jstein has quit [Remote host closed the connection]
HeavyMetal has quit [Ping timeout: 264 seconds]
HeavyMetal has joined #linux-sunxi
HeavyMetal has quit [Changing host]
HeavyMetal has joined #linux-sunxi
NoNicknameLeft has joined #linux-sunxi
<NoNicknameLeft>
What exactly makes allwinner cpus more free than others? Is it the availability of boot0 and allwinner bsps? How does the rockchip rk3288 for example compare to some allwinner cpu?
lurchi_ is now known as lurchi__
lurchi__ is now known as lurchi_
<icenowy[m]>
In fact rk3288 is more free :-)
<NoNicknameLeft>
Could you please elaborate?
<willmore>
NoNicknameLeft, they're cheaper? So, that's closer to 'free'? ;)
<NoNicknameLeft>
Well then, are they more libre in any meaningful way?
<NoNicknameLeft>
I am sorry if this question is overly simplistic, I really don't know.
<willmore>
There are almost no meaningful docs. The BSP code serves as the most authoritative reference in that sense. They use Mali blob based drivers. They provide no info on how the video acceleration works....
<willmore>
So, I don't see that as being very 'libre'. On the other hand, a lot of people have spent a lot of effort to reverse engineer the important bits and free things up. So, the chips are pretty libre now, but that's not because Allwinner did anything to help it.
Gerwin_J has joined #linux-sunxi
Net147 has quit [Ping timeout: 260 seconds]
<wens>
smaeul: could it be something in ATF? IIRC on ARMv7 secure world is supposed to reset the arch timer counter to a correct value before returning control to non-secure
Net147 has joined #linux-sunxi
<NoNicknameLeft>
<willmore> Ok, how come they have so such massive support from users then? Was that a purely random decision?
<NoNicknameLeft>
willmore: Sorry for mis-reference
<willmore>
NoNicknameLeft, Just how it happened. Someone with a chip got interested. Things like communities tend to nucleate around progress. That often leads to more progress.
<NoNicknameLeft>
Huh, I never knew. I thought there was some singular thing one could point to that differentiates allwinner cpus from others. Many thanks
<willmore>
Most progress in free software is because someone wanted to do something. Progress is a byproduct. Then there's the commercial people--their motives are more predictable.
cnxsoft has joined #linux-sunxi
<wens>
also cheap devboards (and devices in general) with allwinner socs came out first
<wens>
then rockchip, then amlogic
cnxsoft has quit [Read error: Connection reset by peer]
<wens>
but rockchip does mainline officially now, and amlogic has baylibre working on it
<wens>
allwinner has... well... us
cnxsoft has joined #linux-sunxi
<willmore>
wens, did allwinner being common in cheap tablets play a role?
Gerwin_J has quit [Quit: Gerwin_J]
<willmore>
exynos has some samsung people working on it.
<smaeul>
wens: no, if it was ATF-related, it would be much less often and much more random
<smaeul>
there's a very obvious pattern to it. 99% of cases are the timer going 0x0000ffff → 0x0001ffff → 0x00010000 (jump forward) or 0x0001ffff → 0x00010000 → 0x00020000 (jumping backward) for various bit positions
<smaeul>
the other 1 percent are larger jumps, where usually not all of of the lesser-significant bits are seen as set, but it's the same principle
<wens>
willmore: seems likely
<wens>
I think it was cheap tablets and tv boxes to begin with, then came the boards
<wens>
but that was before my time
<willmore>
Same here. icenowy[m] seems more into the tablet side of things, maybe she can remember back further. I'm very new here.
lurchi_ is now known as lurchi__
Gerwin_J has joined #linux-sunxi
NoNicknameLeft has quit [Quit: Leaving]
NoNicknameLeft has joined #linux-sunxi
NoNicknameLeft has quit [Client Quit]
popolon has quit [Quit: WeeChat 1.9.1]
kaspter has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
megi has quit [Quit: megi]
megi has joined #linux-sunxi
<DonkeyHotei>
willmore, wens: prior to the A10, allwinner did not release linux BSP's. but at the time of the A10, an engineer worked for them who was a linux enthusiast, and he worked behind the scenes to free things up until he left there. the kernel version at that time was 3.3 or 3.4 iirc, and their releases since have been based on that. there was plenty they did not allow him to free up at that time and some of it has been freed up
<DonkeyHotei>
since
megi has quit [Client Quit]
gnarface has quit [Quit: Leaving]
lrusak has quit [Quit: lrusak]
gnarface has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
JohnDoe_71Rus has quit [Client Quit]
<buZz>
DonkeyHotei: is that guy from allwinner who did it, the guy who started cubietech?
<DonkeyHotei>
the name suggests yes (Tom Cubie)
<buZz>
ah yes indeed
ninolein has quit [Ping timeout: 252 seconds]
<buZz>
who made the cubieboard
ninolein_ has joined #linux-sunxi
<buZz>
funny who similar that origin story is to raspberry pi :D
<DonkeyHotei>
the difference is that broadcom requires secret boot code in the GPU and requires an NDA just to reveal what language the code is in, while the boot code for allwinner is burned into the die
<DonkeyHotei>
the latter was important to people seeking FSF approval, which snowballed the reverse engineering efforts
<buZz>
yeah and allwinner sells their cpus to whoever
<buZz>
whereas broadcom needs special hidden agreements nonsense
lrusak has joined #linux-sunxi
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-sunxi
kaspter1 has joined #linux-sunxi
kaspter has quit [Ping timeout: 240 seconds]
kaspter1 is now known as kaspter
kaspter has quit [Client Quit]
[7] has quit [Ping timeout: 240 seconds]
kaspter has joined #linux-sunxi
TheSeven has joined #linux-sunxi
hardfalcon1 has joined #linux-sunxi
hardfalcon has quit [Ping timeout: 248 seconds]
JohnDoe_71Rus has joined #linux-sunxi
dave0x6d has quit [Quit: Connection closed for inactivity]
TheSeven has quit [Ping timeout: 258 seconds]
TheSeven has joined #linux-sunxi
dave0x6d has joined #linux-sunxi
rexxster has quit [Remote host closed the connection]
jbrown has quit [Ping timeout: 248 seconds]
<KotCzarny>
hmm, opi+2e is back in stock, at higher price which might mean they use same, speedy emmc modules
lurchi_ has joined #linux-sunxi
lurchi__ has quit [Ping timeout: 248 seconds]
reinforce has joined #linux-sunxi
awais has joined #linux-sunxi
tllim has joined #linux-sunxi
<tllim>
@Sunxi folks, I have meet up with Allwinner BU3 vice General manager last week and he has put H6 open source support as a key factor.
<tllim>
The H6 datasheet and user manual will be release as no confidential watermark.
leviathanch has joined #linux-sunxi
<tllim>
I have already consult with Icenowy as Sunxi contact windows with BU3.
<tllim>
BU3 already assigned one engineer name wink as the contact window.
<tllim>
We will start from H6 SoC due to new release.
<DonkeyHotei>
too bad there was no mali concession
<DonkeyHotei>
but that's arm, not allwinner
<tllim>
On H6, Allwinner will release a driver fro MALI T720
<DonkeyHotei>
binary, i'm sure
<tllim>
Allwinner has setup a wechat discussion group and I have already invite icenowy.
<tllim>
hopefully, this is a good start :-)
yann has quit [Ping timeout: 240 seconds]
<awais>
hoping the binary mali T720 blob be compiled with opencl support
leviathanch has quit [Remote host closed the connection]
leviathanch has joined #linux-sunxi
<tllim>
@awais, should be opencl support :)
xes_ has joined #linux-sunxi
<tllim>
@DonkeyHotei, it is binary. The source only can be released by ARM.
xes has quit [Ping timeout: 260 seconds]
tllim has quit [Quit: Leaving]
gnarface has quit [Quit: Leaving]
lemonzest has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
cnxsoft has quit [Quit: cnxsoft]
cnxsoft has joined #linux-sunxi
gnarface has joined #linux-sunxi
awais has quit [Ping timeout: 240 seconds]
dave0x6d has quit [Quit: Connection closed for inactivity]
awais has joined #linux-sunxi
efraim has quit [Remote host closed the connection]
efraim has joined #linux-sunxi
fl__0 is now known as fl_0
anarsoul has joined #linux-sunxi
yann has joined #linux-sunxi
matthias_bgg has joined #linux-sunxi
jack_ has quit [Quit: jack_]
msimpson has joined #linux-sunxi
tom_nov has joined #linux-sunxi
xes_ is now known as xes
chlorine_ has joined #linux-sunxi
tom_nov has quit [Ping timeout: 248 seconds]
kilobyte has quit [Ping timeout: 264 seconds]
jstein_ has joined #linux-sunxi
jstein_ is now known as jstein
kilobyte has joined #linux-sunxi
chlorine_ has quit [Ping timeout: 248 seconds]
benettig has joined #linux-sunxi
chlorine has joined #linux-sunxi
libv_ is now known as libv
leviathanch has quit [Remote host closed the connection]
kaspter has quit [Remote host closed the connection]
chlorine has quit [Ping timeout: 248 seconds]
kaspter has joined #linux-sunxi
<benettig>
Hello, I don't know if it's offtopic, but did someone manage r6p2 x11 to work? I have an X log here https://pastebin.com/sEuQv0yK
montjoie has quit [Ping timeout: 255 seconds]
montjoie has joined #linux-sunxi
tom_nov has joined #linux-sunxi
<benettig>
^ got answer from mripard by e-mail.
chlorine has joined #linux-sunxi
chlorine has quit [Remote host closed the connection]
hardfalcon1 has quit [Ping timeout: 260 seconds]
<willmore>
DonkeyHotei, I didn't know that story. It's good to know. Cudos to Mr. Cubie!
<willmore>
tllim, that's wonderful news!
fl_0 has quit [Ping timeout: 250 seconds]
tuxillo has quit [Quit: leaving]
fl_0 has joined #linux-sunxi
popolon has joined #linux-sunxi
tuxillo has joined #linux-sunxi
hardfalcon has joined #linux-sunxi
hardfalcon1 has joined #linux-sunxi
tuxillo has quit [Remote host closed the connection]
kaspter has quit [Read error: Connection reset by peer]
kaspter has joined #linux-sunxi
_hp197 is now known as hp197
hp197 has quit [Changing host]
hp197 has joined #linux-sunxi
hardfalcon has quit [Ping timeout: 248 seconds]
matthias_bgg_ has joined #linux-sunxi
matthias_bgg has quit [Ping timeout: 268 seconds]
<lvrp16>
yum h6 with opencl
<lvrp16>
time to lay a board
afaerber has quit [Quit: Leaving]
benettig_ has joined #linux-sunxi
msimpson_ has joined #linux-sunxi
kloczek has quit [Remote host closed the connection]
kloczek has joined #linux-sunxi
msimpson has quit [Read error: Connection reset by peer]
benettig has quit [Ping timeout: 248 seconds]
afaerber has joined #linux-sunxi
lkcl has joined #linux-sunxi
The_Loko has joined #linux-sunxi
megi has joined #linux-sunxi
kaspter1 has joined #linux-sunxi
fdcx has quit [Read error: Connection reset by peer]
kaspter has quit [Ping timeout: 248 seconds]
kaspter1 is now known as kaspter
<willmore>
lvrp16, agreed. That's good news. Save the money on the BSP, vendors, provide good docs and a mali blob. :)
<lvrp16>
lets hope, bsp is still necessary especially for android
<willmore>
Do any of the Allwinner chips have multiple simultaniously useable I2S busses?
<lvrp16>
but amlogic and rockchip have been really cool
* willmore
has no love for android on these boards.
<willmore>
Yeah, Samsung as well, but not nearly as agressive as they could be.
<lvrp16>
samsung has been mainlining their stuff??
<lvrp16>
889x?
chewitt has quit [Quit: Zzz..]
awais has quit [Ping timeout: 240 seconds]
<willmore>
lvrp16, #linux-exynos. Mostly older stuff, but yeah, they're working on it.
matthias_bgg has joined #linux-sunxi
awais has joined #linux-sunxi
matthias_bgg_ has quit [Ping timeout: 248 seconds]
matthias_bgg has quit [Read error: Connection reset by peer]
matthias_bgg has joined #linux-sunxi
lkcl has quit [Ping timeout: 248 seconds]
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<wens>
willmore: multiple simultaniously useable?
<willmore>
wens, yes. I need to interface to a lot of audio inputs.
<wens>
I think the maximum is two on the outside
<willmore>
That's better than One. :)
<wens>
though a few chips support TDM, which would give a few more slots
<wens>
not supported in mainline though
<willmore>
That might be the way to go.
<willmore>
I remember looking at TDM busses a decade or so ago and some of them could handle 16 or more CODECS.
<wens>
I think Allwinner ones have 8 slots
<willmore>
Hmm, that's not going to be enough. I guess I'll look into something else. Thanks, wens.
<wens>
H3/H5 has 8 slots for TDM, plus another I2S controller
<willmore>
I'm looking for multiples of 16. :)
<wens>
you doing a mixing board?
dave0x6d has joined #linux-sunxi
<willmore>
No, a microphone array for location sensing.
<willmore>
Maybe MEMS microphones are the way to go. They already have a digital output.
matthias_bgg has quit [Ping timeout: 240 seconds]
<wens>
doesn't DMIC need some hardware support as well?
<ullbeking>
miasma: yes, this is one of my ultimate goals
<miasma>
ullbeking: also beware, the chinese dc/dc converters might supply >5V when the soc shuts down
<miasma>
and chinese 5A = 2A :)
<ullbeking>
miasma: and going direct to gpio pins is how i intend to power my opi zeros
<miasma>
works for pc/one and some other boards too
<ullbeking>
and also opi r1 (i don't know what the hell it is, it was new when i placed the order and i thought,eh, why not ty it out)
<lurchi__>
for anything but the RPi powering is not to critical
<ullbeking>
22:53 <miasma> ullbeking: also beware, the chinese dc/dc converters might supply >5V when the soc shuts down
<ullbeking>
22:54 <miasma> and chinese 5A = 2A :)
<ullbeking>
can you expand on the issue here?
<lurchi__>
RPi uses linear regulators, while most (all?) allwinner boards use switching converters
<ullbeking>
lurchi__: i know that, but having a proper power supply is one factor/variable i don't want to have to worry about
<ullbeking>
22:56 <lurchi__> RPi uses linear regulators, while most (all?) allwinner boards use switching converters
<ullbeking>
what's the difference in practice?
<miasma>
ullbeking: there are posts on some arduino/raspberry forums about those. some converters don't regulate well with no load. the cheap converters don't have heatsink so they get too hot and break
<ullbeking>
miasma: does this have anything to do when powering down the dc/dc converters, e.g., voltage goes up, current goes down, the thing blows up?
<ullbeking>
powering down the soc, i mean
<lurchi__>
ullbeking: the RPi uses cascaded linear regulators, and each has some voltage drop, ~0.5..0.7V each at load
* ullbeking
will be afk for about 20 minutes
<miasma>
ullbeking: you can analyze the voltage with a meter/scope
<ullbeking>
miasma: yup, i just got a meter
<lurchi__>
typically, a 10mA LED is sufficient load
<ullbeking>
miasma: is the xl4015 literally already embedded on the opi pc, or is it part of your dc-dc conversion chain for powering the device from your external source? where are you connecting the power to?
<ullbeking>
22:52 <miasma> ullbeking: if you're doing soldering, you could build a 'hat' and connect directly to gpio pins
<ullbeking>
i've been told about built "hats" before but i don't know what it would look like in this situation. can you point me to an example?
<ullbeking>
sorry for the barrage of questions. i'm really lost!!
<ullbeking>
i still have to find a way to getting power from some not-terrible power supple into a "4.0mm-1.7mm jack"
<lurchi__>
ullbeking: hat in this case just means conneting to the GPIO pin header - form factor is up to you
<lurchi__>
you can also cut one of the USB-barrel-jack cables
<ullbeking>
from this discussion it seems that whatever supply i use i'm going to need to factor in some dc-dc conversion device
<ullbeking>
i already have a few, which i bought on recommendation, but i don't know how to use them
<lurchi__>
only if you main supply is more than 5V
<lurchi__>
plug into phone charger, plug into OPi, done
<ullbeking>
23:56 <lurchi__> you can also cut one of the USB-barrel-jack cables
<ullbeking>
this is looking like the best option so far. either that, or find some stupid converter kit on ali.
<ullbeking>
holy s* yes, a phone charger, maybe i have some around that have the right spec