**** BEGIN LOGGING AT Tue Aug 25 02:59:59 2015
Aug 25 06:28:33 <OpenWrt-Commits> cyrus r46718 trunk/package/network/config/netifd/Makefile * netifd: various updates
Aug 25 06:39:58 <builder> build #75 of mpc52xx is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/75
Aug 25 07:25:50 <OpenWrt-Commits> nbd r46719 trunk/package/network/config/netifd/Makefile * netifd: update to the latest version, adds multicast-to-unicast fixes
Aug 25 07:26:04 <OpenWrt-Commits> nbd r46720 trunk/target/linux/ generic/patches-4.1/645-bridge_multicast_to_unicast.patch generic/patches-3.18/645-bridge_multicast_to_unicast.patch
Aug 25 07:26:04 <OpenWrt-Commits> Revert "kernel: disable multicast-to-unicast translation for ipv6 neighbor solicitation (#17625)"
Aug 25 07:26:14 <OpenWrt-Commits> nbd r46721 trunk/target/linux/ generic/patches-4.1/645-bridge_multicast_to_unicast.patch generic/patches-3.18/645-bridge_multicast_to_unicast.patch * kernel: bridge, multicast-to-unicast: assign src after pskb_may_pull()
Aug 25 07:46:35 <OpenWrt-Commits> nbd r46722 trunk/rules.mk * rules.mk: remove GCC 4.4 and GCC 4.5 conditional
Aug 25 07:47:30 <OpenWrt-Commits> nbd r46723 trunk/scripts/patch-kernel.sh
Aug 25 07:47:30 <OpenWrt-Commits> scripts/patch-kernel.sh: remove -E flag to preserve empty files touched by patches
Aug 25 08:09:14 <builder> build #81 of pxa is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/81
Aug 25 10:50:23 <builder> build #81 of ppc40x is complete: Failure [failed shell_4 shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/81
Aug 25 15:21:43 <OpenWrt-Commits> rmilecki r46724 trunk/package/kernel/mac80211/ (8 files) * mac80211: backport latest patches except for NVRAM support
Aug 25 15:28:08 <rmilecki> oh, i missed "brcmfmac" word
Aug 25 16:35:07 <karlp> bleh, remote gdb is totally failing to follow forks or something.  silly snmp agents
Aug 25 16:45:23 <plntyk> https://sourceware.org/gdb/onlinedocs/gdb/Forks.html
Aug 25 16:45:37 <plntyk> show follow-fork-mode
Aug 25 17:16:01 <plntyk> rmilecki: regarding the mac80211 patches - is it possible that *next* time you add an additional "*upstream*" to the changes - those rt2x00 patches for example arent upstream (not upstreamable?)
Aug 25 17:18:27 <plntyk> so there is some differentiation between upstream/not yet upstream/openwrt SoC or arch patches/openwrt specific "bug" or compatibility hacks
Aug 25 17:59:26 <builder> build #78 of mxs is complete: Failure [failed compile_5]  Build details are at http://buildbot.openwrt.org:8010/builders/mxs/builds/78
Aug 25 19:11:19 <rmilecki> plntyk: i don't understand
Aug 25 19:11:30 <rmilecki> "additional "*upstream*" to the changes
Aug 25 19:11:34 <rmilecki> what do you mean?
Aug 25 19:14:15 <plntyk> rmilecki: so the patchname is composed like <number>-<subsystem>-<upstream|not upstream|(missing)>-<patchname>
Aug 25 19:14:39 <rmilecki> plntyk: 3xx are upstreamed
Aug 25 19:15:25 <plntyk> ah was that documented on the dev wiki ?
Aug 25 19:15:55 <rmilecki> i think it's undocumented
Aug 25 19:16:25 <rmilecki> and i'm not sure if there is a reason to differ this naming schema from target/linux/generic/patches/
Aug 25 19:16:27 <rmilecki> nbd: ?
Aug 25 19:16:29 <plntyk> i remember reading something about the numbers the patches are "supposed" to have but cannot find it either
Aug 25 19:16:43 <rmilecki> maybe we should follow target/linux/generic/patches/ and use 0xx for upstream backports
Aug 25 19:17:18 <rmilecki> plntyk: target/linux/generic/PATCHES
Aug 25 19:21:55 <plntyk> ah k - I c/p it and linked it into http://wiki.openwrt.org/doc/devel/patches too
Aug 25 19:24:11 <plntyk> rmilecki: I was just suggesting *some* change to make it easier for either new people / ppl not knowing *all* internals of various changes to sort through the quite big changeset that openwrt still apply on top of mac80211 WITHOUT reading / digging through commit logs
Aug 25 19:25:06 <rmilecki> sure
Aug 25 19:25:14 <rmilecki> i'm fine with documenting this
Aug 25 19:25:28 <rmilecki> let's first see if we could make mac80211 follow kernel patches naming
Aug 25 19:25:36 <rmilecki> i think nbd me suggest if this is OK or now
Aug 25 19:25:38 <rmilecki> *or not
Aug 25 19:29:30 <builder> build #81 of x86.64 is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/x86.64/builds/81
Aug 25 20:47:11 <nbd> rmilecki: if you want to rework mac80211 patches, please make a detailed proposal and i will review it
Aug 25 20:47:47 <nbd> for patches that i'm tracking in my git (backported + pending patches) i would like to use a single prefix
Aug 25 20:50:30 <nbd> without separating the pending vs upstreamed status
Aug 25 21:05:59 <rmilecki> nbd: what's your git?
Aug 25 21:06:03 <rmilecki> nbd: something private?
Aug 25 21:11:59 <nbd> yes
Aug 25 21:12:25 <nbd> i just use it as staging ground to generate patches
Aug 25 21:13:03 <karlp> plntyk: yeah, it wasn't follow fork mode, I'd already learnt that one :)  I rebuilt it again, and the bug I was trying to look at went away anyway. :<
Aug 25 21:18:39 <rmilecki> nbd: i guess we can share backports & pending anyway
Aug 25 21:18:43 <rmilecki> it makes more sense
Aug 25 21:18:58 <rmilecki> since moving 1xx to 0xx for kernel patches doesn't work too well anyway ;)
Aug 25 21:33:16 <KanjiMonster> rmilecki: nbd: yeah, assigning 0xx or 1xx when the patch is currently pending or already accepted somewhat works, but moving stuff from 1x to 0x or >=2x if it was rejected doesn't
Aug 25 21:33:54 <KanjiMonster> I think most of the patches I removed when updating 4.1 to 6 were from 1xx
Aug 25 22:05:57 <OpenWrt-Commits> jogo r46725 trunk/target/linux/ brcm63xx/patches-4.1/415-MIPS-BCM63XX-export-the-attached-flash-type.patch brcm63xx/patches-4.1/010-4.3-01-spi-bcm63xx-hsspi-add-support-for-dual-spi-read-writ.patch brcm63xx/patches-4.1/411-MIPS-BCM63XX-Register-SPI-flash-if-present.patch * brcm63xx: enable dual rx/tx spi support for hsspi
Aug 25 22:10:14 <builder> build #75 of adm5120 is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/75
Aug 25 22:25:20 <rmilecki> nbd: KanjiMonster: what's up with rt2x00 patches anyway?
Aug 25 22:25:26 <rmilecki> there are plenty of them
Aug 25 22:25:36 <rmilecki> but they were not upstreamed I guess
Aug 25 22:25:39 <rmilecki> who added them?
Aug 25 22:25:43 <rmilecki> can we do sth about this?
Aug 25 22:25:51 <rmilecki> they are there for years I guess
Aug 25 22:25:59 <rmilecki> and it looks like sth upstreamable
Aug 25 22:26:41 <KanjiMonster> no idea
Aug 25 22:26:55 <KanjiMonster> looks like there were from juhosg
Aug 25 22:27:30 <KanjiMonster> rmilecki: I guess they got some change requests, but ultimately weren't resubmitted or so
Aug 25 22:35:57 <rmilecki> uh, I've bad experience pinging Gabor :/
Aug 25 22:37:17 <swalker> has anyone looked into the ar71xx boost bustage?
Aug 25 22:54:04 <plntyk> swalker: seems like a normal musl include issue
Aug 25 22:54:24 <plntyk> regarding first error: include/bits/fenv.h:#define FE_DIVBYZERO  32
Aug 25 22:55:56 <plntyk> all other errors/macros are defined in the same file
Aug 25 22:56:30 <builder> build #83 of malta is complete: Failure [failed shell_4]  Build details are at http://buildbot.openwrt.org:8010/builders/malta/builds/83
Aug 25 22:59:37 <swalker> plntyk: http://buildbot.openwrt.org:8010/broken_packages/octeon/boost/compile.txt also, uClibc
Aug 25 23:00:24 <plntyk> well thats error: #error "platform not supported"
Aug 25 23:00:43 <plntyk> so @!TARGET_octeon or so
Aug 25 23:00:55 <plntyk> :)
Aug 25 23:01:33 <swalker> damn irssi windows :)
Aug 25 23:02:16 <plntyk> might be an issue with boost requiring C++ special sauce and octeon using 4.6 gcc
Aug 25 23:03:35 <swalker> I still haven't see that toolchain requirement, been using 4.{8,9}-linaro on erpoe for awhile now
Aug 25 23:05:20 <swalker> fwiw, boost error isn't seen on x86_64 musl 4.9-linaro
Aug 25 23:05:27 <plntyk> toolchain/gcc/Config.in : default GCC_USE_VERSION_4_6_LINARO if TARGET_octeon
Aug 25 23:05:49 <swalker> right, local edits to remove that :)
Aug 25 23:11:50 <plntyk> swalker: I'm just wondering why octeon has "o64" - mabi=64 is different from mabi=o64 from gcc commandline options
Aug 25 23:13:59 <plntyk> but I found no "special" handling at buildroot which does support mips64/octeon too
Aug 25 23:15:04 <plntyk> http://git.buildroot.net/buildroot/tree/package/boost/boost.mk sets BOOST_ABI ... to sysv
Aug 25 23:19:07 <plntyk> going to bed - n8
Aug 25 23:44:15 <lgierth> are there ideas how to proceed with the wiki, in the future?
Aug 25 23:44:44 <lgierth> building a static page from a git repository might help develop tooling around the docs
Aug 25 23:45:15 <lgierth> device templates, formatting, linting, such things
Aug 25 23:45:36 <lgierth> would also make it trivial to include the documentation with releases
Aug 25 23:47:23 <lgierth> i'm suggesting this because i'd be up to give it a try
Aug 26 01:24:59 <builder> build #89 of ramips is complete: Failure [failed shell_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/89
**** ENDING LOGGING AT Wed Aug 26 02:59:59 2015