Notices


Reply
Thread Tools
Posts: 992 | Thanked: 995 times | Joined on Dec 2009 @ California
#71
Originally Posted by fms View Post
Please, correct me if I am wrong, but I doubt this one makes sense, as flash memory appears to be written in 256kB pages. So, the original number of 5 that translates into 32*4096=128kB makes more sense. You may even increase it to 6!
I just repeat myself (I posted it somewhere near swapiness thread) -

The flash block size is 128KBytes typical. That means - if you write a smaller block it is possible that hardware READS 128KB before, replace your small portion and write back. In this case it has a big sense to write by N * 128KB blocks to eliminate reads.

However, if you write sequentially, a smart hardware may have some buffering and it may determine that your next write is in the same 128KB block and combine it. In this case a size of page cluster less 128KB may have sense because it allows kernel interrupt page write operation and it accelerates a device reaction.

Unfortunately, I don't know - is eMMC hardware smart enough.
 
Posts: 992 | Thanked: 995 times | Joined on Dec 2009 @ California
#72
Originally Posted by fms View Post
I really really think someone should try and package COMPCACHE for the N900. It may make a huge difference, by providing fast intermediate swap area in RAM (where swapped out pages are stored in gzipped form).
It would have a big impact on battery life (1) because CPU usage, and it decreases an available RAM space which is very valuable for N900 Maemo (2). So, it would decrease N900 performance.

But your mileage varies - test yourself.
 
atilla's Avatar
Posts: 1,210 | Thanked: 597 times | Joined on Apr 2010 @ hamburg,germany
#73
for me it works great too.
everything seems much more responsive.
850mhz,mohammads hildon package your swappiness package....
damn my phone is a beast!!
i love you guys.
thanks for your work
__________________


Nobody likes us but we dont care....
 
Posts: 50 | Thanked: 5 times | Joined on Nov 2009
#74
sounds interesting but ill hold off until yall figure all the possible kinks out.
 
arora.rohan's Avatar
Posts: 353 | Thanked: 166 times | Joined on Jun 2010 @ India
#75
This script is amazing...
Just amazing..
look at this...
http://i51.tinypic.com/ac4ayf.png

its been 24 hours since i charged this phone.. Amazing script!!! Amazing work!
850mhz Lv ,mohammads hildon package , this swappiness package
 
Posts: 182 | Thanked: 69 times | Joined on Nov 2009 @ Netherlands
#76
Originally Posted by fms View Post
I know for sure that the swappiness thing helps. You can test it by running Fennec or any other memory hogging application. The kill_oob_* thing looks sane, and so does the laptop_mode (based on its description found with Google). The tcp_* stuff is petty and probably does not make any difference to the network performance (unless somebody proves otherwise). Can't say much about the rest.
I do know that swappiness is not the only thing that helps. I had just that modified first and couldn't notice a big difference, but running this script I most certainly can! (shadowjk, if you think this is placebo effect you probably haven't tried it yet)
The two variables that weren't permanent before are questionable. I didn't notice they didn't stick.
 
Posts: 156 | Thanked: 28 times | Joined on Dec 2009 @ Los Angeles, CA
#77
great optimizations, its way smoother now, thanks!!
 
Posts: 1,418 | Thanked: 1,541 times | Joined on Feb 2008
#78
Originally Posted by egoshin View Post
It would have a big impact on battery life (1) because CPU usage, and it decreases an available RAM space which is very valuable for N900 Maemo (2). So, it would decrease N900 performance.
Not necessarily. Given how expensive MMC access operations are, gzipping most pages to, let us say, 16MB of RAM may well prove to be faster than pushing the same pages to the MMC.
 
Posts: 2,829 | Thanked: 1,459 times | Joined on Dec 2009 @ Finland
#79
 

The Following User Says Thank You to slender For This Useful Post:
Posts: 992 | Thanked: 995 times | Joined on Dec 2009 @ California
#80
Originally Posted by fms View Post
Not necessarily. Given how expensive MMC access operations are, gzipping most pages to, let us say, 16MB of RAM may well prove to be faster than pushing the same pages to the MMC.
I run root FS from /home partition - move root FS from UBIFS which is a compressed file system and I see the battery time increase and also - don't see any significant impact on performance.

So, I guess, a compression of swap space may be not a good thing on N900.
 
Reply

Tags
bada lubes, bada rox, swappiness, tuning

Thread Tools

 
Forum Jump


All times are GMT. The time now is 08:03.