**** BEGIN LOGGING AT Sun Sep 17 03:00:01 2017
Sep 17 20:23:15 <ant_home> bluelightning, hi
Sep 17 20:23:39 <ant_home> I have sent a first set of patches for meta-handheld
Sep 17 20:24:29 <ant_home> please fix s/kexecoot/kexecboot/ in the 2 subjects when you merge, thanks
Sep 17 20:24:31 <bluelightning> hi ant_home
Sep 17 20:24:43 <bluelightning> ok will do, thanks
Sep 17 20:24:47 <ant_home> as always in hurry...
Sep 17 20:25:10 <ant_home> btw I could fix start_getty and kernels 4.4x this weekend :)
Sep 17 20:25:26 <ant_home> 4.4 serie is the last one with lumag patches
Sep 17 20:25:39 <ant_home> (SD for collie etc)
Sep 17 20:25:55 <bluelightning> ah ok... never went upstream?
Sep 17 20:26:04 <ant_home> unfortunately not
Sep 17 20:26:22 <ant_home> 6 or 7 retries then lumag went away
Sep 17 20:26:46 <ant_home> need much work with gpio/spi to adapt
Sep 17 20:27:29 <ant_home> fwiw after some point we have a new bug: lcd is white on boot on 4.10, 4.12, 4.13
Sep 17 20:27:36 <ant_home> must be some silly thing...
Sep 17 20:27:59 <bluelightning> doh
Sep 17 20:28:12 <ant_home> I will commit 4.13 next but I am afraid lumag patches will bitrot
Sep 17 20:28:31 <ant_home> let me fix this backlight stuff..
Sep 17 20:29:41 <ant_home> then we have the gcc7 issue...sigh
Sep 17 20:31:14 <bluelightning> ant_home: did you try asking khem about that?
Sep 17 20:37:02 <ant_home> yes, he's in copy
Sep 17 20:37:10 <ant_home> it's 2 months already...
Sep 17 20:37:54 <ant_home> tried gcc7.1  binutils 2.28 and gcc 7.2 binutils 2.29
Sep 17 20:39:09 <ant_home> bluelightning, what is funny is that back then I did build for versatile/armv5 and it did boot in qemu
Sep 17 20:39:26 <ant_home> then recently the same with spitz: it boots only in qemu
Sep 17 20:39:37 <ant_home> otherwise kernel fails to decompress
Sep 17 20:39:42 <bluelightning> must be triggering some sort of hardware quirk... perhaps some kind of race
Sep 17 20:39:55 <ant_home> I thought alignment
Sep 17 20:40:09 <bluelightning> does QEMU not emulate that?
Sep 17 20:40:25 <ant_home> hehm maybe it tolerates
Sep 17 20:40:33 <ant_home> no precise idea, sorry
Sep 17 20:40:55 <ant_home> I have toinsist in #gcc and #asm
Sep 17 20:41:06 <ant_home> if you see in the patch I posted th elink
Sep 17 20:41:22 <ant_home> there are the decompressed 6/7 artifacts
Sep 17 20:41:42 <ant_home> some good eye would spot it
Sep 17 20:43:36 <ant_home> I tried to find something strange in th edisassembled code but it is beyond my poor asm knowledge
Sep 17 20:43:49 <ant_home> I just checked the addresses to find misalignments
**** ENDING LOGGING AT Mon Sep 18 03:00:01 2017