a certain point and from there it only swaps information in and out.
Since this is a standard install, i will assume that anyone has the same config and ergo the same datas.
Have to read on the memory a little more.
I will track a little more the behavior on my memory and will pos my findings here.
Overall the problem seems to be solved. I am very certain it was digikam/memoryleak.
I'll be back
+++ ATH0
. . . . . . . .
LOGOFF COMPLETE
Flux Capacitor Penguin
Re: Digikam & xorg heavy cpu load
It looks like 1.5 depends on certain KDE 4.5 libraries, although I haven't checked. You must be up to 4.5.2 now. I have that on my test partition, and it appears to run well. I noted your kernel is up to date for the standard release.
I new that digikam had previous form on memory leaks, from googling earlier, but didn't want to over prejudice our conclusions. Well done, you made a lot of progress and I'm glad to hear of the improved performance. Looking forward to reading your findings on the memory.
Posting Permissions
Bookmarks