**** BEGIN LOGGING AT Sun Jul 06 02:59:59 2014 Jul 06 03:44:10 build #523 of iop32x is complete: Failure [failed shell_9] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/523 **** BEGIN LOGGING AT Sun Jul 06 04:03:50 2014 Jul 06 04:18:45 build #551 of au1000 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/551 Jul 06 07:50:29 build #205 of mvebu is complete: Failure [failed shell compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/mvebu/builds/205 Jul 06 08:00:08 build #112 of adm8668 is complete: Failure [failed shell compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/adm8668/builds/112 Jul 06 08:19:01 build #642 of ppc40x is complete: Failure [failed shell_13] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/642 Jul 06 08:29:54 build #222 of imx6 is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/imx6/builds/222 Jul 06 08:52:35 build #511 of mcs814x is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/511 Jul 06 10:11:36 build #568 of kirkwood is complete: Failure [failed shell] Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/568 Jul 06 10:20:08 Hauke: nice hacking yesterday! :) Jul 06 10:20:50 are there any other outstanding ad important issues Jul 06 10:22:09 well, there is one thing I don't understand Jul 06 10:22:20 it seems there is some real poblem with WRT610 rebooting Jul 06 10:22:27 with some weird workaround Jul 06 10:22:42 see https://dev.openwrt.org/ticket/14729 and https://dev.openwrt.org/ticket/15841 (look at 1st point only) Jul 06 10:23:12 i would say it's b53 rebooting device Jul 06 10:23:25 but why dropping "option name 'switch0'" workarounds this issue?! Jul 06 10:23:25 I do not understand it either fully, but now I have a wrt610n again Jul 06 10:23:50 oh, great! Jul 06 10:24:14 maybe removing "option name 'switch0'" from "config switch 'eth0'" stopped b53 reseting the switch? Jul 06 10:24:21 when the switch is named eth0 Ethernet is brought up before the switch is initialized, when it is named switch0 OpenWrt does not ensure that Jul 06 10:24:31 oh Jul 06 10:24:36 when it is named eth0 it works Jul 06 10:25:37 https://dev.openwrt.org/changeset/41042 and https://dev.openwrt.org/changeset/41512 should fix this problem Jul 06 10:26:13 but it looks like a workaround Jul 06 10:26:26 even with different name, router shouldn't reboot! Jul 06 10:28:39 yes Jul 06 10:29:18 I think the Ethernet chips does not like it when we access certain functions when it is not up Jul 06 10:29:20 maybe mdio accesses with a powered down ethernet core cause address errors Jul 06 10:29:26 and that results in an reboot Jul 06 10:29:48 makes sense Jul 06 10:30:41 Hauke: maybe the mdio driver needs some way to ensure the mdio bus is active/enabled when accessing it, regardless of state of the network interface Jul 06 10:36:37 hauke r41523 trunk/package/kernel/mac80211/Makefile * mac80211: update linux-firmware Jul 06 10:42:08 Hauke: i'd be nice to update b43 firmware Jul 06 10:42:16 Hauke: if you have nothing to do ;) Jul 06 10:42:22 to which version? Jul 06 10:49:58 hm, the latest? Jul 06 10:50:26 Hauke: http://bues.ch/gitweb?p=b43-tools.git;a=blob;f=fwcutter/fwcutter_list.h;h=6d87601576062778e51bfbcc5b6c0d7d6f3f7e47;hb=HEAD Jul 06 10:50:41 .ucode_version = "784.2", Jul 06 10:50:44 ok I am at it Jul 06 11:00:17 hauke r41524 trunk/package/kernel/mac80211/Makefile * mac80211: b43: add support for firmware 784.2 Jul 06 11:00:47 nice Jul 06 11:01:21 after nbd updates mac80211 we'll be able to disable G-PHY support in mips74k Jul 06 11:01:31 nice Jul 06 11:03:43 zajec_: why not just merge both cases since they have the same value? (i.e. only add the "case BCM47XX_BOARD_LI ...:" and let it fall through) Jul 06 11:05:29 well, it's just a matter of "luck" they share the same GPIO Jul 06 11:05:40 it's not kind of rule for Linksys devices Jul 06 11:05:52 so I wanted to keep entry per device Jul 06 11:05:59 and let compiler optimize it ;) Jul 06 11:08:39 Hauke: one last thing: we should think about backporting bcma changes in next few days Jul 06 11:08:45 once John pickes up my patches Jul 06 11:09:23 you added initial support for the pcie2 cores? Jul 06 11:09:27 zajec_: well, they are both linksys, so it might be more than luck ;) Jul 06 11:10:11 Hauke: right Jul 06 11:10:23 Hauke: i'll send a patch for PCIe 2 as host in a second Jul 06 11:10:26 (a final one) Jul 06 11:12:08 zajec_: I already wrote a PCIe 2 host conrtoller driver which works on the bcm4708: https://dev.openwrt.org/browser/trunk/target/linux/bcm53xx/patches-3.14/170-pcie2-bcma-add-new-PCIe2-driver-for-bcma.patch Jul 06 11:12:24 I just haven't found the time to clean all the stuff up for mainline submission Jul 06 11:12:33 Hauke: it says "BCMA PCIe2 host controller" Jul 06 11:12:46 and it works on "BCMA_CORE_PCIEG2" Jul 06 11:12:59 ah so you are about the other side? Jul 06 11:13:05 right Jul 06 11:13:49 also, this *may* be wrong: Jul 06 11:13:50 val = bcma_read32(bdev, SOC_PCIE_CFG_DATA); Jul 06 11:13:52 devid = val >> 16; Jul 06 11:14:25 there are no PCI devices with device 0x43a1, 0x43a2, 0x43b2 or 0x43b3 Jul 06 11:14:41 there are only 3 PCI ids: :4360, :43a0, :43b1 Jul 06 11:14:53 the rest are what I call "virtual IDs" Jul 06 11:15:15 what is an virtual id? Jul 06 11:16:03 http://git.kernel.org/cgit/linux/kernel/git/linville/wireless-next.git/commit/?id=773cfc508f4d64c14547ff8751b5cbd473124364 Jul 06 11:16:06 see this commit Jul 06 11:16:45 ah so these are the devids in the sprom or nvram Jul 06 11:17:05 right Jul 06 11:17:20 my WNDR4500 has two 14e4:4331 devices Jul 06 11:17:25 and following nvram entries: Jul 06 11:17:26 pci/1/1/devid=0x4332 Jul 06 11:17:29 pci/2/1/devid=0x4333 Jul 06 11:19:37 zajec_: not necessarily - I have a bcm6368 device with two bcm43222 that have working sproms and use the single-band pciids Jul 06 11:21:18 so I wouldn't rule it out that bcm4360/4352 devices are out there that are single band with working sprom, thus using the single band pciids; especially if they are used on a non-broadcom router Jul 06 11:25:16 hauke r41525 trunk/package/network/services/samba36/ Makefile patches/111-owrt_smbpasswd.patch * samba36: update to minor version 3.6.24 Jul 06 11:27:17 KanjiMonster: not sure if I understand Jul 06 11:27:32 KanjiMonster: what are PCI IDs of your BCM43222 devices? Jul 06 11:27:38 i mean real PCI IDs Jul 06 11:27:51 what do you see in cat /sys/bus/pci/devices/*/device Jul 06 11:28:04 zajec_: 0x4351 / 0x4352 Jul 06 11:28:42 err Jul 06 11:28:44 really? Jul 06 11:29:12 interesting Jul 06 11:29:20 nice to know :) Jul 06 11:29:59 well, yeah, old devices were indeed using different PCI IDs, the same thing it about 14e4:4311, 14e4:4312 and 14e4:4313 Jul 06 11:30:12 you can change the pci id of a Broadcom card through some register writes, I accidentally change the PCI id of all bcm43224 devices to 0xa8d8 (43224) Jul 06 11:30:25 ah, wait, it's 0x4350/0x4351 Jul 06 11:30:26 or 14e4:4328, 14e4:4329 and 14e4:432a - they are are real PCI IDs Jul 06 11:30:44 but I never seen BCM4331 on anything else than 14e4:4331 Jul 06 11:31:00 it could be that there are more pci id, but these devices are not on the normal market Jul 06 11:31:02 Hauke: huh, that's interesting :) Jul 06 11:31:10 yeah Jul 06 11:31:21 5GHz-only wifi cards are likely not high in demand ;) Jul 06 11:31:34 only router manufacturers might be interested in them Jul 06 11:33:54 https://git.kernel.org/cgit/linux/kernel/git/linville/wireless-next.git/commit/?id=1f03bf06e4e3b8ed9a69e7fc4cdb1be4c6c6c819 Jul 06 11:34:14 this fixes a problem were I broke linus wifi Jul 06 11:35:28 KanjiMonster: i guess noone likes my b53 change :P Jul 06 11:35:41 Álvaro complained as well :p Jul 06 11:37:03 eh, you people :p Jul 06 11:37:51 zajec_: so no idea why the 2G-only card uses the 2G device id, but the 5G only card uses the dualband device id (unless the 5g one is actually dual band) Jul 06 11:38:19 KanjiMonster: does this bcm6xxx devices have some normal NVRAM? Jul 06 11:38:28 nope Jul 06 11:39:01 i have no idea how Broadcom wants to know it's 2G only then ;/ Jul 06 11:39:13 i don't think they read any extra registers or sth Jul 06 11:39:20 you mean 5G only Jul 06 11:39:24 right Jul 06 11:39:42 "ssb: Found chip with id 0xA8D6, rev 0x00 and package 0x08" so it must clearly have a working sprom (for both) Jul 06 11:40:13 what do you mean? Jul 06 11:40:19 how SPROM can affect this message? Jul 06 11:40:28 how SPROM is related tot his message? Jul 06 11:41:34 KanjiMonster: maybe your BCM6368 doesn't have arch SPROM, but both cards have SPROM anyway? Jul 06 11:41:52 I mean both :4350 and :4351 having SPROM in PCIe's memory Jul 06 11:42:54 hauke r41526 trunk/target/linux/generic/files/drivers/net/phy/b53/b53_priv.h * b53: hardcode reset GPIO for Linksys WRT300N 1.1 Jul 06 11:46:04 zajec_: yes, that is what I assume Jul 06 11:46:29 well actually that is a funny mini-pci with two bcm43222 chips on it ;) Jul 06 11:46:51 zajec_: can I dump the sprom from somewhere? Jul 06 11:47:05 quite possibly yes Jul 06 11:47:48 http://wireless.kernel.org/en/users/Drivers/b43/developers i never used it Jul 06 11:48:12 there is devid in normal (on board) SPROM as well Jul 06 11:48:23 maybe we need to read it just like we read devid from NVRAM Jul 06 12:06:09 zajec_: I would assume that sprom-equipped card use the devid (and vendor, and subsystem*) fields for the pciid they use, so if there's an sprom, it will always match with the observed pciid Jul 06 12:08:56 not really Jul 06 12:09:01 there is some special entry in sprom Jul 06 13:31:52 accessing the mdio bus when the pci_set_power_state is set PCI_D3hot make the bcm4705 reboot Jul 06 13:33:37 build #112 of mpc83xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/mpc83xx/builds/112 Jul 06 13:39:40 Hauke: nice debugging! Jul 06 13:47:47 hauke r41527 trunk/target/linux/ brcm47xx/patches-3.14/791-tg3-no-pci-sleep.patch brcm47xx/patches-3.10/791-tg3-no-pci-sleep.patch * brcm47xx: fix bcm4705 rebooting when initializing the switch. Jul 06 14:09:46 build #522 of ep93xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/522 Jul 06 14:11:46 build #111 of cns3xxx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/cns3xxx/builds/111 Jul 06 15:23:52 updated openwrt/upstream, https://home.comcast.net/~sdwalker/uscan/index.html Jul 06 17:16:10 hauke r41528 trunk/target/linux/brcm47xx/image/Makefile * brcm47xx: image: prepare for using mips74k subtarget Jul 06 17:16:32 hauke r41529 trunk/target/linux/brcm47xx/image/Makefile * brcm47xx: image: mips74k: take over some device images Jul 06 17:23:40 hauke r41530 trunk/target/linux/brcm47xx/image/Makefile * brcm47xx: image: legacy: take over old devices images Jul 06 17:53:17 hauke r41531 trunk/target/linux/ (5 files in 2 dirs) * gemini: add support for kernel 3.10 Jul 06 18:11:33 jogo r41532 trunk/target/linux/ar7/Makefile * ar7: enable the low-memory feature flag Jul 06 18:12:52 luka r41533 trunk/ (6 files in 4 dirs) * kernel: update 3.14 to 3.14.10 Jul 06 18:23:02 hauke r41534 trunk/target/linux/ (6 files in 4 dirs) * au1000: add support for kernel 3.10 Jul 06 19:54:19 Hauke: will u close https://dev.openwrt.org/ticket/12380 Jul 06 19:57:16 Hauke: u may want to assing https://dev.openwrt.org/ticket/12461 to yourself, but I don't know how to proceed with it Jul 06 20:02:33 hauke r41535 packages/libs/db47 * db47: moved to github Jul 06 20:04:32 hauke r41536 packages/net/softflowd * softflowd: moved to github Jul 06 20:35:45 build #326 of mpc85xx is complete: Failure [failed shell] Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/326 Jul 06 20:46:06 build #494 of mpc52xx is complete: Failure [failed shell_14] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/494 Jul 06 21:32:42 build #510 of gemini is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/510 Jul 06 21:36:45 build #516 of adm5120 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/516 Jul 06 21:56:00 build #657 of cobalt is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/657 Jul 06 22:07:16 hauke r41537 trunk/target/linux/mpc52xx/config-3.10 * mpc52xx: add support for kernel 3.10 Jul 06 22:12:42 build #663 of orion is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/663 Jul 06 22:25:56 hauke r41538 trunk/target/linux/ (9 files in 2 dirs) * mpc83xx: add support for kernel 3.10 Jul 06 22:36:27 hauke r41539 trunk/target/linux/ au1000/config-3.10 mpc52xx/config-3.10 mpc83xx/config-3.10 gemini/config-3.10 * kernel: do not build kernel modules by default Jul 06 22:45:41 build #617 of rb532 is complete: Failure [failed shell_9] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/617 Jul 06 22:48:11 build #617 of ppc44x is complete: Failure [failed shell_9] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/617 Jul 06 22:51:16 build #112 of pxa is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/112 Jul 06 23:03:41 build #7 of brcm63xx.smp is complete: Failure [failed shell] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx.smp/builds/7 Jul 06 23:06:09 build #586 of sibyte is complete: Failure [failed shell_9 shell_14] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/586 Jul 07 00:02:05 build #597 of avr32 is complete: Failure [failed shell compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/597 Jul 07 00:21:51 build #628 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/628 **** ENDING LOGGING AT Mon Jul 07 02:59:59 2014