wha Posted July 5, 2016 Share Posted July 5, 2016 updated to 5.15 (then 5.16) release using aptitude. Kernel update worked. But when I updated u-boot ( linux-u-boot-orangepiplus2e-default ) , the system will no longer boot (see attached). But if I use linux-u-boot-orangepiplus-default , it will boot. (Note that the install script, to write/dd SPL , is hardcoded to mmcblk0, which is usually correct, but for this fix, I start boot on SDHC then pivot-root to eMMC. My running root is mmcblk1 during this "rescue" boot). Both 5.15 & 5.16 exhibit this behavior. Thanks, wha bad-uboot.txt Link to comment Share on other sites More sharing options...
zador.blood.stained Posted July 5, 2016 Share Posted July 5, 2016 Verifying Checksum ... Bad Data CRC Ramdisk image is corrupt or invalid For some reason your uInitrd is corrupt. Since you can boot to similar system, you need to recreate and copy uInitrd from system that can boot to replace corrupt one. Link to comment Share on other sites More sharing options...
wha Posted July 5, 2016 Author Share Posted July 5, 2016 uInitrd is good. Note that ALL I did to get system to boot, is it to re-write uboot block(s) to start of eMMC: ==== wha@orangepiplus2e:~$ sudo dd if=/dev/zero of=/dev/mmcblk1 bs=1k count=1023 seek=1 status=noxfer wha@orangepiplus2e:~$ sudo dd if=/usr/lib/linux-u-boot-orangepiplus_5.16_armhf/u-boot-sunxi-with-spl.bin of=/dev/mmcblk1 bs=1024 seek=8 status=noxfer ==== (just manually typed install script from deb) So I believe that the u-boot [executable] code, for the '2e' is wrong. To verify that assumption, I wrote the SPL from linux-u-boot-orangepiplus2e_5.16_armhf , once again, boot CRC error. Thanks, wha Link to comment Share on other sites More sharing options...
Recommended Posts