| 25 May 2021 |
 novmar | ah thanks a lot man | 00:33:30 |
| novmar | may be i'll have time some other time to try it again ... | 00:33:54 |
| novmar | may be not | 00:33:56 |
| novmar | GN | 00:34:10 |
 samueldr | 'night! | 00:34:21 |
| samueldr | yeah, not worth the buy just to check :/ just over the impulse buy "just for the heck of it" price point | 00:35:39 |
| novmar | samueldr: hexdump of armbian image, i found start of u-boot | 00:46:22 |
| novmar | 00002000 be 02 00 ea 65 47 4f 4e 2e 42 54 30 37 5a 22 a1 |....eGON.BT07Z".|
| 00:46:26 |
| samueldr | armbian doesn't always use mainline U-Boot (mainline meaning from the project) | 00:46:53 |
| samueldr | they sometime use the U-Boot from the board vendor | 00:47:01 |
| samueldr | which is often old, sometimes really old | 00:47:10 |
| novmar | ah main idea was, if it isn\t shiifted ... | 00:48:02 |
| novmar | but it is aligned | 00:48:07 |
| samueldr | the SoC checks for eGON so it wouldn't have tried to use it if it was shifted | 00:48:26 |
| samueldr | exactly for that reason | 00:48:30 |
| novmar | ah ok | 00:49:00 |
| novmar | so it does't matter, where it starts ... :-) | 00:49:24 |
| samueldr | right, yes, the Allwinner family of SoC checks at a location (actually it might be a few, but only one will work with current U-Boot) | 00:49:54 |
| samueldr | exactly 0x2000, or 16*512 | 00:50:27 |
| novmar | :-) | 00:50:36 |
| novmar | ok | 00:50:37 |
| samueldr | ah, bs=1024 seek=8 so 8x1024 | 00:50:56 |
| samueldr | * ah, bs=1024 seek=8 so 8*1024 | 00:51:00 |
| novmar | yeah i did it that way | 00:51:12 |
| samueldr | (yes, the same, but I wanted to show the reason for the numbers) | 00:51:17 |
| samueldr | eGON is 4 bytes later, but it's how it should be | 00:51:28 |
| novmar | yeah i understand :-) | 00:51:34 |
| novmar | there are diferent bits (hat 4 bits) | 00:51:52 |
| novmar | ah but it is mess from sdcard | 00:52:01 |
| samueldr | I wonder if there are differences on h616 that our assumptions don't hold for | 00:52:23 |
