>> >> >> >> > > > > Mine has as of yet failed to boot on any of them. >> >> >> >> > > > >> >> >> >> > > > Mine is currently running 4.8.1-1.fc25 without issues, >> >> >> >> > > > just upgrading it to the latest now to test that too. >> >> >> >> > > >> >> >> >> > > Was running 4.8.2-300.fc25.armv7hl+lpae, upgraded to >> >> >> >> > > 4.8.4-301.fc25 without any issues. Could you post output of >> >> >> >> > > your >> >> >> failure? >> >> >> >> > > >> >> >> >> > > Paul >> >> >> >> >> >> >> >> You might have better luck with a more recent version of U-Boot. >> >> >> >> I'm running U-Boot 2016.11-rc2 with 4.8.4-301.fc25 on sata. >> >> >> >> >> >> >> >> >> >> >> > >> >> >> > That made sense to me so I upgraded to the uboot that you are >> >> >> > using but >> >> >> same problem. >> >> >> >> >> >> Do you have any extra peripherals added like a mPCIe card, camera >> >> >> or other such things or is it just the base board? What device do >> >> >> you have the rootfs on? eMMC, SD or SATA? >> >> >> >> >> >> I have mine running as a bare device with the OS on the eMMC, I >> >> >> think Paul might have the OS on a sata SSD. Just trying to work >> >> >> out what if any differences there are here as there's not much that can >> be different. >> >> >> >> >> > >> >> > Thanks Peter! >> >> > >> >> > Mine is yet another variation. It is a bare board with the boot >> >> > partition on an >> >> SD card and the root filesystem on an SSD. I left the original eMMC >> >> untouched until I upgraded the uBoot. >> >> >> >> The u-boot is on SPI nand not the eMMC so I suspect then you didn't >> >> even touch it. >> >> >> >> > The only thing other than the SDcard and the SSD is the serial console. >> >> > >> >> > Just for kicks I tried an LPAE kernel and the first time I booted >> >> > it got some >> >> kenel oops that I had not seen before but I was not logging the >> >> console output so I turned on capture and tried again but have not >> >> seen that kernel oops since. >> >> > >> >> > For some reason my board is not seeing the sata controller (or not >> >> > getting it >> >> initialized properly) on 4.7 and 4.8 kernels but does on 4.6.7. >> >> >> >> Is it pulled into the initrd? Do you see the driver with a "lsinitrd | grep >> ahci" >> >> >> >> > Tim Krantz >> >> > >> > >> > Well when booting fails with 4.8.4-lape it finally drops into a Dracut >> > command prompt with no lsinitrd >> >> No, do it when booted against a 4.6.x kernel, you can specify the actual initrd >> on the cmd line.... >> >> > But all I see in to console logs is about a half dozen of : >> > >> > tegra-ahci 70027000.sata: couldn't get PHY in node sata: -517 >> >> Looks like it's not getting either appropriate power to the controller or some >> module deps are failing to be pulled into the initrd correctly. >> >> > and that is the only reference to ahci >> > >> > In a good boot on 4.6.7 I do see the driver loaded with lsinitrd | grep ahci. >> > > > I apologize for my ignorance. I think what you are asking is if I run the 4.6.7 kernel with the 4.8.x initrd what do I see related to ahci. No, investigating the 4.8.x initrd while booted with a 4.6 kernel (IE booted fully). See previous commands. > I tried that and get the same issue of not seeing sata and eventually dropping into a Dracut command prompt. > > Ihave been looking into how I can examine the uInitrd and saw something about stripping off the first 64 bytes, gunzipping it and mouting it as a loop device but have as yet been unable to get that work. > > Tim > _______________________________________________ arm mailing list -- arm@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to arm-leave@xxxxxxxxxxxxxxxxxxxxxxx
