rellla 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 - *only registered users can talk*
specing_ has joined #linux-sunxi
specing has quit [Ping timeout: 260 seconds]
specing_ is now known as specing
tuxd3v has quit [Remote host closed the connection]
Mangy_Dog has quit [Ping timeout: 268 seconds]
jstefanop has quit [Remote host closed the connection]
jstefanop has joined #linux-sunxi
jstefanop has quit [Ping timeout: 252 seconds]
apritzel has quit [Ping timeout: 240 seconds]
azend has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
tuxd3v has joined #linux-sunxi
kaspter has quit [Remote host closed the connection]
kaspter has joined #linux-sunxi
jstefanop has joined #linux-sunxi
jstefanop has quit [Ping timeout: 252 seconds]
vagrantc has quit [Quit: leaving]
sunshavi has quit [Ping timeout: 265 seconds]
sunshavi has joined #linux-sunxi
buzzmarshall has quit [Remote host closed the connection]
kaspter has quit [Ping timeout: 246 seconds]
kaspter has joined #linux-sunxi
apritzel has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
Daaanct12 has joined #linux-sunxi
Daanct12 has quit [Ping timeout: 252 seconds]
Daaanct12 has quit [Quit: Quitting - Huong Tram IRC Client 1.54]
Daaanct12 has joined #linux-sunxi
Daaanct12 is now known as Danct12
kaspter has quit [Ping timeout: 252 seconds]
kaspter has joined #linux-sunxi
cmeerw has joined #linux-sunxi
chewitt has quit [Quit: Zzz..]
chewitt has joined #linux-sunxi
hlauer has joined #linux-sunxi
netlynx has joined #linux-sunxi
netlynx has joined #linux-sunxi
choozy has joined #linux-sunxi
Mangy_Dog has joined #linux-sunxi
Mangy_Dog has joined #linux-sunxi
specing_ has joined #linux-sunxi
specing has quit [Ping timeout: 252 seconds]
specing_ is now known as specing
\\Mr_C\\ has quit [Quit: Ping timeout: 999,999,999 years]
jstefanop has joined #linux-sunxi
jstefanop has quit [Ping timeout: 240 seconds]
jstein has joined #linux-sunxi
AF04FB9290474265 has quit [Quit: Leaving]
Daanct12 has joined #linux-sunxi
Danct12 has quit [Ping timeout: 240 seconds]
ats_ is now known as ats
<apritzel> Hi jernej, trying to bringup my Eachlink H6 Mini, and I need video=HDMI-A-1:e on the command line to make it work
<apritzel> could that be that the ddc_en_gpio is wrong?
<apritzel> ddc-en-gpios, really
<apritzel> but that seems to be PH2 everywhere
<jernej> no, if ddc-en-gpios is wrong or missing, you would still get image, just with non-native resolution
<jernej> default is currently set to 1024x768
<jernej> monitor detection is done via HDMI HPD pin
<jernej> do you have Android DT for it?
<apritzel> yes, but it's totally confusing
<apritzel> I scanned for all GPIOs (phandle references)
<apritzel> some GPIOs are mentioned multiple times
<jernej> it's possible they put some kind of circuit between HPD pin on connector and SoC, which need some regulator to be turned on
<jernej> can you check which regulators are turned on in Android?
<jernej> maybe there is some interdependency
<apritzel> well, it's one of those non-AXP boxes
<jernej> ah, that complicates things
<apritzel> so they mention some regulators (ldo<x>), but it's not clear whether those are dummies
<apritzel> so far most things work without GPIO regulators
<jernej> I'm also confused with their regulator setup approach
<jernej> some of those entries don't make sense
<jernej> can you upload Android DT?
<apritzel> yeah, will do in a minute - just mapping the remote keys ;-)
<apritzel> they have a very weird understanding of the DT indeed
<apritzel> and a lot of times this seems to be generic, so including stuff that isn't there, and sometimes even contradictory
<apritzel> jernej: https://pastebin.com/c0Jkyakf ... it's loooong, and it's the decompiled version, so brace yourself
<jernej> if they would even bother to write DT documentation...
<apritzel> the three properties with hpd in their name didn't help
<apritzel> what do you mean with "documentation"? ;-)
<jernej> boot_disp is for U-Boot IIRC
<jernej> so it might be interesting to take a look there
<jernej> I can't find anything special...
<jernej> apritzel: a bit long shot, but you can dump direction and output registers of all GPIOs and compare them to mainline
cnxsoft has quit [Quit: cnxsoft]
<apritzel> yes, was my plan
<apritzel> I think there is debugfs, even
Daanct12 has quit [Quit: Quitting - Huong Tram IRC Client 1.54]
<jernej> apritzel: r_i2c is also enabled, maybe you can scan it
<apritzel> oh, good hint, thanks
<apritzel> you once said that the U-Boot I2C tool is not very reliable?
<jernej> I said that, but maybe I did some mistake, like forgot to enable pull ups for I2C
<jernej> you can always try
ats has quit [Ping timeout: 260 seconds]
victhor has joined #linux-sunxi
elros1 has joined #linux-sunxi
atsampson has joined #linux-sunxi
hlauer has quit [Read error: Connection reset by peer]
<apritzel> so PL2, PL3, PL4 and PL5 are "(?) out lo/hi" in debugfs/gpio
<apritzel> grml, GPIO sysfs depends on EXPERT=y and is =n by default, so need to recompile
jstefanop has joined #linux-sunxi
elros1 has quit [Remote host closed the connection]
kaspter has quit [Ping timeout: 240 seconds]
kaspter has joined #linux-sunxi
<tuxd3v> for what I see nanopi neo Air doesn't have spi flash right?
<tuxd3v> but it does have emmc flash, and it boots from it..
<tuxd3v> I have mine assembled already with heatsink, so I can't check the board..
<tuxd3v> :(
<jernej> apritzel: any luck?
<apritzel> jernej: seems like I hit the first regression of 3.13 :-(
<apritzel> 5.13, I mean
<jernej> what's that?
<apritzel> no video at all, going back to 5.12 now
<jernej> I used linux-next about a week ago and it worked fine
buzzmarshall has joined #linux-sunxi
arnidg has quit [Ping timeout: 258 seconds]
reinforce has joined #linux-sunxi
jstefanop has quit [Remote host closed the connection]
jstefanop has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
kaspter has quit [Ping timeout: 246 seconds]
camus has joined #linux-sunxi
camus is now known as kaspter
jstein has quit [Quit: quit]
atsampson has quit [Ping timeout: 250 seconds]
atsampson has joined #linux-sunxi
kaspter has quit [Remote host closed the connection]
kaspter has joined #linux-sunxi
<apritzel> jernej: it doesn't look like it's any of those four GPIOs
<jernej> too bad
<jernej> PL5 is mentioned as usbc1 drv vbus gpio
<jernej> PL2 is referenced as recovery key, but that would make it input, right?
<jernej> apritzel: Android DT has boot_init_gpio node
<apritzel> PL2 is indeed configured as input
<apritzel> and I configured PL[3,4,5,7] as the BSP did
<jernej> it may be important for something, it references PL7, PL4 and PH2
jstein has joined #linux-sunxi
<apritzel> both within mainline, but also in U-Boot
<apritzel> PH2 is the ddc-enable-gpio
<apritzel> so that's covered by the DT property
<jernej> PL4 is for something called sprite_gpio0, I don't know what that is
<apritzel> yeah, was wondering about this magic fairy bit as well ;-)
<jernej> probably something in U-Boot
<apritzel> jernej: is there a way to enable a monitor from userland?
<jernej> PL6 is speaker amp, it seems
<apritzel> there is /sys/class/drm/cardx-y/enable, but that's read-only
<jernej> I don't know, tbh
<jernej> there is concept of force connection status in framework
<jernej> but I don't know if this mechanism is available through sysfs
<jernej> apritzel: can you try PM3 - it's regulator for wifi, but you never know
netlynx has quit [Quit: Ex-Chat]
<jernej> there are few other PM pins if you feel like testing...
<apritzel> yeah, I saw them, but they seem to be for BT and WiFi
<apritzel> that BSP kernel doesn't support /dev/mem :-(
<jernej> well, there is still a way to read physical memory :)
<jernej> I wrote a simple script, let me find it
azend has quit [Ping timeout: 252 seconds]
<jernej> apritzel: use something like this: http://ix.io/3mdl
<apritzel> I see, /sys/class/sunxi_dump ... what are those guys smoking?
<jernej> why? wouldn't you say it's useful for development purposes? :D
Throwawayname has joined #linux-sunxi
Throwawayname has quit [Max SendQ exceeded]
willmore has quit [Remote host closed the connection]
JohnDoe_71Rus has quit [Quit: KVIrc KVIrc Aria 5.0.1, revision: 5.0.1+git-7433-0df9f22f2, build type: debug, sources date: 20160102, built on: 2019-12-08 19:19:20 UTC 5.0.1+git-7433-0df9f22f2 http://www.kvirc.net/]
<apritzel> jernej: are the AC200 runes the same for the H6? I mean the actual I2C registers? That's still the same IP as the acutal chip in there, right?
<jernej> yes
<jernej> I'm not sure about efuses
<apritzel> I am just interested in the Ethernet for now ...
linkmauve has quit [Ping timeout: 246 seconds]
<apritzel> do you know which PWM and I2C it is? I tried guessing, but U-Boot said: no
<jernej> i2c3
<apritzel> and the S_PWM, by any chance?
<jernej> take a look in above branch
<jernej> it has everything, except properly formed DT
<apritzel> where is the fun in that :-P
<jernej> it seems to be normal pwm
<apritzel> yeah, that's what I tried, together with I2C3
<apritzel> do I need to configure PB19?
<apritzel> I was just curious because the BSP configures PL8 to S_PWM0
<jernej> yes, PB19 should be swithced to pwm functionality
<apritzel> Ah, PB19 is one of those internal pins, right?
<jernej> yes
<jernej> and note that you have to set global ac200 registers first, before ephy
<jernej> well, irq pin is probably not important
<jernej> s/pin/register/
<apritzel> it's just a U-Boot hack for now (tired of swapping this USB key), and for this part I rely on your H616 T95 hack ;-
<jernej> what are you trying to achieve?
willmore has joined #linux-sunxi
<apritzel> have Ethernet with the ePHY in U-Boot. I updated your T95 U-Boot hack and it worked on my X96 Mate box, but now I need it on the H6 box as well
<apritzel> and thanks for the sunxi-dump hint, by the way, the actual pin config is slightly different than expected
<apritzel> for instance R_I2C is gpio out, and PL6 is also "out hi"
<jernej> interesting, PL6 is not referenced in DT
<jernej> by R_I2C you mean PL0 and PL1, right?
<apritzel> yes
tuxd3v has quit [Remote host closed the connection]
azend has joined #linux-sunxi
azend has quit [Ping timeout: 240 seconds]
cmeerw has quit [Ping timeout: 260 seconds]
tuxd3v has joined #linux-sunxi
<tuxd3v> I am trying to boot my NanoPi NEO Air, from sdcard, but mainline uboot doesn't load, intead it loads emmc one :(
<tuxd3v> the one in the emmc is ancient
<tuxd3v> I am trying uboot v2021.01
<tuxd3v> unless H3 cpu uses another boot schemme than traditional allwinner socs
<apritzel> tuxd3v: H3 is very normal in this respect
<apritzel> do you see the new SPL?
<tuxd3v> no
<tuxd3v> I see the one in the emmc flash :(
<apritzel> so it ignore your SD card at all?
<tuxd3v> the allwiner one
<tuxd3v> it seems that is ignoring the sdcard, but I tried armbian and armbian bootloader boots up, I see the SPl 2020.10-armbian
<tuxd3v> I simply can't use the mainline, I don't know why in neo Air
<apritzel> that's odd, that part is the least fragile
<apritzel> at least you should see the SPL banner
<apritzel> dd'ed to bs=1k seek=8?
<tuxd3v> yeah, I should :/
<apritzel> wow, wait!
<apritzel> do you have CONFIG_SPL_SPI_SUNXI defined?
<apritzel> ah, no, scratch that
<tuxd3v> let me check :)
<apritzel> that's the SPI, which you don't have
<apritzel> (I thought for a moment it was missing the SPL symbol)
<tuxd3v> I believe I don't have spi, I don't know...I just assembled it, and now the heatsink is big and doesn't let me see below
<tuxd3v> the wrote the bootloader to the emmc
<tuxd3v> like if it was a sdcard
<tuxd3v> but I do see 2 disks of 4MB each :/
<tuxd3v> 2 spi flash devices?
<apritzel> where do you see them? in Linux?
<tuxd3v> yeah
<apritzel> 4MB smells like the two eMMC boot partitions
<apritzel> that's not the droi^Wpartitions you are looking for
<tuxd3v> but were is the hardware for them? are they in the emmc flash chip?
<apritzel> at least not with that config
<tuxd3v> the emmc flash chip is of 8GB
<apritzel> yes, they are separate storage inside the eMMC chip
<tuxd3v> dd if=neoAir_u-boot-sunxi-with-spl.bin of=/dev/sdb bs=512 seek=16 conv=notrunc,fdatasync
<apritzel> looks alright
<tuxd3v> ho, I didn't knew of that.. I just went crazy, I didn't saw the hardware :D
<tuxd3v> Now its explained
<tuxd3v> I couldn't even right data to them..
<apritzel> the boot partitions are really nice, but they are a bit special
<apritzel> for a start: they are write protected at reset, you have to un-protect them first
<tuxd3v> ho...
<apritzel> then you have to write u-boot-sunxi-with-spl.bin to sector 0
<tuxd3v> that explains a lot
<apritzel> and then you have to define a special config symbol in U-Boot
<apritzel> especially the "Installing mainline U-Boot in a boot partition"
<apritzel> but I recommend you get the SD card working first ;-)
<tuxd3v> apritzel, many many taanks, I am fighting it since yesterday, went crazy in full drive with 2 hardware devices of 4MB that doesn't exist in the board :D
<tuxd3v> apritzel, well that is the problem I tried uboot 2021.01 but it doesn't work :/
<tuxd3v> in the sdcard
<apritzel> why not 2021.04?
<tuxd3v> well, you right, I forgot to update my builder :)
<tuxd3v> thanks
<apritzel> just trying on my OrangePi Zero as we speak
<tuxd3v> your Orange zero also has emmc?
<apritzel> no, but SD card ;-)
<tuxd3v> I am going to test uboot 2021.04 :)
<apritzel> both mainline and 2021.04 work for me on the OPi Zero
<apritzel> and as I said: no output at all from SD card is rare
<tuxd3v> yeah with 2020.01 it is not working with neoAir
<tuxd3v> BUT 2021.04 is :)
<tuxd3v> now I don't know why he doesn't pickup my boot.scr file :/
<tuxd3v> Loading Environment from FAT... Unable to use mmc 0:1... In: serial
<tuxd3v> from fat?I am using ext2..
<apritzel> is that boot.scr on the first FAT partition?
<tuxd3v> I am using ext2
<apritzel> environment must be on FAT
<tuxd3v> not a Fat one
<tuxd3v> maybe emmc has a FAT partition
<apritzel> but that's the environment, not the boot script
<apritzel> mmc0 is current device
<apritzel> ** Unrecognized filesystem type **
<apritzel> starting USB...
<tuxd3v> why he says unrecognized filesystem :/
<tuxd3v> ext2 works fine
<apritzel> what does "mmc dev 0; mmc part" say?
<tuxd3v> I mean have being working fine
<apritzel> but U-Boot looks on the SD card, is there a usable partition there?
<tuxd3v> yeah 2 partitions 1 for /boot and other ext4 for rootfs
<tuxd3v> this is the output
jstein has quit [Quit: quit]
<apritzel> does "ext4ls mmc 0:1" work and show boot.scr?
<tuxd3v> ** Unrecognized filesystem type **
<tuxd3v> i tried 'ext2ls' and 'ext4ls' both fail
<tuxd3v> But if I do 'ext4ls 0:2' it shows rootfs content
<tuxd3v> so its ext2 that doesn't work
<apritzel> what does Linux say about this partition?
<apritzel> maybe you did accidentally overwrite the beginning of it with your "dd"?
<apritzel> which is a cool feature of those boot partitions, btw: they are just for U-Boot, you cannot overwrite other data and U-Boot doesn't need to share
sunshavi has quit [Ping timeout: 265 seconds]
<tuxd3v> maybe I screw the partition yes
<tuxd3v> I run a fsck on it and it presented a lot of problems in inods..
<tuxd3v> inodes
<tuxd3v> apritzel, and then how to you upgrade those boot partitions?
<tuxd3v> I mean if you want too
<tuxd3v> its ninja stuff
<tuxd3v> :D
<tuxd3v> Yeah the Idea of boot partitions is good, in that way you don't need SPI flash devices :)
<apritzel> yes, and basically every eMMC has them
<tuxd3v> why they were not used before and everybody was going with SPI flash in their designs?
<apritzel> and upgrading is really easy: just unlock them, then dd to sector 0
<tuxd3v> I believe those sectors are 4096kB right?
<apritzel> SPI flash has other advantages (better for environment / variable storage, for instance)
<tuxd3v> to make 4MB each
<apritzel> 2 * 4MB is a typical value for those smaller eMMCs
<apritzel> but the 64GB eMMC in the X96 Mate has 32MB partitions, for instance
<tuxd3v> I need to check the emmc better now I will be focused in get again my /boot working, but very nice of you, to tell me about uboot 2021.04
<tuxd3v> it does work :)
<apritzel> also eMMC is quite expensive, compared to the price of a cheap board
<tuxd3v> indeed, nanopi neo Air costs a lot of money for what it is, to be honest..
<tuxd3v> I can do the same with a bananapi m2 zero
<tuxd3v> for almost 20$
<tuxd3v> while nanopi neo Air costs a lot more
<tuxd3v> but it has the flash storage, and ubunto core come preloaded..although its a very old release
<apritzel> Pine64 has those nice sockets on the board, so you can add an eMMC module, if you like
<tuxd3v> but I didn't knew about those 2 partitions..
<tuxd3v> never come up to me :)
<tuxd3v> one maybe is a backup of the other or something else..
<apritzel> yes, that's the idea
<apritzel> you can select which one should boot
choozy has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]