**** BEGIN LOGGING AT Sun Jul 10 02:59:58 2011 Jul 10 10:45:01 blogic * r27575 /trunk/package/ltq-ifxos/Makefile: Jul 10 10:45:01 wrong md5sum in ltq-ifxos Jul 10 10:45:01 Signed-off-by: Luka Perkov Jul 10 10:48:56 blogic * r27576 /branches/backfire/package/ltq-ifxos/Makefile: Jul 10 10:48:56 wrong md5sum in ltq-ifxos Jul 10 10:48:56 backport r27575 to backfire Jul 10 11:20:05 <_trine> http://www.zdnet.co.uk/news/emerging-tech/2011/07/07/dont-kill-fm-for-broadband-build-a-new-internet-40093344/ Jul 10 12:33:16 blogic: i'm looking at difference between adm6996i and adm6996m Jul 10 12:33:54 it looks like all registers are the same Jul 10 12:34:19 only thing that should be different are CI0 and CI1 registers Jul 10 12:35:43 but in target/linux/generic/files/drivers/net/phy/adm6996.c lines 613 to 628 Jul 10 12:36:04 difference between adm6996m and adm6996f is done differently Jul 10 12:37:41 if CI0 and CI1 are used do diferentiate adm6996m, adm6996f and adm6996i that should be way to go, at least that is my opinon Jul 10 12:37:46 what do yout think? Jul 10 12:45:43 luka12345|wiik: ok Jul 10 12:46:01 sound resonable Jul 10 12:51:28 blogic: i have board with adm6996m chip so i can compare the registers... i'll bug you some more if i get stuck :) Jul 10 13:22:46 updated openwrt/upstream, https://home.comcast.net/~sdwalker/uscan/index.html Jul 10 14:12:22 It appears my quilt provides a .quilt_used file, not a .quilt_checked, but during the build of the kernel source, it is erroring out while checking .quilt_checked. There are conditionals in include/quilt.mk, but i am not sure why it is triggering the check for .quilt_checked Jul 10 14:13:36 is this behavior others have seen? Is this related to my quilt version? Jul 10 14:14:00 by commenting out the lines which work with the .quilt_checked I was able to continue Jul 10 14:14:06 but that is not a cleansolution Jul 10 18:27:19 xMff: ping Jul 10 19:09:52 who maintains freeswitch= Jul 10 19:09:54 ? Jul 10 19:09:58 its broken on backfire Jul 10 19:10:24 libre / restund is also broken, its a new record, I never seen it compile since its addition Jul 10 19:10:37 I'm about to kick it out of the feed Jul 10 19:45:00 jow * r27577 /packages/net/dante/Makefile: [packages] dante: fix compilation with backfire Jul 10 19:54:31 jow * r27578 /packages/libs/libcrypto++/Makefile: [packages] libcrypto++: fix compilation on backfire by getting rid of the GNUmakefile which injects invalid -mtune=native arguments Jul 10 19:56:47 oh how I hate this configure crap Jul 10 19:56:48 checking flex version 2.5.4 >= 2.5.31... configure: error: no Jul 10 19:56:51 fail Jul 10 20:00:11 jow * r27579 /branches/backfire/tools/ (Makefile flex/): [backfire] backport r24631, needed for at least gstreamer on backfire Jul 10 20:00:38 jow_laptop: can you look at my PPP and OCF version bumps, or the Perl fix for compilation against eglibc please? Jul 10 20:01:48 the PPP update is fairly trivial... stuff that we had separate patches for in Openwrt got picked up in 2.4.5, so a lot of those patches simply became redundant. Jul 10 20:03:45 later, I'm busy with bringing backfire into release shape Jul 10 20:03:54 I do not intend to do version updates until it is done Jul 10 20:05:36 <^Willie^> jow_laptop: may i ask you something about collectd and some of the plugins ? Jul 10 20:06:13 most of what is left is patches that actually add functionality, such as the "replacedefaultroute" option. Jul 10 20:07:25 philipp64|laptop: after seing pppd fuckups like deleting default routes of foreign interfaces just because it forgot to copy the ifname into the ifreq struct I'm reluctant to bump it without thorougly testing it Jul 10 20:14:14 jow_laptop: yeah, I get that. I'm just saying that from my end, the changes were fairly minor. now am I saying that they didn't introduce major regressions in their new version? no. Jul 10 20:15:02 btw, other than Geos (or a net5501 with a Solos card), what other platforms have DSL (which can then be used to test PPPoA)? Jul 10 20:17:19 jow * r27580 /packages/multimedia/crtmpserver/Makefile: [packages] crtmpserver: properly locate toolchain on backfire Jul 10 21:50:49 jow_laptop: I'm working with the PPP maintainers to get some of these patches upstreamed, but what would help is ordering them differently: first patches that fix bugs, second patches that add functionality, third patches that aid cross-compilation, fourth patches that turn off features that aren't supported on Openwrt Jul 10 21:51:10 because obviously the 4th category will never get upstreamed, so it should be applied last. Jul 10 21:57:08 although frankly I'd rather fix that via a compile-time flag, like HAVE_WTMP, and upstream that as well. Jul 10 22:44:48 jow * r27581 /packages/utils/mdadm/patches/200-uclibc_0.9.30_compat.patch: [packages] mdadm: add missing functions posix_memalign() and canonicalize_file_name() to work with backfire/uClibc 0.9.30 Jul 10 23:15:13 jow * r27582 /packages/utils/mdadm/patches/ (100-cross_compile.patch 200-uclibc_0.9.30_compat.patch): [packages] mdadm: refresh patches Jul 10 23:17:02 jow * r27583 /packages/ipv6/gw6c/Makefile: [packages] gw6c: switch to v5.1 as used by Freetz, the v5.0 seems to have gone missing Jul 10 23:44:01 jow * r27584 /packages/libs/tokyocabinet/patches/200-uclibc_0.9.30_compat.patch: [packages] tokyocabinet: add missing function nanl() to work with backfire/uClibc 0.9.30 Jul 11 00:04:39 jow * r27585 /packages/utils/procps/patches/020_hz_fix.patch: [packages] procps: __attribute__((constructor)) needs an arg for gcc >= 4.3, fixes procps compilation on backfire Jul 11 00:15:21 jow * r27586 /packages/net/weechat/Makefile: [packages] weechat: need to link against libdl, fixes compilation on backfire Jul 11 01:03:53 jow * r27587 /packages/ipv6/mrd6/patches/004-ambiguous_call_fix.patch: [packages] mrd6: fix ambigious call errors exposed during backfire/avr32 compilation Jul 11 02:31:48 jow * r27588 /packages/ (libs/re/Makefile net/restund/Makefile): [packages] re, restund: mark as broken, its buildsystem fails completely, does not support anything besides mipsel and mixes host with target variables, attempts to fix it where unsucessful **** ENDING LOGGING AT Mon Jul 11 02:59:56 2011