narmstrong changed the topic of #linux-amlogic to: Amlogic mainline kernel development discussion - our wiki http://linux-meson.com/ - ml linux-amlogic@lists.infradead.org - Publicly Logged on https://irclog.whitequark.org/linux-amlogic
sputnik_ has quit [Remote host closed the connection]
sputnik_ has joined #linux-amlogic
<gavlee> ok, i've got the air on a different port now can ssh in... in the LE autostart.sh i put a sleep 45s then the ethtool command posted... when reboot I see dhcp release then lots of discover/offer until the sleep times out and then I get an address
<gavlee> i put docker and other tools on there before i moved the box into other room so at least I can do some stuff... I really appreaciate all your help with this
buzzmarshall has quit [Remote host closed the connection]
<gavlee> thanks
<chewitt> gavlee: so it's working with the change of switch port/cable?
<chewitt> no autostart.sh hacks?
<chewitt> the BT firmware is embedded in the latest images, but looks like still an issue to track down
<gavlee> chewitt: only after the ethtool hack I think
<gavlee> one thing puzzle me i don't get an address in textmode?
<gavlee> does that mean single user or the like
<chewitt> nope, the only difference is we don't start the kodi.target in systemd and we route console to screen
<chewitt> hardware wise it's all the same
<gavlee> so should get an address even in textmode is enabled?
<chewitt> yes
<gavlee> hm
<chewitt> what does "connmanctl services" show?
<gavlee> funky
gavlee has quit []
gavlee has joined #linux-amlogic
<gavlee> *AR Wired ethernet_020000132801_cable
<gavlee> some issue here went to check chat log but says the ssl cert expired on whitequark.org
<gavlee> my irc client glitch..
<gavlee> i'm currently without textmode because never seemed to get an address when I booted with that, takes some cpu and headless while in that so left it
<gavlee> just glad it's running and can poke around really :D
<chewitt> can you test https://www.dropbox.com/s/y07eldrtr823qjc/meson-sm1-x96-air.dtb?dl=1 without any autostart.sh hacks
<gavlee> Sure but it's 5am here and if it doesn't boot I will be crashing around lol. can do it tomorrow
<gavlee> i mean boots but doesn't get network
<gavlee> can't wait to test it now. nap time. cheers
Barada has joined #linux-amlogic
Barada has quit [Quit: Barada]
<chewitt> sure, take your time
Barada has joined #linux-amlogic
ldevulder_ is now known as ldevulder
Elpaulo has quit [Read error: Connection reset by peer]
Elpaulo has joined #linux-amlogic
Barada has quit [Quit: Barada]
sputnik_ has quit [Ping timeout: 272 seconds]
<chewitt> jbrunet: I'm still trying to resolve this on VIM3 after the routing change you pushed
<chewitt> [ 40.104819] fe.dai-link-0: ASoC: no backend DAIs enabled for fe.dai-link-0
<chewitt> this is what I had been guessing for mixer settings https://pastebin.com/raw/2imeJYDh
<chewitt> can you help please
random_yanek has quit [Quit: random_yanek]
random_yanek has joined #linux-amlogic
gavlee has quit []
gavlee has joined #linux-amlogic
Barada has joined #linux-amlogic
Barada has quit [Client Quit]
TheAssassin has quit [Remote host closed the connection]
TheAssassin has joined #linux-amlogic
<gavlee> tried the last dtb posted connected to the gigabit port without autostart.sh hacks. no address. latest dmesg http://dpaste.com/1WYFSP1 moved it back on other net for now but don't see why that stop getting address, other that it not reaching gigabit speeds on this segment
ldevulder has quit [Quit: Leaving]
<gavlee> it may do but connecting it directly to that port I know for sure
vagrantc has joined #linux-amlogic
<chewitt> ^ this attempts to set 100-BaseT so it at least works without hacks
<chewitt> as this contains a patch to set the SoC ID correctly, and if it works I can upstream it
<gavlee> chewitt: sure. downloading. btw can i drop these in /storage/.update or do I have to dd them?
<chewitt> dtb needs to be placed in the right folder on the SD card
<chewitt> the .img.gz (which has the same dtb) can go in /storage/.update
<gavlee> ah ok that's great, thanks
drieschel has joined #linux-amlogic
drieschel has quit [Quit: drieschel]
drieschel has joined #linux-amlogic
<gavlee> chewitt: http://ix.io/2ocs boots fine, straight into network and ssh without delay, no autostart.sh hacks :D
<chewitt> [ 0.152081] soc soc0: Amlogic Meson SM1 (S905X3) Revision 2b:c (10:2) Detected
<chewitt> ^ and that worked too
<chewitt> it's a hack though, the dts sets "max-speed = <100>;"
<gavlee> well it wfm and it makes things so much easier rather than swapping my battered sd card reader !!!
<xdarklight> chewitt: if your CPU is idle once again I'm interested in the output of this additional debug line: https://pastebin.com/6iHGeiui
<xdarklight> background: I found this in the 4.9 kvim3l dts: https://github.com/khadas/linux/blob/khadas-vims-4.9.y/arch/arm64/boot/dts/amlogic/kvim3l_linux.dts#L1665 - cali_val is punched directly into the PRG_ETH1 register
<xdarklight> also kvim3l inverts the (RGMII) RX_CLK signal, which is also not supported on mainline yet
<chewitt> rebuilding..
<chewitt> put it in /storage/.update and reboot
<chewitt> and then dmesg
<chewitt> I left the max-speed change in, so it should work
<chewitt> xdarklight: I also noticed https://patchwork.kernel.org/patch/10468005/
<xdarklight> chewitt: what exactly in there? ;)
<chewitt> if you look at the first set of gavlee dmesg logs it was showing an "invalid config magic" error, which was added in that patch
<chewitt> in the latest logs it appears to get further, but still fails to init the BT
<chewitt> any suggestions?
<chewitt> i'm wondering if it's a baud rate issue similar to the QCA9337 chip "we" fixed recently?
<xdarklight> chewitt: what is the realtek Bluetooth config blob that you are using on that board?
<xdarklight> and more importantly: what is the encoded baudrate inside it
<xdarklight> the code inside btrtl_convert_baudrate defaults to a baudrate of 115200 - if this is incorrect then the firmware download won't work
armoon has joined #linux-amlogic
<xdarklight> and is there any RTL8822CS Bluetooth repo out there in the wild with vendor init code?
<xdarklight> that is the code of the wifi part, but since the baudrate blob value is the same as on RTL8723BS (which translates to 1.5Mbit/s) I'm going to assume that this is all the same
<chewitt> all the broadcom chips, and that qualcom device, seem to run at 200000
<xdarklight> the part I find worrying is that the RTL8822CS config blob doesn't have "flow control" enabled
<chewitt> how do you decode the config blob?
<xdarklight> two steps: 1) there's a "rtlfw" tool in bluez-utils
<xdarklight> that prints for the config blob RTL8822CS (extract only): len=16 offset=000c,{ 02 80 92 04 50 c5 ea 19 e1 1b fd af 5b 01 a4 0b },UART_CONFIG
<xdarklight> then I follow btrtl_get_uart_settings - the first four bytes are little endian 32-bit value "encoded baud rate", and at index 12 BIT(2) decides whether flow control is enabled
<xdarklight> this is for example from RTL8723BS: len=16 offset=000c,{ 02 80 92 04 50 c5 ea 19 e1 1b f1 af 5f 01 a4 0b },UART_CONFIG
<xdarklight> see 0x5f (flow control enabled) on RTL8723BS is 0x5b (bit 2 unset, flow control disabled) on RTL8822CS
armoon has quit [Ping timeout: 245 seconds]
drieschel has quit [Quit: drieschel]
sputnik_ has joined #linux-amlogic
<gavlee> chewitt: latest one http://ix.io/2od5
<xdarklight> interesting, PRG_ETH1 = 0x000f0000
<gavlee> didn't see message was making dinner...
<xdarklight> unfortunately these bits are all undocumented... :-(
<gavlee> not sure what to say :-S
<xdarklight> gavlee: a good answer would be "I just saw an updated datasheet from Amlogic which documents these registers" ;)
<gavlee> ha
<xdarklight> gavlee: do you have the original .dtb or .dts of your board somewhere?
<gavlee> I was trying to get that the day before when chewitt messaged asking what make and model it was that I was trying o get working
<gavlee> i had beckup of the firmware in twrp before started messing around was looking at the contents of the partitions seeing what's in there
<xdarklight> gavlee: I typically extract it from vendor firmware updates. so if you have a link to that we can look into it as well
<xdarklight> gavlee: also do you have the kernel log output from the Android kernel the board came with?
<gavlee> i didn't find one for this P3 until i made the backup but it was on russian site and i had to use translator which didn't work too good... lol
<gavlee> i think there were like 4 different firmware of varying quality
<gavlee> some don't work on others,, from what i was reading
<gavlee> i'll have a look and see what I can find
<gavlee> i may have saved the link
<gavlee> already made a backup with twrp by the time I found that, didn't download it so not sure if it works or is actually there. the firemware version currently on mine is the P3 version
<gavlee> i don't even know where 'official' stock rom are to be found. not that I was expecting updates...
ldevulder has joined #linux-amlogic
ldevulder_ has joined #linux-amlogic
ldevulder has quit [Ping timeout: 240 seconds]
<xdarklight> I'll try to look into it tomorrow, I'm too tired today
<gavlee> Appreciate that, thanks. I would be happy the box running like it is I can do what I wanted to do on it now the network is up.
warpme_ has quit [Quit: Connection closed for inactivity]
vagrantc has quit [Quit: leaving]