**** BEGIN LOGGING AT Tue Apr 19 03:00:00 2016 Apr 19 07:14:27 bluelightning: gm Apr 19 07:23:49 hi ant_work Apr 19 07:23:55 sorry still haven't merged that stuff Apr 19 07:25:19 np, do that when you can Apr 19 07:26:22 I'm puzzled by one last issue...the zaurus-legacy tar ka6ox uploaded in OE mirrors is not gzipped so the unpacker fails :/ Apr 19 07:26:58 I wrote him, let's wait a bit or I'll have to hack the recipe temporarly Apr 19 07:27:36 see...he uploaded as zaurus-legacy.tar.gz (note the two dots) Apr 19 07:28:19 I've asked him to rename to zaurus-legacy-tar.gz and to compress it ofc :) Apr 19 07:29:54 anyway I'd just modify the do_deploy and install it as gnu-tar as it was Apr 19 07:30:33 bluelightning, one last thing: pls do tag meta-handheld once Yocto is released Apr 19 07:31:20 there is smthg wrong with the tags of old releases i.e. dora and dylan Apr 19 07:31:39 ok... Apr 19 07:31:46 wrong with tags in meta-handeld you mean? Apr 19 07:31:49 both pointing to the same commit Apr 19 07:31:55 yes Apr 19 07:31:56 http://cgit.openembedded.org/meta-handheld/ Apr 19 07:32:07 hi Apr 19 07:32:16 I noticed it when I was doing time-travel to rebuild 3.x kernels Apr 19 07:32:20 hi VartiWork Apr 19 07:32:56 hmm I think I probably just picked an arbitrary point to tag that release, perhaps it was a bad choice Apr 19 07:33:27 ant_work: no luck trying the kernel yesterday evening (real life demands :), will try this evening Apr 19 07:33:43 VartiWork, thanks Apr 19 08:50:13 ant_work : Hi ant, had no time to test yet, hopefully later this week. Apr 19 08:50:24 what is going on at OESF ? still down.. Apr 19 08:51:37 hi dall Apr 19 08:51:40 daal Apr 19 08:51:45 (sorry) Apr 19 08:51:59 Hi Varti Apr 19 08:52:43 check your PM on archlinuxarm, there's an ongoing conversation with me and Archi, I have added you in CC Apr 19 08:52:58 hi daal Apr 19 08:53:45 Michael Hubbard's email adress you found seems to be not valid anymore, it has bounced back to me Apr 19 08:53:54 daal, I have been in touch with anarsoul and he has maybe the fixes for sound Apr 19 08:53:57 https://github.com/anarsoul/linux-2.6/commits/v4.4-anarsoul Apr 19 08:54:14 I'll test some of these patches later tonight Apr 19 08:54:44 but I think I have found his tweet and blog... https://twitter.com/rikosintie http://mwhubbard.blogspot.it/ Apr 19 08:54:51 great, some progess ! Happy to test too, just give me some time Apr 19 08:55:44 Varti : ask Archi if he has some backups of the site. If it stays down, well that would be bad, but a backup of the phpBB would be good Apr 19 08:57:46 ant_work : the new build kexeboot is based on what kernel and gcc ? Apr 19 08:58:19 kernel 4.4.7 gcc 5.3.0 kexecboot OE-master (o.6 + my patches) Apr 19 08:59:11 ok, I will give it a go. Apr 19 08:59:30 I didn't flash it yet but it boots if put on a SD card Apr 19 08:59:54 hm, I did not check the size yet but should be fine ? Apr 19 09:00:52 yes, there is a size_check on OE builds Apr 19 09:01:38 I am at the limits with 1MB for collie: I had to disable F2FS Apr 19 09:02:24 on Cxxx we can disable much stuff and gain more space (libs, crypto, ...) but atm it still fits Apr 19 09:02:31 daal: ok, I'll let you know Apr 19 09:03:25 ant_work : great thanks. Apr 19 09:04:03 VartiWork : I did PM ArchiMark, too. Hope we can get hold of a dump of the database at least. Apr 19 09:10:00 daal, fwiw 4.6-rc4 contains rework of gpio-pca953x. I hope akita does work with 4.4.7 Apr 19 09:10:34 there are 9 patches between the two releases... Apr 19 09:19:57 ant_work: ok I will keep this in mind Apr 19 09:20:31 but I can not test akita, VaritWork would be the test candidate :) Apr 19 09:21:28 ant_work: does your latest kexecboot parse NAND for UBIFS only ? Apr 19 09:21:41 those days with the transition to devicetree there are too many changes to clock, gpio, dma.. :/ Apr 19 09:22:13 daal, not just that, basically all FS for SD/CF/MTD Apr 19 09:23:34 ant_work, good to know, I have not used NAND for ages and disabled it for ArchLinuxARM as it delayed the boot a bit Apr 19 09:23:56 would be good to have the feature Apr 19 09:24:04 hmm, how so? Apr 19 09:24:07 https://github.com/LinuxPDA/meta-handheld/tree/master/recipes-kernel/linux/linux-kexecboot-4.4/spitz Apr 19 09:24:20 the parsing of NAND delayed the boot a bit I believe Apr 19 09:24:26 ^ note that I compiled for my 3200 so there is the mtd patch applioed Apr 19 09:24:47 jffs2 is rather slow on first boot, yes Apr 19 09:24:58 jffs2.summary is much better Apr 19 09:25:10 ubifs is surely faster Apr 19 09:25:29 ok, my 3100 has some empty UBIFS partitions, that was fast Apr 19 09:26:05 argh..you have 3100 ...I can rebuild the specific (unpatched) version Apr 19 09:26:43 please do I you can Apr 19 09:27:11 * daal has borzoi only Apr 19 09:27:19 I am postponing it but there is the need for a kernel patch detecting mtdparts Apr 19 09:27:26 as we do in kexecboot Apr 19 09:28:15 then we'd treat spitz-borzoi-terrier with one kernel Apr 19 09:35:47 I see, previously spitz-borzoi used the same kernel ? Apr 19 09:38:17 yes, terrier/3200 is unknown in linux kernel Apr 19 09:39:35 I do 2 linux-kexcecboot: one unpatched for 3000 and 3100 and the other one patched for 3000 and 3200 Apr 19 09:40:53 we would simplify kexecboot doing the detection of mtdparts in kernel Apr 19 10:00:18 * daal just powered up his c3100 Apr 19 20:40:49 bluelightning: in the patchset there is a little typo in one comment in linux-kexecboot_4.4.bb Apr 19 20:41:25 should I add a patch or can you manually fix it? Apr 19 20:41:49 -# only for SL-C3200 (terrier): uncomment for SL-C3100 (borzoi) Apr 19 20:42:07 +# only for SL-C3200 (terrier): comment for SL-C3100 (borzoi) Apr 19 20:42:41 it seems 3200 had more diffusion so I'd take care of its case Apr 19 20:43:36 and add as default SRC_URI_append_spitz += " file://3200-mtd.patch" Apr 19 20:52:44 ant_home: hi Apr 19 20:52:57 I can fix that Apr 19 20:54:14 thanks! Apr 19 21:23:56 bluelightning: as sequel for the 'no cruft in my images' saga: Apr 19 21:24:22 - how do we opt-out from eudev 7MB hwdb? BAD_RECOMMENDS? Apr 19 21:25:21 - manifest of initramfs-kexecboot images shows extras... Apr 19 21:25:33 kexec-klibc armv5e 2.0.2-r1 Apr 19 21:25:33 kexecboot-klibc spitz 0.6-r0 Apr 19 21:25:33 run-postinsts all 1.0-r9 Apr 19 21:25:33 ubiattach-klibc armv5e 1.5.1-r0 Apr 19 21:25:33 update-rc.d all 0.7-r5 Apr 19 21:26:11 I thought we nuked update-rc and postins long ago...zombies Apr 19 21:26:44 IIRC recently the logic changed to remove them only when the rootfs was readonly rather than based on whether packaging was enabled Apr 19 21:26:59 damn, another thing I missed in the migration guide :/ Apr 19 21:27:05 heh Apr 19 21:27:08 re the hwdb, not sure, let me check Apr 19 21:30:08 yeah, BAD_RECOMMENDATIONS should be able to get rid of it Apr 19 21:30:49 in MACHINE conf files then? Apr 19 21:31:26 why didn't split the package in two... Apr 19 21:31:35 that's just it, it was split Apr 19 21:31:46 then one rdepends on other Apr 19 21:31:49 eudev-hwdb is in the RRECOMMENDS for eudev Apr 19 21:31:53 ah, ok Apr 19 21:32:06 still unsatisfactory imho Apr 19 21:32:28 do you have a better suggestion on how it should work? Apr 19 21:32:51 I'd say who needs it should install it separately Apr 19 21:33:36 in our case we badly need the 6-7MB Apr 19 21:35:27 * ant_home is finally uploading up-to-date linux-kexecboot_4.4 precompiled installers Apr 19 21:41:07 BAD_RECOMMENDATIONS can be set in the image recipe btw Apr 19 21:43:11 hm, we should need it generally Apr 19 21:43:21 for any image Apr 19 22:21:43 I will enter a bug about the other packages at least Apr 19 22:32:56 https://bugzilla.yoctoproject.org/show_bug.cgi?id=9491 Apr 19 22:32:59 Varthall: anarsoul solved right now the no-sound-at-boot issue :) Apr 19 22:33:08 bluelightning: thanks Apr 19 22:33:17 I've also updated this one: https://bugzilla.yoctoproject.org/show_bug.cgi?id=6478 since the list of files has changed slightly since that was filed Apr 19 22:48:21 heh..the saga Apr 19 23:00:07 gn **** ENDING LOGGING AT Wed Apr 20 02:59:58 2016