2 hours ago, balbes150 said:

version 20220916 with the change proposed by tkaiser

 

Well, I was talking about 3 different issues:

• switching with ASPM from powersave to default (you chose performance now, no idea what this does wrt consumption, switching from powersave to defailt is 100mW)
• io_is_busy (that's addressed now though in a redundant way but doesn't matter since only delaying the armbian-hardware-optimization service by a few ms). Both these changes should help with NVMe (and PCIe device) performance in general
• setting the dmc_governor to performance instead of dmc_ondemand (results in 600mW higher consumption though). This is reponsible for better overall performance since the dmc_ondemand governor often does not ramp up memory clock fast enough. A comparison between clocking LPDDR4 with only 528 MHz vs. the current maximum of 2112 MHz is here: https://browser.geekbench.com/v5/cpu/compare/17009700?baseline=17009078

The latter isn't addressed at all but I posted multiple times how this can be changed from userspace:

echo performance >/sys/class/devfreq/dmc/governor

 

32 minutes ago, blondu said:

Does the change help NVMe SSD speed?

@balbes150 did (only) two changes and the io_is_busy tweak will help with every storage device (USB and SATA too). No need for a new image since you can simply edit /usr/lib/armbian/armbian-hardware-optimization. Or overwrite it with the contests of http://ix.io/4aFe

Version 20220916 contains an error in the installation files, do not use it to install on eMMC (if you start the installation process, then you will have to restore u-boot using a USB-C cable). There will be a new version with the correct files in the near future.

Ver 20220917 fix install to eMMC\NVMe\SATA

https://forum.radxa.com/t/rock-5b-debug-party-invitation/10483/472?u=tkaiser

 

Latest sbc-bench version 0.9.9 has also an additional service for ignorant people: reporting probably performance relevant governors...

All of this (maybe except USB PD negotiations) applies to M3 as well: https://github.com/ThomasKaiser/Knowledge/blob/master/articles/Quick_Preview_of_ROCK_5B.md#important-insights-and-suggested-optimisations

27 минут назад, tkaiser сказал:

All of this (maybe except USB PD negotiations) applies to M3 as well

Interestingly, perhaps by the end of the week, when my instance is released, I will try to perform these steps and run the tests.

On 9/16/2022 at 5:02 PM, tkaiser said:

I was talking about 3 different issues:

• switching with ASPM from powersave to default (you chose performance now, no idea what this does wrt consumption, switching from powersave to defailt is 100mW)

 

Today I had the chance to retest again. The good news: there's no real difference whether ASPM is set to default or performance. The bad news is that the idle consumption difference is quite higher. With dmc governor enabled (and therefore idling at 528 MHz DRAM clock) the difference between powersave and default/performance is almost 250 mW, at the highest DRAM clock it's around 150mW (back then when measuring the consumption difference there was no dmc governor and most probably the 100mW I was talking about were the result of rounding down).

 

More details here: https://forum.radxa.com/t/rock-5b-debug-party-invitation/10483/508?u=tkaiser

 

From an energy efficiency point of view a script line in armbian-hardware-optimization checking for PCIe devices being present at boot, then setting either powersave or performance would be perfect but as I recently learned Armbian isn't about such optimisations any more

Kernel 6.0 (mainline + patches) 

 

the details

https://bbs.stationpc.com/forum.php?mod=redirect&goto=findpost&ptid=323&pid=1190&fromuid=636914

30.09.2022 в 17:19, tkaiser сказал:

From an energy efficiency point of view a script line in armbian-hardware-optimization checking for PCIe devices being present at boot, then setting either powersave or performance would be perfect but as I recently learned Armbian isn't about such optimisations any more

I would formulate it differently, "optimization is performed as far as possible, but is not a top priority (due to lack of qualifications\time)" (this is my personal opinion)

Hello, today I enabled the update and lost the graphic mode. To check, I booted a new system from the sd card, it worked and I intentionally updated it, after which it started only in text mode. Can you help me fix it, I have a lot of data and I arranged the graphic mode with many preferences.
Thanks.

16 часов назад, blondu сказал:

today I enabled the update and lost the graphic mode. To check, I booted a new system from the sd card, it worked and I intentionally updated it, after which it started only in text mode. Can you help me fix it, I have a lot of data and I arranged the graphic mode with many preferences.

Details are not specified - the kernel\system version before and after the update. It is not specified what you have changed in the system.

Thank you, I recovered the data and reinstalled after which I blocked the update.

41 минуту назад, blondu сказал:

Thank you, I recovered the data and reinstalled after which I blocked the update.

You have not answered the questions, it is important for other users so that they do not encounter such a situation.

I know for sure that it's not from the kernel because it's in the boot folder that's on the sd card and the rest is on the 254GB ssd, I don't even know that another kernel appeared other than version 5.10.66-station-m3 for rk3588s.
I can try to update manually and find out...

16 часов назад, blondu сказал:

kernel because it's in the boot folder that's on the sd card and the rest is on the 254GB ssd

Hmm, you use SSD for the whole system, but why is /boot on the Sd card? Why not follow the standard installation procedure (via armbian-config) and have a full-fledged system?

Due to the fact that you use a non-standard file location, the system may collapse during the update (there will be a wrong kernel update)

I use the sd card because when I remove it, it boots from the internal emmc where the STATION M3 operating system is.

Is there another dual boot method?

Edited October 27, 20223 yr by blondu

I updated, how can I export the list?

downloadable.txt

Edited October 27, 20223 yr by blondu

20 часов назад, blondu сказал:

I use the sd card because when I remove it, it boots from the internal emmc where the STATION M3 operating system is.

Is there another dual boot method?

I have not tested the compatibility of the Android bootloader (the StationOS system is Android) with Armbian in multi-boot mode. In theory, it is possible to configure the system\bootloader to run different systems, but so far this can only be done through the UART console (video output and USB keyboard are not working yet).

 

16 часов назад, blondu сказал:

I updated, how can I export the list?

I don't understand the question.

I didn't know how to save the list of updates, but I managed and attached it.

Hello, no news about STATION M3 or why it is not included in the official download?

17 часов назад, blondu сказал:

Hello, no news about STATION M3 or why it is not included in the official download?

In the near future, M3 will be added to the download page.

20 часов назад, blondu сказал:

not included in the official download?

https://www.armbian.com/station-m3/

Thank you, it's a good board, it's worth it.

Ver 20221125 kernel 5.10.110

1.What difference is there?

2.Why is it called M4

3.I tried to boot from the SD card but I have no video signal...

Edited November 25, 20223 yr by blondu

15 часов назад, blondu сказал:

1.What difference is there?

Other source code (for details, see the rockchip website).

 

15 часов назад, blondu сказал:

2.Why is it called M4

So that there is no overlap on the previous kernel.

 

15 часов назад, blondu сказал:

3.I tried to boot from the SD card but I have no video signal...

Where is the output connected ? What device ?

It is about Firefly Station M3 and I tried on hdmi and on usb3, the old kernel works on usb3 as well.

 

45 минут назад, blondu сказал:

It is about Firefly Station M3 and I tried on hdmi and on usb3, the old kernel works on usb3 as well.

My HDMI output works without problems. I checked several TV\monitors in 1080p mode.

I am interested in which monitor you are using. Are you writing about USB 3 - meaning USB-C in DP mode for output to the monitor?

Which HDMI cable do you use?

I replaced the hdmi cable and it works, the previous version works with both cables!

For USB-C I use a CABLETIME 4k/60hz cable http://www.amazon.com/Adapter-CABLETIME-Thunderbolt-Compatible-ChromeBook/dp/B082XQL1WS?th=1 and the monitor is a Benq G2420HD, Full HD, DVI, HDMI,
Senseye+photo. It does not work on the DVI input, but on the HDMI it works.