martinayotte
-
Posts
3892 -
Joined
-
Last visited
Content Type
Forums
Store
Crowdfunding
Applications
Events
Raffles
Community Map
Posts posted by martinayotte
-
-
The command you mentioned above is erasing the first sector of the u-boot. This command is only to make sure that SD Card won't stay bootable for people wishing to boot from eMMC with still the SDcard inserted.
It should not have been executed until you are sure that you eMMC is actually bootable and working properly.
-
Ok ! but thing 88 means the count of gpio definition lines, not the gpio number. This means, like you've shown in your first post, gpio_pin_1 to gpio_pin_88.
Did you try the commands I've provided above ?
-
You can build your own image, but be aware that WiFi isn't working yet ...
-
From where this 88 number comes from ?
Did you tried to do the following commands directly into a shell ?
echo 225 > /sys/class/gpio/export echo out > /sys/class/gpio/gpio225/direction echo 1 > /sys/class/gpio/gpio225/value -
Maybe adding mmc2 in the DT would be enough ...
-
Be aware than we are still using Apritzel's master branch, while he suggested us (at least to me) to shift to his newest branch. But "time is the missing ingredient" ...
-
I've never seen such error ...
But your code seem to read in half-duplex mode.
You should try in full duplex mode by sending the same amount of dummy bytes you wish to read using spi.xfer()
rx_data = spi.xfer([0,0,0,0,0,0,0,0], 8) print rx_data -
Your first log shows clearly that it was in readonly mode until you rebooted.
Once rebooted, system has done an fsck to recover it, but probably there are tons of files that are not recovered to their original state...
It is suggested to try a new SD installed from scratch !
-
Some news : The trick is also working with the fw_bcm43438a1.bin for Non-A of my BananaPiM2Plus ...
(The only small problem is they ship me this board without antenna, fortunately, it still work since it sit by few feet of the AP)
-
Yes, you're right, let polish this stuff. And we also need to figured out why 64 bits are not working.
About '_apsta', I presume/guess that allows firmware to run in both mode, while the other can only work in STA or AP, but not both at the same time.
-
I'm glad that some of you got it working on ZeroPlus-H3 !
Now, seeing the advance, I retried it on my side too and it is now working.
Last time I did on H3 was 2 weeks ago, when we didn't knew about the firmware workaround.
Those recent days, when I was pulling my hairs, it was all on A64 or H5 boards, and the workaround didn't work there.
Also, yesterday, my trials was on BPiM2Plus, but this one is effectively a Non-A version.
-
** File not found /boot/dtb/sun8i-h3-orangepi-pc.dtb **The DTB file is missing ...
You can grab this file from another image and add it to your current sdcard.
-
That is a bit strange since this line shows that it there since months :
https://github.com/armbian/build/blob/master/scripts/nfs-boot.cmd.template#L14
-
Quote
BSP kernel (not BSD, that's something different)
You're right, sometime, I'm suffering of dyslectic ...
-
Which version has that "console=tty1" ?
I think it was always been "console=ttyS0,115200" ...
-
Now that I've done builds for NanoPiM1Plus2 and BananaPiM2Plus, along with the previous series OPiZeroPlus-H3/OPiZeroPlus-H5/OPiWin/OPiPrime, and that all of those boards has the same symptoms of missing AP6212 (+/- some have mmc-sdio appear, some not), I've continue my search for hints.
To my discovery, I found the following quote in http://linux-sunxi.org/Sinovoip_Banana_Pi_M2%2B
QuoteMainline kernel
(everything working except of WiFi/BT due to lack of interest)
I hope that is not true any more ...
-
Ok ! I've added the new add_nanopim1plus2_dts.patch, and committed it along with few tweaks for Makefile since patches are applied in alphabetic order.
Now, this image boot properly, but as expected, like other AP6212, it doesn't load up, and even mmc-sdio isn't fully initialized.
Back to SquareOne with this AP6212 issue ...
EDIT : I've added MMC2 for eMMC into this ... GoodBye, BSD ! I'm now over-writing you !
-
-
BSD is not using DeviceTree DT, it is using FEX/BIN.
So, you will need to use bin2fex to decompile script.bin (after having done a backup first), edit the the resulting script.fex to enable I2C buses, and recompile using the fex2bin into script.bin.
But be aware that people on Armbian forum isn't inclined to give much support of builds that are not from Armbian, we consider BSD as crappy builds !
-
Apologize for some previous noise ...
Ok ! I did the right build, now the u-boot is Ok, but it missing the sun50i-h5-nanopim1plus2.dtb.
This means the WIP target is incomplete ! I will have to finish it, probably like I did for other H5 recently.
But this means that my whole goal of trying to see if AP6212 is working on this board can't help, because I will probably end up with same things as other H5.
-
Ohhh !
Confusing ... Also, checking the silkscreen on the board, I see it has small checkboxes with 1 and 2, but none of them are mark. So same PCB for both like ZeroPlus ...
I'm redoing the new right build
-
Yes ! I've got the idea while you were answering : why not trying to boot an OPiZeroPlus-H5 on this board ?
Success ! ... This means the Armbian_5.27_Nanopim1plus_Debian_jessie_dev_4.10.11.img is complete crap !
We will have to check this target completely ...
EDIT : I will redo a build and inspect all the steps of u-boot compile (since I didn't keep the log, and have built some other H5) ...
-
Exactly, nothing shows in picocom.
-
19 minutes ago, sn0wblind said:
@martinayotte, I'll post my boot log tomorrow.
If it is Mainline, it will be appreciated, thanks.
If not, it is useless.
Can someone clear this GPIO thing?
in Allwinner sunxi
Posted
I'm not using Legacy, but always Mainline, so my previous post was wrong.
Zador was right, Legacy image still using the old numbering scheme.
So the commands would be more likely be for PH1 :
echo 16 > /sys/class/gpio/export echo out > /sys/class/gpio/gpio16_ph1/direction echo 1 > /sys/class/gpio/gpio16_ph1/value