<TheLinuxBug>
so this is fun, while I used that rootmydevice technique to get root (uid 0, gid0) the user while being root does not have the ability to mount/remount /system or do system operations, and instead provides a 'Operation not permitted" so still trying to figure out how to fully root it...
<TheLinuxBug>
(Android)
<TheLinuxBug>
gonna boot arbian and just manually edit things and see how that goes
<TheLinuxBug>
armbian*
vin_ has joined #linux-sunxi
<vin_>
mainline kernel on pine64
<lennyraposo>
yes
<lennyraposo>
what about it ;)
<vin_>
is everything working
<vin_>
or whr can i see the activities?
<vin_>
to see how far it is working
<lennyraposo>
mainline is without display usb audio
<lennyraposo>
check the wiki ;)
vin_ has quit [Quit: Page closed]
<wens>
jrg: that's ShenZhen city, GuangDong province
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
ninolein has quit [Ping timeout: 260 seconds]
ninolein has joined #linux-sunxi
<jrg>
wens: it is the way it is spelled that caught my eye
<jrg>
the orange pi runs a lot cooler than the bpi
<jrg>
bpi with 4 cores shut off hovers around 80C
<jrg>
under load
cnxsoft has joined #linux-sunxi
lvmc has joined #linux-sunxi
Shirasaka-Hazumi has quit [Ping timeout: 244 seconds]
Shirasaka-Hazumi has joined #linux-sunxi
reev has joined #linux-sunxi
arossdotme-planb has joined #linux-sunxi
arossdotme has quit [Ping timeout: 276 seconds]
reev has quit [Read error: Connection reset by peer]
reev has joined #linux-sunxi
reev has quit [Max SendQ exceeded]
reev has joined #linux-sunxi
lvmc has quit [Quit: Page closed]
IgorPec has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
bamvor has quit [Ping timeout: 250 seconds]
arnd has quit [Ping timeout: 250 seconds]
jernej has quit [Ping timeout: 276 seconds]
zerotri has quit [Ping timeout: 260 seconds]
Tartarus has quit [Ping timeout: 260 seconds]
lvrp16 has quit [Ping timeout: 260 seconds]
bamvor has joined #linux-sunxi
nihcas has quit [Ping timeout: 276 seconds]
nihcas has joined #linux-sunxi
lvrp16 has joined #linux-sunxi
Tartarus has joined #linux-sunxi
zerotri has joined #linux-sunxi
arnd has joined #linux-sunxi
dearfibonacci has joined #linux-sunxi
reinforce has joined #linux-sunxi
ssvb has quit [Remote host closed the connection]
arnd has quit [Ping timeout: 250 seconds]
arnd has joined #linux-sunxi
disik has joined #linux-sunxi
jrg has quit [Ping timeout: 252 seconds]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
jrg has joined #linux-sunxi
IgorPec has quit [Ping timeout: 252 seconds]
IgorPec has joined #linux-sunxi
apritzel has joined #linux-sunxi
bfree has quit [Ping timeout: 276 seconds]
massi has joined #linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
peepsalot has quit [Ping timeout: 250 seconds]
yann|work has quit [Ping timeout: 260 seconds]
popolon has joined #linux-sunxi
peepsalot has joined #linux-sunxi
matthias_bgg has joined #linux-sunxi
Mr__Anderson has joined #linux-sunxi
apritzel has joined #linux-sunxi
<Amit_T>
Hello, where value of this macro REG_PMU_UNK_H3 drived from ?
<Amit_T>
I mean from where value of this macro REG_PMU_UNK_H3 is drived from ?
<tecc>
anyone can help me out on how to boot from SD card in linux on my Mele A1000G Quad?
cnxsoft1 is now known as cnxsoft
lemonzest has joined #linux-sunxi
tecc has quit [Quit: gonna try later]
reev has quit [Read error: Connection reset by peer]
merbanan has joined #linux-sunxi
<TheLinuxBug>
*sigh* after hours of messing with this android to root it, it turns out that the kernel used in the Android image for Plus 2E happens to have secure mount in the kernel which prevents you remounting /system as rw from in the kernel. So looks like only way to truly root the system and fix the DRAM stuff is gonna be to manually build my own Android image from bottom up.
<TheLinuxBug>
Am I the only one that thinks when you release a dev board you should provide an Android image that is actually rootable withotu having to tear it all apart and rebuild it?
<TheLinuxBug>
huge waste of time.
<KotCzarny>
its dev board, not user board
<KotCzarny>
:P
<KotCzarny>
also, android :P
<TheLinuxBug>
all I really want is to have root and fix the DRAM config, both which seem like I am gonna have to basiaclly tear down the android image and rebuild it just to insert
<TheLinuxBug>
all the convient programs like DragonFace arn't compatible with the image they are providing
<KotCzarny>
just go to the nearest phone store amd grab 2-4GB microsd?
<KotCzarny>
much faster
<TheLinuxBug>
i have one, but I want this to run from EMMC
<TheLinuxBug>
that why I bought a board with emmc ;p
<KotCzarny>
i meant to fix dram setting only :P
<TheLinuxBug>
I already got armbian on a sdcard and loaded it
<TheLinuxBug>
however, again, you still have to deconstruct and reflash if you want to change things
<TheLinuxBug>
it is not as simple as mounting the available partitions
<TheLinuxBug>
you have to extract boot.img
<TheLinuxBug>
then you have to decompress it
<TheLinuxBug>
change the setting
<TheLinuxBug>
etc
<TheLinuxBug>
I was hoping it would be an easier change
<KotCzarny>
same
<TheLinuxBug>
I guess I will have to spend some real time on this later and make my own non-shitty image
<KotCzarny>
dont forget to update the wiki
<TheLinuxBug>
;)
<TheLinuxBug>
well its just a shame that these guys released it with poor dram settings
<TheLinuxBug>
if I try to playback full 1080p in android
<TheLinuxBug>
it craps its self
<KotCzarny>
o.O
<TheLinuxBug>
:(
<TheLinuxBug>
same with teh GPU test, any testing of GPU leads to freeze/reboot
<TheLinuxBug>
cause those settings are obviously too high
<KotCzarny>
they are or not
<KotCzarny>
many boards pass 672 (at least on opipc)
<TheLinuxBug>
well I confirmed its set to 672
<TheLinuxBug>
but it doesn't seem to work out for me :(
<KotCzarny>
yeah, my opipic fails limamemtester @672
<KotCzarny>
;)
dearfibonacci has quit [Remote host closed the connection]
<TheLinuxBug>
so this weekend it looks like I will be making a better androidimage with lower DRAM freq, I think I will stick to 600 to be on the safe side
<oliv3r>
ssvb well it look slike even the di 8 doesn't work; U-Boot SPL 2014.10+dfsg1-5 (Apr 07 2015 - 21:52:20)
<oliv3r>
ohhh wait, this might be a rare a20 lime1 board i have
<oliv3r>
CPU: 912000000Hz, AXI/AHB/APB: 3/2/2
<oliv3r>
duh, it is an a20
ssvb has joined #linux-sunxi
<oliv3r>
oh heh, was talking to thin air :)
bmeneg has joined #linux-sunxi
<oliv3r>
ssvb: i think i figured out my problem, a20 soc on an lime1 board (stupid heatsink covering the soc)
<ssvb>
oliv3r: ok, it's good that this has been resolved :)
cptG has joined #linux-sunxi
<ssvb>
we can probably add a safety check for the SoC type in the SPL
* jonkerj
has been adding stuff to the H3/OrangePi+ DTS lately, enabling a lot of stuff
<bmeneg>
guys, let me ask you a single question, FEX files are specific to sunXi SoCs? It isn't related to old Kernel version at all?
<jonkerj>
some of this "should" be in mainline, like the enabling OHCI1 for instance
<oliv3r>
ssvb: it 's just stupid mistake but yeah adding a check in the SPL might not be a bad idea
<jonkerj>
other stuff (uart/i2c on GPIO header) I'm not so sure
<oliv3r>
'you are running on but this was compiled for'
<oliv3r>
bmeneg: both! old sunxi kernels :)
Worf has quit [Quit: Konversation terminated!]
<jonkerj>
should I push these patches somewhere, so others can benefit from this?
<cptG>
I saw that sun4i-ts also includes code for temperatture information. Can I read the values via sysfs? I can find i2c0 but this seems to be the temp of the power management chiip (AXPfoo)...
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<oliv3r>
cptG: yes, you get a rough cpu temperature from the sun4i-ts
<oliv3r>
cptG: you need to enable it in the kernel (if it is not done so)
<bmeneg>
oliv3r, yeah, but it isn't present on any other SoC's vendors, right?
<oliv3r>
bmeneg: exactly
<bmeneg>
oliv3r, thx
tkaiser has joined #linux-sunxi
<cptG>
oliv3r: I see. I saw device tree stuff got added to mainline but I'm on 3.4.103 and can't find anything related in config.gz at first glance. You happen to know if it's possible at all on "the old" kernel?
<TheLinuxBug>
tkaiser: You know what, you did and I overlooked it, tbh, my grandmother died yesterday afternoon (I cared for her last 1.5 years she was 93) and I was just trying to find something to keep me busy and my mind was all over the place, I woke up this morning and just started toying with it again, thanks for again providing that. I will for sure go through that now.
dearfibonacci has joined #linux-sunxi
<TheLinuxBug>
ahh
<TheLinuxBug>
need to find my serial cable
<TheLinuxBug>
its in one of like 80 boxes
<TheLinuxBug>
:(
<KotCzarny>
just buy new one?
<oliv3r>
cptG: not really no
<NiteHawk>
cptG: config.gz support has been around for ages - check CONFIG_IKCONFIG=y and CONFIG_IKCONFIG_PROC=y are set in your kernel config
<TheLinuxBug>
the hunt is on!
<oliv3r>
cptG: 3.4 = fex; 3.9 or 3.10+ devicetree
<tkaiser>
TheLinuxBug: Not necessarily. That's just how wens extracted the fex. So you can download it, modify to your needs, convert back to bin and then overwrite the approriate sectors on eMMC when running from SD card / in Armbian
<NiteHawk>
oh, nvm - you were talking about device tree - that's indeed a post 3.4 kernel thing
<tkaiser>
oliv3r: One exception: Allwinner's 3.10.65 BSP kernel: A crude mixture between sys_config.fex and .dts ;)
<oliv3r>
tkaiser: ugh
<tkaiser>
oliv3r: But in the end you can drop all the fex stuff there and completely rely on .dts
<oliv3r>
tkaiser: i wonder what they'll do if they ever have to upgarde to a 4.4+ kernel; that will be an interesting monster
<TheLinuxBug>
ahh I see
<tkaiser>
oliv3r: I think the next Allwinner SoCs will get 3.10.65 (at least until 2018 or so)
<cptG>
NiteHawk: I do have config.gz What I meant was that zgrep SUN /proc/config.gz did not show anything that seems related to TS, temparature or alike. Just that touch is enabled...
<cptG>
oliv3r: yeah, I know that dtb is not for me unfortunately. Assuming you're on mainline and the needed changes in the device tree are present. How would you go about querying the temp? Maybe I can find sth relevant from that direction ;)
<tkaiser>
cptG: Basically you search for a tool to read out A20 SoC temperature with an old kernel, yes?
<tkaiser>
cptG: On A20 use it with ' 1447'. And don't be surprised if values are random :)
<cptG>
tkaiser: you're my hero!
<tkaiser>
cptG: People love numbers without meaning. Applies to thermal readouts too. A20 thermal sensor is not that reliable if you ask me
<TheLinuxBug>
tkaiser: now to ask the very stupid question -- So I patch the fex to my needs, fex2bin it and I have boot armbian on the board, how exactly do I find the correct sectors to overwrite/what would be the correct procedure - if you don't have time to explain a link to any docs on learning that info would be appriciated.
<cptG>
tkaiser: I have a thermal camera at hand so I'll see if I can find some correlation in the randomness :)
<TheLinuxBug>
tkaiser: I will also include your suggested changes here, but I do have all the tools to get to that point and could have fex and have it fex2bin in next 20 so minutes.
<tkaiser>
cptG: Especially the link to the forum thread
<tkaiser>
TheLinuxBug: I don't know the offset unfortunately. Give me a minute
<cptG>
What puzzles me is the fact that they somehow thought Temperature and Touch interface are related :) On the other hand - they're both some sort of ADC... Who knows.
<cptG>
tkaiser: either way. Thanks for the pointers, it does seem that the values might actually relate to the temperature. Hogging the CPU as we speak ;)
dearfibonacci has quit [Quit: Leaving]
jstein has joined #linux-sunxi
cnxsoft has quit [Remote host closed the connection]
<GeneralStupid>
hi, i tested two different usb WLAN sticks on my orangepi pc and i guess its a kernel / USB problem because its very very slow.
<TheLinuxBug>
tkaiser: I assume I should be able to dd the fex to the emmc with same or similar settings?
<GeneralStupid>
the access point is 1.5 meters away
<GeneralStupid>
sometimes it reaches 2 Mb/s but normally i stuck around 100 kb/s
<TheLinuxBug>
GeneralStupid: you sure you don't have any heavy network traffic causing the slow down (contention), I ran into an issue on my RPi3 that I had 2 security cameras on same network I slapped it on and it just couldn't get enough throughput?
<tkaiser>
cptG: The thermal sensor is there 'by accident'. Allwinner included a TP controller, this controller has a thermal sensor and values can be read out. Where this sensor is or how accurate and what needs to get a proper calibrated temperature curve out of it: Who knows?
<TheLinuxBug>
GeneralStupid: it acted really weird and I got very poor throughput until I placed it on a less busy network
<GeneralStupid>
TheLinuxBug: i have only 2 WLAN devices right now
<GeneralStupid>
TheLinuxBug: my laptop and the orangepi
<TheLinuxBug>
hm
merbanan has quit [Ping timeout: 240 seconds]
<TheLinuxBug>
have you attepmted to tweak sysctl to see if it helps at all?
<GeneralStupid>
TheLinuxBug: which variable exactly?
<TheLinuxBug>
GeneralStupid: pm
nove has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
<tkaiser>
TheLinuxBug: No idea, you would need such an u-boot.fex, a hex editor and your .bin file to get an idea about the offsets.
<tkaiser>
TheLinuxBug: Maybe it's written somewhere in linux-sunxi wiki at which offset a binary sys_config.fex has to be written to
yann|work has joined #linux-sunxi
Amit_t_ has joined #linux-sunxi
merbanan has joined #linux-sunxi
cosm has joined #linux-sunxi
merbanan has quit [Ping timeout: 246 seconds]
<GeneralStupid>
i guess it is caused by the sd card (wa)
<cptG>
tkaiser: Kelvin, Celsius, Fahrenheit, Allwinners - doesn't really matter as long as I can name some "too hot" value. And it seems to correlate with the actual temp of the SoC, depending on which point of the SoC I'm focussing on with the IR camera, I'm getting a readout of ~73% of the actual temperature. Works for me ;)
<cptG>
tkaiser: User space works fine for me. The device has a broker service running with root privileges that I can query via UNIX domain sockets. That service can run the application on demand and I feels safe enough with that. I prefer that over recompiling the kernel ;)
<tkaiser>
There you find the iozone call you could try to verify (we've seen many cards with pretty poor random IO at 16K writes, maybe it's even worse with 24K which seems the chunk size your wget processes try to write)
<GeneralStupid>
tkaiser: there no units in that graph
<tkaiser>
Also some recommendations. Samsung EVO/EVO+ with 32 or 64 GB seem to be best buy for sunxi boards since here is sequential performance limited to 23 MB/s anyway
<tkaiser>
GeneralStupid: Should be KB/s
<GeneralStupid>
tkaiser: iam using a 16 Gb Samsung Evo
<GeneralStupid>
tkaiser: i already read, allwinners kernel is shitty too :)
<GeneralStupid>
tkaiser: but mainline is not really useable with hdmi and stuff
<tkaiser>
GeneralStupid: iozone -e -I -a -s 100M -r 24k -i 2
<GeneralStupid>
tkaiser: is your armbian image with video and audio and stuff?
<tkaiser>
Or add '-0 -1' also to get the whole picture
<GeneralStupid>
in which packet is iozone?
<tkaiser>
Sorry, '-i 0 -i 1'
<tkaiser>
GeneralStupid: Which distro you're using?
<GeneralStupid>
debian
<tkaiser>
iozone3
<GeneralStupid>
...
jernej has joined #linux-sunxi
<tkaiser>
GeneralStupid: In case you let the wget tasks run this might take ages (same without, so maybe limiting to 10M as first step would be an idea)
<tkaiser>
GeneralStupid: Looks like broken SD card or at least fsck is needed urgently ;)
<tkaiser>
Or maybe iozone is somewhat weird. Better try to prefix '-i 2' with '-i 0 -i 1' and try again
jemk_ is now known as jemk
<GeneralStupid>
tkaiser: ill try later got to go now. but thanks
merbanan has joined #linux-sunxi
yann|work has quit [Ping timeout: 260 seconds]
disik has quit [Ping timeout: 240 seconds]
merbanan has quit [Ping timeout: 260 seconds]
zuikis has joined #linux-sunxi
reinforce has joined #linux-sunxi
<oliv3r>
ssvb: where would you put this code more generically (with some defines): http://git.denx.de/?p=u-boot/u-boot-sunxi.git;a=blob;f=arch/arm/mach-sunxi/board.c;h=bd15b9bfb056c217c68b956a5aa173e037b3a8e7;hb=refs/heads/next#l154
<oliv3r>
ssvb: e.g. a sunxi_get_version()
<oliv3r>
i always forget why we have 2 board.c (march and board) :S
lamer14654877111 has joined #linux-sunxi
lamer14654877111 has quit [Client Quit]
tkaiser has quit [Ping timeout: 272 seconds]
tkaiser has joined #linux-sunxi
merbanan has joined #linux-sunxi
cptG_ has joined #linux-sunxi
cptG has quit [Ping timeout: 250 seconds]
pmattern has joined #linux-sunxi
pmattern has quit [Client Quit]
Nacho_ has quit [Ping timeout: 244 seconds]
Nacho_ has joined #linux-sunxi
phipli has joined #linux-sunxi
vagrantc has joined #linux-sunxi
caog has quit [Quit: Ex-Chat]
popolon has quit [Quit: WeeChat 1.3]
libv__ has joined #linux-sunxi
libv has quit [Ping timeout: 244 seconds]
libv__ has quit [Ping timeout: 244 seconds]
libv has joined #linux-sunxi
The_Loko has quit [Quit: Leaving]
enrico_ has quit [Quit: Bye]
JohnDoe_71Rus has joined #linux-sunxi
oneinsect has joined #linux-sunxi
<oneinsect>
friends how is the progress on mainline for Allwinner H3 coming?
<montjoie>
oneinsect: see the mainlining wiki page, I dont know what feature you expect
<KotCzarny>
the most important, ths and friends
<montjoie>
i have an idea for mainlining page, how about a feature matrix (SoC x feature) (with direct link to the wiki info) and beautifulcolor like green(mainline) yellow(driver done) red(nothing)
<KotCzarny>
black(never)
<montjoie>
:)
<montjoie>
like my feature matrix for Crypto on sunxi.montjoie.ovh
<oneinsect>
usbs working and ethernet montjoie:
<oneinsect>
one of the usbs used to crash
<KotCzarny>
oneinsect: without power management it will get hot as hell
<oneinsect>
has it been worked on?
<oneinsect>
oooh
<montjoie>
oneinsect: ethernet driver was submitted and I have mostly finished corrections (soon v2)
<oneinsect>
ooh i thought power management was done?
<montjoie>
no, and the last patch is old
<tkaiser>
oneinsect: Patches exist but not sent upstream. On Orange Pi that doesn't matter that much, the only H3 boards that overheat way too easy are BPi M2+ and....... your NanoPi M1
<oneinsect>
its good you remember me tkaiser: and my board
<oneinsect>
btw loverpi sent me
<oneinsect>
two orange pi;s
<KotCzarny>
which ones?
<oneinsect>
one pc and another opi one....so i have shifted now
<KotCzarny>
:)
<oneinsect>
i keep asking but is there is a place where all these patches were collected and being tested?
<TheLinuxBug>
tkaiser: it looks like I need the SDK to be able to do that, is the SDK the same between the 2 and 2E?
<TheLinuxBug>
Cause I didn't see one for 2E yet
<tkaiser>
You can even grab that for Banana Pi M3 (A83T) since you just need a few crappy binaries inside
<TheLinuxBug>
hmm now to find a source that doesn't go 20kb/sec :(
<TheLinuxBug>
ugh I take that back find a source even
<TheLinuxBug>
baidu is unacceptable
libv has quit [Ping timeout: 244 seconds]
<TheLinuxBug>
worst place ever to host files
<TheLinuxBug>
also why does it take forever for oragepi.org to authorize new sign-ups and why the heck do they make it so you can't even browse the forum when your in the waiting auth group
<tkaiser>
TheLinuxBug: Might be, then take a look into SinoVoip's M2P repo whether this stuff is there. But sys_config.fex might be in reality the fex wens extracted (won't check now, bandwidth is too low here)
<bjoernb>
solving captchas when using torbrowser on linux-sunxi.org is not nice.
<bjoernb>
thanks though for the link
tkaiser has quit [Ping timeout: 244 seconds]
ricardocrudo has joined #linux-sunxi
jstein has joined #linux-sunxi
tkaiser has joined #linux-sunxi
Amit_t_ has quit [Quit: Page closed]
IgorPec2 has quit [Ping timeout: 258 seconds]
vetkat has quit [Quit: Bye bye]
<bjoernb>
got screen and uart running. thx.
phipli has quit [Ping timeout: 264 seconds]
mpmc has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
mpmc has quit [Quit: ZNC 1.6.3+deb1+jessie0 - http://znc.in]
nove has quit [Quit: nove]
mpmc has joined #linux-sunxi
mosterta has joined #linux-sunxi
<TheLinuxBug>
tkaiser: before pulling all my hair out, I decided to test something before trying to modify the DRAM freq. and instead I installed a Aluminum heat sink (pulled it off my RPi 2, so not a very agressive one) but just adding it has results in the GPU stability test running now for 7 minutes 63 GPU passes 10 CPU passes and is still running currently without freezing/rebooting. So... hmm
<TheLinuxBug>
LOL I am gonna eat my hat, 8 minutes 72 gpu passes, red light of death :(
<vagrantc>
anyone have links to the H3 patches for mainline linux regarding CPU throttling?
staplr has joined #linux-sunxi
* vagrantc
vaguely recalls someone being nudged to submit them
<apritzel>
ssvb: not yet, just could read the manufacturer ID yesterday with this spidev-test from the kernel source
<apritzel>
but I am going to test flashrom now
al1o has joined #linux-sunxi
<megi>
I've collected all the patches that are needed over the vanilla 4.6/4.7 kernels to run Orange Pi PC and Orange Pi One here: https://xff.cz/orange-pi-dvfs/