terra854 has quit [Quit: Connection closed for inactivity]
egbert has joined #linux-sunxi
whaf has joined #linux-sunxi
kaspter has joined #linux-sunxi
ninolein has quit [Ping timeout: 256 seconds]
ninolein has joined #linux-sunxi
gumblex has joined #linux-sunxi
Seppoz has quit [Ping timeout: 246 seconds]
gumblex has quit [Remote host closed the connection]
gumblex has joined #linux-sunxi
bugzc has quit [Ping timeout: 260 seconds]
gumblex has quit [Remote host closed the connection]
deskwizard has joined #linux-sunxi
gumblex has joined #linux-sunxi
KB3VGW has joined #linux-sunxi
chomwitt has quit [Ping timeout: 256 seconds]
<KB3VGW>
ok who was i talking with about the openwrt img and pine64
pg12 has quit [Ping timeout: 268 seconds]
pg12_ has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
Andy-D has quit [Ping timeout: 268 seconds]
bugzc has joined #linux-sunxi
terra854 has joined #linux-sunxi
IgorPec has joined #linux-sunxi
HeavyMetal has quit [Ping timeout: 250 seconds]
JohnDoe_71Rus has joined #linux-sunxi
HeavyMetal has joined #linux-sunxi
HeavyMetal has joined #linux-sunxi
HeavyMetal has quit [Changing host]
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
muvlon has joined #linux-sunxi
jernej has joined #linux-sunxi
mpmc has quit [Ping timeout: 244 seconds]
mpmc has joined #linux-sunxi
leviathanch has joined #linux-sunxi
deskwizard has quit [Ping timeout: 256 seconds]
reinforce has joined #linux-sunxi
Putti has quit [Ping timeout: 256 seconds]
JohnDoe_71Rus has quit [Read error: Connection reset by peer]
JohnDoe_71Rus has joined #linux-sunxi
jernej has quit [Quit: Konversation terminated!]
muvlon has quit [Ping timeout: 260 seconds]
gpgreen has quit [Quit: Konversation terminated!]
muvlon has joined #linux-sunxi
terra854 has quit [Quit: Connection closed for inactivity]
KB3VGW has quit [Read error: Connection timed out]
KB3VGW has joined #linux-sunxi
<freemangordon>
ok, got 4.9-rc6 booting, lets see if I can boot devuan
Putti has joined #linux-sunxi
<freemangordon>
great, devuan boots, mission accomplished, thanks guys and girls for all the help :)
mihlit has joined #linux-sunxi
<freemangordon>
MoeIcenowy: while I was building the kernel, make modules produced almost no .kos. Do I need to tweak config to enable something? also, it seems USB keyboard is not recognized, is that expected?
f0xx has joined #linux-sunxi
kaspter has quit [Ping timeout: 250 seconds]
Putti has quit [Ping timeout: 250 seconds]
Putti has joined #linux-sunxi
<tuxillo>
hi
IgorPec has quit [Ping timeout: 250 seconds]
IgorPec has joined #linux-sunxi
<jelle>
freemangordon: usb keyboard via otg?
IgorPec2 has joined #linux-sunxi
<freemangordon>
jelle: yes
<freemangordon>
tablet came with a kbd in the box
<jelle>
you need to modprobe a few modules g_serial and others
<freemangordon>
jelle: it needs gadget loaded?
<jelle>
yup
<freemangordon>
hmm, and it won;t work even if I change kernel config to host-only?
<freemangordon>
that's bad, because I need keyboard to do modprobe :)
IgorPec has quit [Read error: Connection reset by peer]
<freemangordon>
I don;t have those build, seems defconfig needs some tweaking
<jelle>
freemangordon: did you figure out what touchscreen you have?
<freemangordon>
jelle: no, will do it as soon as I have devuan shell with kbd
<jelle>
freemangordon: oh I'd also enable serial over usb but not sure how that works on non-systemd
<freemangordon>
already did it, but it seems I have to do what you said ^^^ first
<freemangordon>
(rc.local)
<freemangordon>
jelle: how's kenel modules support related to systemd?
<freemangordon>
*kernel
<jelle>
not
<freemangordon>
serial-over-usb works on n900 with upstart :). Mainline kernel that is
<jelle>
sure
<freemangordon>
anyway, thanks
<jelle>
btw you could have detected with touchscreen controller you had with adb shell
<jelle>
also you might need the android kernel module for the firmware later
<freemangordon>
jelle: I have no experience with android tools
<freemangordon>
but sure, will do whatever is needed when it comes to it
<freemangordon>
I guess having login prompt via fel in 2-3 hours is a good progress so far :)
<freemangordon>
time to go to work, see you later, night
<miasma>
omg, i just realized why one of my orange pi pcs was failing. the boards have huge differences in power handling capacities. if you power the board without an sd, the ethernet port has lights on by default. now i added two usb devices. the lights went totally dim. this one board can't handle load at all
<miasma>
i managed to help it a bit by increasing dc plug voltage to 5,3V but it probably won't handle a lot more
<jelle>
freemangordon: good luck :)
<miasma>
the problem was, I was using an usb ethernet dongle and it was using too much power -> board won't boot
lemonzest has joined #linux-sunxi
paulk-collins has quit [Ping timeout: 268 seconds]
terra854 has joined #linux-sunxi
tkaiser has joined #linux-sunxi
yann|work has joined #linux-sunxi
ErwinH has joined #linux-sunxi
fkluknav has joined #linux-sunxi
<dgp>
MoeIcenowy: I know the root cause of the firmware crash. Sometimes the driver get's confused and things there is data to read, that causes a data error on the sdio bus.. the driver sees that and tries to read again (it tries up to 3 times) and the retry probably triggers an assert in the firmware that checks if the host is reading when there is nothing to readf
<dgp>
The block size thing just seems to hide the issue with the driver thinking there is data to read when there isn't
lkcl has quit [Read error: Connection reset by peer]
lkcl has joined #linux-sunxi
Ultrasauce has quit [Ping timeout: 244 seconds]
bugzc has quit [Ping timeout: 260 seconds]
Ultrasauce has joined #linux-sunxi
Worf has joined #linux-sunxi
BuddyZhang1 has quit [Quit: Page closed]
BuddyZhang1 has joined #linux-sunxi
massi has joined #linux-sunxi
lkcl has quit [Read error: Connection reset by peer]
lkcl has joined #linux-sunxi
leviathanch has quit [Remote host closed the connection]
LargePrime has quit [Ping timeout: 268 seconds]
f0xx has quit [Quit: terminated!]
f0xx has joined #linux-sunxi
fl_0 has quit [Ping timeout: 258 seconds]
fl_0 has joined #linux-sunxi
LargePrime has joined #linux-sunxi
mzki has joined #linux-sunxi
Putti has quit [Ping timeout: 244 seconds]
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 265 seconds]
fkluknav has quit [Ping timeout: 258 seconds]
mhlavink has quit [Ping timeout: 260 seconds]
lemonzest has quit [Ping timeout: 250 seconds]
fkluknav has joined #linux-sunxi
lemonzest has joined #linux-sunxi
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 258 seconds]
whaf has quit [Ping timeout: 246 seconds]
foodev has joined #linux-sunxi
libv_ has joined #linux-sunxi
whaf has joined #linux-sunxi
mhlavink has joined #linux-sunxi
libv has quit [Ping timeout: 248 seconds]
BenG83 has joined #linux-sunxi
smashr_ has joined #linux-sunxi
<smashr_>
my dev kernel's package name is wrong, it doesn't include "-dev" and hence it can't be installed during the sd card image build
<smashr_>
did anyone experience the same issue?
<KotCzarny>
are you aware this channel is distro agnostic mostly?
<KotCzarny>
people are using different distros, but overall kernel is distro independent
<smashr_>
hm right :)
<smashr_>
gonna ask on armbian's forums, sry
fkluknav has quit [Ping timeout: 268 seconds]
dan0_0 has quit [Ping timeout: 252 seconds]
lkcl has quit [Read error: Connection reset by peer]
lerc has quit [Read error: Connection reset by peer]
dan0_0 has joined #linux-sunxi
Amit_T has quit [Ping timeout: 260 seconds]
lerc has joined #linux-sunxi
mhlavink has quit [Ping timeout: 246 seconds]
mhlavink has joined #linux-sunxi
Worf has quit [Quit: Konversation terminated!]
smashr_ has quit [Quit: Page closed]
Amit_T has joined #linux-sunxi
lkcl has joined #linux-sunxi
libv has joined #linux-sunxi
msevwork has joined #linux-sunxi
libv_ has quit [Ping timeout: 256 seconds]
mhlavink has quit [Ping timeout: 248 seconds]
mhlavink has joined #linux-sunxi
yann-kaelig has joined #linux-sunxi
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 265 seconds]
paulk-collins has joined #linux-sunxi
fkluknav has joined #linux-sunxi
dizz74 has joined #linux-sunxi
<dizz74>
Good day. Somebody know what different allwinner from R16 than A33? In hardware and in software. Can I build u-boot for R16 from A33 sources? Thank
<KotCzarny>
as for uboot: yes
<KotCzarny>
the rest is marketing
<dizz74>
And kernel too? from a33
<KotCzarny>
same
<dizz74>
okay, thank you
terra854 has quit [Quit: Connection closed for inactivity]
libv has joined #linux-sunxi
fkluknav has quit [Remote host closed the connection]
fkluknav has joined #linux-sunxi
libv_ has quit [Ping timeout: 245 seconds]
Putti has joined #linux-sunxi
yann-kaelig has quit [Quit: Leaving]
yann-kaelig has joined #linux-sunxi
pg12_ has quit [Ping timeout: 244 seconds]
deskwizard has joined #linux-sunxi
pg12 has joined #linux-sunxi
afaerber has joined #linux-sunxi
libv_ has joined #linux-sunxi
BuddyZhang1 has quit [Ping timeout: 260 seconds]
libv has quit [Ping timeout: 256 seconds]
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
chomwitt has joined #linux-sunxi
hojnikb has joined #linux-sunxi
<hojnikb>
guys
<hojnikb>
if i want to use armbian on h5
<hojnikb>
what do i have to do
<hojnikb>
use a64 image and replace the kernel from xunlong debian image
<hojnikb>
anything else ?
<KotCzarny>
i think there was one experimental
<KotCzarny>
but you can just wait if you dont do dev work to make it work
<KotCzarny>
it shouldnt be long
jonkerj has quit [Quit: reboot]
<hojnikb>
KotCzarny: great
<hojnikb>
because xunlong image is pretty much useless
<hojnikb>
cant even change resolution without fideling with fex files
<KotCzarny>
and if you are impatient just search armbian forums from time to time
<tkaiser>
hojnikb: And be assured that every time someone asks 'when will Armbian for OPi PC 2 be ready?' we delay everything one more week. If you want something that works _now_ choose different hardware.
<wens>
qschulz: i'll try to make time this week to review the axp gpio patches you sent
<qschulz>
wens: thanks, that'd be great!
KB3VGW has quit [Ping timeout: 258 seconds]
KB3VGW has joined #linux-sunxi
<dgp>
tkaiser: got to love it when people pressure others to do something for free right now
libv has joined #linux-sunxi
<tkaiser>
dgp: Exactly, especially at this time where providing an 'desktop image' is BS anyway.
libv_ has quit [Ping timeout: 240 seconds]
msevwork has quit [Quit: Leaving]
naobsd has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
calimero_82 has joined #linux-sunxi
<calimero_82>
hi
orly_owl has quit [Quit: leaving]
lkcl has quit [Read error: Connection reset by peer]
JohnDoe_71Rus has joined #linux-sunxi
gumblex has quit [Quit: Bye.]
hojnikb has joined #linux-sunxi
<hojnikb>
dgp: I dont think anyone is pressuring anybody to do something :)
<hojnikb>
but it is nice to know whats the state of such images for us not doing dev work
hojnikb has quit [Client Quit]
popolon has joined #linux-sunxi
libv_ has joined #linux-sunxi
libv has quit [Ping timeout: 245 seconds]
apritzel has joined #linux-sunxi
<tkaiser>
Hmm... by looking at the 'model' description of some sunxi/sun8i devices I got the impression that these do not always follow strict rules? http://pastebin.com/eGmA00XY
<tkaiser>
Are there rules like 'vendor device-name' or something like that?
f0xx has quit [Ping timeout: 248 seconds]
reinforce has joined #linux-sunxi
Christos_ has joined #linux-sunxi
fkluknav has quit [Ping timeout: 250 seconds]
<Christos_>
Does by any chance someone knows which headers should I install if I use H5 ? currently having the Xunlong image and trying to do some compile but need headers (as it seems)
<calimero_82>
hi guys i've bought orange pione, is it normal with openelec temperature is 65 ? or is too hot?
jernej has joined #linux-sunxi
<calimero_82>
jernej do u have orange pi one?
<tkaiser>
jernej: You had a tool around to read out this sys_config.fex stuff from memory, right? From memory?
<jernej>
calimero_82: no, why do you ask?
<jernej>
tkaiser: no, I only extracted it from android images
<Christos_>
..Hi tkaiser: .. lol you are indeed pretty much pissed off, just looked at your remark about PC2. lol
<tkaiser>
jernej: I thought in the beginning you had a tool running that extracted some information from DRAM? Back in Feb or March IIRC we talked about...
<KotCzarny>
he is not pissed, he is just tkaiser
<calimero_82>
jernej: about temperature, is hot 65 70 c ?
<calimero_82>
i use openelec
<jernej>
calimero_82: if you shutdown kodi, then it is around 45
<jernej>
it also depends on your ambient temperature
<jernej>
one thing you can do to lower temperature in home screen is disabling RSS feeds
fkluknav has joined #linux-sunxi
<calimero_82>
i've bought also heatsink
<calimero_82>
openelec starts with 65 c
<jernej>
tkaiser: Then that was certainly not me
<jernej>
as I said, disable RSS feeds
<calimero_82>
jernej: t
<calimero_82>
sorry
<calimero_82>
maybe i've downloaded the wrong version?
bmeneg has joined #linux-sunxi
<jernej>
uh, I don't test much on boards with only two cpu voltage options
<bmeneg>
hello everyone, i'm using cubietruck to built a personal project where I was trying to use PB16 and PB17 pins as GPIOs but it didn't work.
<calimero_82>
jernej: i've developement openelec, isn't ok for the one?
<jernej>
calimero_82: buy better board... if you ask me, boards like one, lite, bpi m2+ and bx2 are not best suited for OE
<bmeneg>
these pins just doesn't respond to any action (in/out, high/low) either accessing it through sysfs or directly through memory mapping
<tkaiser>
jernej: Hmm... almost sure, IIRC I proposed something stupid like 'do bin2fex' and you said you've a tool lying around that extracts specific sysconfig bits from DRAM. But then I'm wrong here
<jernej>
actually, they are not best suited for many thing
<calimero_82>
thanks jernej
bmeneg has quit [Client Quit]
<jernej>
tkaiser: Isn't something like that described on the wiki?
bmeneg has joined #linux-sunxi
<dgp>
bmeneg: what kernel? have you got the pinmux setup? I'm not sure if the kernel does that when you export the gpio but you can check in debugfs
<bmeneg>
dpg: I tried with kernel 4.3 (dts config file) and 3.4 (fex config file) and on both didn't work.
<bmeneg>
dgp, at first I thought it could be something related to some kernel module and than tried to access pin configuration and data registers directly through memory mapping (mmap). I could read and write values to it, checked each write to its stored value, and so on, everything seems correct, but the values weren't driven to the physical pin itself
fkluknav has quit [Ping timeout: 244 seconds]
<dgp>
bmeneg: There is a mux after the gpio controller that selects what is connected to the physical pin. If you look in the pinctrl nodes in debugfs it should tell you want pinmux settings look like
<bmeneg>
dgp, any other pin work perfectly. but PB14, 15, 16 and 17 don't.
<jernej>
tkaiser: I honestly completely forgot on that program
<dgp>
it's possible they are configured to something other than gpio
<bmeneg>
dgp, wow. I'll take a look. I didn't know about that.
<bmeneg>
dgp, these pastes has just the lines related to the pins I mentioned
<bmeneg>
dgp, I checked if any other modules (JTAG for instance) is using them, but everything could conflict are disabled
IgorPec has joined #linux-sunxi
<tkaiser>
jernej: Don't know yet and hope to delegate the task to Pete Scargill. I've no idea how all this low-level stuff works so just tried to prepare a bit and collect information.
<freemangordon>
jelle: adding insmod stuff in rc.local doesn't seem to help. Any idea what else is needed to have USB working?
leviathanch has joined #linux-sunxi
<freemangordon>
jelle: I fixed it :)
<KotCzarny>
what was the problem?
<freemangordon>
KotCzarny: USB keyboard was not running
<freemangordon>
it runs now, along with LXDE :)
<KotCzarny>
i mean, what was missing piece to make it work
<freemangordon>
USB gadget support and USB OTG
<freemangordon>
in kernel config that is
<KotCzarny>
mhm
[Awaxx] has quit [Quit: "Hack at your level, don't try to be someone else."]
<freemangordon>
KotCzarny: now it's wifi's turn
<freemangordon>
I guess I'll need some firmware
deskwizard has quit [Ping timeout: 240 seconds]
pekka10 is now known as pekka_afk
mzki has quit [Quit: leaving]
cptG_ has joined #linux-sunxi
cptG has quit [Ping timeout: 256 seconds]
massi has quit [Quit: Leaving]
Christos_ has quit [Quit: Page closed]
<mihlit>
I've extracted sys_config.fex from android image for Anichips PhoenixA20. I plan to create device tree based on that, but some dram values seem wrong. dram_size, dram_io_width, dram_bus_width, dram_rank_num and dram_chip_density are set to 0xffffffff. Is it some special magic value, is it wrong or are those unused variables with nothing cares what's their value?
iamfrankenstein has quit [Quit: iamfrankenstein]
afaerber has quit [Quit: Ex-Chat]
f0xx has joined #linux-sunxi
libv has joined #linux-sunxi
[Awaxx] has joined #linux-sunxi
libv_ has quit [Ping timeout: 248 seconds]
[Awaxx] has quit [Client Quit]
[Awaxx] has joined #linux-sunxi
[Awaxx] has quit [Client Quit]
[Awaxx] has joined #linux-sunxi
<miasma>
tkaiser: btw i found a thread about usb otg & ethernet gadgets in armbian forums. did you try the different ethernet modes or just rndis?
<tkaiser>
miasma: Only rndis. You get ~120 Mbits/sec with H3 BSP kernel
<miasma>
tkaiser: i was planning to write a wiki page about usb otg networking. it seems the armbian forums are full of gems, but it takes some time to find them all
libv_ has joined #linux-sunxi
<miasma>
tkaiser: there probably isn't any kind of index about the more useful threads :P ?
<tkaiser>
miasma: I always use google search, IgorPec implemented 'custom search' when you click eg on H2/H3 forum.
[Awaxx] has quit [Quit: bleh]
libv has quit [Ping timeout: 256 seconds]
<miasma>
tkaiser: many of the threads seem useful to sunxi users in general. i just thought that some kind of link list with short descriptions might be helpful
<tkaiser>
miasma: Please go ahead. I don't expect the forum to disappear anytime soon and it allows 'permalinks' even for threads that are renamed or moved.
deskwizard has joined #linux-sunxi
florianH has quit [Quit: Connection closed for inactivity]
libv has joined #linux-sunxi
libv_ has quit [Ping timeout: 260 seconds]
iamfrankenstein has joined #linux-sunxi
<BurtyB>
heh searching on the arbian forum (and others) is a pain when you want to look for fel or otg as they're too short without turning to google
<BurtyB>
armbian even
|Jeroen| has joined #linux-sunxi
Nacho has quit [Remote host closed the connection]
<tkaiser>
(still searching for tx-delay and rx-delay since 2/0 seems to be best value with production Pine64+)
libv has quit [Ping timeout: 260 seconds]
<tkaiser>
On the upper OTG port nothing happens but on the lower USB port I get 41/42 MB/s sequential read/write and dmesg reports proudly 'scsi host0: uas' every time I attach the disk :)
<KotCzarny>
tkaiser, you should measure iops too, i guess
IgorPec has quit [Ping timeout: 265 seconds]
<KotCzarny>
it would show how much cpu intensive it is
iamfrankenstein1 has joined #linux-sunxi
<KotCzarny>
(and how effective is the protocol/chip)
<tkaiser>
And when I attach one of the 'bad' ASMedia chips I get what I deserve: 'usb 1-1: UAS is blacklisted for this device, using usb-storage instead'
iamfrankenstein has quit [Ping timeout: 248 seconds]
iamfrankenstein1 is now known as iamfrankenstein
<tkaiser>
Bus 001 Device 004: ID 174c:55aa ASMedia Technology Inc. ASM1051E SATA 6Gb/s bridge, ASM1053E SATA 6Gb/s bridge, ASM1153 SATA 3Gb/s bridge
libv has joined #linux-sunxi
<tkaiser>
KotCzarny: Testing IOPS is hard since it totally depends on the disk in question. You need a test setup that remains the same for every device you test.
|Jeroen| has quit [Quit: dada]
<tkaiser>
KotCzarny: I might do some test the next days though simply comparing a few USB-to-SATA bridges with one of the test SSDs that are lying around.
<tkaiser>
KotCzarny: Next problem: SSD performance can also be altered over time. And so on.
<KotCzarny>
not if you trim before test (or just never trim)
<tkaiser>
KotCzarny: BS. We monitor a bunch of SSDs and you simply never know.
<tkaiser>
BTW: [ 1539.078737] BTRFS: error (device sda) in btrfs_commit_transaction:2227: errno=-5 IO failure (Error while writing out transaction)
<KotCzarny>
well, there is cell degradation too
yann|work has joined #linux-sunxi
<tkaiser>
So I wouldn't call this first test that successful. OTOH Pine64+ is not that great when looking at manufacturing quality and is a total fail regarding powering (I was lazy and powered through the shitty Micro USB port)
Mr__Anderson has quit [Remote host closed the connection]
<apritzel>
tkaiser: just add them to the emac node
<apritzel>
tkaiser: I guess nobody cared so far about these settings
<tkaiser>
apritzel: How should I ever do the ultimate Pine64+ benchmark then?!§!!1!!11!
<apritzel>
I couldn't tell a difference while typing via ssh ;-)
<KotCzarny>
tkaiser: you need mali for that
<KotCzarny>
;)
<beeble>
tkaiser: sysbench –test=cpu –cpu-max-prime=20000 –num-threads=4 run :)
<tkaiser>
apritzel: Yeah, I know the problem ;) Anyway: quite surprised, with both UASP capable enclosures I exceeded 41/42 MB/s (write/read) using those 2 small tweaks learned from you.
<tkaiser>
beeble: Am running with 864 MHz and Xenial so I can lookup the result with Spotlight already ;)
<tkaiser>
beeble: execution time (avg/stddev): 10.6418/0.01
freemangordon has quit [Quit: Leaving.]
<tkaiser>
So with UASP enabled A64 is the fastest sunxi SoC so far (even if just clocked with 864 MHz). Looks promising regarding H5 :)
<beeble>
actually, i really only care for spec results for cpu benchmarks
<muvlon>
wait what
<muvlon>
how does it compare to the A80?
<KotCzarny>
H5 is very likely to be easily mainlined, a80 not so
<tkaiser>
apritzel: which cpufreq do you set? Since I fear the 864 MHz I use now (1.1V?) might be too much.
<apritzel>
tkaiser: 0x80001010, of course
freemangordon has joined #linux-sunxi
<tkaiser>
Yep, just deciphered
<apritzel>
tkaiser: which (as everyone knows) is 816 MHz ;-)
cptG has joined #linux-sunxi
<tkaiser>
apritzel: You write to much comments, it's written in dec above ;)
<lennyraposo>
good to be back from hiatus
<lennyraposo>
had casts removed from a fall I took back in August
<lennyraposo>
broken wrist on th eright
<lennyraposo>
ring middle and pointer on the left broken
<lennyraposo>
:S
<apritzel>
lennyraposo: eek
<KotCzarny>
o.O
<lennyraposo>
got rehab next week
<lennyraposo>
for left hand
<apritzel>
lennyraposo: so you could only type with the pinky?
<lennyraposo>
on my left yes
<lennyraposo>
haven't touched a keyboard really up until last week
deskwizard has quit [Ping timeout: 248 seconds]
<lennyraposo>
the worst thing was I had ordered an elitebook
<lennyraposo>
and not until monday Ihave a I got a chance to work on it
Ntemis has quit [Remote host closed the connection]
<calimero_82>
hi guys
<calimero_82>
does someone use orangepi one?
<lennyraposo>
I hear that mainline is coming along nicely with the pine
<lennyraposo>
kudos to everyone
<KotCzarny>
on linux-sunxi, yes
<KotCzarny>
on pine64.org NO
<KotCzarny>
;)
<lennyraposo>
lol
<lennyraposo>
sounds about right
<lennyraposo>
not much has changed with bsp stuff
<lennyraposo>
either
<lennyraposo>
I will try to connect with Jet from AllWinner
<muvlon>
KotCzarny: what about kernel.org?
<KotCzarny>
kernel.org is the result of submitted work of various people. but mostly linux-sunxi
<lennyraposo>
if it's AW linux-sunxi
freemangordon has quit [Ping timeout: 260 seconds]
<lennyraposo>
is where to get the insight
<lennyraposo>
last I heard Linus isn't much of a fan of most sbc's
<lennyraposo>
mainly because of the binary blobs and licensing
<muvlon>
wouldn't he be a fan of projects trying to change this?
freemangordon has joined #linux-sunxi
<lennyraposo>
surely
<lennyraposo>
jus tnot a fan of the companies that push out crap
<lennyraposo>
AW is chalk full of it
<apritzel>
so for the records: the Pine64 DTs are in linux-next since about a week or so
<lennyraposo>
tkaiser can fill you in on that portion
<muvlon>
oh, I'm well aware of the kind of shit allwinner pulls
mzki has joined #linux-sunxi
<apritzel>
tkaiser: so mainline boots fine on the H5 and USB works already, just tested
<apritzel>
tkaiser: don't ask me about benchmarks, though
<tkaiser>
apritzel: Good to hear.
<tkaiser>
lennyraposo: What?
<muvlon>
okay so I'm a little new to this whole sunxi thing
<muvlon>
I'm not sure I understand the mali situation
<muvlon>
is it correct that there are only blob drivers for android, and you can use them via libhybris?
<tkaiser>
muvlon: Are you the kind of retro gamer keen on Quake 3 fps?
<muvlon>
not really
<apritzel>
muvlon: admit it: you are a glxgears gamer, right?
<tkaiser>
muvlon: Great, so simply don't care about 'the mali situation' that much
<muvlon>
can you get some kind of hdmi output without any blobs?
<muvlon>
that's really all I care about
<tkaiser>
muvlon: Are we talking about some special 3D displays?
<muvlon>
no, simple 2D desktop
<apritzel>
muvlon: mali and gfx output are totally separate things
<muvlon>
ah, ok
<muvlon>
so if you don't need acceleration you can ignore the mali?
<apritzel>
muvlon: you need the DisplayEngine (DE) and the HDMI PHY to get a picture
<apritzel>
muvlon: AFAIK there is some basic 2D accel in the drivers (fbturbo)
<apritzel>
muvlon: though I believe the DE doesn't help as much as on earlier SoCs
<apritzel>
and also video decoding is separate
<apritzel>
muvlon: so mali is only needed for 3D accell
<muvlon>
ah right
<tkaiser>
muvlon: GPU as known in x86 land == 2D, 3D and video acceleration. GPU in ARM land == 3D acceleration only
<muvlon>
yes, I think that's why I was confused
<apritzel>
though some 2D apps use OpenGL for acceleration these days
<muvlon>
in x86 land, the GPU is usually *also* responsible for output
<KotCzarny>
well, in early x86 land there was cpu, 2d gpu and separate 3d gpu card
<apritzel>
tkaiser: I think technically it's not that far from ARM land, just nobody looks at those components separately
<apritzel>
tkaiser: is there any real shutdown on those AXP-less boards?
<lennyraposo>
I was saying that you are knowledgeable about the ad these manufacturers ut out there n terms of linux tkaiser
<lennyraposo>
ad=bad
<tkaiser>
lennyraposo: Nope
<lennyraposo>
I stand corrected
<lennyraposo>
but I must say you are pretty knowledgeable mate ;)
<tkaiser>
apritzel: I don't really know honestly. Rather clueless here. I thought some time it's just drivers that crash when powering down that prevent the board really powering off. But lost track there
<apritzel>
ERROR: PSCI system shutdown: still alive ...
<tkaiser>
lennyraposo: Maybe I just talk to much in the wrong places ;=
<lennyraposo>
lol
<lennyraposo>
I fall victim to that too mate ;)
<apritzel>
mmh, there is an enable bit in the SY8106A regulator VOUT_COM register
<tkaiser>
apritzel: You ask the wrong one when you're asking me :) All that stuff I discovered by accident. Do some stuff I don't do and start to wonder.
<apritzel>
tkaiser: ah, the H5 is too easy, no real challenges ;-)
<apritzel>
tkaiser: what about the R40? Do you have one already?
<tkaiser>
apritzel: Nope, only board so far made by the wrong vendor
<apritzel>
I just figured, because the Wiki falls silent on R40 ;-)
yann-kaelig has quit [Quit: Leaving]
yann-kaelig has joined #linux-sunxi
<tkaiser>
apritzel: Want me to put a "don't buy (yet)" page there? ;)
<tkaiser>
KotCzarny: OMG, now I know where I remember A20 from. PC crap from last century. A20-Gate and stuff like that ;)
libv has quit [Ping timeout: 260 seconds]
<KotCzarny>
hehe
<apritzel>
KotCzarny: atm the SPL only briefly enables it very early, checks the level and un-configures it again
<KotCzarny>
apritzel, so the only way is to write openrisc app i guess. hum.
<KotCzarny>
can uboot run openrisc app?
<apritzel>
KotCzarny: you could have something in Linux
<apritzel>
KotCzarny: could even be a userland app
<tkaiser>
apritzel: Do you know whether the OTG port on H5 also can be turned into a 'true' USB host port?
<apritzel>
tkaiser: it's really an H3 in this respect, so no, I guess
<apritzel>
tkaiser: you can use the normal USB OTG as-a-master config, though
<muvlon>
what's the difference between OTG and "true host"?
<apritzel>
but lemme check the manual ...
<apritzel>
muvlon: apparently the performance of the OTG controller in host mode isn't great
<tkaiser>
apritzel: Ok, performance on H3 in this mode doesn't differ that much from true host ports (compared to older Allwinner SoCs)
<apritzel>
also unstable
<calimero_82>
guys have u problems with the temperature?
<Pepe>
put there heatsink and you can put there even fan
<KotCzarny>
you can only have that many manipulation techniques ;)
<tkaiser>
muvlon: depends on the controller and stuff. With Pine64 the upper USB port is an OTG port but there are magic bits that can turn that into a real host port with an own PHY.
<tkaiser>
muvlon: So same performance when used as 'real' host port. You can also use OTG ports in host mode but then it depends on the implementation.
<tkaiser>
muvlon: Older Allwinner SoCs weren't that great here, at least with H3 it's ok-ish. And H5 seems to be like H3 here.
<apritzel>
tkaiser: mmh, the diagrams look the same between H3 and H5, but still are a bit confusing
<muvlon>
hmm, it seems I was confused again
<tkaiser>
apritzel: Anyway, as long as H5 devices are around that expose all 3 USB host ports I don't care that much about using OTG as host anyway :)
<muvlon>
I thought OTG-USB meant having a Type-B port that optionally acts as a host
<apritzel>
muvlon: true, but apparently you can still mess up the implementation of that part ;-)
<apritzel>
so it's not as good as a dedicated host controller
<muvlon>
wait, so the Pine64 has that on a Type-A port??
<apritzel>
so AW put really two different controllers on that one top port, but you can use only one at a time, of course
<apritzel>
muvlon: yup
<muvlon>
weirdness
<apritzel>
muvlon: and you are not the first one to freak out on that
<muvlon>
I'm not very used to this brave new arm world :)
<Wizzup>
INET-D70B-REV01 is what's on the PCB
<apritzel>
it's just to help those eBay sellers which offer male-A <-> male A cables
<tkaiser>
muvlon: Did you know that the USB implementors forum wants to rule the world? By releasing specs that make everyone mad who reads it? Did you know that Micro USB A/AB/B exist? ;)
<muvlon>
apritzel: so can you plug an male-A <-> male-A cable into that OTG port and have the Pine64 act in device mode?
<apritzel>
muvlon: that's what
<apritzel>
I use everyday for FEL mode
<muvlon>
I did know that micro A/AB/B exist
<muvlon>
which is why it's so weird to me that there are now devices which have a "Micro-B OTG-only" port
<muvlon>
why didn't they just put a micro-A port?
<muvlon>
if it only acts as host anyway
<apritzel>
muvlon: because the A64 has only one real host controller
<apritzel>
so you would end up with only one proper USB port
<tkaiser>
muvlon: Go grab a 'Lemaker Guitar', that's the only USB3 device that uses also a different pinout on the Micro USB connector. So you can't do anything with USB3 here ;)
<freemangordon>
hmm, TS FW load fails with error -6
<apritzel>
freemangordon: that's an improvement, wasn't it -2 before? ;-)
<freemangordon>
chip id is 0xB4820000
<freemangordon>
apritzel: yeah :)
<apritzel>
muvlon: and people put "number of USB ports" in those comparison tables, you know ;-)
<apritzel>
freemangordon: or not, that's now: ENXIO vs. ENOENT ;-)
<freemangordon>
the problem was that rootfs was not mounted by the ime driver was requesting FW
<freemangordon>
as it was built-in
<freemangordon>
now with driver as a module, it finds the FW wile, but fails with -6 :(
<freemangordon>
FW file that is
muvlon has quit [Quit: Leaving]
<freemangordon>
when I extract FW from android module, do I need to do some kind of post-process it?
<freemangordon>
MoeIcenowy: ^^^
<tkaiser>
freemangordon: Right, save it somewhere.
<freemangordon>
tkaiser: well, it is saved in /lib/firmware/silead/
mihlit has quit [Ping timeout: 260 seconds]
yann-kaelig has quit [Quit: Leaving]
<freemangordon>
driver finds it, but i2c_smbus_write_i2c_block_data fails with -6
<apritzel>
freemangordon: ENXIO hints at "no match"
<tkaiser>
freemangordon: MoeIcenowy asked for specific firmware bits to be able to extract them before you trash Android on your device :)
<freemangordon>
tkaiser: android is not touched
<freemangordon>
I am booting devuan through fel
<KotCzarny>
fmg: try adding all those firmware files into kernel?
<freemangordon>
apritzel: I tested all the files that were extracted by fw-extractor, with the same result
<freemangordon>
KotCzarny: add FW files into kernel? what do you mean?
<KotCzarny>
embed in kernel vs fs
<freemangordon>
KotCzarny: should make no difference for modprobe, ain't?
<KotCzarny>
who knows, maybe fs read fails for some reason?
<freemangordon>
ls -al
<freemangordon>
oops
gumblex has joined #linux-sunxi
gumblex has quit [Client Quit]
gumblex has joined #linux-sunxi
<tuxillo>
hi
<tuxillo>
NiteHawk: I'm still far from home but I'll try it when I get back
<tuxillo>
probably on Friday. do you need it earlier?
jstein has quit [Remote host closed the connection]
<calimero_82>
with heatsink is normal to have 70 c of temperature?
<KotCzarny>
calimero_82: what conditions? soc? kernel/config?
<tkaiser>
apritzel: Please elaborate what's wrong with
<apritzel>
you change the output of /proc/cpuinfo?
<apritzel>
I mean the original patch, not the patch to the patch
<apritzel>
tkaiser: let me guess: some bloody looserland tool relies on this ...
<tkaiser>
apritzel: Don't know why but yes, I think Martin explained the reason 400 years ago in pine64 forum (backwards compatiblity)
<apritzel>
tkaiser: I am not in charge, but I guess this has zero change of getting upstream ...
<tkaiser>
apritzel: I know /proc/cpuinfo only from BogoMIPS complaints
<apritzel>
that's a good example why this is broken
<apritzel>
basically you shouldn't use cpuinfo for that
<tkaiser>
apritzel: I think it's a temporary measure. But Martin is one of the brave guys who always tries to push the envelope and tries to do all the fancy stuff with mainline kernel :)
<apritzel>
technically this isn't mainline, then ;-)
<apritzel>
if there are tools that look for arm(32) entries in /proc/cpuinfo, those should be fixed
<apritzel>
I think you can do this easier than hacking the kernel
<NiteHawk>
tuxillo: no, that's not in a hurry - whenever you find the time
<tkaiser>
apritzel: thanks, I dropped a note.
<apritzel>
tkaiser: sorry if that sounds harsh, but that is nothing compared to the replies you'd get on the list for that ;-)
<apritzel>
tkaiser: and please be aware that the stuff MoeIcenowy's branch is different from what's in next
<apritzel>
tkaiser: the DTs have changed slightly last minute
<apritzel>
it's quite a mess, but a bit expected in that state
<tkaiser>
apritzel: I've no problems with harsh comments at all :) I don't even remember what that change was made for. And the whole idea to provide 'mainline OS images' is maybe just braindead anyway. Since we don't get any useful feedback from end users so far.