flat assembler
Message board for the users of flat assembler.

Index > Heap > Strange netbook problem.

Author
Thread Post new topic Reply to topic
JohnFound



Joined: 16 Jun 2003
Posts: 3500
Location: Bulgaria
JohnFound
So, testing the performance of some of my programs, I found very strange problem with my netbook ASUS X102B. It runs 64 bit Linux Mint 17.

The problem is that after the screensaver is started (or lid closed, or suspend and back on, or screen locked) for the first time after the boot, the performance of the whole computer decrease by 30..50%.

The clock frequency does not changes. There is no active running processes different than before the speed decrease. The CPU load does not changes according to the task manager and "top" utility. Everything looks fine, but the speed is lower.

After the restart of the computer everything is fine again, until the first start of the screensaver.

What can be the reason for such strange behavior and how to search the problem? Is it possible to be some rootkit? In SMM mode, fully transparent for the OS?

_________________
Tox ID: 48C0321ADDB2FE5F644BB5E3D58B0D58C35E5BCBC81D7CD333633FEDF1047914A534256478D9
Post 12 Feb 2015, 07:19
View user's profile Send private message Visit poster's website ICQ Number Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17287
Location: In your JS exploiting you and your system
revolution
What does a (the) screensaver do that is different from a normal program?

You could write a program to do each part of what a screensaver will do and see when the problem begins.
Post 12 Feb 2015, 08:56
View user's profile Send private message Visit poster's website Reply with quote
JohnFound



Joined: 16 Jun 2003
Posts: 3500
Location: Bulgaria
JohnFound
I am not sure it is the screensaver itself. Because disabling the screensaver actually does not fix the problem after the lid close and suspend mode. IMHO, it is something in the power management mode that causes the performance degradation.

But what is more interesting for me is why this degradation happens, if there are no programs running and the clock frequency is not changed.
Post 12 Feb 2015, 09:00
View user's profile Send private message Visit poster's website ICQ Number Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17287
Location: In your JS exploiting you and your system
revolution
There are some common things that can cause low performance. The most common I know of are changed SDRAM timings and disabled cache. If you can run some tests you might be able to discover if one or both are responsible.

1) Run a pure CPU test to eliminate cache or memory timing differences.
2) Run a CPU and cache test.
3) Run a memory test.

If you have kernel access then you can directly examine the cache enable registers, and the SDRAM controller timings.
Post 12 Feb 2015, 09:14
View user's profile Send private message Visit poster's website Reply with quote
JohnFound



Joined: 16 Jun 2003
Posts: 3500
Location: Bulgaria
JohnFound
What "pure CPU" test means? Every test program will run in RAM, so the memory performance and the cache will affect the execution time, even if the program does not read and write in memory.
Post 12 Feb 2015, 09:25
View user's profile Send private message Visit poster's website ICQ Number Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17287
Location: In your JS exploiting you and your system
revolution
Depending upon your CPU something like "loop $" will run entirely out of the prefetch buffers. You can compare timings for some loop values and see if there is a difference.
Code:
mov ecx,1 shl 30 ;1G iterations
loop $
;...    
Post 12 Feb 2015, 09:29
View user's profile Send private message Visit poster's website Reply with quote
JohnFound



Joined: 16 Jun 2003
Posts: 3500
Location: Bulgaria
JohnFound
Ah, it is clear now. I will try.
Post 12 Feb 2015, 09:31
View user's profile Send private message Visit poster's website ICQ Number Reply with quote
JohnFound



Joined: 16 Jun 2003
Posts: 3500
Location: Bulgaria
JohnFound
Very strange! I tested it with this code and the effect disappeared. But what is more strange, I can't reproduce it right now even for programs that were affected earlier (with heavy work in memory).

The only difference is that I am not connected to WiFi network here on my work. I will test it again this evening at home with WiFi connected.
Post 12 Feb 2015, 09:59
View user's profile Send private message Visit poster's website ICQ Number Reply with quote
gens



Joined: 18 Feb 2013
Posts: 161
gens
suspend-resume can mess with some device drivers states

idk, look at powertop if something is going wack (should give more data if ran as root)

you could also restart X to make sure it's not a DE/toolkit problem
(ctrl+alt+backspace should do the trick)
Post 12 Feb 2015, 12:41
View user's profile Send private message Reply with quote
HaHaAnonymous



Joined: 02 Dec 2012
Posts: 1180
Location: Unknown
HaHaAnonymous
[ Post removed by author. ]


Last edited by HaHaAnonymous on 28 Feb 2015, 17:58; edited 1 time in total
Post 12 Feb 2015, 13:45
View user's profile Send private message Reply with quote
JohnFound



Joined: 16 Jun 2003
Posts: 3500
Location: Bulgaria
JohnFound
Well, the problem disappeared and I am not able to reproduce it anymore... Maybe it was some driver problem fixed with the later updates, or I didn't properly noticed the conditions causing the problem.

Or the people [paranoia] spying me stopped the rootkit for a while after I open this topic. Laughing Wink
Post 13 Feb 2015, 07:00
View user's profile Send private message Visit poster's website ICQ Number Reply with quote
revolution
When all else fails, read the source


Joined: 24 Aug 2004
Posts: 17287
Location: In your JS exploiting you and your system
revolution
JohnFound wrote:
Or the people [paranoia] spying me stopped the rootkit for a while after I open this topic. Laughing Wink
I wish it was always so easy to stop rootkits. Maybe I should pre-emptively open a topic just to make sure no future rootkits will run. Wink
Post 13 Feb 2015, 08:43
View user's profile Send private message Visit poster's website Reply with quote
Display posts from previous:
Post new topic Reply to topic

Jump to:  


< Last Thread | Next Thread >
Forum Rules:
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You can attach files in this forum
You can download files in this forum


Copyright © 1999-2020, Tomasz Grysztar. Also on YouTube, Twitter.

Website powered by rwasa.