**** BEGIN LOGGING AT Sun Jan 09 02:59:58 2011 Jan 09 04:23:38 03Roman I Khimov  07org.openembedded.dev * r3d030ffa57 10openembedded.git/recipes/perl/ (perl.inc perl_5.10.1.bb): (log message trimmed) Jan 09 04:23:38 perl: blacklist perl-module-cpanplus-internals-source-sqlite for perl-modules Jan 09 04:23:38 It has runtime dependencies on libdbd-sqlite-perl and libdbix-simple-perl Jan 09 04:23:38 which doesn't get built automatically if not required as build time dependency Jan 09 04:23:38 for something. Jan 09 04:23:39 It's not possible to add them as build time dependencies for Jan 09 04:23:40 perl itself since they're in turn depend on perl making a circular dependency Jan 09 08:04:02 gm Jan 09 08:15:17 is hostap-utils, hotstap-daemon enough for managing WAP on Linux? Jan 09 08:18:12 it is for WPA Jan 09 08:19:34 eFfeM: thanks Jan 09 08:20:52 I can for sure say, that hostap-modules doesn't complete. Uses some obsolete kernel config options Jan 09 08:28:28 eFfeM: You mean WPA/WPA2, or just WPA? Jan 09 08:33:14 davidlt: oops, I think i gave wrong info (didn't get my first coffee yet). Jan 09 08:33:25 I was mixing up with wpa-supplicant etc Jan 09 08:34:43 iirc hostap is for running an access point on your system so it is the other side Jan 09 08:35:01 http://hostap.epitest.fi/hostapd/ Jan 09 08:35:43 no idea how well it works today with usb wifi, usb historically was less supported or more difficult than minipci wifi cards Jan 09 08:35:54 There are 5 recipes for hostap Jan 09 08:36:10 3 of them works only on older kernels. Jan 09 08:36:23 1 on 2.4 and two other on <2.6.20 I think Jan 09 08:36:43 but hostap-daemon and utils at least compiles. Jan 09 08:37:02 so if I have a board config that "almost" works for a different board do I create a new "config" for the new board or mod the old one? Jan 09 08:38:50 ka6sox: do you mean machine.conf file? Jan 09 08:39:09 yes Jan 09 08:39:20 eFfeM: I have integrated WLAN module on by board. The problem is that current driver doesn't support AP/Master mode at all. So, trying to move to XO-1.5 Alpha version of driver. Jan 09 08:39:29 ka6sox: while you're at it, can you restore a.c.neal (if I recall the name right) page that has been accidently deleted on the wiki (see ML) Jan 09 08:39:48 the omapzoom2 is "close" but not quite the what I need. Jan 09 08:39:52 davidlt: USB wifi ? Jan 09 08:40:06 what is a.c.neal? Jan 09 08:40:14 a user or ??? Jan 09 08:40:23 a user Jan 09 08:40:48 eFfeM: it uses libertas_sdio Jan 09 08:40:55 ka6sox: see this thread: http://article.gmane.org/gmane.comp.handhelds.openembedded/41230 Jan 09 08:41:31 davidlt: and you want to make an AP out of it, guess your driver needs fixing Jan 09 08:41:45 eFfeM sledz deleted it and I think its banished. Jan 09 08:41:47 so I dunno Jan 09 08:42:22 eFfeM: the current driver doesn't support AP mode, but XO-1.5 uses the same chipset and people are reporting it working on my dev board. Jan 09 08:42:43 eFfeM: if it doesn't work I probably gonna have to replace some USB with external WLAN module. Jan 09 08:43:04 ka6sox: i think it can be undeleted, see http://www.mediawiki.org/wiki/Manual:Administrators#Deletion, 2nd dot Jan 09 08:43:24 davidlt: then talk to the people that got it working Jan 09 08:44:02 eFfeM: already did, now putting all the required changes. Jan 09 08:44:03 ka6sox: typically I see people create a different machine for different hw, see the various omap3 boards Jan 09 08:44:39 unless the changes are very small, beagle and beagle-xm seem to share the same conf (but then again xm is a bb with some addons bolted onto it) Jan 09 08:44:52 ka6sox: so it also seems to depen on what changes you need to make Jan 09 08:45:44 well, nothing I've seen is for a OMAP3621 (which the board I'm working on has) so a beagle/panda/etc doesn't have the changes I need. Jan 09 08:47:32 for Angstrom I am using the omapzoom36xx but that is more 3630 specific Jan 09 08:48:28 so I guess the answer is that I need to create a new config for it and submit that. Jan 09 08:48:58 and sledz needs to restore that page...it was an error for him to delete it. Jan 09 08:52:47 Any ideas how stable is linux-omap3 2.6.35 kernel? Jan 09 08:53:37 davidlt, I'm using it with no problems. Jan 09 08:53:49 ka6sox: thanks Jan 09 08:57:08 eFfeM do I use its common name or the board name Jan 09 08:58:46 ka6sox: I'd say board name (actually the panda guys do somethign like chip-board) Jan 09 08:58:54 don't think there is a real rule for it Jan 09 08:59:31 most machine configs seem to use the common name (e.g. sheevaplug, openrd-client etc) Jan 09 08:59:42 okay so in my case its called 3621_ev1ta Jan 09 08:59:51 (thats its defconfig name) Jan 09 09:00:06 or OMAP3621_evt1a Jan 09 09:00:49 (aka Nook Color) Jan 09 09:01:04 ka6sox: seems like a good choice Jan 09 09:01:37 so board or common? Jan 09 09:01:39 i'd go for omap3621-evt1a Jan 09 09:01:43 kk Jan 09 09:01:52 we do not have upper case in the names Jan 09 09:02:02 okay nookcolor Jan 09 09:02:05 and no _ that might cause probems with overrides Jan 09 09:02:07 also ok Jan 09 09:02:27 we have it mostly working trying to finish it up. Jan 09 09:08:59 eFfeM: is it better to write patch for deconfing (kernel configuration) or just change the file? Jan 09 09:09:53 davidlt: change it, this is machine specific add the defconfig file in linux*/yourmachine/defconfig Jan 09 09:09:58 see how others do this Jan 09 09:10:13 This is what I did right now Jan 09 09:11:26 eFfeM the kernel we have is 2.6.29 can we pin the board to that or do we have to bring it up to latest? Jan 09 09:17:34 you can pin the board if you want to, if you are the user you are in control Jan 09 09:18:13 although maybe people like koen would prefer that you do not pin the kernel version, but just select the preferred recipe for virtual/kernel and use default preference to get the version you want Jan 09 09:18:36 afk 4 tea Jan 09 09:19:00 as far as I know we are the first to work on linux on 3621...android so far. Jan 09 19:44:36 hi hrw Jan 09 19:45:27 woglinde: hey Jan 09 19:45:35 he zecke Jan 09 19:45:43 * woglinde is watching basketball Jan 09 19:45:47 hey zecke did anyone answer your gnuradio question? Jan 09 19:46:07 Crofton|work: not that i know of. Is the question too stupid? Jan 09 19:46:16 I do not think so Jan 09 19:46:17 Crofton|work: how is your neon code coming along? Jan 09 19:46:28 I'll know in a few minutes Jan 09 19:46:42 may have outline for ccf fir filter working Jan 09 19:46:45 Crofton|work: I was good to myself and bought a USRP1 (i should have just bought the WBX frontend but I thought it does not work...) Jan 09 19:46:48 will you be at fosdem? Jan 09 19:46:53 Crofton|work: yes Jan 09 19:46:57 awesome Jan 09 19:47:05 I will have an E100 Jan 09 19:47:24 Crofton|work: Signal Processing is incredible... Jan 09 19:47:33 zecke hm Jan 09 19:47:39 you are now at fosdem Jan 09 19:47:43 you want something that has a work function in python Jan 09 19:47:44 or going Jan 09 19:47:51 yes, it is Jan 09 19:47:57 Crofton|work: i spend some time with just a bit wise CRC... and I felt so stupid Jan 09 19:48:00 it is math with a purpose Jan 09 19:48:35 I'm not sure I can make crc work Jan 09 19:48:35 hi florian Jan 09 19:48:52 crofton is learning asm Jan 09 19:49:12 I prefer to think of it as "remembering asm" Jan 09 19:49:22 Crofton|work: we had Reed-Muller in a math class, but now people use a generalized reed muller that is not documented in any literature. :) Jan 09 19:49:43 zecke hm oh Jan 09 19:50:11 I need to find a coding theory for dummies book Jan 09 19:50:11 woglinde: and i am too stupid to remember the decoding rule. :) Jan 09 19:50:36 Crofton|work: well, the math is easy... you have proofs for stuff.. but it does not help in practice. :) Jan 09 19:50:42 hi all Jan 09 19:50:45 exactly Jan 09 19:50:59 Crofton|work: now I need to figure out IQ... Costa-Loops... Muller-Muller.. :} Jan 09 19:51:04 I feel stupid once again Jan 09 19:51:05 a friend of mine teaches that class, but I only want to make it work, not prove it :) Jan 09 19:52:01 you'll be smarter than me soon :) Jan 09 19:56:01 hi, is midori considered core? Jan 09 19:56:10 I want to add a midori_git Jan 09 19:56:15 that fixes a segfault Jan 09 20:00:57 03Thomas Zimmermann  07org.openembedded.dev * r22b30a19a0 10openembedded.git/recipes/shr/shr-e-gadgets_git.bb: Jan 09 20:00:57 shr-e-gadgets: update SRCREV Jan 09 20:00:57 *update SRCREV to get GSM gadget fix Jan 09 20:00:57 Signed-off-by: Thomas Zimmermann Jan 09 20:01:40 woglinde, hi Jan 09 20:05:49 JaMa, hi can I push newer midori and xkeyboard-config (git versions) to oe ? Jan 09 20:05:51 or should I send to the ml? Jan 09 20:06:15 for midori it fixes huge segfaults: Jan 09 20:06:32 you load a page and click on a link while it's loading and it segfaults Jan 09 20:06:37 and various other segfaults Jan 09 20:06:47 for xkeyboard-config it adds htcdream support Jan 09 20:07:31 I did quite some work to upstream the htcdream keyboard....it's not complete yet(no extra keys like the arrows) but the alt keys like _ - ? | are mapped Jan 09 20:07:39 tab has an issue with gtk+ Jan 09 20:08:28 GNUtoo|laptop: imho it would be good idea to go through ML Jan 09 20:08:46 ah ok Jan 09 20:08:47 or commit with DP=-1 then post request to ML about promoting to default Jan 09 20:08:59 ok I'll send to ml Jan 09 20:09:23 push with dp-1 first Jan 09 20:09:24 I push some other stuff that only touch htcdream, and get thoses on ml Jan 09 20:09:29 ok Jan 09 20:11:21 maybe I put DP -1 for all but not for SHR Jan 09 20:44:28 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * r4092253536 10openembedded.git/conf/machine/htcdream.conf: Jan 09 20:44:28 htcdream: load the wl1251_sdio (the wifi driver) module by default. Jan 09 20:44:28 Signed-off-by: Denis 'GNUtoo' Carikli Jan 09 20:44:30 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * r967a3dde43 10openembedded.git/recipes/shr/shr-splash-theme-logo_git.bb: Jan 09 20:44:31 shr-splash-theme-logo: bump SRCREV(fixes htcdream splash screen) and PR Jan 09 20:44:31 Without that SRCREV bump, the htcdream has a wrong boot splash Jan 09 20:44:31 screen, in fact the old splash screen was made for the palm pre Jan 09 20:44:31 which doesn't have the exact same framebuffer(while having the Jan 09 20:44:31 same resolution). Jan 09 20:44:32 Signed-off-by: Denis 'GNUtoo' Carikli Jan 09 20:44:33 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * rc5b4df259e 10openembedded.git/recipes/xkeyboard-config/xkeyboard-config_git.bb: (log message trimmed) Jan 09 20:44:33 xkeyboard-config: add git version(has htcdream keyboard support) Jan 09 20:44:33 This version add the (upstream) support for htc dream. Jan 09 20:44:34 The support for the htcdream is composed the 'us','it','de' Jan 09 20:44:34 keyboards and has the extra-keys like | - _ etc... mapped. Jan 09 20:44:35 But no custom keys(keys that weren't printed on the keyboard Jan 09 20:44:54 both are DP -1 Jan 09 20:45:11 the midori and the xkeyboard-config Jan 09 21:38:17 GNUtoo|laptop: yes go ahead Jan 09 21:43:58 03Khem Raj  07master * r8d51dd82e8 10openembedded.git/recipes/eglibc/ (eglibc-options.inc eglibc.inc): Jan 09 21:43:58 eglibc: Make it configurable by tuning eglibc options Jan 09 21:43:58 * Look for disabled features in DISTRO_FEATURES and disable them while building eglibc Jan 09 21:43:58 * By default all features are built Jan 09 21:43:58 * One can disable a feature by specifying DISTRO_FEATURES += "noinet6 nobsd ...." e.g. Jan 09 21:43:58 Signed-off-by: Khem Raj Jan 09 21:44:08 03Khem Raj  07master * r8151bc5dd1 10openembedded.git/conf/distro/include/sane-toolchain.inc: Jan 09 21:44:09 sane-toolchain.inc: GCC 4.5 works for SH3/SH4 now Jan 09 21:44:09 * Dont set PREFERRED_GCC_VERSION for SH Jan 09 21:44:09 Signed-off-by: Khem Raj Jan 09 21:44:10 03Khem Raj  07master * r3aa8afe97e 10openembedded.git/recipes/gcc/ (gcc-4.5.inc gcc-4.5/sh4-multilib.patch gcc-cross4.inc): Jan 09 21:44:10 gcc-4.5: Fix toolchain builds for SH4/SH3 Jan 09 21:44:10 Signed-off-by: Khem Raj Jan 09 21:44:31 khem, I'll send a mail, sorry I was busy Jan 09 21:44:36 now I'm not busy anymore Jan 09 21:45:30 GNUtoo|laptop: ok np Jan 09 21:45:49 by GNUtoo|laptop: yes go ahead you mean about what? Jan 09 21:45:56 sending to ml and pushing dp -1? Jan 09 21:46:01 or pushing without dp -1 Jan 09 21:46:02 ? Jan 09 21:50:07 hmm... my arm build still fails Jan 09 22:00:38 GNUtoo|laptop I always encourage to test well and not use DP -1 Jan 09 22:00:52 and use DP = -1 for needed arches only Jan 09 22:01:41 khem, I tested well but only under what I had Jan 09 22:01:52 that is to say htcdream+SHR Jan 09 22:02:05 midori was also tested on nokia n900 Jan 09 22:02:39 (I'm afraid of breaking something core and beeing pm-ed by koen :) ) Jan 09 22:12:07 03Klaus Kurzmann  07org.openembedded.dev * r8e739658fb 10openembedded.git/recipes/freesmartphone/libfso-glib_git.bb: Jan 09 22:12:07 libfso-glib: bump SRCREV Jan 09 22:12:07 Signed-off-by: Klaus Kurzmann Jan 09 22:12:11 03Klaus Kurzmann  07org.openembedded.dev * rf1ac34beae 10openembedded.git/recipes/freesmartphone/cornucopia.inc: Jan 09 22:12:12 cornucopia.inc: bump SRCREV Jan 09 22:12:12 Signed-off-by: Klaus Kurzmann Jan 09 22:17:19 03Khem Raj  07master * r198ced43fd 10openembedded.git/recipes/apex/ (apex-env_1.5.14.bb files/invalid-conversion.patch): Jan 09 22:17:19 apex-env_1.5.14.bb: Fix compile error discovered with gcc 4.5 Jan 09 22:17:19 Signed-off-by: Khem Raj Jan 09 22:17:21 03Paul Menzel  07master * ra38f097831 10openembedded.git/recipes/xbmc/xbmc_svn.bb: (log message trimmed) Jan 09 22:17:22 xbmc: add `python` to `DEPENDS` Jan 09 22:17:22 XBMC needs Python to be built properly [1][2]. Python is pulled in implicitly [3], so no error is seen. Make it explicit though to be safe. Jan 09 22:17:22 [1] https://github.com/xbmc/xbmc/blob/e9e7990b428863b8f2daad4dc5fd2b0a320aa965/Makefile.in#L475 Jan 09 22:17:22 [2] http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/xbmc/xbmc/0006-Hardcode-python2.6-for-now.patch?id=0cc02ded7bd429208ccb0e0e0aa3e43aec8a4722 Jan 09 22:17:22 [3] http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/boost/boost-with-bjam.inc?id=d470cfde5bd1d4c9ab23a37d286456726711229f#n67 Jan 09 22:17:24 Signed-off-by: Paul Menzel Jan 09 23:52:30 03Khem Raj  07master * rf0a4b900ea 10openembedded.git/recipes/uclibc/ (8 files in 2 dirs): Jan 09 23:52:30 uclibc_git.bb: Add patches for sub arch variant removal. Jan 09 23:52:30 * Also Bump to latest revision Jan 09 23:52:30 Signed-off-by: Khem Raj Jan 10 00:08:29 03Khem Raj  07master * r6d58ac4a1c 10openembedded.git/recipes/qemu/ (3 files in 2 dirs): Jan 10 00:08:29 qemu_0.12.5.bb: Backport patches from upstream to fix qemu for SH4 Jan 10 00:08:29 * Now we can boot SH4 qemu again with these patches and gcc 4.5 compiled Jan 10 00:08:29 kernel/root file system Jan 10 00:08:29 Signed-off-by: Khem Raj Jan 10 00:12:25 03Andrea Adami  07org.openembedded.dev * rb15fe5d14e 10openembedded.git/recipes/kexecboot/ (kexecboot-klibc_git.bb kexecboot.inc kexecboot_git.bb): Jan 10 00:12:26 kexecboot: bump to 92403e305a57154c7c779d6d47600387116a7988 Jan 10 00:12:26 * update list of configure args Jan 10 00:12:26 * bump PR Jan 10 00:12:26 Signed-off-by: Andrea Adami Jan 10 00:12:40 03Andrea Adami  07org.openembedded.dev * rc33e7bb2b6 10openembedded.git/recipes/linux/linux-kexecboot.inc: Jan 10 00:12:40 linux-kexecboot: bump PR to follow kexecboot changes. Jan 10 00:12:40 Signed-off-by: Andrea Adami **** ENDING LOGGING AT Mon Jan 10 02:59:58 2011