**** BEGIN LOGGING AT Mon Sep 05 02:59:57 2011 Sep 05 08:50:24 hi bluelightning Sep 05 08:50:31 hi ant_work Sep 05 08:52:09 so, we just we have a couple of fixes for meta-hh remaining (Zaurus-centric), then you'll be a free man ;) Sep 05 08:54:55 ant_work: heh, I'm not sure I believe that :) Sep 05 08:55:15 he eh Sep 05 09:02:26 I'm looking for instructions about use of PACKAGES-DYNAMIC but can't find any, only rare examples grepping in the source Sep 05 09:04:31 it's unclear to me the relation between PACKAGES and PACKAGES-DYNAMIC Sep 05 09:05:11 well, or how to extend PACKAGES =+ while using PACKAGES_DYNAMIC Sep 05 09:09:46 ant_work: PACKAGES_DYNAMIC really just provides a hint for dependency resolution Sep 05 09:12:15 see, we have klibc_xy.bb declaring PACKAGES = "libklibc libklibc-dev". This is wrong, we need to add the various klibc-utils-$PN Sep 05 09:12:27 but, doing PACKAGES =+ "libklibc libklibc-dev" Sep 05 09:12:37 pulls in a bougus klibc package Sep 05 09:13:02 (named as $PN ofc) Sep 05 09:13:57 the issue is the klibc_xy.bb recipe tries to package the klibc-utils too. Sep 05 09:14:13 if I package those separately like for klibc-static-utils, then no issues Sep 05 09:14:56 but I'm not sure such split is desired, even the contrary Sep 05 09:16:58 but... things are complicated by the fact we split the utils one by one Sep 05 09:17:05 I'm not quite following your explanation, but PACKAGES_DYNAMIC won't help with that I don't think Sep 05 09:17:29 it avoids a long list of declarations in the recipe Sep 05 09:18:34 such it was in oe-dev http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/klibc/klibc-utils.inc?h=org.openembedded.dev Sep 05 09:19:48 so, instead of having Sep 05 09:19:51 PACKAGES += "${KLIBC_UTILS_PKGNAME}-sh \ Sep 05 09:20:04 ... Sep 05 09:20:06 FILES_${KLIBC_UTILS_PKGNAME}-sh = "${base_bindir}/sh" Sep 05 09:20:23 we have some python code Sep 05 09:20:56 and this works for klibc-static-utils_xy.bb Sep 05 09:26:33 morning! Sep 05 09:26:51 hello lumag_work Sep 05 09:34:30 hi lumag_work **** BEGIN LOGGING AT Mon Sep 05 11:28:16 2011 Sep 05 13:11:07 ant_work: ping Sep 05 13:11:34 pong, hi Sep 05 13:11:52 ant_work: I've added devtmpfs Sep 05 13:12:14 w/o fiddling with filesystems code as before Sep 05 13:12:23 just mounting before procfs Sep 05 13:12:35 and --disable-devtmpfs to disable :) Sep 05 13:13:08 will push today Sep 05 13:13:32 lumag: your code is commited as well Sep 05 13:13:50 * Jay7 is looking now at pfalcon's patches Sep 05 13:14:51 I did some progress splitting the things but there are really lot of #ifdef Sep 05 13:15:52 may be things should be shuffled to reduce ifdefs Sep 05 13:16:02 nothing done about fallback Sep 05 13:16:35 we need cleanup our terminal processing code.. Sep 05 13:17:17 disable evgrab and flush input on stdin at least Sep 05 13:17:48 pfalcon done this but I'm unsure it's enough Sep 05 13:41:47 bbl Sep 05 14:55:20 * Jay7 gone Sep 05 20:16:38 evening Sep 05 22:15:31 lumag: night :) Sep 05 22:16:07 * Jay7 is fighting with saa7134-alsa Sep 05 22:16:17 Hi Jay7 Sep 05 22:16:34 something inside is switching sound off periodically when recording from composite.. Sep 05 22:16:47 strange. Sep 05 22:17:02 may be automute AI is wrong Sep 05 22:17:20 bad thing - seems there is no way to completely disable it Sep 05 22:18:05 now I have workaround :) Sep 05 22:18:13 watch --interval=1 amixer .... :) Sep 05 22:18:52 v4lctl/v4l2-ctl does not affect volume/audio inputs at all Sep 05 22:19:02 well.. offtopic anyway Sep 05 22:32:16 well.. time to push changes Sep 05 22:55:53 kexecboot: jay4mail master * rv0.5-183-gec06876 / (configure.ac kexecboot.c): Sep 05 22:55:54 kexecboot: Mount devtmpfs after starting as init process Sep 05 22:55:54 kexecboot: Devtmpfs is enabled by default now. To disable use --disable-devtmpfs Sep 05 22:55:54 kexecboot: configure option. Sep 05 22:55:56 kexecboot: dbaryshkov master * rv0.5-179-gfc5ad2b /.gitignore: Sep 05 22:55:56 kexecboot: Add lots of autogenerated files to .gitignore Sep 05 22:55:56 kexecboot: Signed-off-by: Dmitry Eremin-Solenikov Sep 05 22:55:57 kexecboot: jay4mail master * rv0.5-184-g122708b / (evdevs.c util.c): Don't grab evdev's exclusively. Flush input on start/exit instead. Sep 05 22:55:57 kexecboot: dbaryshkov master * rv0.5-180-ga246445 /configure.ac: Sep 05 22:55:57 kexecboot: configure.ac: add a little bit of cross-compilation glue Sep 05 22:55:58 kexecboot: Signed-off-by: Dmitry Eremin-Solenikov Sep 05 22:55:58 kexecboot: dbaryshkov master * rv0.5-182-gf24fbf3 /devicescan.c: Sep 05 22:55:59 kexecboot: devicescan: correct partitions size priting Sep 05 22:55:59 kexecboot: Partition size is stored in unsigned long long. Kexecboot uses %d, which Sep 05 22:56:00 kexecboot: assumes plain int. So, in some circumstances the partition size will be Sep 05 22:56:00 kexecboot: printed incorrectly. Fix that by using %llu format specification. Sep 05 22:56:02 kexecboot: Signed-off-by: Dmitry Eremin-Solenikov Sep 05 22:56:03 kexecboot: dbaryshkov master * rv0.5-181-g9a364c6 /evdevs.c: Sep 05 22:56:03 kexecboot: Add some debug code to evdevs Sep 05 22:56:47 ant__: kexecboot should be devtmpfs compatible now :) Sep 05 22:57:16 btw, I've not tested that code in runtime.. Sep 05 22:57:23 only compiled once Sep 05 23:00:09 great, thx Sep 05 23:00:14 we will test Sep 05 23:00:41 \o/ klibc packages almost perfectly now... Sep 05 23:01:38 nice :) Sep 05 23:03:26 I still have a doubt... debian packages sh.shared (dash)and kinit.shared *plus* the 2 static in the shared utils.. Sep 05 23:03:40 should I rebase jay7/debug branch? Sep 05 23:03:47 why the two statics? for convenience? Sep 05 23:03:51 or just drop it? Sep 05 23:03:54 hm.. Sep 05 23:04:00 something hanging? Sep 05 23:04:16 there was only menu debugging code Sep 05 23:04:24 to catch that NULL Sep 05 23:04:36 maybe you could merge that patch too... Sep 05 23:05:25 the code is triggering some strange bug, maybe the input event Sep 05 23:05:45 now seems correct but we know there is resident evil in th ebasement :) Sep 05 23:07:02 I have other private branch with menu rewrite.. Sep 05 23:07:09 but it looks ugly.. Sep 05 23:07:14 I mean code :) Sep 05 23:07:30 so I'm in doubt now Sep 05 23:19:18 * Jay7 -> sleep Sep 05 23:21:42 gn **** ENDING LOGGING AT Tue Sep 06 02:59:56 2011