<alexvf>
philippe_fouquet: i'm no expert but ... i would say you can use them at your own risk
<alexvf>
it should be alternative community sources
<alexvf>
and i think that you can even use mainline kernel if you don't need hd video
<alexvf>
but hopefully someone else here can confirm you this
<philippe_fouquet>
ok thank for your reply
pmattern has quit [Quit: Genug für heute.]
<philippe_fouquet>
I would try this source because for the moment I use source where I din't have all source (for exemple nand driver) and I cloudn't use sunxi team source (for the hw video driver)
<Skylark>
alexvf: i've successfully integrated this 3.4 branch with the A20 SDK android, except for cedar which for some reason stops after a few seconds
<Skylark>
heh
<Skylark>
so if you don't need that, it's actually perfectly workable
philippe_fouquet has quit [Remote host closed the connection]
<rellla>
iirc there is no display driver in allwinners 3.4 repo. there is no git history at all. i doubt, that anybody diff'ed that tree against our 3.4 tree. so probably nobody knows, what the difference is.
<codekipper>
I'm waiting for H3 code to become available to start adding to the head of linux-sunxi. Did some work with the A31 but f/&ked my board up before being able to test
Black_Horseman has quit [Remote host closed the connection]
<codekipper>
philippe_fouquet: but follow rellla's advice...use linux-sunxi if mainline isn't suitable...there are a lots of clangers in the allwinner code.
<codekipper>
mripard: ping
<mripard>
codekipper: pong
<codekipper>
I saw that firnsy had an issue with the IR TX pin yesterday
<mripard>
ok
<codekipper>
should we remove this from pin-mux as there is no code to support it.
<codekipper>
A10 has it removed on IR0
<codekipper>
saves having to build transmit/receive capabilitiies into the driver
<mripard>
what was the issue?
<codekipper>
I still have this MK808C patch outstanding...
<codekipper>
the issue was that he had both pins used as IR was enabled and the TX pin was assigned on the Banana pi for powering the SATA
<codekipper>
concerning MK808 just need a vendor name..or suggestion
Andy-D__ has quit [Ping timeout: 250 seconds]
philippe_fouquet has quit [Remote host closed the connection]
philippe_fouquet has joined #linux-sunxi
<mripard>
codekipper: one thing at a time please
<mripard>
so, that IR stuff
<mripard>
what is this issue you were telling me about, what do you want to do?
<codekipper>
OK....IR first
<codekipper>
Just remove the pin-settings for IR TX on sun4i-a10.dtsi and sun7i-a20.dtsi
philippe_fouquet has quit [Ping timeout: 264 seconds]
<mripard>
why?
<mripard>
because the bananapi alone has it wired differently ?
<mripard>
that seems backward, what about the other boardS?
<codekipper>
A1000 uses that pin for an external audio amp
<codekipper>
I've not looked at any others
<mripard>
then they should have a different muxing for it, or not handle IR at all?
<codekipper>
IR0_TX isn't declared in sun4i-a10.dtsi only in sun5i-a20.dtsi
<codekipper>
well that's a little bit more work and I've got my lazy hat on
<mripard>
well, the nodes contain two pins on both DTSI
<codekipper>
I saw in sun7i_a20 that both pins were declared for both IR0 and IR1
<mripard>
yeah
<mripard>
just like in the A10 DTSI
<codekipper>
in sun4_a10.dtsi on RX is declared on IR0, both on IR1
<codekipper>
only
paulk-collins has quit [Read error: Connection reset by peer]
<alexvf>
leviathancn: did you have time to ask allwinner about dram host ports documentation?
<leviathancn>
alexvf: the guy hasn't answered yet who is responsible for the graphics part
<leviathancn>
a friend there was looking and saying he was very busy with another project right now and hadn't seen my email
<leviathancn>
he tries to look into it tomorrow
<leviathancn>
but we both don't understand the question fully frankly
<alexvf>
leviathancn: ok, thank you, let me know if you get some answer please :)
<codekipper>
there must be something in the water!....now I remember I made the dif ontop of my audio patches and then pushed to my upstream branch
<codekipper>
the things is TX is being enabled.....and it's never being used.
<leviathancn>
alexvf: can you maybe describe in an email to me (david.lanzendoerfer@o2s.ch)
<leviathancn>
alexvf: exactly what you are trying to do and what exact informations you need to have?
<leviathancn>
alexvf: it would simplify things a lot, because we communicate with emails in simplified Chinese, so having an email I can just translate would help a lot
<alexvf>
sure, i will email you later today
<leviathancn>
:)
<leviathancn>
alexvf: thanks!
<alexvf>
leviathancn: :) you are welcome
<mripard>
codekipper: if it's not used, what's the issue?
<mripard>
it might be a bit under-optimal, but that's bikeshedding at this point
pmattern has joined #linux-sunxi
<codekipper>
the issue if that we've identified more than one board that uses ir rx and pin allocated for tx is assigned for something else.
<codekipper>
s/if/is....I need to buy some glasses
pmattern has quit [Client Quit]
<mripard>
ah, so you mean it's using PB4 as a GPIO, and PB3 as IR0 ?
<codekipper>
yep...
<mripard>
why didn't you say so....
<mripard>
ok, then split it in half, take the one or two pins depending on the board in the pinctrl node, create a patch with a nice commit log explaining all this, send the patch
<codekipper>
OK......will do. Can we chat about the MK808c patch now :)
<mripard>
what about it?
<codekipper>
I was missing the vendor name and I couldn't find one
<codekipper>
you pinged Arnd and he sort of suggested using upstreamer organisation.
<codekipper>
I'm not bother...so whatever you say..I'll wack into the patch and resubmit
<mripard>
which we don't have
<mripard>
so it's not an option
<mripard>
he was kind of agreeing with having allwinner as the vendor, so let's use that
<libv>
but i am getting quite worked up about our two whatever pi producers
<libv>
they are rather sizable fraudes
<a1d3s>
with script.bin from fex the 7" works
<libv>
i should get back to working device pages and properly marking these shallow rip-offs
<ssvb>
a1d3s: the point is that I have a simple web spider script, which can automatically retrieve the information from the linux-sunxi wiki and crosslink the fex, dtb, board names and other information
<ssvb>
thanks to libv, the wiki pages can be very easily automatically parsed because they all are presented in a standardized way
<NiteHawk>
libv: i hope "pi producers" doesn't relate to the wiki contributors, but the manufacturers instead ;)
<libv>
NiteHawk: of course
<libv>
ssvb: :)
<ssvb>
a1d3s, libv: however we need to decide how to list many fex files for a single board in a consistent way (banana, banana + 7" lcd, banana + 10" lcd, etc.)
<libv>
ssvb: perhaps list the changes in a special subsection in the wiki?
<libv>
lcd changes like 25-30 lines
<libv>
which is borderline, but still managable
<a1d3s>
lcd_bl_en = port:PH08<1><0><default><1>
<a1d3s>
lcd_power_used = 1
<a1d3s>
lcd_power = port:PH12<1><0><default><1> these the things i need?!
<ssvb>
yes
<libv>
also, there are several banana pi versions
<libv>
and the lcd changes might be somewhat generic
<ssvb>
libv: maybe instead of "The .fex file can be found in sunxi-boards as <fex link>" we can have statements like "The .fex file for the <xyz> variant can be found in sunxi-boards as <fex link>"
<libv>
ssvb: i hate to have 25 fex files for 5 fex files + different lcd sections
<libv>
imho the olimex fex files should be done differently
<ssvb>
the vendors still provide separate fex files for different configurations
<libv>
but whatever offers a quick solution for now
<ssvb>
of course, you can decompose them and extract the lcd sections to list them in the wiki, but this needs either some manual labour or the use of another script
<ssvb>
and it would be difficult to do this in a consistent way without doing any mistakes in the process
<ssvb>
so if the vendor provides let's say 5 fex files for the board, then we can just add them all to the sunxi-boards repository and add links to the wiki with some annotations to describe the differences
<a1d3s>
banana m1 / pro and m2 are all the same way on con2
premoboss has quit [Ping timeout: 255 seconds]
<ssvb>
сщт2,
<ssvb>
con2?
domidumont has quit [Read error: Connection reset by peer]
<a1d3s>
lemaker named the port for lvcs so
<a1d3s>
lvds
JohnDoe_71Rus has joined #linux-sunxi
<libv>
ssvb: whatever offers a quick solution, we could always go change it later :)
<ktemkin>
ssvb: I can't quite recall-- were you the one looking for touchscreen support with the mainline kernel on the Primo 73?
<ssvb>
ktemkin: no, it was not me
<a1d3s>
ssvb touchscreen still black
<ssvb>
ktemkin: but I remember somebody contacting me via e-mail and asking to test something touchscreen related
<ssvb>
ktemkin: unfortunately I did not have time for that
<ssvb>
a1d3s: but it does work properly in linux, right?
<a1d3s>
i have only debian 8 with kernel 4.0 mainline and console
<a1d3s>
i didnt install any xserver
<ktemkin>
Ah, okay. Pity my leaky brain can't remember who it was.
<a1d3s>
i thought there is no need of an full x to display something, hdmi works
<ssvb>
a1d3s: I mean, does it work with the 3.4 kernel? just to make sure that it is connected correctly
<a1d3s>
yes this works
<ssvb>
good
<ssvb>
try to disconnect hdmi
<a1d3s>
i did , then i see i fast flash
<a1d3s>
but no blinking cursor
<ssvb>
do you have a serial console?
<ssvb>
if yes, then try to interrupt u-boot
<a1d3s>
i have one but dont know where ;) i was 10 weeks on travel and my wife cleared my place ;)
<ssvb>
it is possible that the lcd is working correctly, but the kernel just boots too fast and you don't have time to see the u-boot output on screen
<a1d3s>
u-boot says it is dirty
<a1d3s>
after adding the line for pin definition
<ssvb>
can you make the kernel non-bootable then by modifying boot.scr?
<a1d3s>
yes
<ssvb>
to check if you can get to the u-boot command prompt
<ssvb>
a1d3s: if you still don't get correct output, one of the things to try is a change of "depth:24" to "depth:18"
<a1d3s>
i got only tftp boot now
<a1d3s>
ok i change this and test again
Night-Shade has quit [Read error: Network is unreachable]
Night-Shade has joined #linux-sunxi
<a1d3s>
ssvb same again , but is 55000hz the real clock or 22000?
<ssvb>
a1d3s: how do you see tftp boot?
<a1d3s>
on hdmi
<a1d3s>
i test with an without hdmi
<a1d3s>
i must bring my daughter to bed now ;)
<ssvb>
hdmi has higher priority than lcd, so lcd will never work with hdmi connected
massi_ has quit [Remote host closed the connection]
<a1d3s>
lcd_dclk_freq = 55 or lcd_pwm_freq = 22000 ?
<ssvb>
ok, let me check something
sdschulze has quit [Ping timeout: 240 seconds]
MackBoy has joined #linux-sunxi
<a1d3s>
ssvb what about the ctp_para in fex file? there are some more ports configured
<ssvb>
a1d3s: maybe that's a touch panel
<ssvb>
a1d3s: in any case, can you try adding CONFIG_VIDEO_LCD_BL_PWM_ACTIVE_LOW=n line to the defconfig?
robertz has quit [Ping timeout: 252 seconds]
CMDann has joined #linux-sunxi
khuey|away is now known as khuey
<CMDann>
Hey guys, I am having trouble with the Banana Pro running Bananian Linux, the WiFi driver (ap6210) is only allowing 5 clients when connected to the pro as an AP.
<ssvb>
a1d3s: does the linux kernel work with your lcd correctly now?
<a1d3s>
no
<a1d3s>
i found the driver info in fex file
<Omega_>
This is a bit of a long shot, but maybe someone here might know something about it... I'm using the AP6210 driver for wifi and am encountering a hard limit of 5 clients when running in access point mode.
<Omega_>
Would anyone happen to know if this is a limitation of the hardware, or perhaps some kind of inadvertent bug/setting?
<a1d3s>
omega sorry i used the banana only as client
<a1d3s>
omega_ but toni posted somewhere a package of drivers in lemaker forum, maybe one of the other .bin file would do a better job
<a1d3s>
ssvb funny thing the display stop kernel booting
<a1d3s>
i got no network if lvds TS is in use , if i boot with hdmi network works too
<Omega_>
a1d3s: Is there any way these alternative drivers could make it into more standard channels?
<Omega_>
a1d3s: (as opposed to forum posts)
<a1d3s>
ssvb forget all what i write lol. i used otg port for power and this went all down . switched power supply and usb kabel on right power port and i got linux booting with login screen
<a1d3s>
omega_ must take a look for drivers because i was 10 weeks away
<a1d3s>
a this can happen i think because raspian use 3.4 kernel
<a1d3s>
or bananian
<a1d3s>
i play around with 4.0 kernel on debian jessie
<Omega_>
hah, lucky :)
WarheadsSE has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
<CMDann>
In regards to the AP6210 driver, I've been in communication with LeMaker and they have identified it an issue with the driver, not the hardware (which is good). I saw your post the other day a1d3s, seems we are both having issues. It would be awesome if Bananian was using the 4.0 kernel haha
<Omega_>
Well, it's brand new - so let's not be hasty ;)
<Omega_>
So long as we can get a backport or working drivers, we'll all be happy :)
<a1d3s>
CMDann, if the basics are working i will upload my image :)
<CMDann>
Thanks, can you shoot a link to me to where you upload it?
<a1d3s>
yea if its up :)
FR^2 has quit [Quit: Connection reset by peer]
<TheLinuxBug>
Hey if you guys need a place to store some images I have a little CDN I setup for linux-sunxi related files and I would be happy to host it for you
<TheLinuxBug>
just provide a link to the file and preferably a md5 for it
<TheLinuxBug>
and I will be happy to put it up
RSpliet has quit [Read error: Connection reset by peer]
Netlynx has quit [Quit: Leaving]
RSpliet has joined #linux-sunxi
<a1d3s>
CMDann, ok will give you the link later , now i go to bed
<a1d3s>
TheLinuxBug, thanks
<CMDann>
Sounds good, take care
<juser123>
ssvb: Thanks for the help a few days back. i think i understand the video/gpu/cedar stuff a bit better now.
<juser123>
TheLinuxBug: hey, you still getting the X11 session going away with the erors you were showing?
<juser123>
ssvb: so if you were to recommend the lesser evil of blobs whould you go with r3p0 or something different? I was thinking that r3p2 was “better” as both GPU cores are recognized. but given I’m not using a compositing desktop window manager, it makes little sense to try to push it. the video playback is far more important for my use. thanks
<TheLinuxBug>
juser123: tbh I have been super busy this week and the BPi has just been sitting on the corner of my desk being ignored, lol. Hopefully will get some more time here to mess with it. Thanks for asking.
<juser123>
TheLinuxBug: no problem at all. was just curious and saw you on. :)
<juser123>
TheLinuxBug: btw, I’ve totally dorked up the load on my BPi trying to get the “latest” mali stuff running. holy crap that was stupid. almost have it working again, sort of.. silly me… now I’m in the same boat you were of not wanting to re-image :)
<TheLinuxBug>
:D
<TheLinuxBug>
ahaha
<TheLinuxBug>
yeah
<TheLinuxBug>
what I eneded up doing is using my one cubbie and a 320gb 2.5" drive and I use it just to make backup images of my sdcard for the BPi etc so that I can nuke em and if I get mad enough reload the last image
<TheLinuxBug>
hehe
<TheLinuxBug>
more efficient than trying to use Windows to do it
kaspter has quit [Ping timeout: 250 seconds]
<juser123>
thought I figured out something, but just learned the hard way that the drivers everyone is using are for a darn good reason. still trying to get back to fully functioning on the old drivers, which is where i started
<juser123>
indeed. i sorta do the same thing with my mac. dd it to an image file for safe keeping. which I’d done that about 3 kernel rebuilds back…
<TheLinuxBug>
:)
physis has quit [Remote host closed the connection]
arokux1 has quit [Remote host closed the connection]
premoboss has joined #linux-sunxi
sdschulze has joined #linux-sunxi
nove has quit [Quit: nove]
CMDann has quit [Ping timeout: 244 seconds]
CMDann has joined #linux-sunxi
domidumont has quit [Ping timeout: 246 seconds]
iamfrankenstein has quit [Quit: iamfrankenstein]
bonbons has quit [Quit: Leaving]
iamfrankenstein has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
CMDann has quit [Ping timeout: 244 seconds]
fredy has quit [Excess Flood]
Nyuutwo has quit [Ping timeout: 272 seconds]
awe00 has joined #linux-sunxi
fredy has joined #linux-sunxi
pmattern has quit [Read error: Connection reset by peer]