Jump to content

Albert Barcelona

  • Posts

  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. @root-stas I hope your had a great resting time Have you checked about the GPU progress?
  2. For now, I'll leave the box 'as is'. It works fine with your .dtb even GPU acceleration isn't working. I am not using it for video or even desktop. I've installed some server stuff and I'm happy as long as it is stable. I shall keep an eye on this thread to see if progress is being made or if somebody needs help. Thanks a lot @root-stas. Can't wait to see your .dtb progress!
  3. @root-stas I don't even know how to start with that but.. anything we can learn from ODROID c4 hardkernel's own dts? (same SOC) https://paste.ubuntu.com/p/HQ5Wfz7mZp/
  4. Now that almost everything seems to work.. I'd like to run some performance benchmarking tests. I am curious about the Mali G31 MP2 performance as the dtb evolves together with the lima / meson driver. This is a cheap SoC but graphic performance should definitely do better than what I can see now.
  5. Just as an example, docker won't work on focal for now. I don't have a preferred distro... pros & cons everywhere...
  6. @root-stas thanks. Your dtb works very well! I couldn't figure out what was wrong. Also, focal isn't my preferred choice as of today because of some packages limitations. Great news, however. I installed today's Buster image with 5.7.0-rc5 and HDMI audio works! Ethernet speed is 3x faster.!! I am not sure whether Bluetooth works. I will test and update. There are a few errors on dmesg about wifi and HDMI audio but both work.
  7. Hey @root-stas: I've installed focal from scratch and tried your new dtb file. Ethernet works fine both wired (1Gbps) and wireless. Weirdly enough; HDMI audio works now when booting from SDCard but after burning to emmc, something happens and: I get audio when booting from SD but when booting from emmc, the issue appears again!! Here are the two dmesgs: Booting from SD: https://paste.ubuntu.com/p/xH5JXP8MmV/ Booting from emmc: https://paste.ubuntu.com/p/Yw7myxWj2B/ How is that possible? Any clue?
  8. Ok, thanks!. Will try now.. I'll post results in the other thread. I believe it is better to keep the discussion in the H96 Max X3 thread.
  9. @root-stas: Thanks for sharing! Seems you really know about this! Unfortuntely HDMI audio continues not to work. Apparently no changes for me on behaviour from test8 file. Here's my last dmesg (with the new h96max-x2-test-bus80-gpu-volt.dtb) https://paste.ubuntu.com/p/KYNQPPvwMn/ And here the audio devices list: albert@asl-h96:~$ aplay -l **** List of PLAYBACK Hardware Devices **** card 0: AMLG12 [AMLG12], device 0: fe.dai-link-0 (*) [] Subdevices: 1/1 Subdevice #0: subdevice #0 card 0: AMLG12 [AMLG12], device 1: fe.dai-link-1 (*) [] Subdevices: 1/1 Subdevice #0: subdevice #0 card 0: AMLG12 [AMLG12], device 2: fe.dai-link-2 (*) [] Subdevices: 1/1 Subdevice #0: subdevice #0 albert@asl-h96:~$ Again; thanks for keeping up!
  10. Actually.. when 'decompiling' your test8 dtb to dts I get the following warning: albert@asl-h96:~/dtbs$ dtc -I dtb -O dts -o test8.dts /boot/dtb/h96max-x3-test8.dtb test8.dts: Warning (unit_address_vs_reg): /soc/bus@ff600000/bus@34400/pinctrl@40: node has a unit name, but no reg property test8.dts: Warning (unit_address_vs_reg): /soc/bus@ff800000/sys-ctrl@0/pinctrl@14: node has a unit name, but no reg property test8.dts: Warning (unit_address_format): /soc/bus@ff800000/rtc@0a8: unit name should not have leading 0s test8.dts: Warning (power_domains_property): /soc/vpu@ff900000: Missing property '#power-domain-cells' in node /soc/bus@ff600000/bus@34400/pinctrl@40/bank@40 or bad phandle (referred from power-domains[0]) test8.dts: Warning (gpios_property): /meson-vfd: Missing property '#gpio-cells' in node /emmc-pwrseq or bad phandle (referred from gpios[4]) What is this #power-domain-cells property? seems to be the key. Yet, the vpu@ff9900000 reads exactly as in the meson-610: vpu@ff900000 { reg = < 0x00 0xff900000 0x00 0x100000 0x00 0xff63c000 0x00 0x1000 >;
  11. Thanks @root-stas I've used your test8 dtb and compared my dmesg with yours and I have found where the hdmi problem happens: [ 2.898291] OF: /soc/vpu@ff900000: could not get #power-domain-cells for /soc/bus@ff600000/bus@34400/pinctrl@40/bank@40 [ 2.905345] meson-drm ff900000.vpu: Queued 2 outputs on vpu [ 2.911406] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). [ 2.917483] meson-drm ff900000.vpu: CVBS Output connector not available [ 2.923982] meson-drm ff900000.vpu: failed to bind ff600000.hdmi-tx (ops meson_dw_hdmi_ops): -517 [ 2.932738] meson-drm ff900000.vpu: Couldn't bind all components [ 2.938842] meson-drm ff900000.vpu: master bind failed: -517 The mesage: 20.005163] fe.dai-link-0: ASoC: no backend DAIs enabled for fe.dai-link-0 continues to be there with test8 file. I suppose the issue is related to the HDMI. Any clue?????
  12. Hi there, Assuming the process you follow is correct, the box won't boot unless you properly configure the uEnv.txt (or equivalent in other OS) to config the kernel loading. The initial challenge is to find and select the right DTB. If you don't do that, the boot will just not happen. If when you don't do the AV button, system boots from emmc, it means that the AV thing is really bypassing the normal boot, that is, it is working.
  13. (1) My experience with SD cards aren't any good. Besides lower speed, SD cards tend to degrade and are more likely to failures. If I was you, I'd partition the emmc (you've got a lot with the 128Gb) and try to build a dual boot although I have no clue whether that is possible under this architecture.. (2) I suppose the reason why we're getting different messages is because we used different DTBs, and that's the key. I am testing a different one now that fixed the Ethernet to 1Gbps, Wifi works and theoretically bluetooth (but still no luck with the BT).
  14. @dgm78: Could you please elaborate a little further? Did you edit the dts line? From what to what? Thank you!!
  15. Hello, While there are a bunch of vnc and x11 options to access the box remotely, I found chrome remote desktop as a very useful tool to integrate access to different hosts wherever you are. chrome remote desktop 'host' runs on many platforms including, of course, debian. Installation does not work because package gets downloaded from Google directly and it is compiled for x86. The .deb package can't therefore be installed on armbian. Any chance chrome remote desktop could be compiled for armbian? I believe source code is under Chromium project but I could not find it. Ideas?
  • Create New...