Turl changed the topic of #linux-sunxi to: Allwinner/sunxi /development discussion - did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait! - https://github.com/linux-sunxi/ - Logs at http://irclog.whitequark.org/linux-sunxi
berkutta has joined #linux-sunxi
dh1tw has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
jernej has quit [Ping timeout: 260 seconds]
Pepe has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
lkcl has quit [Ping timeout: 250 seconds]
lkcl has joined #linux-sunxi
jstein has quit [Remote host closed the connection]
florianH has quit [Quit: Connection closed for inactivity]
bugzc has joined #linux-sunxi
honx has quit [Ping timeout: 252 seconds]
marvs has quit [Ping timeout: 260 seconds]
honx has joined #linux-sunxi
marvs has joined #linux-sunxi
nashpa has quit [Ping timeout: 244 seconds]
nashpa has joined #linux-sunxi
BuddyZhang1 has quit [Quit: Page closed]
tlwoerner_ is now known as tlwoerner
Ntemis has quit [Remote host closed the connection]
dlan has joined #linux-sunxi
<victhor> hi, I'm taking a look at this "Aoson M751s" tablet board, from the wiki it says there is no UART found on the board. I wonder if anyone with one of these managed to find one on those pads on the bottom of the board, and the wiki's just not updated... I went through them with a oscilloscope, I found a lot of pins at 3.3V, but no activity, and some pins, which would show some activity when the power button is pushed, but it's a 5V signal, and those
<victhor> pins would not respond to echo to the serial ports, and I could not decode the signal with my USB-serial adapter.
<victhor> I guess I'm out of luck and my only choice for debugging is to plug the LCD in?
cnxsoft has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
terra854 has joined #linux-sunxi
chomwitt has quit [Ping timeout: 256 seconds]
muvlon_ has joined #linux-sunxi
muvlon has quit [Ping timeout: 258 seconds]
ninolein has quit [Ping timeout: 258 seconds]
ninolein has joined #linux-sunxi
kaspter has joined #linux-sunxi
BuddyZhang1 has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
<wens> victhor: it's possible that the pin isn't muxed to uart, so you don't get any traffic
yann-kaelig has quit [Quit: Leaving]
<victhor> hmm, right
<victhor> since u-boot and kernel work, I could try changing parameters on the device tree... it's quite annoying to not be able to interact with the device, short of using a keyboard, but I don't have any spare :/
<wens> victhor: u-boot doesn't look at the devicetree
<wens> iirc the default is to not use the microSD pins for UART
<victhor> right
Andy-D has joined #linux-sunxi
victhor has quit [Read error: Connection reset by peer]
marvs has quit [Ping timeout: 260 seconds]
marvs has joined #linux-sunxi
leviathanch has joined #linux-sunxi
leviathanch has quit [Read error: Connection reset by peer]
leviathanch has joined #linux-sunxi
orly_owl has joined #linux-sunxi
_whitelogger has joined #linux-sunxi
kaspter has quit [Ping timeout: 250 seconds]
foxx_ has joined #linux-sunxi
pg12 has quit [Ping timeout: 260 seconds]
pg12 has joined #linux-sunxi
IgorPec has joined #linux-sunxi
Putti has quit [Quit: Leaving]
JohnDoe_71Rus has joined #linux-sunxi
_whitelogger has joined #linux-sunxi
TheSeven has quit [Ping timeout: 260 seconds]
[7] has joined #linux-sunxi
Putti has joined #linux-sunxi
Putti has quit [Ping timeout: 245 seconds]
jernej has joined #linux-sunxi
reinforce has joined #linux-sunxi
f0xx has joined #linux-sunxi
foxx_ has quit [Remote host closed the connection]
IgorPec has quit [Ping timeout: 240 seconds]
scream has joined #linux-sunxi
<MoeIcenowy> jernej: is your HDMI register addresses obfuscated now?
<wens> got my bpi m2 ultra
sunxi_fan1 has joined #linux-sunxi
<MoeIcenowy> jernej: your new commit works well on Orange Pi One.
<jernej> MoeIcenowy: They were always obfuscated
<jernej> thanks for testing
<jernej> This obfuscation thing is something I wanted to ask in RFC
<MoeIcenowy> jernej: will you care to add https://github.com/Icenowy/u-boot-1/tree/h3_hdmi_simplefb 's last commit to your RFC?
<MoeIcenowy> it's simplefb support for H3
<jernej> should we go with vendor provided code (obfuscation) or with nicer, but unproven, deobfuscated registers
<jernej> also magic was found in blob, not sure if this is ok or not
<jernej> ok, I will add as separate patch
<jernej> but I must to go on work now :)
<jernej> bye
<MoeIcenowy> bye
<miasma> how stable is the h3's ethernet driver for 24/7 use?
<wens> miasma: seems ok
jernej has quit [Ping timeout: 244 seconds]
<montjoie> :)
<MoeIcenowy> have you heard "Lichee Pi"?
<MoeIcenowy> A13-based SBC
<MoeIcenowy> ;-)
DullTube has joined #linux-sunxi
sunxi_fan1 has quit [Ping timeout: 250 seconds]
scream has quit [Remote host closed the connection]
IgorPec has joined #linux-sunxi
<wens> ssvb: SID address changed in R40 :/
<ssvb> wens: SID or SoC ID?
<wens> SID
<ssvb> well, the SID address is already different in A10/A20/A23/A33 and H3/H5/A64, so we are kinda prepared for this
<wens> SoC ID is in the same place
<ssvb> it's good, let's hope that A80 will remain the only oddball
<ssvb> do you know the new SID address for A40?
<ssvb> R40
<MoeIcenowy> ssvb: check the device tree in the SDK
<wens> pull request for sunxi-tools already sent :)
<ssvb> thanks!
<wens> each time i get a new board from foxconn i also get a new power adapter... they're piling up
lennyraposo has quit [Ping timeout: 260 seconds]
<MoeIcenowy> ;-)
<ssvb> wens: while you are at it, could you also try to add proper a80 support to uart0-helloworld?
<wens> ssvb: sure, let me first finish the wiki page
tkaiser has joined #linux-sunxi
Putti has joined #linux-sunxi
<ssvb> wens: no secure SRAM in R40 looks like an extra inconvenience
<ssvb> I guess, the PSCI code needs to live somewhere else?
<wens> ssvb: we can put it in DRAM, like we do for A23 and H3
<wens> but yeah, it sucks
<MoeIcenowy> I think someone should write a wiki page for fake RAM size...
lennyraposo has joined #linux-sunxi
<wens> should we just drop the "sunxi support" section from boards that have newer SoCs?
silviop has joined #linux-sunxi
<MoeIcenowy> wens: I think so
<MoeIcenowy> as we can only get "not supported" there
Mr__Anderson has joined #linux-sunxi
Mr__Anderson has quit [Client Quit]
ErwinH has joined #linux-sunxi
<KotCzarny> MoeIcenowy: lol @ lichee pi
yann has quit [Ping timeout: 250 seconds]
<KotCzarny> ssvb: there is a 'reserve first 64k on x86'
<KotCzarny> maybe push an initiative to do the same on arm? or that option is available with arch=arm too?
<ssvb> KotCzarny: leecher pi?
<ssvb> and what do you mean 'reserve first 64k'?
<KotCzarny> ssvb: so you can have area that nothing else would touch?
<ssvb> it's already done
<KotCzarny> :)
<tkaiser> MoeIcenowy: This one: https://zepan.gitbooks.io/lichee-pi/content/ ?
<KotCzarny> tkaiser: its empty still
<MoeIcenowy> tkaiser: yes
<jelle> hmm what is xyz?
<KotCzarny> answer to what is xyz.
<MoeIcenowy> lichee pi is on sale on taobao for ¥59
<KotCzarny> MoeIcenowy: link?
<jelle> KotCzarny: the answer is 42
<KotCzarny> jelle: that is only answer to universe and everything. not for xyz
<tkaiser> MoeIcenowy: The name is great, you'll never find anything (or at least you'll find a lot related to Orange *Pi* and H3 *lichee*)
<MoeIcenowy> KotCzarny: I'm not sure whether they are on aliexpress
<MoeIcenowy> and they do not have official english name
<jelle> I need a hyperloop between me and aliexpress
<KotCzarny> just curious, you can paste taobao link
<MoeIcenowy> lichee pi is only a translation
<MoeIcenowy> the real name is "荔枝派"
<KotCzarny> what does 'lichee' mean then?
<jelle> it's a fruit right?
<MoeIcenowy> yes
<MoeIcenowy> lichee is some kind of fruit planted in Guangdong
<MoeIcenowy> very tasty ;-)
* jelle waits on the durian pi
<MoeIcenowy> jelle: ;-)
<ssvb> leecher pi, pain64, onrage pi, ...
<tkaiser> ssvb: Well, the 'Lemon Pi' guys realized their mistake and went Roseapple later ;)
<KotCzarny> ehehe
<MoeIcenowy> jelle: you will be hurt by GPIO pins on durian pi ;-)
<jelle> lol
<KotCzarny> cactee pi
<KotCzarny> cacti
<KotCzarny> all pins, no ports ;)
<jelle> MoeIcenowy: but this is a new board, with an a13?
<MoeIcenowy> jelle: yes
cosm has quit [Ping timeout: 245 seconds]
<MoeIcenowy> here's a series of video of its design, if you can listen to Chinese: http://www.bilibili.com/video/av6069742/
cosm has joined #linux-sunxi
<jelle> listening is easy, understanding is harder :)
<wens> ssvb: the uart0 binary is supposed to be univeral right?
<ssvb> wens: yes
<MoeIcenowy> wens: right.
<wens> yeah... going to have to rework a bunch of constant addresses :(
<ssvb> which constant addresses?
<wens> ssvb: the addressing scheme on the a80 is different
<wens> so the ccu, uart, pio base addresses are all different
<ssvb> yeah, that's a pita
<wens> for the others i could add if (soc_is_a80()) {} else {}
<wens> but i need to think abou the SoC ID part
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
<ssvb> wens: you can find the BROM address by checking the V bit in SCTRL
<ssvb> I think you suggested before that we can get the SoC ID from the BROM header
florianH has joined #linux-sunxi
<wens> given that the BROM is only in 1 of 2 locations, and has a distinguishing header, should be easy to check
f0xx has quit [Quit: terminated!]
diego_r has joined #linux-sunxi
massi has joined #linux-sunxi
BuddyZhang1 has quit [Ping timeout: 260 seconds]
f0xx has joined #linux-sunxi
f0xx has quit [Ping timeout: 248 seconds]
<wens> interesting, the bsp still has a fex file
<MoeIcenowy> wens: it's a stub, contains only some flashing parameters
tuxillo has quit [Remote host closed the connection]
<wens> dram controller might be the same as the h3
<wens> nope, looks like the a64
yann has joined #linux-sunxi
<ssvb> wens: does it work?
<ssvb> I mean the a64 dram init code
<wens> ssvb: haven't tried it yet
Worf has joined #linux-sunxi
Worf has quit [Quit: Konversation terminated!]
leviathanch has quit [Remote host closed the connection]
Worf has joined #linux-sunxi
fkluknav has joined #linux-sunxi
LargePrime has quit [Ping timeout: 256 seconds]
kotky has joined #linux-sunxi
f0xx has joined #linux-sunxi
tkaiser has joined #linux-sunxi
<tkaiser> wens: Any specific reason why you wrote DRAM clocked @ 672 MHz and CPU at 1.0 GHz to the BPi M2 Ultra page?
<wens> tkaiser: it's the default setting from their dts file
<MoeIcenowy> Lichee Pi guys are going to make an A33 board and *a V3s board*
<tkaiser> wens: Hmm... CPU cores are there allowed to clock with 1200 MHz. And based on the 'budget cooling' settings DRAM will be clocked with 576 MHz max currently
LargePrime has joined #linux-sunxi
<wens> :/
<tkaiser> wens: DRAM seems ok to me but IMO we should change 1.0 GHz to 1.2 GHz on the wiki page.
<tkaiser> wens: When looking through the various settings it seems that Allwinner is pretty aware of thermal issues (also GPU throttling). But I believe that's due to R40 being also a V40 and T3 (automotive use, GPU and video engine)
<MoeIcenowy> V series is used for cameras
<tkaiser> MoeIcenowy: But you can't compares V40 with V3/V3s (based on product brief)
<MoeIcenowy> yes... maybe V40 is for higher usage
<tkaiser> BTW: V40 product brief mentions EMAC+EMAC while those of R40 and T3 say GMAC+EMAC -- everything else seems to be the same as R40
<wens> tkaiser: the A31 manual also uses EMAC instead of GMAC
<tkaiser> wens: Ok, and that's not even a 'manual' but a 'product brief' so why bother anyway :)
<wens> tkaiser: my point was they use the terms interchangeably :)
<tkaiser> wens: I know but I fear the marketing folks responsible for product briefs have no idea regarding the meaning of all those 4 letter words anyway ;) So they use what looks reasonable (more or less)
popolon has joined #linux-sunxi
<tkaiser> wens: A decade ago I wrote an article for an IT magazine about ZeroConf and Multicast DNS. The copy editor then did a last-minute replacement of all 'mDNS' occurences with 'DNS' -- that was fun since this version went to press :)
<wens> :|
<tkaiser> wens: Another question: Did they gave you a battery too? AFAIK no schematics are released for the M2 Ultra so no one knows whether a SATA disk would be powered when running on battery or not.
<wens> nope
<wens> i'm quite sure i can't buy one from taobao though
<wens> something about couriers not liking li-ion batteries
<tkaiser> wens: ok, then lets wait until schematics are released. Hope it's faster than with Lamobo R1 (took them almost one year) ;)
<wens> ok, got my opi pc plus working (needed to fetch the mac address)
<wens> they changed the polarity of the network leds :|
muvlon_ has quit [Quit: Leaving]
<wens> i think one of my h3 boards is screwing with my ethernet switch...
<wens> montjoie: see the last 2 patches # https://github.com/wens/linux/commits/h3-audio-emac
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<montjoie> wens: I will add them to V5:)
hojnikb has joined #linux-sunxi
<wens> and for some reason my opi-pc+ only runs at 100M half duplex
hojnikb has quit [Quit: Page closed]
mcan_ has joined #linux-sunxi
<dgp> Is there a way to set the pinmux settings within uboot at runtime?
<wens> don't think so
<wens> you can play with the registers directly though
<MoeIcenowy> even in u-boot source we played with the registers directly, with some helper functions
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
Nacho has joined #linux-sunxi
IgorPec has quit [Quit: Nettalk6 - www.ntalk.de]
<dgp> Getting JTAG going on the orange pi zero was easier than I thought it would be :)
<mripard> dgp: if you could document that on the wiki somewhere, that would be great
<dgp> Was just thinking that
<MoeIcenowy> dgp: you used MicroSD Breakout?
<dgp> MoeIcenowy: no, JTAG pins on the two row header
<dgp> OpenOCD can see the target but can't do anything with GDB yet
mcan_ has quit [Quit: by]
victhor has joined #linux-sunxi
lkcl has quit [Read error: Connection reset by peer]
lemonzest has joined #linux-sunxi
lkcl has joined #linux-sunxi
chomwitt has joined #linux-sunxi
Da_Coynul has joined #linux-sunxi
Da_Coynul has quit [Client Quit]
Nacho has quit [Remote host closed the connection]
<terra854> Hey guys, did anyone got hold of the NES Classic?
<terra854> edition
Nacho has joined #linux-sunxi
<jelle> someone posted a u-boot patch
<MoeIcenowy> mripard: do you know who's working on A13/R8 CSI?
Pepe has joined #linux-sunxi
<mripard> I am, but with a low priority
<MoeIcenowy> mripard: have you made any progress?
ldearquer has quit [Remote host closed the connection]
f0xx has quit [Quit: terminated!]
f0xx has joined #linux-sunxi
<MoeIcenowy> mripard: another question, is anyone working on porting axp288-xx drivers to axp20x generic?
<mripard> I haven't made any progress since september or so
<MoeIcenowy> may I do it :-)
<mripard> and I'm not sure what your question about the axp288 is
<MoeIcenowy> I just do not want to duplicate work
<MoeIcenowy> there's many drivers named axp288-xxx in kernel now, which is written by Intel
<MoeIcenowy> is there anyone working on make them not axp288-specified but axp20x-generic?
tuxillo has joined #linux-sunxi
tuxillo has quit [Client Quit]
tuxillo has joined #linux-sunxi
<mripard> not as far as I'm aware
<MoeIcenowy> I think I may do it :-)
<mripard> go ahead
<mripard> I think doesn't care now it's upstream
<wens> intel seems to have abandoned them :(
<scelestic> absolutely unrelated to sunxi but as a debian user since 2.x i just developed some serious hate towards systemd (yes i've been living underneath a rock for some time0
<mripard> and we don't really either since it's not really sunxi related
<MoeIcenowy> wens: someone give them back to life
<mripard> so I you feel like it, then it's definitely something worth doing
<KotCzarny> scelestic: you are not alone, though.
<wens> MoeIcenowy: if you want to, go ahead. I don't have hardware for it
<wens> having my name pop up as maintainer doesn't help :|
<MoeIcenowy> wens: I do not mean doing anything AXP288-related
<MoeIcenowy> but make the drivers working for axp209 and 223
<MoeIcenowy> you at least will care axp223, right?
<wens> yes
<mripard> MoeIcenowy: qschulz has been working on an ADC driver for the 209 / 223
<mripard> and on the battery support as well
<MoeIcenowy> I see nothing in his github
<mripard> he was planning on submitting his patches this week or the next
<MoeIcenowy> oh got a WIP patch
<wens> MoeIcenowy: not everyone pushes to github :)
<wens> mripard: i think the main headache point is that a lot of the pin functions are very generic and flexible
<MoeIcenowy> wens: yes I sometimes also forgot to push
tsuggs has quit [Remote host closed the connection]
<mripard> wens: he also did a pinctrl driver
<MoeIcenowy> but I think qschulz cannot get his patch merged, since there's already axp288-adc and axp288-fuel-gauge :-(
<wens> for example the battery temp. sensor can also be used as an ADC
<wens> and the charging LED can be used like an output GPIO
<wens> MoeIcenowy: honestly, i don't care about axp288
<MoeIcenowy> yes
<MoeIcenowy> but I think at least some kernel maintainers won't allow two similar drivers to exis
<MoeIcenowy> exist
<wens> i would like someone to at least look into whether the 2 axp288 drivers can be reused
<MoeIcenowy> I'm trying to modify axp288-adc to axp20x-adc
<wens> if not, why? is it a mismatch with the split of functions for what we have planned for the axp20x drivers?
<MoeIcenowy> what's planned?
<wens> well, we already have usb-power-supply and gpio
<hramrach> it seems the axp288 functions are hooked into axp20x already
<wens> acin-power-supply is just a copy of usb-power-supply away
<hramrach> it's just not other chip revision has support for that function
<MoeIcenowy> and I do not like the axp288-charger driver
<mripard> wens: and he did acin too
<MoeIcenowy> it's a mixture of usb and acin
<wens> and who knows how those 2 are supposed to interact with the battery charger driver
<wens> mripard: cool
<MoeIcenowy> so axp288-charger, at lesat, won't be used by me
<hramrach> it depends on chip revision :/
<MoeIcenowy> reused *
<MoeIcenowy> but fuel-gauge may be reused
<hramrach> some are more featureful than others with regards to possible use of acin/usbin/battery
<wens> hramrach: what chip revision?
<hramrach> axp152/axp209/whatever
<hramrach> I don't reacall exact details but looked at the registers once and there were some which you would think there must be to make sane use of the PMIC .. bu sometimes they were missing
<hramrach> like registers telling you which power source you are on
<wens> they are the same family
<hramrach> or at least they were not documented well enough
deskwizard has joined #linux-sunxi
bmeneg has quit [Ping timeout: 260 seconds]
bmeneg has joined #linux-sunxi
<willmore> What was wrong with "Lemon pi"? It's my favorite kind of pi. Mmmm.....
<willmore> KotCzarny, now I like it more. ;)
<KotCzarny> it's free then. ever though of creating allwinner sbc? ;)
<willmore> LOL, sure, why not? Everyone else and their dog has.
<willmore> But, no. I'm not up for something like that. I have never designed a board that complex.
<KotCzarny> but usually it just means failure or something bad
<KotCzarny> 'i bought new iphone from ebay but i got a lemon'
<willmore> KotCzarny, "thing with a chronic problem" is a common useage in the USA.
<willmore> The anime meaning hasn't gotten much use, though.
<willmore> Outside of that little group at least.
<KotCzarny> ever seen lemon party website?
<willmore> no. Pretty sure I'm not going to. :)
<KotCzarny> ;)
Ntemis has joined #linux-sunxi
<KotCzarny> good choice
<willmore> I'll add that to the list of things that I won't google for.
<willmore> This is the most complex board I have ever done: http://midnightdesignsolutions.com/dds60/
<KotCzarny> allwinner based one shouldnt be much harder
<KotCzarny> see opi0
<willmore> It's a two layer board with high speed data, low speed data, low level RF, high level RF, and a need to isolate all of them and their separate power runs.
<willmore> I like my Opi Z. I keep thinking it would be better with no ethernet.
<mripard> MoeIcenowy: and for the camera, I had most of the driver written
<mripard> I was able to capture some frames
<mripard> at the right size
mcan_ has joined #linux-sunxi
<mripard> but the image itself wasn't good
<mripard> I have to setup a board to boot with 3.4 and compare
<jelle> nice!
<ErwinH> Why is the root account not locked out after creating a user account?
<MoeIcenowy> ErwinH: it's a distro-related problem
<MoeIcenowy> please ask it with your distro
<jelly> (and it's not a problem at all, root account is independent of normal user accounts)
<ErwinH> Mea culpa!
yann-kaelig has joined #linux-sunxi
<hramrach> hm, USB3
<hramrach> what chip is that S500 Lemon Pi uses?
DullTube has quit [Quit: Leaving]
<willmore> KotCzarny, google "creamy lemon". Do it. :)
<KotCzarny> not going to
<KotCzarny> food time ;)
<willmore> I get a screen full of recipies.
<willmore> Most of them sound good.
afaerber has joined #linux-sunxi
<willmore> No anime at all. even the suggestions are all for different recipies. Maybe google knows me too well? ;)
<hramrach> google's self-censorship is sometimes subtle and sometimes very blatant when you compare results with competition
<hramrach> I have no idea if it applies in this case, though
cnxsoft has quit [Quit: cnxsoft]
afaerber has quit [Ping timeout: 260 seconds]
afaerber has joined #linux-sunxi
f0xx has quit [Ping timeout: 248 seconds]
Jacmet has quit [Ping timeout: 260 seconds]
Jacmet has joined #linux-sunxi
<tuxillo> NiteHawk: thanks
Worf has quit [Quit: Konversation terminated!]
hojnikb has joined #linux-sunxi
<hojnikb> anyone figured out how to change hdmi resolution on h5 ?
<KotCzarny> the same as for any other h3 legacy kernel?
<hojnikb> nope
<hojnikb> its aarch64
<hramrach> so does it have script.bin?
<MoeIcenowy> no
<hramrach> does it have fbdev?
<NiteHawk> tuxillo: if you encounter any problems on Dragonfly, let me know - but the patch seemed pretty "natural", so i finally merged it
<hramrach> does it have simplefb?
<hojnikb> no idea
<hojnikb> i'm running ubuntu image from xunlong site
<tuxillo> NiteHawk: yeah, it built on dragonfly. there are a couple warnings about usleep and getimeofday
<hramrach> then look in /sys to get some idea
<tuxillo> related to the posix chosen for the build
<tuxillo> but that's not so important :)
<tuxillo> and they were there since I tried the first time
tkaiser has joined #linux-sunxi
<NiteHawk> tuxillo: yes, i'm seeing those too on windows builds, but apparently they're not breaking anything important :D
<tkaiser> hojnikb: As the one who released the OS image you use. They currently rely on Allwinner's BSP so for every little bit of changed stuff you need to overwrite large portions of the image's start. In other words: Try to find someone as brave as longsleep back then who did the dirty work to clean up A64 BSP for H5 now. Until this is resolved not that much will happen anyway.
<NiteHawk> btw: wens, thanks for the R40 and A80 patches
<hojnikb> tkaiser: i see
<hojnikb> well at least board is somewhat useful
<longsleep> tkaiser: i might do it again when they release a 2GB version and mainline situation does not suit my needs then :)
<tkaiser> hojnikb: Allwinner BSP issues aren't that on-topic here either. Your only chance at the moment is to check out that H5 'SDK' and try your luck. Awful experience anyway.
<tkaiser> longsleep: In case you want an OPi 3 sample, just ask IgorPec ;)
<hojnikb> tkaiser: i think i'll skip that :)
<hojnikb> besides my little h3 boards are plenty useful now that armbian support is somewhat mature
<longsleep> tkaiser: OPi3 is what exactly?
<hojnikb> orangepi 2e with h5 i'm guessing
<tkaiser> hojnikb: Then simply sit and wait until someone did the work. I really don't understand why people buy new hardware and expect it should work as mature hardware.
<hojnikb> tkaiser: obviously nobody expects thats, especially not from chinese vendors
<tkaiser> longsleep: H5 with 2 GB DRAM and a bit larger. And the same Wi-Fi/BT combo as Pine64 uses
<tkaiser> longsleep: In other words, Pine64 in better quality and A64 exchanged with H5 (and 2 more real USB ports) ;)
<tkaiser> longsleep: Half the size
<hojnikb> yeah whats up with the huge size of pine64
<wens> looks like the r40's pio is compatible with the a20
<longsleep> tkaiser: sounds almost useful :) is the wifi optional?
reinforce has quit [Quit: Leaving.]
<hojnikb> any tests of r40 sata speed ?
<hojnikb> is it still the same crappy implementation as a20 ?
<MoeIcenowy> wens: I think for R40 there's also libdram, right?
<wens> MoeIcenowy: there are sources, which seems the same as a64
<tkaiser> longsleep: I've only seen pictures, it's onboard. But I don't know much about it since not released. Just a 'leak' some weeks ago.
<MoeIcenowy> sources?!
<wens> MoeIcenowy: see the bsp
<longsleep> tkaiser: all right, lets wait for more details then :)
<tuxillo> NiteHawk: in linux you don't get them?
<NiteHawk> tuxillo: nope. linux and osx don't complain - see the build logs at https://travis-ci.org/n1tehawk/sunxi-tools
<tkaiser> longsleep: I would believe Xunlong just waits a bit so in case you want to have a look asking Igor is the best idea (he's in touch with Steven on a regular basis)
<NiteHawk> tuxillo: might also be related to compiler version, i think Travis is a bit conservative there (gcc 4.x)
<hojnikb> is xunlong affiliated with allwinner in any way ?
<hojnikb> everything they put out is based on their hw
<wens> very nice, r40 pinctrl is completely the same as a20
<wens> only new stuff is the I/O disabled state, which we don't care about
<NiteHawk> errr... https://travis-ci.org/linux-sunxi/sunxi-tools of course, not my fork :)
<MoeIcenowy> wens: yay!
<KotCzarny> wens: collect differences,
clonak has quit [Ping timeout: 240 seconds]
<MoeIcenowy> and there's source for H5 libdram!
<MoeIcenowy> apritzel: ^
<wens> KotCzarny: it's what I'm doing :)
<wens> KotCzarny: not so easy to do when there's no datasheet
<KotCzarny> run a20 kernel and check what stopped working? ;)
f0xx has joined #linux-sunxi
<tuxillo> NiteHawk: let me try something
<MoeIcenowy> KotCzarny: many things would stop working :-)
<tkaiser> hojnikb: http://linux-sunxi.org/User:Tkaiser#First_steps_with_R40 (use google translate, everything in german)
<wens> MoeIcenowy: that's not helpful
<wens> looking at the code is easier, and has more info
<MoeIcenowy> do R40 have a CPUs ?
<wens> don't think so
<wens> no secure sram either
clonak has joined #linux-sunxi
<wens> hmm, maybe it does, but the bsp doesn't exactly scream it
<MoeIcenowy> it's really an A20 with some IP block updates :-)
<wens> which makes it extra tricky to work into the existing u-boot structure
<hojnikb> looking at the their product portfolio (allwinner sure has a lot of stuf) i'm guessing they only really make 2 or 3 dies and fuse off stuff that a particular segment doesn't need. For example h3 and v3 seem very similar, even though they are mean for completly different market.
<MoeIcenowy> hojnikb: ARE YOU SURE THAT H3 IS SIMILAR TO V3?
<KotCzarny> wens: maybe there shouldnt be socs categories, only IP blocks autodetect routine
<wens> KotCzarny: we're kind of past that :p
<hojnikb> MoeIcenowy: well both use a7 and mali 400
<MoeIcenowy> V3 is single-core
<hojnikb> you can disable 3 cores
<KotCzarny> wens: code refactoring, ahoy!
<wens> hojnikb: so does half of allwinner's products :p
<hojnikb> wens: thats why i'm suspecting that they only design a few dies and just disable stuff that doesn't work or is not needed
<wens> hojnikb: but they don't tell you this :p
<hojnikb> wens: obviously not
<hojnikb> but if this is true, this would mean possibility to unlock certain ip blocks. Unless they are laser cut.
<MoeIcenowy> I now doubt that the known whole T series are all renamed SoCs.
<MoeIcenowy> T2 - A20, T3 - R40, T8 - A83T
<mripard> hojnikb: usually, the pins are simply not routed
<MoeIcenowy> mripard: I now believe A13 is the same as R8
<MoeIcenowy> someone made a board that uses A13's TVout
<MoeIcenowy> and it's tested
<mripard> who ?
<MoeIcenowy> the "Lichee Pi" I said this afternoon in UTC+8
<MoeIcenowy> (things are in Chinese
<MoeIcenowy> I talked with Zepan several hours ago, and have known his project to make two new boards, with A33 and *V3s*
<wens> shit, need a new clock driver, ever so slight different :(
<MoeIcenowy> wens: you mean a new ccu driver?
<wens> MoeIcenowy: yeah
<wens> hoping that all types are covered
<hojnikb> isn't v3s meant for dashcams and sports cams ?
<MoeIcenowy> The A33-based "Lichee Pi 2" is now still as a task, but I know the V3s-based "Lichee Pi 0" is now being designed.
<MoeIcenowy> hojnikb: yes
<MoeIcenowy> there's still now no development boards of V3s
<MoeIcenowy> although someone posted a link to V3s datasheet (user manual included)
<hojnikb> so v3s is v3 with onpackage ram correct ?
<MoeIcenowy> hojnikb: I think so
<hojnikb> so basicaly you could make a very simple and very cheap dev board with this
<hojnikb> no need for ram or ram routing :)
<MoeIcenowy> yes
<MoeIcenowy> and it's in TQFP
<MoeIcenowy> so it's even possible to solder the board without BGA :-)
<hojnikb> chop chop xunlong, make it happen :)
<hojnikb> rpi zero is still cheaper >:)
<jelle> not avaliable ;-)
<MoeIcenowy> Zepan is a new allwinner board designer
<MoeIcenowy> the project Lichee Pi is started in Augest this year
<hojnikb> lichee pi ???
<MoeIcenowy> hojnikb: a new allwinner pi ;-)
<hojnikb> heh
<hojnikb> don't we have enough fruits allready ? :)
<jelle> hojnikb: none as tasty
<hojnikb> jello: touché :)
<MoeIcenowy> Let's check our community's fruits storage ;-)
<MoeIcenowy> we have oranges, bananas and lichees now ;-)
<NiteHawk> and we were obviously missing one that creates a nice amount of confusions with the "Lichee SDK" :D
<hojnikb> dont forget bananas
<KotCzarny> we dont talk bananas here.
<KotCzarny> spoiled fruit.
<jski> yum oranges
<hojnikb> WHHY
<KotCzarny> to all: new device howto shouldnt contain 'support status', only a link to http://linux-sunxi.org/Linux_mainlining_effort#Status_Matrix
<KotCzarny> otherwise most of the device pages are horribly outdated
<MoeIcenowy> KotCzarny: soon we may need support to V3s
<MoeIcenowy> V3{,s} is Single-Core Cortex-A7
<MoeIcenowy> designed for cameras and (ab)used by Zepan for Lichee Pi Zero
<jelle> but you like the abuse, right?
f0xx has quit [Ping timeout: 258 seconds]
<NiteHawk> KotCzarny: may still be somewhat relevant for the (3.4) BSP kernels - but yes, mainline tends to get outdated quickly. maybe adjust the ND template accordingly... libv?
<KotCzarny> nitehawk: and what's worst, people see device page, see a13, see 'unsupported' and leave
<KotCzarny> etc
<MoeIcenowy> I think http://linux-sunxi.org/index.php?title=Allwinner_SoC_Family needs a big refactoring
<libv> start fixing device pages then.
<libv> it's a wiki.
<KotCzarny> i was ASKING first
<KotCzarny> :P
<MoeIcenowy> I want to discuss the Allwinner_SoC_Family refactoring
<libv> i just noticed how stupid mainline only, never mind docs or users, is pretty stupid by poking at rockchip for a few d
<ssvb> MoeIcenowy: what's wrong with this page other than missing a bit of information about newer SoCs?
<libv> if you actually want users, and feedback, and better code, you have to provide some documentation as well.
JohnDoe_71Rus has joined #linux-sunxi
<MoeIcenowy> ssvb: the reference links are dying
<libv> and that information in turn should be updatable.
<libv> not some blog entry, or something on a website of a vendor
<MoeIcenowy> and now bigger diversity appears in AW SoCs
<MoeIcenowy> some SoCs' pages disappeared from allwinnertech.com
<ssvb> we can't help it
<ssvb> but there is also https://archive.org/web/
<MoeIcenowy> yes
<MoeIcenowy> We may link the died SoC pages to archive.org
<libv> which is why stuff should be documented on the wiki. vendors will move on rather quickly
<ssvb> we should make sure to teach it about every external web page we are linking to
<MoeIcenowy> and the still alive ones to allwinnertech.com
<MoeIcenowy> and maybe we need some colors to identify whether a SoC is still on the website or not
<MoeIcenowy> for example, A10 disappeared, and A20 is still there
<terra854> They disappeared probably because it is discontinued
<MoeIcenowy> terra854: you got it
<KotCzarny> MoeIcenowy: just add 'dcnd' on the http://linux-sunxi.org/index.php?title=Allwinner_SoC_Family ?
<ssvb> just first fix whatever is obvious and easy
<ssvb> if some page has moved, just update the link
<ssvb> if it disappeared completely, then we can link to its archive.org copy
<MoeIcenowy> ssvb: will you care add (dcnd) to disappeared SoCs?
<ssvb> what is dcnd?
<MoeIcenowy> discontinued
<jelle> A10 is gone?
<MoeIcenowy> and we need something to indicate SoC renames (e.g. H8 to A83T, R8 to A13, R16 to A33
<MoeIcenowy> jelle: yes, gone
<MoeIcenowy> as it can be fully replaced by A20
<jelle> why?
<MoeIcenowy> wow 50000 A10 ;-)
<wens> i assume the designs, masks and packaging directions will live forever, you just have to give them enough incentive to do a run for you
<hojnikb> roseapple pi
<hojnikb> found another fruit board :)
<ssvb> I guess Allwinner A10 is more like NRND (not recommended for new designs)
avph has quit [Ping timeout: 246 seconds]
<KotCzarny> hmm, looking at online stores, there are usually raspberries and oranges
<MoeIcenowy> raspberry have the biggest community
<MoeIcenowy> and oranges are cheapest
<KotCzarny> yeah, whats new is that oranges have penetrated the market
<terra854> What about bananas?
<KotCzarny> bananas suck
<ssvb> MoeIcenowy: btw, here is the first link from the Allwinner_SoC_Family page, which is now broken - http://web.archive.org/web/20160510052432/http://www.allwinnertech.com/clq/processorh/AllwinnerH8.html
<MoeIcenowy> ssvb: I will also try to add all reference pages to the wiki page
<MoeIcenowy> from web archive
<hojnikb> woah roseapple pi actually has usb3
<hojnikb> noice
reinforce has joined #linux-sunxi
avph has joined #linux-sunxi
<scelestic> oh and a powervr
<tuxillo> NiteHawk: tried building it with gcc47 but it produced the same warning.
<hojnikb> s500 just needs a gigabit mac and it would be perfect
<tuxillo> it is related to the system includes
<tuxillo> apparently usleep and gettimeofday are not included in that posix
ErwinH has quit [Ping timeout: 260 seconds]
<NiteHawk> tuxillo: it's related to our "progress" display code. i'll eventually get around to take a closer look at it, and possibly fix the warnings
<NiteHawk> not considered 'critical' for now (aka "that can wait" :P)
hojnikb has quit [Quit: Page closed]
<tuxillo> sure, just mentioning
<tuxillo> I'd have to guess why that posix define is set
<KotCzarny> anyone can paste mainline status link into 'status' paragraph?
silviop has quit [Remote host closed the connection]
<miasma> hm the soc pages are probably a bit outdated too. e.g. on the H3 SoC page "Currently, there is no sunxi support available yet. SDK has been released in May-2015, and devices are slowly becoming available."
<MoeIcenowy> I think for some SoC I should do "citation needed"
<MoeIcenowy> e.g. R18
<MoeIcenowy> R18 never exists on allwinnertech.com...
<MoeIcenowy> and allwinner seems to made newer F-series SoCs..
<MoeIcenowy> renewer many references on Allwinner_SoC_Family
<tuxillo> besides this pine64, are there more armv8 boards?
<tuxillo> SoCs
<KotCzarny> a64, h5
<tuxillo> like which one?
<MoeIcenowy> SoC: H5
<MoeIcenowy> in Orange Pi PC2
<tuxillo> ok
<MoeIcenowy> Banana Pi M64 is also A64-based
<tuxillo> worth buying?
<MoeIcenowy> how to show SoC rename on Allwinner_SoC_Family is also a problem...
<KotCzarny> never buy bananas
<tuxillo> you know, my drawers are empty :)
<KotCzarny> pine64 is better
<KotCzarny> and you can wait for more oranges if you need more ram
<KotCzarny> MoeIcenowy: redirect page
<KotCzarny> or add as notes under table
<MoeIcenowy> I will do some try on R8/A13 pair
<terra854> Was hoping Allwinner made a SoC with 8 core arm64 and a better Mali GPU
<terra854> and not using the crappy Mali 4xx
<terra854> and outdated
<wens> terra854: you'll probably have better luck with mediatek or rockchip?
<MoeIcenowy> do not try mediatek
<MoeIcenowy> they're covered with heavy NDA
<tuxillo> hmm, I thought ODROIDs had Allwinner
<MoeIcenowy> As OtakuNekoP said, Banana Pi R2 used a MediaTek SoC, and a lot of the price of R2 is for the document fee
<terra854> ODROIDs have exynos
<MoeIcenowy> s/document/SDK
<terra854> and amlogic
<MoeIcenowy> MediaTek SDK is even difficult to retrieve
<libv> miasma: that is so because nothing related to c.h.i.p. ever seeps back into the wiki.
tkaiser has quit [Read error: No route to host]
tkaiser has joined #linux-sunxi
<tuxillo> so besides the pine64 what else would you recommend?
<KotCzarny> opipc2, but h5 is not yet supported, so might take a while
<tkaiser> MoeIcenowy: First Pine64 prototypes were made with R18 and SinoVoip also did a few BPi M3 with R58 and H8 instead.
<MoeIcenowy> tkaiser: yes... do you have the link to the Pine64 prototypes now?
<tuxillo> orange pi pc2 from aliexpress 22eur, comes with H5
<tuxillo> and yeah, 1GB
gnufan has joined #linux-sunxi
<MoeIcenowy> and there's also a problem for slightly different SoCs
<MoeIcenowy> for example, A64/H64, H3/H2+
<tuxillo> what is it
<MoeIcenowy> I'm talking about the wifi
<MoeIcenowy> wiki *
<tuxillo> ah
<tuxillo> :)
<MoeIcenowy> I created a redirect from R16 to A33, R8 to A13
<KotCzarny> you should also a note on a13/a33 pages if they dont have one regarding R version
<MoeIcenowy> KotCzarny: note added
<MoeIcenowy> and added R8/R16 devices categories' show to A13/33 pages
<MoeIcenowy> now... should we merge H64 and A64, merge H2+( not exist now) and H3 ?
<MoeIcenowy> they're proven to have the same SoC ID, but slight differences
<tkaiser> MoeIcenowy: H64 is dead. And H2+ slightly different.
<MoeIcenowy> tkaiser: there's many dead SoCs on our wiki :-)
<tkaiser> What about an own page for H2+ that just outlines the differences and then contains a link to H3 page? Same with H64
<MoeIcenowy> OK
<KotCzarny> tkaiser, why not just a paragraph on h3 page?
<MoeIcenowy> yes it's a problem
<tkaiser> I'm fine with that but OPi Zero page already links to a H2+ page. We could add a 'variants' section on the 'main' SoC pages and the link to the anchor?
<tkaiser> So H2+ is a link to [[H3|#Variants]]
<MoeIcenowy> we can make redirections
<KotCzarny> its like cataloguing 'products' made from lego bricks
<tkaiser> Can a redirect also target an anchor? H2+ is a redirect to [[H3|H2+]]?
<MoeIcenowy> I think it's available on Wikipedia
<MoeIcenowy> and I think for one SoC allwinner have 3 kind of IDs
<MoeIcenowy> we cared most on product names (A33, A64)
<MoeIcenowy> then on sequence id (sun8iw5, sun50iw1)
<MoeIcenowy> but we nearly ignored the develoment code (aster, tulip)
<KotCzarny> fruits made from flowers
<MoeIcenowy> and from Tina SDK we proved that R16 comes with the same development code with A33 (aster)
<MoeIcenowy> The development code usually is also the start of "Build Number" in stock Android
LargePrime has quit [Ping timeout: 265 seconds]
<yann-kaelig> Hi. This is not really clear for me on the sunxi wiki, but can I use the sunxi-next branch to get the 3D graphics and playback acceleration ?
<MoeIcenowy> yann-kaelig: no.
lemonzest has quit [Quit: Leaving]
<yann-kaelig> Arf. There is no way to get 3D graphics and acceleration instead the old sunxi-3.4 branch ?
<KotCzarny> heh, files in lichee/tools/daily_build/ contain plain email addresses if anyone wants to spam allwinner guys
<MoeIcenowy> KotCzarny: ;-)
KB3VGW has joined #linux-sunxi
<MoeIcenowy> but I think at least one of the mail addresses are not available ;-)
LargePrime has joined #linux-sunxi
<yann-kaelig> Why is there no up-to-date of 3D graphics and acceleration with newer kernel, at least on the longterm 4.4.x kernel version ?
<KotCzarny> because licenses
<KotCzarny> and moneys
<yann-kaelig> really, the licence close the use on a specific kernel ?
<KotCzarny> licence closes use on any kernel if its not permissive/unknown
jstein_ has joined #linux-sunxi
deskwizard has quit [Ping timeout: 248 seconds]
jstein_ is now known as jstein
<MoeIcenowy> to be honest there's 3d support for some forked 4.x
<MoeIcenowy> using blobs from NextThingCo
<yann-kaelig> that really strange for me this sort of licence. that mean all the work who has done in the past is for ever lost
<wens> yann-kaelig: the license is for the vendor, not the community
<wens> yann-kaelig: it is not redistributable
leviathanch has joined #linux-sunxi
<tuxillo> what's the CatA errata?
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
massi has quit [Quit: Leaving]
afaerber has quit [Quit: Ex-Chat]
leviathanch has quit [Read error: Connection reset by peer]
ldearquer has joined #linux-sunxi
leviathanch has joined #linux-sunxi
tkaiser has joined #linux-sunxi
<willmore> scelestic, PowerVR graphics on a chip? Cool, then noone will have to worry about if it'll get 3D graphics support! The answer will be a solid "no". That makes things a lot easier.
<willmore> KotCzarny, aren't all fruits made from flowers?
<KotCzarny> willmore: what about fruits of the loins?
<willmore> You going to name boards off of my children?
<willmore> Can we use their nicknames? That'd be cool. :)
paulk-collins has joined #linux-sunxi
<KotCzarny> tbh, child pi could be cool marketing name
<KotCzarny> or kid pi
<willmore> Moo pi and Bear pi. That'll work.
<willmore> Moo pi might be taken badly in English speaking countries.
<willmore> Bear pi just sounds confusing. Is it pie made from a bear? Is it a pie for a bear? Is it a spherical bear with uniform density?
<KotCzarny> or a pi board fitting in a beer
<willmore> Beer pi?
<KotCzarny> great hiding spot
<willmore> Too bad we can't sell them to anyone under 21/18. Then again, that might simply support in the community. ;)
<willmore> Did you mean Bear?
<willmore> All boards will fit in a bear. You just have to get the bear to hold still for a bit.
<KotCzarny> IoA, internet of animals
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
<KotCzarny> now run them around in the forest with mesh networking configured
<KotCzarny> and power off body heat
<willmore> Piezo implants in the feet harvesting energy from locomotion.
<KotCzarny> too many cables, it should fit in a small package
<BurtyB> hmm beer
<willmore> The board wouldn't have to be open hardware. Because it only has to be free as in beer.
<KotCzarny> well, once implanted, you set the bear free
cptG has joined #linux-sunxi
<willmore> That's why we can't call it bear, because then it would have to be free as in bear.
cptG_ has quit [Ping timeout: 260 seconds]
* NiteHawk thinks that bear-ly makes a difference
<MoeIcenowy> this channel is good at making me laugh ;-)
* willmore finally found a way to help linux-sunxi
<longsleep> cheers!
* longsleep just got a fresh beer
<longsleep> mut it was not free :/
<willmore> Make sure it's not a bear!
<willmore> That would be unplesant for all involved.
<longsleep> yes but in german luckily those two sound different :)
IgorPec has joined #linux-sunxi
<willmore> What is bear in German?
<willmore> I know beer. :)
<KotCzarny> ba"r
<longsleep> Bier
<longsleep> Baer
<longsleep> i got english keyboard hold on
<longsleep> Bär
<longsleep> umlaut ftw
* willmore remembers to be careful when he orders a drink next time in Germany.
<willmore> Wait, umlats are used for something other than heavy metal band names?
<longsleep> just say it as you would in english as those two sound the same
<KotCzarny> willmore: yup
<willmore> Back on topic.. How's the H5 hacking going?
* willmore looks over at his shiny new PC2.
|Jeroen| has joined #linux-sunxi
florianH has quit [Quit: Connection closed for inactivity]
<MoeIcenowy> oh H5 user manual seems to be now available
<MoeIcenowy> hooary
<KotCzarny> MoeIcenowy: you mean that 800pages copypastegoodfornothing thing?
Ntemis has quit [Remote host closed the connection]
<willmore> KotCzarny, link?
<KotCzarny> on opipc2 page
<KotCzarny> hrm. or where did tkaiser upload it
<willmore> I saw the datasheet. Didn't see the user manual. Downloading...
<willmore> Uhh, it's the same file as the datasheet. ;)
<KotCzarny> H5_User_Manual.pdf
<KotCzarny> this one
<MoeIcenowy> it's should be called a "detailed datasheet"
<willmore> KotCzarny, same file as the allwinner-h5-datasheet-v1.0.pdf file.
<willmore> MoeIcenowy, yeah, not quite user manual level of detail.
Pepe has quit [Ping timeout: 252 seconds]
<KotCzarny> willmore: nope, one file is 2MB, the other is 23MB
<willmore> KotCzarny, Hmm, I have a datasheet file that's 24.9MB--byte identical to the user manual link there.
<MoeIcenowy> willmore: it's just the file.
<MoeIcenowy> but what we get at first is a 2MB datasheet
<willmore> So many renamed things..... If the user manual isn't detailed enough, I question the value of having a 'datasheet' at all.
<willmore> Understood.
<MoeIcenowy> this file is detailed enough, at the perspective of allwinner user manual
* willmore is spoiled by good user manuals.
<KotCzarny> willmore: interesting (still unknown) bits are missing
<willmore> Is the SPI driver for the H3 (and related chips) stable and full featured?
* willmore is trying to find some code he might be able to help with
tkaiser has joined #linux-sunxi
vagrantc has joined #linux-sunxi
Err has quit [Read error: Connection reset by peer]
lennyraposo has quit [Ping timeout: 250 seconds]
yann has quit [Ping timeout: 265 seconds]
terra854 has quit [Quit: Connection closed for inactivity]
yang has joined #linux-sunxi
ldearquer has quit [Remote host closed the connection]
JohnDoe_71Rus has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/]
pg12 has quit [Ping timeout: 240 seconds]
pg12 has joined #linux-sunxi
lamer14804481979 has joined #linux-sunxi
tkaiser has quit [Ping timeout: 250 seconds]
BenG83 has quit [Ping timeout: 260 seconds]
Err has joined #linux-sunxi
leviathanch has quit [Remote host closed the connection]
bonbons has joined #linux-sunxi
KB3VGW has quit [Ping timeout: 240 seconds]
KB3VGW has joined #linux-sunxi
lennyraposo has joined #linux-sunxi
lemonzest has joined #linux-sunxi
Ntemis has joined #linux-sunxi
jernej has joined #linux-sunxi
BenG83 has joined #linux-sunxi
<jernej> Can someone upload datasheet and user manual to wiki from here: https://github.com/Zepan/ilichee/tree/master/%E8%B5%84%E6%BA%90%E6%96%87%E4%BB%B6
<jernej> tkaiser: ^^^
reinforce has quit [Quit: Leaving.]
bugzc has quit [Ping timeout: 260 seconds]
BenG83 has quit [Quit: Leaving]
iamfrankenstein has quit [Quit: iamfrankenstein]
dh1tw has joined #linux-sunxi
|Jeroen| has quit [Quit: dada]
<NiteHawk> jernej: you mean the A13 one?
yann has joined #linux-sunxi
yann-kaelig has quit [Quit: Leaving]
IgorPec has quit [Ping timeout: 258 seconds]
<NiteHawk> looking at https://github.com/allwinner-zh/documents/tree/master/A13, those v1.30 aren't the most recent ones either?
<jernej> NiteHawk: yes, version 1.3 is not yet on the wiki
<jernej> NiteHawk: Ok, I missed those
KB3VGW has quit [Ping timeout: 265 seconds]
<jernej> wiki has outdated versions
<NiteHawk> could you cross-check them against those allwinner ones on github? if those are more recent, i can add them to the wiki
KB3VGW has joined #linux-sunxi
<jernej> I can, but I wonder if linking to github is best thing to do
<NiteHawk> nah, allwinnner-zh has been pretty unreliable in updating and/or providing docs for newer SoCs there too - at least that's my impression
<jernej> it seems that github versions are the newest, but strangely page count is much lower
<jernej> they claim that they changed document format
<jernej> which means comparing is tedious
<NiteHawk> just check the creation date, should be a good indicator
<KotCzarny> keep both versions then
<jernej> that is newer, for sure
<KotCzarny> just in case they removed info
<jernej> githoub documents have newer creation date
bonbons has quit [Quit: Leaving]
smooker has joined #linux-sunxi
smooker has quit [Remote host closed the connection]
KB3VGW has quit [Ping timeout: 240 seconds]
KB3VGW has joined #linux-sunxi
lemonzest has quit [Quit: Leaving]
KB3VGW has quit [Ping timeout: 265 seconds]
apritzel has joined #linux-sunxi
KB3VGW has joined #linux-sunxi
<apritzel> willmore: you were asking about H5 support?
Mr__Anderson has joined #linux-sunxi
KB3VGW has quit [Read error: Connection reset by peer]
KB3VGW has joined #linux-sunxi
<apritzel> willmore: and I have a U-Boot branch as well, but DRAM support is not ready yet (though some work has started)
sunxi_fan1 has joined #linux-sunxi
alex______ has joined #linux-sunxi
paulk-collins has quit [Quit: Leaving]
jstein has quit [Remote host closed the connection]
<alex______> Hi
<alex______> I would like to enable eMMC on my M64, anyone has tried it with success?
<apritzel> alex______: BananaPi-M64?
<apritzel> which kernel?
<alex______> yes
<alex______> longsleep's kernel
<apritzel> too bad ;-)
<alex______> i have red your comment abouteMMC
<KB3VGW> longsleep sounds like a knap
<alex______> apritzel, do you know if it is 1.8v or 3.3v?
<apritzel> 3.3V
<apritzel> alex______: so mripard has found a proper fix, now it works nicely with a mainline kernel
<apritzel> the DMA buffer size is limited to 8K on that MMC2 controller (and only there, MMC0 & MMC1 can take 64K)
<alex______> it does not work on 3.10.x?
<apritzel> alex______: no idea and really don't care
<alex______> i see
<apritzel> but it should
<alex______> i configures with 1.8v regulator
<alex______> how can i grab your kernel?
lennyraposo has quit [Quit: Leaving.]
<apritzel> if you have used 3.10 before, you need to grab more than the kernel
<alex______> u-boot you mean?
<apritzel> yes
<apritzel> and ATF
<apritzel> but I don't think I have published much about the BPi-M64
<apritzel> it's not much, really, mostly just adding the eMMC to .dts and defconfig
<alex______> last thing i read from your patch you say you had some problems with eMMC
<apritzel> upstream U-Boot works fine
<apritzel> alex______: yeah, that's now fixed thanks to mripard
<apritzel> give me some minutes to push that fix to my kernel branch
<alex______> oh, great
<apritzel> do you desperately need to _boot_ from eMMC?
<alex______> no, not really
<apritzel> otherwise you could grab my firmware image (SPL, ATF, U-Boot)
<apritzel> this has USB and Ethernet support, so you can then load the kernel with the fix and the .dtb
<alex______> i would prefer building from source
<apritzel> as you like, but it's not a walk in the park, unfortunately
<apritzel> as it's WIP in the moment
<alex______> fine for me, i am always learning...
<apritzel> this even has BPi-M64 support
<alex______> i am not in a rush, if you can update about the eMMC, please do so, i can wait, no problem
<alex______> i will try, thank you
<apritzel> alex______: so as I said: U-Boot is fine, and you can be my guinea pig and check the updated README.pine64
<alex______> i will, thanks
<apritzel> alex______: regarding the voltage: I can't tell, AFAIK there is no schematics
<apritzel> but I doubt that it's 1.8V
<alex______> i received the fex from them using 1.8v
<alex______> if it is 1.8v and i use 3.3v could damage it?
<apritzel> since I believe the SPI flash can't take this, and it's connected to the same port
<apritzel> argh, wrong board, sorry
<apritzel> the BPi has no SPI flash ;-)
<apritzel> let me try to measure the voltage on PortC
<apritzel> alex______: atm there is no explicit voltage setting with mainline kernel
<apritzel> so it just defaults to whatever is the reset value
<alex______> ok, will do some experiments and see how it works
<apritzel> the "mainline" firmware does not touch the AXP rail for PortC, so it should be safe
<apritzel> and it worked for me
<apritzel> as the _board_ is configured to one voltage already
<apritzel> and the upstream MMC driver does not support HS400 anyway, so no need for 1.8V
lamer14804481979 has quit [Quit: jIRCii - http://www.oldschoolirc.com]
<apritzel> and HS400 is not needed since the actual flash ship is limited to 100MB/s for reads anyway (per datasheet, probably lower in real life)
KB3VGW- has joined #linux-sunxi
KB3VGW has quit [Ping timeout: 244 seconds]
<alex______> ok, thanks appreciate your help...
alex______ has quit [Quit: Page closed]
KB3VGW- has quit [Read error: Connection reset by peer]
dh1tw has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
Mr__Anderson has quit [Remote host closed the connection]
Seppoz has quit [Remote host closed the connection]
Da_Coynul has joined #linux-sunxi