jstein has quit [Remote host closed the connection]
<apritzel>
Re: Pine64 AXP: (dumping here what I found out)
<apritzel>
I tried to enable the S_TWI in Linux, which seemed to work
<apritzel>
however scanning the I2C bus gives me nothing, also takes a while
<apritzel>
looking at the bus gating and reset registers, I see that RSB is enabled
<apritzel>
which apparently must have been done by boot0 (or even the BROM?)
<apritzel>
the AXP resets in I2C mode, but has most likely been switched to RSB mode (via register 0x3E: interface mode select)
<apritzel>
so trying to access it via I2C does no longer work, unless we switch it back to I2C mode: which of course requires a command to be send via RSB
<apritzel>
so we would need to use the RSB, at least for a single "revert to I2C" command
<apritzel>
but it looks like that the RSB has been initialized already, so we may skip this step and just go ahead with a read (or write) request
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
apritzel1 has quit [Ping timeout: 244 seconds]
apritzel has quit [Ping timeout: 244 seconds]
VargaD has quit [Ping timeout: 260 seconds]
willmore has quit [Ping timeout: 250 seconds]
VargaD has joined #linux-sunxi
willmore has joined #linux-sunxi
ninolein has joined #linux-sunxi
ninolein_ has quit [Ping timeout: 276 seconds]
avph has quit [Ping timeout: 240 seconds]
avph_ is now known as avph
avph_ has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
cnxsoft has quit [Read error: Connection reset by peer]
<luoyi>
codekipper: and sun7i-a20-olimex-som-evb.dts is the demo dts file ?
cnxsoft has joined #linux-sunxi
<codekipper>
you have to tailor it to your needs..what codec and target are you thinking of.
Andy-D has joined #linux-sunxi
<luoyi>
I didn't need to config the codec. just config the board's i2s output first.
FLorian has joined #linux-sunxi
<luoyi>
I can't find any dts file which contain : allwinner,sun4i-a10-dai this compatiable string
<luoyi>
how do you load the module ?
JohnDoe_71Rus has quit [Quit: KVIrc KVIrc Aria 4.9.2, revision: git-6627-gf708225, sources date: 20160102, built on: 2016-05-12 17:32:15 UTC git-6627-gf708225 http://www.kvirc.net/]
<codekipper>
oh yeah..you also need all the dtsi stuff related like the clocks...what tag you working off?
<codekipper>
quickest thing to do would be to add my github as a remote to your git, fetch then cherry pick what you need.
<luoyi>
linux-sunxi-audio_for_next
<codekipper>
is this Maxime's branch?
FLorian has quit [Remote host closed the connection]
<KotCzarny>
i'm compiling hans' kernel now to see if it works
<dave0x6d>
I thought there was a repo that was a bit newer than those. Let me check my scrollback...
<tkaiser>
KotCzarny: When OPi Plus 2E arrives please use Armbian image for OPi PC and this fex: http://pastebin.com/jKK17CkV (or in case you want to start using SSH then use image for OPi Plus instead since then some USB ports won't work until you exchange script.bin)
<KotCzarny>
tkaiser, remind me when it arrives, my memory is awful, and i guess it will be few days till it arrives
<tkaiser>
dave0x6d: I would use wens' -- he updated it yesterday.
<dave0x6d>
wens repo isn't listed on the wiki, is it?
<tkaiser>
KotCzarny: Nope, that was fire&forget ;)
<KotCzarny>
;)
<KotCzarny>
tkaiser: does wens' tree have usb working?
<tkaiser>
KotCzarny: AFAIK yes
<dave0x6d>
tkaiser: yesterday?
<KotCzarny>
k, gotta grab it too
<tkaiser>
dave0x6d: Or the day before, please see IRC log
<dave0x6d>
tkaiser: er, I mean I'm looking at the commits and the last one is from April 4th :p
IgorPec has joined #linux-sunxi
<tkaiser>
dave0x6d: Well some patches are flying around since ages, at least he updated stuff 2 days ago and I would assume it works
<dave0x6d>
ah
<dave0x6d>
holy crap, is there really 600 patches, or are some of those just merges from mainline again?
iamfrankenstein1 has joined #linux-sunxi
<tkaiser>
The latter
<tkaiser>
And BTW: the last time I played with it I tried to combine USB, Ethernet and megi's THS patches (https://github.com/megous/linux/tree/orange-pi -- necessary for thermal protection) and failed totally :)
<KotCzarny>
umkay, lets see if hans' kernel boots
apritzel has joined #linux-sunxi
iamfrankenstein1 has quit [Ping timeout: 260 seconds]
<wens>
apritzel: unless you want to use some gpio sequence to mimic it
<KotCzarny>
how come arp packets work and icmp dont?
<wens>
montjoie: ^
<apritzel>
I would have felt better if we are using a well known and documented interface
<apritzel>
instead of some dodgy, undocumented and proprietary one
<wens>
it's documented enough actually
<wens>
all the command sequences and stuff
<apritzel>
but in the end you are right that is probably does not matter
<KotCzarny>
and the cable is crossed, so its not an mdi/mdix issue
<wens>
you just need to combine the info from multiple datasheets
<apritzel>
;-)
IgorPec has joined #linux-sunxi
<wens>
apritzel: anyway, we have drivers for it in both the kernel and u-boot
<apritzel>
I wonder if we should publish a linux-sunxi editition of those datasheets
<wens>
the u-boot one is really basic though
<apritzel>
wens: the point is: I don't want to control the regulator from there
<wens>
i know :)
<apritzel>
but instead rather from ATF
<apritzel>
well anyway, I was hoping for I2C to work to make it easier
<wens>
and what about kernel <-> ATF part? custom SMC calls?
<apritzel>
and I tried U-Boot's RSB code som weeks ago
<apritzel>
and it didn't work as expected
<apritzel>
wens: not custom, but either ARM's SCPI protocol or something similar
<wens>
ok
<wens>
wonder what it would look like in the dT
<apritzel>
SCPI support is already in mainline kernel, along with proper binding, IIRC
tkaiser has quit [Read error: Connection reset by peer]
<apritzel>
the point is: the DT part is pretty simple, as you query the clocks at runtime
<apritzel>
there is a call to give the number of clocks and a call to get per-clock information
<apritzel>
*to get the number of clocks*
<apritzel>
per-clock information is pretty limited at the moment (basically just a string)
<apritzel>
and then you set the frequency
<apritzel>
so the kernel enumerates all clocks and picks the one with the right name, then sets the frequency, if needed
<apritzel>
not sure this is enough for this sophisticated clock setup we have with sunxi boards, though
<wens>
so ATF exports all clocks to the kernel? then the CCF would still be in control?
<wens>
bed time
raknaz has joined #linux-sunxi
<apritzel>
wens: I have not much experience with the CCF, but that was my plan, yes
<apritzel>
same with regulators, by the way: seems to be more useful there, as they are more dangerous and potentially more board dependent
medvid has quit [Ping timeout: 260 seconds]
medvid has joined #linux-sunxi
tkaiser has joined #linux-sunxi
cajg has quit [Ping timeout: 276 seconds]
raknaz has quit [Quit: raknaz]
tkaiser has quit [Ping timeout: 276 seconds]
cajg has joined #linux-sunxi
<NiteHawk>
longsleep: i'm currently experimenting with gentoo-arm64 on your kernel 3.10.101-0-pine64-longsleep. in doing so, i noticed that CONFIG_BLK_DEV_BSG is not set - something that eudev compilation warns about. you might consider enabling it, as it seems to be used by recent udev (see the help text in ./block/Kconfig)
tkaiser has joined #linux-sunxi
IgorPec has quit [Ping timeout: 276 seconds]
valkyr1e has quit [Quit: Bye.]
IgorPec11116 has joined #linux-sunxi
iamfrankenstein1 has joined #linux-sunxi
valkyr1e has joined #linux-sunxi
iamfrankenstein has quit [Ping timeout: 260 seconds]
iamfrankenstein1 is now known as iamfrankenstein
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
<jelle>
hmm how do you usually dump clock settings from a BSP image?
<jelle>
I'm guessing I'm not setting some clocks up in u-boot for hdmi h3 supprot
Netlynx has quit [Ping timeout: 260 seconds]
IgorPec11116 has quit [Ping timeout: 246 seconds]
apritzel has quit [Ping timeout: 244 seconds]
Netlynx has joined #linux-sunxi
IgorPec11116 has joined #linux-sunxi
IgorPec11116 has quit [Ping timeout: 250 seconds]
Gerwin_J has quit [Quit: Gerwin_J]
IgorPec11116 has joined #linux-sunxi
IgorPec has joined #linux-sunxi
IgorPec has quit [Ping timeout: 260 seconds]
tkaiser has quit [Ping timeout: 246 seconds]
tkaiser has joined #linux-sunxi
barish has joined #linux-sunxi
<jemk>
jelle: h3 hdmi only needs pll3 and hdmi_slow_clk, the hdmi_clk is unused, instead the divider in the hdmi block itself has to be set correctly
<jelle>
jemk: hmmm the ddc_clk I guess
<jelle>
or is that pll3 *or* hdmi_slow_clk
<jemk>
for ddc it uses hdmi_slow_clk, for pixeldata pll3
<jelle>
aha
<jelle>
so I'm enabling the ppl3 clock already
<jelle>
HPD detection works
<jelle>
but reading edid returns 00000 :-)
barish has left #linux-sunxi [#linux-sunxi]
sirblackheart has quit [Quit: sirblackheart]
libcg has joined #linux-sunxi
MY123 has joined #linux-sunxi
<MY123>
libv: ping
apritzel has joined #linux-sunxi
<jemk>
jelle: no timeouts or errors?
<longsleep>
NiteHawk: ok, thanks for the hin - i will add it
<jelle>
jemk: nope, all I've done so far is copy the code from francois mainline driver
<jelle>
or I'm failing at printing the edid header ofcourse
<jemk>
jelle: if you use the original code pll3 has to be 297mhz, because the dividers are hardcoded for the modes, but i don't think its necessary for ddc
<jelle>
jemk: hmm yup I'm basically rewriting sunxi_display.c and removed everything not relevant to the h3. So I re-used the clock_set_pll3
<jelle>
to 300mhz
jernej_ has quit [Quit: Konversation terminated!]
zuikis has left #linux-sunxi [#linux-sunxi]
<longsleep>
NiteHawk: yes, everyone can add that later - i left the journal on by default as people are not aware that they should shutdown and not just unplug power
<NiteHawk>
longsleep: ah, i see - so recovery can be done gracefully on the next startup. that's a good argument