**** BEGIN LOGGING AT Thu Mar 16 03:00:02 2017 Mar 16 09:14:15 bluelightning: hi Mar 16 09:14:57 I am trying to debug the ncurses-native vs. kernel menuconfig error but I think I am at a dead point... Mar 16 09:15:28 as it is now, do_menuconfig is ok for i.e. busybox Mar 16 09:16:47 if you read the log history of cml1.bbclass you'll see two distinct fixes specific for for ncurses Mar 16 09:17:16 b036adf cml1: set and export TERMINFO, to handle ncurses-native relocation Mar 16 09:17:50 889e026 ncurses, busybox, cml1.bbclass: Fix menuconfig display corruption Mar 16 09:19:07 well, I am pretty sure the issue is that with shared_workdir kernel is in STAGING_KERNEL_BUILDDIR Mar 16 09:20:36 hi ant_work Mar 16 09:20:52 hello Mar 16 09:21:40 STAGING_KERNEL_DIR = "${TMPDIR}/work-shared/${MACHINE}/kernel-source" Mar 16 09:21:42 vs. Mar 16 09:21:46 STAGING_DATADIR_NATIVE = "${STAGING_DIR_NATIVE}${datadir_native}" Mar 16 09:22:06 ( Mar 16 09:22:06 STAGING_DIR_NATIVE = "${RECIPE_SYSROOT_NATIVE}" ) Mar 16 09:23:04 bluelightning, it's very complicated interaction between bitbake.conf, kernel.bbclass and cml1.bbclass changes :/ Mar 16 09:24:06 afais if we do fix kernel menuconfig we'd break busibox menuconfig... Mar 16 09:24:56 VartiWork, sorry I am wasting time at menuconfig stage and have not yet finished to build 4.10.2 :) Mar 16 09:25:15 np :) Mar 16 09:25:25 every time I stumble in the same bug and I'd like to see it fixed now Mar 16 09:26:27 I'm slowly continuing to test kexec, I have checked and neither the SD I use to flash it nor the NAND on my 1000 have any bad block Mar 16 09:29:43 next time I'll try again the NAND restore but with no initrd.bin and home.bin flashing (in order to make kexec find JFFSv2 partitions on mtd2 and mtd3) Mar 16 09:31:01 since AFAIR I have seen the kexec menu (at least twice) only when I didn't flash those two partitions Mar 16 09:31:29 good luck with menuconfig btw Mar 16 09:36:34 there is something I'm forgetting then, when the partitions were 'dirty' it took long bot th emenu did show Mar 16 09:37:02 this on other Z, never tested on akita Mar 16 09:44:55 once you get to console please do ubiformat /dev/mtd2 and /dev/mtd3 Mar 16 09:45:12 maybe there are some old cleanmarkers from the original restore image Mar 16 09:45:26 so we know they are clean Mar 16 09:50:26 ok, I'm writing this down to remind me Mar 16 09:58:24 I've put all here: http://cgit.openembedded.org/meta-handheld/tree/README.zaurus Mar 16 10:03:52 thanks Mar 16 10:05:52 feel free to update it :) Mar 16 10:11:48 ok :) Mar 16 10:13:36 one thing I could already update is that the OK+PowerON flashing procedure require an SD card which is <=1GB Mar 16 10:13:51 (which are pretty rare to come by nowadays) Mar 16 10:14:14 didn't check with the CF card, I only have one which is 1GB Mar 16 10:15:15 I have also tried to format a 2GB one with a 512MB FAT16 partition only, with no avail Mar 16 10:17:54 it would be interesting to try to make a new NAND restore image with a new kernel for the OK+Power ON flashing menu, with an updated SD card driver (assuming that the kernel for flashing is inside the NAND restore) Mar 16 15:09:39 VartiWork, see you soon Mar 16 15:09:40 bye Mar 16 21:18:11 bluelightning, I'm trying to lure kergoth on the issue... Mar 16 21:19:18 hi ant Mar 16 21:20:49 hi ant_home Mar 16 21:21:19 sorry I can't really help much, but if RSS makes the issue more apparent hopefully others will be able to see and help out with fixing it Mar 16 21:32:46 bluelightning, it's a mess, keep away ;) **** ENDING LOGGING AT Fri Mar 17 03:00:01 2017