**** BEGIN LOGGING AT Wed Jun 24 02:59:58 2015 Jun 24 05:44:53 ddggttff3_: yes there are issues reported on mailing list in may/june see thread : https://lists.openwrt.org/pipermail/openwrt-devel/2015-June/033414.html with replies Jun 24 06:00:27 build #40 of ipq806x is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/ipq806x/builds/40 Jun 24 06:23:27 Hi guys Jun 24 06:24:00 sorry, exiting Jun 24 06:41:47 build #39 of mpc52xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/39 Jun 24 07:12:00 Trying to fix dts for my device, is anyone who understand how it works here? Jun 24 07:24:42 seems like any device with m25p80 compatible flash uses same address Jun 24 07:24:48 so my fix maybe should work Jun 24 07:50:23 build #39 of octeon is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/39 Jun 24 07:55:58 with fixed dts it now seems to find mtdblock0, but still has some problems with rootfs Jun 24 07:56:08 if here anyone who know how it must work? Jun 24 07:57:04 is this an existing dts? Jun 24 07:58:24 i took part of Y1.dts Jun 24 07:58:44 my device has M25P128 flash Jun 24 07:58:56 seems like there is this or compatible flash Jun 24 07:59:46 i mean, is your device one that already has its own dts? Jun 24 08:00:02 no Jun 24 08:00:04 it don't Jun 24 08:00:12 but it is basic RT3052F device Jun 24 08:00:21 there is many of them supported by openwrt Jun 24 08:02:25 do you know the flash layout of your device? Jun 24 08:02:40 yes Jun 24 08:03:09 http://pastebin.com/P2DjXFgi Jun 24 08:03:15 seems like it close to boot and work Jun 24 08:04:18 lol, maybe it just wants root=mtdblock0 ? Jun 24 08:06:02 [ 0.600000] mtd: partition "u-boot" extends beyond the end of device "spi32766.0" -- size truncated to 0x8000 Jun 24 08:06:10 there's your problem Jun 24 08:06:14 [ 0.550000] m25p80 spi32766.0: found mr25h256, expected w25q128 [ 0.560000] m25p80 spi32766.0: mr25h256 (32 Kbytes) Jun 24 08:06:17 or here Jun 24 08:06:39 that too Jun 24 08:06:41 it detected it like wrong small chip Jun 24 08:10:37 do you know how to fix it? Jun 24 08:12:48 build #40 of brcm47xx is complete: Failure [failed shell_14] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/40 Jun 24 08:14:15 i do not know why the chip is detected wrong Jun 24 08:16:02 how do you think if i finally get it work, should i try to commit support patch to openwrt? Jun 24 08:16:42 if the device is available, yes Jun 24 08:54:24 ok, i found fault place Jun 24 08:54:29 it was my mistake =) Jun 24 08:55:46 maybe it will boot this time Jun 24 09:14:08 that fix of dts does not help Jun 24 09:14:45 spi deice id: 20 20 18 0 0 (20180000) find flash: M25P128 Jun 24 09:14:52 it is from original firmware Jun 24 09:15:03 how to make it work? Jun 24 09:21:27 NermaN: m25p80 spi32766.0: mr25h256 (32 Kbytes) Jun 24 09:21:42 NermaN: your flash was detected incorrectly probably Jun 24 09:21:45 don't know why Jun 24 09:21:55 did you do some hacking on that Jun 24 09:21:57 ? Jun 24 09:22:54 NermaN: well, maybe there is some wrong addess Jun 24 09:22:58 or some initialization missing Jun 24 09:23:04 and the ID read from flash is a garbage? Jun 24 09:23:46 NermaN: that log from original firmware makes sense Jun 24 09:23:47 { "m25p128", INFO(0x202018, 0, 256 * 1024, 64, 0) }, Jun 24 09:24:04 you need to find out why flash replied with some wrong ID Jun 24 09:24:09 when using OpenWrt Jun 24 09:33:09 rmilecki: probably my dts is just mess http://paste.debian.net/259485/ Jun 24 09:35:12 maybe this address is wrong? spi@b00 Jun 24 09:35:25 or the frequency? Jun 24 09:35:29 i don't know... Jun 24 10:57:28 cyrus r46119 trunk/include/hardening.mk * hardening: make override variables more intuitive Jun 24 10:57:30 cyrus r46120 trunk/scripts/feeds * scripts/feeds: observe -p flag for preferential feeds Jun 24 11:10:06 build #39 of ar71xx.mikrotik is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx.mikrotik/builds/39 Jun 24 11:29:36 plntyk: Awesome, thanks for the info. I will have a read through then. :) Jun 24 13:23:51 plntyk: Thanks for the link, but I am not 100% sure the issue I am seeing is the same. For me, the device at multiple speeds (1000,100) will not send traffic, even when tx and rx delay are enabled. I may mess with triggering it on and off as referenced in the chat to see if it resolves the issue. Jun 24 13:30:04 in some reason original fw don't use m25p80 driver in same way like openwrt try Jun 24 13:30:25 seems like address is configured like this in original: CONFIG_MTD_PHYSMAP_START=0xBF000000 CONFIG_MTD_PHYSMAP_LEN=0x1000000 CONFIG_MTD_PHYSMAP_BUSWIDTH=2 Jun 24 13:37:14 is this parts of config now just moved to dts? Jun 24 13:37:31 openwrt don't use it by default Jun 24 13:39:33 <_trine> can anyone help with why in an old revision of openwrt reaver fails to compile yet everything else does Jun 24 13:39:37 <_trine> http://paste.debian.net/260075/ Jun 24 13:43:21 _trine: something wrong with pcap? Jun 24 13:47:25 <_trine> yes I may have fixed it now Jun 24 13:47:43 <_trine> It was an old version of libpcap Jun 24 13:48:29 NermaN: the original fw uses older stuff for sure Jun 24 13:50:10 Kaloz: how to fix my dts? Jun 24 13:51:51 <_trine> nope it did not fix it Jun 24 13:53:54 NermaN: pastebin the dts, the factory bootlog and the owrt one and I take a look Jun 24 13:55:32 Kaloz: dts: http://paste.debian.net/259485/ Jun 24 13:55:55 Kaloz: original bootlog (my is same) http://wiki.openwrt.org/toh/asus/dsl-n12u Jun 24 13:56:30 i don't need dsl to work, so i'm trying to get work only other parts Jun 24 13:57:36 Kaloz: openwrt boot: http://pastebin.com/P2DjXFgi Jun 24 13:58:17 i'm now reading original fw driver Jun 24 14:02:52 Kaloz: seems like original driver uses this mapping: http://pastebin.com/1hXPpvMK Jun 24 14:05:30 what's weird is that it finds a different device id Jun 24 14:07:19 is palmbus@10000000 { spi@b00 { same to 0xB0000B00 ? Jun 24 14:08:06 maybe there is other chip, which is on this wrong address Jun 24 14:13:46 Kaloz: i even finded from what driver is string about 202018 device id Jun 24 14:17:21 m25p80 driver from openwrt has { "m25p128", INFO(0x202018, 0, 256 * 1024, 64, 0) }, and asus driver has { "M25P128", 0x20, 0x20180000, 256 * 1024, 64 }, this statement Jun 24 14:17:39 that's fine Jun 24 14:19:01 on boot it says with my dts that it finded that chip: { "mr25h256", CAT25_INFO( 32 * 1024, 1, 256, 2) }, Jun 24 14:19:13 seems like it does not finded anything actually Jun 24 14:21:26 if it would be parallel flash, I would have said the buswindth can be the issue Jun 24 14:23:48 Kaloz: is palmbus@10000000 { spi@b00 { same to 0xB0000B00 ? Jun 24 14:24:05 original fw uses everywhere 0xB0000B00 for spi address Jun 24 14:24:14 maybe my dts is wrong Jun 24 14:24:29 build #39 of adm5120 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/39 Jun 24 14:31:03 original fw released by asus is just a mess Jun 24 14:31:33 seems like they don't heard about good coding style Jun 24 14:36:30 build #39 of omap is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/omap/builds/39 Jun 24 14:38:04 Kaloz: with palmbus B0000000 it does not show anything about rom Jun 24 14:38:12 and now i am out if ideas =( Jun 24 14:53:56 NermaN: B0000000 is a virtual address mapped to the physical address 10000000 Jun 24 14:54:43 and dts files expect physical addresses, so B0000000 as a physical address wouldn't work Jun 24 14:55:43 KanjiMonster: rt3050 has predefined section about spi at b00, seems like everything should work but don't Jun 24 14:56:44 NermaN: physmap implies parallel flash, while spi is serial flash Jun 24 14:56:51 two different things Jun 24 14:57:34 KanjiMonster: this dts is not for spi? http://paste.debian.net/259485/ Jun 24 14:58:00 NermaN: this is for spi Jun 24 14:58:49 KanjiMonster: booting with it fails in weird way, bootlog: http://pastebin.com/P2DjXFgi Jun 24 15:00:21 NermaN: did you write the email to linux-mtd abou the same issue? Jun 24 15:00:24 *about Jun 24 15:00:49 NermaN: you may just need a support for DT in physmap driver Jun 24 15:01:19 but it seems to have support Jun 24 15:01:30 [ 0.470000] rt2880-pinmux pinctrl: pin io3 already requested by pinctrl; cannot claim for 10000b00.spi Jun 24 15:01:36 rmilecki: look at the bootloader output, this is spi flash, not parallel flash (-> spi deice id: 20 20 18 0 0 (20180000)) Jun 24 15:01:42 maybe it is just something like driver overlap? Jun 24 15:01:52 the pinmux thingy might point to the actual issue Jun 24 15:01:58 KanjiMonster: i'm confused with SPI vs. using CONFIG_MTD_PHYSMAP_START Jun 24 15:02:43 rmilecki: the MTD_PHYSMAP_START is from the vendor sdk, it's probably set regardless whether parallel flash is actually used Jun 24 15:02:53 KanjiMonster: ok Jun 24 15:05:47 NermaN: mr25h256 is the very first flash in the supported flashs without any id (so an id of "0"), so reading the id likely returned all zeros, i.e. spi doesn't work at all, and it just defaulted to the first flash with the id of "0" Jun 24 15:07:20 (3.18 doesn't ignore 0-length flash ids while autodetecting) Jun 24 15:07:23 i'm checking now theory of spi address being already blocked by pinctrl Jun 24 15:08:53 NermaN: in state_default you assign the function "gpio" to the group "spi" Jun 24 15:09:13 in the dts you pasted Jun 24 15:11:02 NermaN: so just remove the "spi" group from the list, and then the group should be muxed to spi automatically Jun 24 15:13:50 wow Jun 24 15:14:03 now it seems way closer to working condition Jun 24 15:14:44 KanjiMonster: http://pastebin.com/1SC8PtHg Jun 24 15:15:31 looks like it worked ;) Jun 24 15:15:45 and now its new problem around mtd partitions Jun 24 15:16:35 is this an openwrt kernel or a vanilla linux kernel? Jun 24 15:17:17 openwrt kernel Jun 24 15:17:27 maybe it is because wrong mtd partitions mapping? Jun 24 15:17:39 i don't really sure how it works at all Jun 24 15:18:01 and how it should guess system partition =( Jun 24 15:19:43 usually you just need to make sure you have a "firmware" partition that points to the place where you flash an image, code with then check for image header there on scan and create kernel/rootfs(_data) partitons automatically Jun 24 15:19:50 *will Jun 24 15:21:11 NermaN: so according to u-boot your firmware partition should start at 0xc0000, not 0x50000 Jun 24 15:22:53 <_trine> KanjiMonster, can you help porting opwrt to one of these checking for pcap_open_live in -lpcap... no Jun 24 15:22:53 <_trine> error: pcap library not found! Jun 24 15:23:03 <_trine> opps Jun 24 15:23:11 <_trine> http://www.aliexpress.com/item/300Mbps-high-power-long-range-wireless-n-cpe-repeater-ATHEROS-AR9341-Comfast-CF-E316NV2-0-with/32259599919.html?s=p Jun 24 15:23:13 will try to fix it now, i'm thinked about it like just MBR, so it can be remapped like i want, so, don't checked that part Jun 24 15:24:09 _trine: you probably need to pass the configure script the location where it can find the pcap library Jun 24 15:24:30 NermaN: there is no such thing on embedded, only hardcoded offsets in flash (mostly) Jun 24 15:24:48 <_trine> KanjiMonster, my last message to was garbled sorry Jun 24 15:24:58 <_trine> KanjiMonster, my last message to you was garbled sorry Jun 24 15:25:18 <_trine> I have 2 things I am working on at the same time Jun 24 15:25:53 <_trine> first is the pcap error and second is that I would like to be able to flash opwrt to that new device Jun 24 15:26:17 <_trine> as it looks a very capable device for the price Jun 24 15:28:53 <_trine> cheaper price here http://www.aliexpress.com/item/COMFAST-WIFI-router-wireless-bridge-outdoor-CPE-300Mbps-AR9341-wi-fi-access-point-router-with-dual/32336878271.html Jun 24 15:29:32 <_trine> just noticed that doe snot have free postage Jun 24 15:31:47 KanjiMonster: http://pastebin.com/HVMDW0qv maybe you can help with guessing how it should be? Jun 24 15:32:02 KanjiMonster: first log part is from original fw Jun 24 15:32:46 or just check that i will write now =) Jun 24 15:33:09 NermaN: well, Bootlader => u-boot, Config => u-boot-env, Factory => factory and kernel => firmware ;) Jun 24 15:35:46 _trine: regarding the device - I don't have a crystall ball to see what you need to do to flash openwrt to it. Likely effort/possibility depends largely on whether you can get a GPL tarball. Jun 24 15:37:03 KanjiMonster: http://pastebin.com/vJDyaKnC i did it, can you check do this looks correct? Jun 24 15:37:34 NermaN: looks plausible Jun 24 15:46:29 WHOA thos thing runs Jun 24 15:47:02 btw it is first time when i'm trying to use openwrt in any way =) Jun 24 15:48:26 KanjiMonster: big thanks to you, and to everyone on this channel Jun 24 15:48:33 it runs Jun 24 15:48:49 yay \o/ Jun 24 15:49:51 NermaN: next step is obviously -> https://dev.openwrt.org/wiki/SubmittingPatches ;D Jun 24 15:52:57 KanjiMonster: even leds seems to work as it should Jun 24 15:54:01 KanjiMonster: is there any reason why openwrt don't even try to support dsl devices? There is big amount of people who are upgraded from dsl to just RJ45, and dsl support does not seems that hard Jun 24 15:54:21 NermaN: did the changed offset of the factory partition also fix the random mac address for ethernet (and likely wmac)? Jun 24 15:55:11 KanjiMonster: seems like yes, and it goes very well now Jun 24 15:55:19 wireless seems to work perfectly Jun 24 15:56:23 NermaN: dsl support requires drivers for the dsl chip, which aren't very popular - only lantiq has them. Broadcom, mediatek, ikanos and realtek have closed source drivers which require extensive changes in the network stack, and aren't usable with modern kernels. Jun 24 15:57:34 And writing new drivers is far from easy, DSL is quite complex Jun 24 15:58:30 apart from the fact that you can't test easily as you can't just connect two dsl modems, but you need a "CO" modem, which costs you several hundred bucks (at least) Jun 24 15:59:45 and then there's the ATM/PTM part, and a extra management channel, etc etc Jun 24 16:00:23 and the ITU documentation is also not an easy read (I tried, and it usually takes me several times until I get an understanding how the standard is supposed to be) Jun 24 16:06:20 i think idea of making good router from stupid dsl router is quite good Jun 24 16:06:30 so should i really try to make patch for this device? Jun 24 16:06:35 it is popular dsl here Jun 24 16:12:07 awesome, it really works and shows good perfomance Jun 24 16:22:14 KanjiMonster: it don't wanted to flash through recovery mode Jun 24 16:22:18 there was some check Jun 24 16:22:21 <_trine> Hey KanjiMonster thanks but no thanks begging letters will not be answered Jun 24 16:22:22 <_trine> <_trine> can you tell me how to do it Jun 24 16:22:22 <_trine> please don't query me with stuff that belongs in the channel Jun 24 16:22:22 <_trine> <_trine> IC Jun 24 16:22:22 <_trine> unless you intend to pay for the time Jun 24 16:22:42 that don't give to flash "wrong" firmware Jun 24 16:24:34 _trine: it's just like with webcam girls, if you want a private show you need to pay for it ;P (else keep it public so others have something from it too) Jun 24 16:25:01 <_trine> KanjiMonster, go flog your cunt somewhere else then Jun 24 16:25:58 <_trine> dont worry I shall be remind to never ask for your help again Jun 24 16:26:15 NermaN: you probably need to add some extra header to the image, but don't ask me what you need to do Jun 24 16:26:24 _trine: you can ask me, but please do it in the channel Jun 24 16:26:32 that's all what I'm asking for Jun 24 16:27:07 <_trine> look lets just forget I ever asked,, it's not a problem Jun 24 16:27:26 <_trine> and its only a hobby Jun 24 17:16:18 wat Jun 24 18:52:21 cyrus r46121 trunk/package/utils/px5g-standalone/src/Makefile * Revert "px5g-standalone: compatibility with fortify-headers" Jun 24 18:52:23 cyrus r46122 trunk/toolchain/fortify-headers/patches/101-fix-pedantic.patch * fortify-headers: fix compatibility with -pedantic Jun 24 20:10:12 build #41 of cobalt is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/41 Jun 24 20:23:27 build #41 of pxa is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/41 Jun 24 20:23:31 build #41 of cns21xx is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/cns21xx/builds/41 Jun 24 20:24:13 build #41 of orion is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/41 Jun 24 20:24:42 build #41 of ppc40x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/41 Jun 24 21:31:13 build #40 of netlogic is complete: Failure [failed compile_3] Build details are at http://buildbot.openwrt.org:8010/builders/netlogic/builds/40 Jun 24 21:40:51 build #41 of x86.64 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/x86.64/builds/41 Jun 24 22:19:03 build #40 of avr32 is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/40 Jun 24 22:21:42 build #40 of x86.kvm_guest is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/x86.kvm_guest/builds/40 Jun 24 23:04:57 plntyk: No go, but I believe I have a different issue. Mine is unable to send packets under every situation, which makes me wonder if I am doing something wrong to init the device. Jun 24 23:22:46 build #40 of x86 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/x86/builds/40 **** ENDING LOGGING AT Thu Jun 25 02:59:59 2015