#linux-sunxi on 2020-08-11 — irc logs at freenode.irclog.whitequark.org

2018-08-03 10:18 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*

00:08 sunshavi has quit [Ping timeout: 264 seconds]

00:26 <karlp> ok... I think I just need to get the regulator tweaked. I've got https://termbin.com/imwl as an overlay, but none of those three regulators show up in /sys/class/regulator, and the gpio doesnt show upin /sys/kernel/debug/gpio so I don't think it's being "overlayed" correctly.

00:26 <karlp> without power enabled, it's kinda expected that the ov5640 driver complains about register read failures.

00:26 <karlp> but I'm not sure what I'm missing to make the overlay build those regulators?

00:29 gaston1980 has quit [Quit: Konversation terminated!]

00:45 ChriChri_ has joined #linux-sunxi

00:47 ChriChri has quit [Ping timeout: 256 seconds]

00:47 ChriChri_ is now known as ChriChri

00:59 obbardc has quit [Ping timeout: 265 seconds]

00:59 obbardc has joined #linux-sunxi

01:18 RichardG867_ has joined #linux-sunxi

01:22 RichardG867 has quit [Ping timeout: 272 seconds]

01:35 sunshavi has joined #linux-sunxi

02:39 random_yanek has quit [Ping timeout: 264 seconds]

02:40 sunshavi has quit [Remote host closed the connection]

02:41 RichardG867_ is now known as RichardG867

02:41 sunshavi has joined #linux-sunxi

03:04 wigyori has quit [Ping timeout: 256 seconds]

03:04 wigyori has joined #linux-sunxi

03:04 sunilmohan has quit [Ping timeout: 256 seconds]

03:04 buZz has quit [Ping timeout: 256 seconds]

03:05 mripard has quit [Ping timeout: 260 seconds]

03:06 mripard has joined #linux-sunxi

03:06 buZz has joined #linux-sunxi

03:06 buZz is now known as Guest37465

03:06 sunilmohan has joined #linux-sunxi

03:07 tuxd3v has quit [Quit: Leaving]

03:10 tuxd3v has joined #linux-sunxi

03:24 lurchi_ has joined #linux-sunxi

03:29 tuxd3v has quit [Quit: Leaving]

03:38 victhor has quit [Quit: Leaving]

03:40 TheSeven has quit [Ping timeout: 260 seconds]

03:40 TheSeven has joined #linux-sunxi

03:44 sunshavi has quit [Ping timeout: 265 seconds]

03:48 lurchi_ is now known as lurchi__

03:54 _whitelogger has joined #linux-sunxi

03:54 sunshavi has joined #linux-sunxi

03:58 sunshavi has quit [Remote host closed the connection]

04:02 sunshavi has joined #linux-sunxi

04:22 <wens> karlp: not all camera modules require MCLK. some use an external crystal

04:23 <wens> also , MCLK is not part of CSI. it's more of a direct output from the CCU

04:24 <wens> karlp: use dtc included with the kernel?

04:26 <wens> karlp: try having the regulators in a fragment, with the root ( / ) as the target (target = <&{/}>) ?

04:32 lurchi_ has joined #linux-sunxi

04:35 lurchi__ has quit [Ping timeout: 260 seconds]

05:06 ldevulder has quit [Ping timeout: 240 seconds]

05:13 netlynx has joined #linux-sunxi

05:13 netlynx has quit [Changing host]

05:13 netlynx has joined #linux-sunxi

05:23 JohnDoe_71Rus has joined #linux-sunxi

05:25 diego71 has quit [Ping timeout: 256 seconds]

05:34 lkcl__ has joined #linux-sunxi

05:37 lkcl_ has quit [Ping timeout: 256 seconds]

05:42 jbrown has quit [Remote host closed the connection]

06:01 reinforce has joined #linux-sunxi

06:27 Guest37465 has quit [Changing host]

06:27 Guest37465 has joined #linux-sunxi

06:27 Guest37465 is now known as buZz

07:08 AneoX has joined #linux-sunxi

07:17 cmeerw has joined #linux-sunxi

07:54 ldevulder has joined #linux-sunxi

07:56 diego71 has joined #linux-sunxi

08:00 ldevulder_ has joined #linux-sunxi

08:02 ldevulder has quit [Ping timeout: 256 seconds]

08:03 ldevulder__ has joined #linux-sunxi

08:07 ldevulder_ has quit [Ping timeout: 265 seconds]

08:08 AneoX has quit [Ping timeout: 264 seconds]

08:12 AneoX has joined #linux-sunxi

08:18 random_yanek has joined #linux-sunxi

08:32 warpme_ has joined #linux-sunxi

08:57 matthias_bgg has joined #linux-sunxi

09:03 tnovotny has joined #linux-sunxi

09:44 <karlp> wens: great thanks for both.

09:45 <karlp> yeah, I'm more used to a plain dtc, but this armbianbuild root is... more complicated than I'mused to, and the "just add an overlay" worked out very well for the lcd touch screen, so was tryign to get it for this camera too.

09:45 <karlp> your patches for pinetab and example for some bpi model have proved very informative :)

10:02 Mangy_Dog has joined #linux-sunxi

10:12 jbrown has joined #linux-sunxi

10:55 <TiD91> Hello, with my BPI-M1 I'm on kernel 5.8.0 with Archlinux Arm (I guess it's mainline) and u-boot from latest commit on master. Booting up works as expected, but I can't shutdown or reboot. Uart shows a kernel panic when shutting down... Any help on what to look for to fix shutdown and reboot?

11:10 cmeerw has quit [Ping timeout: 244 seconds]

11:28 cmeerw has joined #linux-sunxi

11:30 jbrown has quit [Ping timeout: 272 seconds]

11:36 cnxsoft has joined #linux-sunxi

11:36 cnxsoft1 has quit [Read error: Connection reset by peer]

11:50 jbrown has joined #linux-sunxi

12:33 andy25225 has quit [Ping timeout: 260 seconds]

12:53 Tenkawa has joined #linux-sunxi

12:53 <martinayotte_> tuxd3v: the 1500000 is the maximum possible, but the default is been determine by kernel argument, such as "console=ttyS0,115200"

12:58 andy25225 has joined #linux-sunxi

13:07 specing has joined #linux-sunxi

13:14 <willmore> It's the base baud, i.e. the clock that supplies the serial port and is divided down to get the other baud rates.

13:14 andy25225 has quit [Ping timeout: 256 seconds]

13:21 andy25225 has joined #linux-sunxi

13:27 JohnDoe_71Rus has quit [Ping timeout: 246 seconds]

13:37 <mru> no, that clock is usually 24 MHz on sunxi

13:38 <mru> dividing that by 16 gives the highest bit rate of 1.5 MHz

13:39 <Ashleee> hmm I found quite interesting boards... https://www.alibaba.com/product-detail/Helper-a64-Low-Price-A64core-board_62589177873.html?spm=a2700.pc_countrysearch.main07.57.157817f3orkG7r

13:39 <Ashleee> better than RPi Zero

13:40 <Ashleee> for when you are scared to route your own (LP)DDR3 :)

13:41 <Ashleee> how hard is it to route (LP)DDR3 on some small board? :)

13:41 JohnDoe_71Rus has joined #linux-sunxi

13:51 diego71 has quit [Read error: Connection reset by peer]

13:54 diego71 has joined #linux-sunxi

14:02 <mru> if you have to ask, too hard

14:11 <DuClare_> I'd like to see more chips with ram in the package

14:11 <DuClare_> Like the S3

14:11 <DuClare_> It's pretty convenient

14:20 <Ashleee> mru, I mean, everybody had to start somewhere once :)

14:20 <mru> I've never done it

14:21 <mru> I assume it's a PITA

14:27 reinforce has quit [Quit: Leaving.]

14:34 kaspter has joined #linux-sunxi

14:45 cnxsoft has quit [Quit: cnxsoft]

14:46 ldevulder__ has quit [Quit: Leaving]

14:47 kaspter has quit [Ping timeout: 256 seconds]

14:48 kaspter has joined #linux-sunxi

15:26 netlynx has quit [Quit: Ex-Chat]

15:31 <jernej> Ashleee: ping

15:57 hexdump0815 has joined #linux-sunxi

15:57 <hexdump0815> jernej: some test update - setting al values here to zero did not help: https://gitlab.denx.de/u-boot/u-boot/-/blob/master/arch/arm/mach-sunxi/dram_sun50i_h6.c#L645

15:59 <hexdump0815> jernej: and your patch does not change anything for my box - same result as without patch und the usual rank=2 - "Error while initializing DRAM PHY!"

16:00 <jernej> hexdump0815: Thanks for testing. Can you also give me values printed before this error message?

16:00 <jernej> are values printed with and without patch same?

16:09 <Ashleee> jernej, sorry cannot test today :( maybe tomorrow

16:10 <jernej> np

16:13 ndufresne has quit [Quit: The Lounge - https://thelounge.chat]

16:13 <hexdump0815> jernej: sure can add the regdump back with your patch - not near the box right now, but will test later

16:14 <jernej> no, I didn't mean regdump

16:14 <jernej> ah, sorry

16:14 ndufresne has joined #linux-sunxi

16:14 <hexdump0815> what else? i think the patch itself did not print anything

16:15 <jernej> please add "#define DEBUG 1" at the very first line in the dram_sun60i_h6.c file

16:15 <jernej> it should produce some additional output

16:15 <hexdump0815> ah - ok - right - will do an rerun later

16:21 lerc has quit [Ping timeout: 260 seconds]

16:24 lerc has joined #linux-sunxi

16:25 jbrown has quit [Ping timeout: 244 seconds]

16:28 jbrown has joined #linux-sunxi

16:32 lkcl__ is now known as lkcl

16:44 tnovotny has quit [Quit: Leaving]

16:48 vagrantc has joined #linux-sunxi

17:02 andy25225 has quit [Ping timeout: 256 seconds]

17:13 andy25225 has joined #linux-sunxi

17:26 kaspter has quit [Quit: kaspter]

17:38 Tenkawa has quit [Quit: Leaving.]

17:45 tuxd3v has joined #linux-sunxi

17:46 Tenkawa has joined #linux-sunxi

17:52 AneoX has quit [Quit: Textual IRC Client: www.textualapp.com]

18:12 sunshavi has quit [Ping timeout: 256 seconds]

18:27 jbrown has quit [Ping timeout: 272 seconds]

18:28 jstein has joined #linux-sunxi

18:39 florian_kc has quit [Quit: Leaving]

18:45 jbrown has joined #linux-sunxi

18:55 arti has quit [Ping timeout: 244 seconds]

18:56 arti has joined #linux-sunxi

19:22 jbrown has quit [Ping timeout: 246 seconds]

19:49 TheSeven has quit [Disconnected by services]

19:49 [7] has joined #linux-sunxi

19:50 lurchi_ is now known as lurchi__

19:53 gaston1980 has joined #linux-sunxi

19:57 tuxd3v has quit [Quit: Leaving]

20:08 <hexdump0815> jernej: i guess this is the output you were asking for: https://pastebin.com/raw/UZzDRcsC

20:10 <hexdump0815> jernej: the pll looks quite different than https://forum.armbian.com/topic/12372-h6-allwinner-tv-box-need-help/?do=findComment&comment=92499 or https://forum.armbian.com/topic/12372-h6-allwinner-tv-box-need-help/?do=findComment&comment=92690

20:11 andy25225 has quit [Ping timeout: 264 seconds]

20:12 <hexdump0815> jernej: btw. i put the patches we tried back then here just in case: https://github.com/hexdump0815/u-boot-misc/tree/master/misc.h6/misc-tests

20:12 andy25225 has joined #linux-sunxi

20:15 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/]

20:18 andy25225 has quit [Ping timeout: 240 seconds]

20:19 andy25225 has joined #linux-sunxi

20:26 jbrown has joined #linux-sunxi

20:32 <willmore> mru, good point, sorry to be unclear. There is a manditory /16 in the hardware of the 16550 type UART.

20:37 [7] has quit [Ping timeout: 260 seconds]

20:38 andy25225 has quit [Ping timeout: 240 seconds]

20:38 TheSeven has joined #linux-sunxi

20:41 msimpson has joined #linux-sunxi

20:49 andy25225 has joined #linux-sunxi

20:58 <hexdump0815> jernej: maybe better ignore my last output - here is one after a full make clean and remake which i think looks more real: https://pastebin.com/raw/6XhSSXkm

21:03 jbrown has quit [Ping timeout: 246 seconds]

21:17 <hexdump0815> jernej: i just gave the old patch from https://github.com/hexdump0815/u-boot-misc/blob/master/misc.h6/misc-tests/test-04.patch another try from which the two armbian forum output example from january were from and i'm able to get the same behaviour: after a few minutes of resets it sometimes works

21:17 <hexdump0815> jernej: https://pastebin.com/raw/6BWJc1kR

21:21 <jernej> hexdump0815: I already forgot that it worked sometimes

21:22 <jernej> anyway, to find out something new, you'll have to expand regdump from mainline to include whole range, not only PHY

21:22 <jernej> e.g. that it starts from 0x04002000

21:27 cmeerw has quit [Ping timeout: 244 seconds]

21:33 <hexdump0815> jernej: so this one? - https://pastebin.com/raw/3V5SiinU

21:34 <hexdump0815> jernej: ah or do you mean to adjust the regdump from ashleee to that range?

21:42 <jernej> yes, second one

21:43 <hexdump0815> jernej: on it - i combine it with the test-04 patch and maybe we see a difference in the dumps between the reset case and the case when it works - would this make sense?

21:54 <jernej> not sure if it would help much

21:54 <jernej> anyway, are you using this patch? https://github.com/hexdump0815/u-boot-misc/blob/master/misc.h6/h6-dram-register-dumping.patch

21:55 <jernej> please remove dtcr[1] write

21:55 <hexdump0815> jernej: this is interesting - i have now combined https://github.com/hexdump0815/u-boot-misc/blob/master/misc.h6/misc-tests/test-04.patch with https://freenode.irclog.whitequark.org/linux-sunxi/2020-08-09#27685489 - on that i only changed it to start at 0x04002000 and run for 4 * sizeof(struct sunxi_mctl_phy_reg) to cover the full range

21:55 <hexdump0815> jernej: and with that it works reliable each time :)

21:57 <hexdump0815> jernej: can you maybe make sense out of this? it does not print any registers though ... maybe the patch from ashleee is setting the rank to 1 combined with your test-04 patch from january make it work together?

21:57 <jernej> can you make diff between mainline and changes you have?

21:57 <hexdump0815> sure

21:59 <hexdump0815> jernej: here it is - https://pastebin.com/raw/REEtTgxf

22:01 <hexdump0815> jernej: only memory size is detected wrong - but i think Ashleee: had that problem too initially with rank=1

22:02 <hexdump0815> jernej: i tried it many times - this version works absolutely reliable for some reason ...

22:02 <jernej> I see only two potential reasons - #define DEBUG - it may delay some operations or writel(0x00010237, &mctl_phy->dtcr[1]);

22:02 <jernej> I bet on second one

22:03 <jernej> can you test which one is the reason?

22:03 <hexdump0815> sure - the write only would be test-04 patch plus rank=1 in the end - right?

22:05 <jernej> just remove lines in current sources...

22:05 <hexdump0815> without DEBUG it still works

22:06 <hexdump0815> will switch to test-04 plus rank=1 ... after making a backup of the current source :)

22:09 <jernej> If that works reliably then please test combination of test-04 and patch I made yesterday for rank detection

22:12 <hexdump0815> jernej: still working and now even the memory size is correct :) - output and current diff: https://pastebin.com/raw/FzZtfjZ7

22:13 <jernej> great!

22:14 <jernej> now the only question remains how to fix rank detection

22:14 <jernej> hopefully patch from yesterday will help with that

22:14 <hexdump0815> shall i add it on top?

22:17 <jernej> revert rank change

22:17 <jernej> and add it on top of test-04

22:18 <jernej> DRAM driver must always start with rank=2

22:18 <jernej> it tests if that works or not. If not, it changes to rank=1

22:22 <jernej> hexdump0815: to avoid misunderstanding, here is the patch: http://ix.io/2tVW

22:22 <jernej> it would be great if that works on top of clean mainline

22:27 <hexdump0815> jernej: looks like rank detection does not work - i reverted rank back to 2 and added your patch from two days ago

22:27 <hexdump0815> DRAM:Error while initializing DRAM PHY!

22:28 <jernej> ok then, thanks for testing

22:28 <jernej> I'll look into rank detection code again

22:28 <hexdump0815> my diff is matching yours exactly

22:29 <jernej> but at least now you have workaround :)

22:29 <hexdump0815> thank you for getting this forward ... time to go to bed for today :)

22:29 <jernej> me too

22:30 <hexdump0815> do you have a plan how to integrate test-04.patch into regular mainline or is it ok for the normal case as well as it is?

22:30 <jernej> not sure what you mean

22:30 <hexdump0815> for my box i still need that test-04 hack - can that go as it is into the u-boot tree or is it just a specific hack?

22:31 <jernej> it should go to mainline but it needs to be double checked

22:31 <jernej> some changes may not be needed

22:31 <hexdump0815> cool - let me know when you have some new rank detection code to test tomorrow or another day

22:32 <jernej> for example, tcke should probably stay as it is already in mainline

22:32 <jernej> I'm not certain about trrd and txp

22:32 <hexdump0815> ok - we can reduce it further step by step and see when it breaks ... its always easier to debug from the working end :)

22:33 <jernej> exactly

22:35 <jernej> anyway, if you have time, try to figure which changes are needed from test-04 patch and which are not

22:36 <hexdump0815> i can try during the next days - tomorrow most probably not

22:36 <jernej> no hurry :)

22:37 <jernej> just ping me when you'll have result

22:38 hexdump0815 has quit [Remote host closed the connection]

22:59 Mangy_Dog has quit [Ping timeout: 256 seconds]

23:06 msimpson has quit [Remote host closed the connection]

23:20 ganbold has quit [Ping timeout: 240 seconds]

23:20 ganbold has joined #linux-sunxi

23:27 ullbeking has quit [Ping timeout: 246 seconds]

23:29 ullbeking has joined #linux-sunxi

23:43 iyzsong has quit [Quit: ZNC 1.7.5 - https://znc.in]

23:44 warpme_ has quit [Quit: Connection closed for inactivity]

23:44 jstein has quit [Quit: quit]

23:45 iyzsong has joined #linux-sunxi