**** BEGIN LOGGING AT Mon Aug 12 02:59:58 2013 Aug 12 10:18:37 blogic ping Aug 12 10:48:27 luka r37755 trunk/ package/kernel/imx target/linux/imx6/Makefile * kernel: remove imx-sdma-fw because of license issues Aug 12 10:48:27 Devastatr: hello Aug 12 10:54:25 nerniv6: Ping Aug 12 10:54:35 berniv6: Ping Aug 12 11:17:26 build #316 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/316 Aug 12 11:21:45 build #316 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/316 Aug 12 11:44:45 nbd r37756 trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_main.c * ar71xx: add missing return statement in ethernet mtu change op Aug 12 11:44:49 nbd r37757 trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_main.c * ar71xx: ethernet: do not update the unused dev->last_rx field Aug 12 11:44:52 nbd r37758 trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_main.c Aug 12 11:44:53 ar71xx: ethernet: skip calls to netdev_completed_queue and netif_wake_queue if no tx cleanup was done Aug 12 12:21:27 build #349 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/349 Aug 12 12:42:10 nbd r37759 trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_main.c * ar71xx: fix typo Aug 12 14:15:03 build #330 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/330 Aug 12 14:17:48 build #300 of sibyte is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/300 Aug 12 15:34:32 blogic hi, I've tried to send my patches in git format but this is driving me nuts, for whatever reason, it is stripping whitespaces, making it unreadable, can I pastebin it instead? or if you have a suggestion to fix this.. Aug 12 16:11:43 Devastator: what are your patches? Are you subscribed to openwrt-devel with the email address? Aug 12 16:12:35 KanjiMonster yes, it's to support a ramips based router Aug 12 16:13:16 the dir 610 one? Aug 12 16:13:19 yes Aug 12 16:14:23 when I send via e-mail, text is all aligned left, but I've just pastebin it and seems fine.. Aug 12 16:15:02 Devastator: https://lists.openwrt.org/pipermail/openwrt-devel/2013-August/020989.html looks correct to me (and applies with a bit of fuzz) Aug 12 16:15:02 Devastator: gmail strips spaces, unless you just use it as an SMTP server and let git handle the rest Aug 12 16:15:27 * MartijnVdS used git send-email, which seemed to work Aug 12 16:15:56 KanjiMonster odd, I've received that e-mail blank Aug 12 16:16:42 Devastator: there is a second one, https://lists.openwrt.org/pipermail/openwrt-devel/2013-August/020990.html Aug 12 16:17:26 MartijnVdS I'm using my isp's smtp Aug 12 16:17:35 KanjiMonster so I did screw up heh Aug 12 16:18:11 Devastator: did you maybe accidentially have two 0*.patch, one with the correct one and one empty one so git sent out both? Aug 12 16:18:44 hummm, that's it! Aug 12 16:18:50 according to the headers the empty one is a reply to the first one Aug 12 16:19:16 I tried to send a flashchip patch as well, but that wasn't in git format Aug 12 16:20:14 sorry guys Aug 12 16:20:19 no problem Aug 12 16:20:52 looks like patchwork just ignored the empty one, so no harm done Aug 12 16:21:05 with this thing solved, I guess I just have to wait for review then Aug 12 16:22:34 btw, next time use "RFC" instead of "PATCH" for the subject, to show that you are looking for comments Aug 12 16:22:51 oh, ok Aug 12 16:22:58 (you can tell git format-patch to use a custom subject with --subject="foo") Aug 12 16:23:31 any other tips or hints? :) Aug 12 16:24:37 jow_laptop: uploading photos to the wiki seems currently broken, still Aug 12 16:27:03 KanjiMonster I spent two days reading how to make it right and I almost did heheh, I don't want to piss off the devs with an ugly patch or something like that.. Aug 12 16:28:28 many many thanks for your patience, I hope John understands as well Aug 12 17:26:04 nbd r37760 trunk/target/linux/generic/files/drivers/net/phy/rtl8367.c * rtl8367: add support for configuring the VLAN FID Aug 12 17:26:17 nbd r37761 trunk/target/linux/ ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_main.c ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx.h Aug 12 17:26:17 ar71xx: ethernet: cache skb->len in the tx function to avoid accessing it again in completion Aug 12 17:26:21 nbd r37762 trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx.h * ar71xx: ethernet: reduce tx and rx DMA ring size to improve cache footprint Aug 12 17:26:26 nbd r37763 trunk/target/linux/ generic/patches-3.10/131-mips_inline_dma_ops.patch generic/patches-3.10/300-mips_expose_boot_raw.patch generic/patches-3.10/130-mips_remove_plat_dma_functions.patch Aug 12 17:26:26 kernel: add patch to inline mips dma mapping functions - reduces code size and improves performance Aug 12 19:40:03 zajec: ping Aug 12 19:41:21 zajec: i'm just reading that bcm47xx ethernet performance thread (about the alignment issue) Aug 12 19:41:29 i've dealt with that sort of thing a few times before Aug 12 19:42:43 so first question: did you guys test if the MAC is really unable to deal with buffers that have a 2-byte offset after 4-byte alignment? Aug 12 19:43:31 that would really surprise me, since that sort of thing would be a really stupid design flaw for a router Aug 12 19:43:46 i'd expect broadcom to know better than that Aug 12 19:43:54 especially since they didn't make that mistake in earlier design Aug 12 20:37:42 nbd: can you please explain how the "mode" key here is supposed to work? https://lists.openwrt.org/pipermail/openwrt-devel/2013-August/020887.html Aug 12 20:38:49 i have no idea Aug 12 20:39:32 i mean, is it supposed to generate a button event? Aug 12 20:40:19 * nbd repeats: i have no idea Aug 12 20:42:39 the kernel module differentiates a "button" from a "switch" with no further explanation Aug 12 20:42:57 your name is on the module iirc Aug 12 20:45:35 well, a button acts like a button and a switch like a switch ;) Aug 12 20:53:32 larsc: i am trying to verify that the gpio specified in the patch actually does something, because i see no feedback Aug 12 20:54:53 you'll get a similar event to button events Aug 12 20:55:32 evtest on the event file will output something when you flip the switch (if everything is working) Aug 12 20:55:40 evtest? Aug 12 20:58:37 root@OpenWrt:/# evtest Aug 12 20:58:52 /bin/ash: evtest: not found Aug 12 21:02:45 larsc: care to elaborate? Aug 12 21:06:38 a cat will also do Aug 12 21:09:07 larsc: where is the event file? Aug 12 21:09:20 /dev/input/eventX Aug 12 21:09:44 there is no /dev/input at all Aug 12 21:10:16 this is openwrt Aug 12 21:11:00 my openwrt has /dev/input Aug 12 21:11:19 must be a menuconfig option somewhere Aug 12 21:12:36 build #310 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/310 Aug 12 21:16:22 build #265 of iop32x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/265 Aug 12 21:27:22 nbd: input-utils under Utilities has a newer upstream, and you last touched the Makefile Aug 12 21:28:17 no, i didn't Aug 12 21:28:50 and i don't care about that package ;) Aug 12 21:44:21 gah, no udev **** BEGIN LOGGING AT Mon Aug 12 22:07:04 2013 Aug 12 22:30:54 tmp/.config-package.in:18558:error: recursive dependency detected! Aug 12 22:30:54 tmp/.config-package.in:18558: symbol PACKAGE_liblzma is selected by FS_WITH_LZMA Aug 12 22:30:54 tmp/.config-package.in:35519: symbol FS_WITH_LZMA depends on PACKAGE_liblzma Aug 13 00:27:42 build #268 of ep93xx is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/268 **** ENDING LOGGING AT Tue Aug 13 02:59:59 2013