**** BEGIN LOGGING AT Sat Feb 29 02:59:59 2020 Feb 29 03:01:10 why is kergoth still online at 3am? Feb 29 03:01:22 sleep is for the weak Feb 29 03:01:25 :) Feb 29 03:02:31 well, so much for that theory Feb 29 03:03:49 how to waste another half a day... I hope not Feb 29 03:05:31 WTF at-spi2-atk why are you ignoring the patch? Feb 29 03:05:43 ^-native Feb 29 03:06:29 also does not work: http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/commit/?h=timo/a11y-upgrades&id=1395901596a3bb32d93848483628831b11895922 Feb 29 04:15:51 smoking gun: SRC_URI = "${GNOME_MIRROR}/${GNOMEBN}/${@gnome_verdir("${PV}")}/${GNOMEBN}-${PV}.tar.${GNOME_COMPRESS_TYPE};name=archive" Feb 29 04:16:16 so SRC_URI += before that is overwritten Feb 29 04:16:37 ^gnomebase.bbclass Feb 29 04:16:48 * moto-timo shakes fist Feb 29 04:17:04 why on earth did it work for class-target though? Feb 29 12:55:12 smurray: sorry about the delay, I've fixed up the psplash recipe to work with latest git Feb 29 12:55:18 (in -next) Feb 29 14:23:49 RP: okay, I'll rebase off that. Heh, I see you made the same fix to the psplash Makefile.am as I was going to propose Feb 29 14:25:12 RP: one thing that I had a question about was that there's still a psplash-poky-img.h in psplash that maybe should be removed if we're always regenerating? Feb 29 14:26:16 RP: one benefit of doing that is the .h from the recipe could then be copied to $B instead of $S, which seems cleaner Feb 29 14:41:09 smurray: yes, that sounds like it makes sense Feb 29 14:41:45 smurray: I'd been meaning to try and sort this out for a while, just so many different fires to fight :( Feb 29 14:42:05 RP: I noticed that the original patch to the ml removed that .h, but that bit was left out of the actual commit, and only the bar .h was removed Feb 29 14:43:13 RP: I'll test a bit more here and post changes later today or tomorrow Feb 29 14:44:11 smurray: I think I had 'fun' getting various patches to apply together so something probably got lost :/ Feb 29 14:44:24 smurray: I appreciate the help, thanks! Feb 29 14:44:53 RP: no problem, this is at least something I can wrap my brain around ;) Feb 29 15:25:30 RP: sent few fixes to ml based on your email it shoould help fiix ltp and make/mips issues Feb 29 15:26:42 khem: thanks. I did hack my fix into -next which is in a build that is running. I'll then get to the others after that Feb 29 15:27:52 khem: we should drop the make workaround to test I guess Feb 29 15:30:16 yes Feb 29 15:30:38 works on qemumips Feb 29 15:31:22 I was running into riscv failures as well n qemu so this should be fixed too Feb 29 15:32:45 RP: since one fix is glibc patch upgrade perhaps you should cancel build and restart Feb 29 15:32:54 since it will rebuild everything Feb 29 15:33:04 will save some time Feb 29 15:41:34 khem: well, the current build isn't rebuilding everything! :) Feb 29 15:42:49 ok I guess it will test rest of fixes so not bad Feb 29 15:52:31 khem: had I know the other patch was coming... :) We'll get there. Things are looking better for the M3 state Feb 29 15:55:08 yeah I was sort of not following up ml so well yesterday Feb 29 15:55:42 was trying to get get to riscv issue at hand Feb 29 15:57:15 which is in bochs-drm crash when x11 starts so was trying glibc syscall patch which did not fix bochs issue but did fix the make issue so good byproduct Feb 29 16:20:47 khem: queued up that build with some other fixes in too Feb 29 16:23:20 perfect Feb 29 17:11:18 Hi guyss, do you happen to know if InnoDB on warrior's mariadb actually works? Feb 29 18:53:57 RP: we need to relax GLIBCVERSION as well, I send another patch Feb 29 18:54:10 to fix that alone, please apply it on top of glibc patch Feb 29 21:21:01 Can we use golang new modules system inside a Yocto recipe? Feb 29 21:21:21 To avoid specify all the dependencies in the Yocto recipe Feb 29 22:07:00 kanavin_home: the problem is worse than I thought. I left memres enabled in my local.conf and was wondering why selftest was behaving very oddly :( Feb 29 22:08:27 completely unusable there :/ Feb 29 22:08:45 * RP wishes he hadn't seen this Feb 29 22:12:08 RP: right, so memres is altogether broken, and maybe the patch should go in after all to at least resolve the selftest failures? Feb 29 22:12:46 bitbake issues are understood by very few people, and I'm not one of them :( Feb 29 22:13:05 kanavin_home: right, that worries me quite a bit :( Feb 29 22:13:34 kanavin_home: I think if we can find the right place to add the right reset call we should be able to fix this Feb 29 22:13:56 If the right reset function exists Feb 29 22:14:11 RP: what happens when tinfoil shuts down? Feb 29 22:14:28 the challenge is a loaded system can behave like memres is enabled :/ Feb 29 22:14:43 kanavin_home: the bitbake server side should see a new tinfoil clients as a new controlling UI Feb 29 22:18:56 kanavin_home: it looks like cooker's clientComplete() should be called between tinfoil sessions. I wonder if that actually happens though Feb 29 22:19:30 and cooker.clientComplete() doesn't call cooker.reset() which may be an oversight **** ENDING LOGGING AT Sun Mar 01 02:59:57 2020