[Announce] kernel-power stable v49 in Extras-Testing - Page 240 - maemo.org

Active Topics

The N900 is now 15 years old! (9)
to Nokia N900 by Arno_11 - 1 day, 1 hr ago
Sailfish OS for the Motorola Moto G7 Power (XT1955-5) - (ocean) (4)
to SailfishOS by Maemish - 6 days, 13 hrs ago
more...

Page 240 of 300

240

Thread Tools

	x61	2011-12-01 , 04:32
	Posts: 932 \| Thanked: 278 times \| Joined on Sep 2009 @ Kentucky	#2391

Originally Posted by knobby

firsly u must install kernel power 49..there's a conflict b'coz u r tying to install bootimage of kp49 but kp49 is not installed instead u have kp48

And why does it show an available update when it knows the program is not yet installed?

Quote & Reply |

Joseph9560	2011-12-01 , 16:19
Posts: 440 \| Thanked: 160 times \| Joined on Aug 2010 @ Las Vegas, NV	#2392

There is 19+MB folder in /usr/src/ Is it required or not (because folder name is src). Don't want to delete it without confirmation.

Quote & Reply |

Laughingstok	2011-12-01 , 16:34
Posts: 472 \| Thanked: 442 times \| Joined on Sep 2007	#2393

Running kp49 without issue. Don't forget to do your apt-get upgrade folks.

__________________
If you don't know how to check your N900's uptime, you probably shouldn't own it.

Quote & Reply |

	Estel	2011-12-01 , 17:06
	Posts: 5,028 \| Thanked: 8,613 times \| Joined on Mar 2011	#2394

Don't forget to *not* use apt-get upgrade, as far as You want to keep any custom-installed packages (i.e. 720p video recording/playback from freemangordon). Just update manually, what is needed to be updated.

/Estel

__________________
N900's aluminum backcover / body replacement
-
N900's HDMI-Out
-
Camera cover MOD
-
Measure battery's real capacity on-device
-
TrueCrypt 7.1 | ereswap | bnf
-
Hardware's mods research is costly. To support my work, please consider donating. Thank You!

Quote & Reply |

The Following User Says Thank You to Estel For This Useful Post:
vi_

Megaltariak	2011-12-01 , 17:52
Posts: 346 \| Thanked: 271 times \| Joined on Jan 2010	#2395

You can also use the dpkg --set-selections for holding packages, e.g: (as root)
dpkg --set-selections << EOF
camera-ui hold
libgdigicam-gst-camerabin0 hold
libgdigicam0 hold
EOF

Quote & Reply |

The Following 6 Users Say Thank You to Megaltariak For This Useful Post:
Estel, impeham, maacruz, maartenmk, moepda, nkirk

StocChr	2011-12-01 , 20:44
Posts: 306 \| Thanked: 38 times \| Joined on Dec 2009 @ Austria	#2396

hi guys,

some people have big troubles with kernel-power-settings

with kp-settings three processes stay always ON ( per, core, iva2 )
so the battery drains very fast ( 8 hours from 100%-0% )

You will notice while the results are not that significantly different, there is a shitload more DMA activity and IRQing going on when kp-settings is installed. While the CPU is not hammering some hardware on the N900 is. Notice the extra 100% ONs in the power domain table for with kp-settings.

There is an ancient bug in kp-settings. I had issues with kp-settings since version 26. I never figured out why it caused all these wake ups I just know in some circumstances it does.

have someone an idea what can be wrong with the kernel-power-settings? i would really like to use kp-settings, maybe someone can fix it?

with kp-settings:

Code:

C#      | Ratio  | Avg/dura | Frequency | Ratio
--------+--------+----------+-----------+--------+
     C0 |   0.2% |          |  1150 MHz |   nan% |
     C1 |   0.0% |    0.3ms | 
     C2 |   3.3% |    9.4ms | 
     C3 |   4.2% |  104.3ms | 
     C4 |  92.3% | 1730.5ms |

Code:

Power domain activity breakdown
Domain  | % of time spent in states
--------+---------+---------+---------+---------+----------
usbhost |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
    sgx |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
    per |OFF:   0%|RET:   0%|INA:   0%| ON: 100%| now:(ON)
    dss |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
    cam |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
   core |OFF:   0%|RET:   0%|INA:   0%| ON: 100%| now:(ON)
   neon |OFF:  92%|RET:   4%|INA:   3%| ON:   0%| now:(ON)
    mpu |OFF:  92%|RET:   4%|INA:   3%| ON:   0%| now:(ON)
   iva2 |OFF:   0%|RET:   0%|INA:   0%| ON: 100%| now:(ON)

without kp-settings:

Code:

C#      | Ratio  | Avg/dura | Frequency | Ratio
--------+--------+----------+-----------+--------+
     C0 |   0.2% |          |  1150 MHz |   nan% |
     C1 |   0.0% |    0.2ms | 
     C2 |   0.2% |    2.4ms | 
     C3 |   2.9% |  215.9ms | 
     C4 |  96.7% | 2903.8ms |

Code:

Power domain activity breakdown
Domain  | % of time spent in states
--------+---------+---------+---------+---------+----------
usbhost |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
    sgx |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
    per |OFF:  96%|RET:   3%|INA:   0%| ON:   0%| now:(ON)
    dss |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
    cam |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
   core |OFF:  90%|RET:   5%|INA:   0%| ON:   3%| now:(ON)
   neon |OFF:  91%|RET:   6%|INA:   1%| ON:   0%| now:(ON)
    mpu |OFF:  91%|RET:   6%|INA:   1%| ON:   0%| now:(ON)
   iva2 |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)

Quote & Reply |

	Estel	2011-12-01 , 21:56
	Posts: 5,028 \| Thanked: 8,613 times \| Joined on Mar 2011	#2397

Originally Posted by StocChr

You will notice while the results are not that significantly different, there is a shitload more DMA activity and IRQing going on when kp-settings is installed. While the CPU is not hammering some hardware on the N900 is. Notice the extra 100% ONs in the power domain table for with kp-settings.

There is an ancient bug in kp-settings. I had issues with kp-settings since version 26. I never figured out why it caused all these wake ups I just know in some circumstances it does.

Maybe not 100 percent related, but those aren't Your words - You just copied and pasted parts of vi_ posts from other thread. While it does not invalidate report itself, I think it's better (and more fair), to mark quotations with proper credits, especially, that You're mixing it with Your own findings and requests in-between. And it make people think really *You* got issues dating from kp-settings 26 (in fact, it seems that You haven't had N900 then), which may result in directing questions to You, that should end up in vi_'s "box". Not to mention using language and terms quite out of line for Your actual knowledge/understanding of what is going on here. No offense in this last sentence, just preserving "quality and assurance" of discussion here.

/Estel

Last edited by Estel; 2011-12-01 at 22:02.

Quote & Reply |

The Following 4 Users Say Thank You to Estel For This Useful Post:
don_falcone, handaxe, strange1712, tlex26

vi_	2011-12-01 , 22:35
Posts: 1,680 \| Thanked: 3,685 times \| Joined on Jan 2011	#2398

LoLing
________________

__________________

N900: One of God's own prototypes. A high-powered mutant of some kind never even considered for mass production. Too weird to live, and too rare to die.

Quote & Reply |

The Following 3 Users Say Thank You to vi_ For This Useful Post:
Estel, strange1712, tlex26

StocChr	2011-12-02 , 07:05
Posts: 306 \| Thanked: 38 times \| Joined on Dec 2009 @ Austria	#2399

muhahah, estel, you are the man!!

it's correct, that all the credits from my posting #2397 in this thread are going to vi_, we have discussed the issue in an other thread within 40! replies
why reinvent the wheel? i hope and i think so, that vi_ is not angry about my procceding!

without vi_ i had not come to the idea, that kp-settings fu*k my battery up!
it's fact that kp-settings cause massive problems, and the question is why?

screenshot with kp-settings from my n900:

screenshot without kp-settings from my n900:

Quote & Reply |

xes	2011-12-02 , 08:58
Posts: 638 \| Thanked: 1,692 times \| Joined on Aug 2009	#2400

After a few days of tests and reboots i think i have found something....

There are many posts of people using kp49 without any battery issue...

In the meanwhile, there are many others asking why kp49 drains the battery a lot faster if compared to kp48...

Observing my N900 i have found both of the situations, BUT.... WHEN?

When the battery drains fast, the cpu is always at full clock or remains locked on an high clock profile never changing from that value also if the is apparently there is no load.
Very often, also rebooting, the situation does not change.

In other situations, same phone, same profile (usually i use the "default" one) there is almost no drain and the batteryeye draws a orizontal line.

So what changes?

Look at dmesg... if you can find a
"DSPTrace DSP MMU FAULT "
the battery drain will be like a waterfall!!!!

Many times, also during the boot process you can find that error.
If you start with that, you can be sure, say bye bye to your battery....

If you start the phone without errors, and the battery drain is ok, if while playing media file or running something a DSP FAULT message appears on dmesg, from that moment the cpu will remain locked at an high clock and the battery drain will go to the stars!

Any suggestion or confirm to my teory is welcome!

Thanks to all for this great comunity work and forum!