**** BEGIN LOGGING AT Mon Mar 01 02:59:57 2010 Mar 01 04:26:57 xMff: would've been easier to just consolidate all of those redundant netconfig entries :) Mar 01 05:14:58 build #18 of ar7 is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/18 Mar 01 05:44:14 nico * r19923 /trunk/package/mac80211/Makefile: mac80211: add 2 missing config symbols needed to build rt2x00-lib & rt2800-lib modules Mar 01 06:20:47 <{Nico}> cshore: ping Mar 01 06:22:44 nico * r19924 /trunk/target/linux/generic-2.6/config-2.6.32: kernel: add 2 config symbols preventing ppc40x & ppc44X from being configured with all modules enabled Mar 01 06:25:47 nico * r19925 /trunk/package/kernel/modules/block.mk: kernel: mark ata-magicbox-cf & ide-magicbox as BROKEN for now (they don't build on 2.6.32) Mar 01 06:36:49 xMff: arghh glibc built - but kernel failed: configure: error: compiler support for __thread is required Mar 01 06:36:54 er no. Mar 01 06:37:01 cc1: error: unrecognized command line option "-fno-tree-ch" Mar 01 07:34:38 juhosg * r19926 /trunk/target/linux/ar71xx/ (3 files in 3 dirs): ar71xx: add QinQ tagging format for the DSA driver Mar 01 07:34:43 juhosg * r19927 /trunk/target/linux/ar71xx/ (3 files in 3 dirs): ar71xx: add DSA driver for the AR7240 built-in ethernet switch Mar 01 07:34:45 juhosg * r19928 /trunk/target/linux/ar71xx/files/arch/mips/ar71xx/dev-ap91-eth.c: ar71xx: register DSA switch for the AP91 based boards Mar 01 07:34:47 juhosg * r19929 /trunk/target/linux/ar71xx/base-files/etc/defconfig/dir-600-a1/ (. network): ar71xx: add default network configuration file for the DIR-600 rev. A1 board Mar 01 07:34:50 juhosg * r19930 /trunk/target/linux/ar71xx/base-files/etc/defconfig/tl-wr741nd/ (. network): ar71xx: add default network configuration file for the TL-WR741ND board Mar 01 08:00:06 kaloz * r19931 /trunk/package/kernel/modules/block.mk: [kernel]: partly revert [19925] as the ata version builds fine Mar 01 08:04:16 kaloz * r19932 /trunk/ (2 files in 2 dirs): [kernel]: drop support for the old style magicbox-cf driver Mar 01 09:19:48 hello Mar 01 09:44:08 <_lore_> Hi all Mar 01 09:44:37 <_lore_> I have an issue compiling the latest version of hplip driver Mar 01 09:44:44 <_lore_> 3.10.2 Mar 01 09:45:14 <_lore_> I am using the latest trunk version Mar 01 09:46:22 <_lore_> but during the linking stage I obtain the error "lcupsimage not found" Mar 01 09:48:51 <_lore_> is this library missing?? Mar 01 10:00:05 ~take-a-number does anybody know if mod_sofia uses apr library to manage network sockets? Mar 01 13:02:47 hi Mar 01 13:03:15 does someone know how to fix the build error when building a new toolchain with a 2.6.33 kernel Mar 01 13:17:08 simple, do not build with it Mar 01 13:17:25 thats not a fix, thats a workaround ;) Mar 01 13:27:10 Kaloz :) Mar 01 13:27:27 yeah but what is the problem - and which file should be fixed Mar 01 13:28:38 i think i ported brcm47xx to 2.6.33 but i cant build the toolchain - i deleted to toolcahin that was created with 2.6.32.9 Mar 01 13:39:44 tripolar_: forget .33 Mar 01 13:43:30 <[florian]> tripolar_: there are also brcm47xx patches for 2.6.33 Mar 01 13:43:45 <[florian]> tripolar_: but for now do not hope it will work better than it already does with older kernels :) Mar 01 13:54:24 [florian]: no i dont hope it will work better then with the 2.6.32.9 kernel it works nice on 32 but id like to use .33 as my desktop sys Mar 01 13:55:36 Kaloz: why should i forget it? Mar 01 13:55:47 openwrt wont use .33 and wait for 34 ;) Mar 01 14:15:41 <[florian]> tripolar_: I will have a fix for toolchain building with 2.6.33 in a couple of minutes Mar 01 14:16:40 nice Mar 01 14:17:28 [florian] i tried to fix it on my own by it didnt get very far :( Mar 01 14:32:47 the idea would be to have th eunit somewhere where i get access to it and we can test new devices. Mar 01 14:32:51 alternatively we decide, yes it is worth the money and we buy a unit. not sure. i just found out they exist :) Mar 01 14:33:18 grrrrrrrrrrr Mar 01 14:34:07 <[florian]> -EWRONGCHAN ;) ? Mar 01 14:37:03 c&p Mar 01 14:37:04 :) Mar 01 14:37:33 [florian]: btw, that link i sent early uses this project as a base Mar 01 14:37:35 http://www.sump.org/projects/analyzer/ Mar 01 14:40:17 blogic how much does it cost? Mar 01 14:42:10 blogic: this logic sniffer is avaiable for about 45 bucks if i read it right Mar 01 14:42:13 http://www.gadgetfactory.net/gf/project/butterflylogic/ Mar 01 14:46:11 KanjiMonster: I read you are working on a ar8316 switch driver ? Mar 01 14:46:25 stintel: yeah Mar 01 14:46:38 KanjiMonster: nice, can I ask you something about it ? :_) Mar 01 14:46:42 sure Mar 01 14:47:09 would it solve the problem of not receiving multicast ? Mar 01 14:47:29 tripolar_: 45% Mar 01 14:47:31 tripolar_: 45$ Mar 01 14:47:49 tripolar_: http://www.seeedstudio.com/depot/preorder-open-workbench-logic-sniffer-p-612.html?cPath=61_76 Mar 01 14:49:19 stintel: I don't know if it does currently, I did not test it yet, but I am sure it can be made to Mar 01 14:50:29 KanjiMonster: nice, I guess I will wait some more days before ordering the normal routerstation Mar 01 14:50:59 stintel: the normal routerstation has a different switch than the pro, so my work would solve nothing on it Mar 01 14:51:26 KanjiMonster: I know. I have 2 rs pro Mar 01 14:51:31 KanjiMonster: they are almost useless Mar 01 14:52:12 I was planning to replace them with the non pro version to be able to receive multicast is what I meant Mar 01 14:52:42 can you recommend a way how I can test if it works? I have no experience with multicast ;) Mar 01 14:52:59 hrrm Mar 01 14:53:23 create an ospf network between the two devices for example Mar 01 14:53:43 iirc ospf routers communicate with eachother via mcast Mar 01 14:54:41 then just tcpdump on both, see them leaving at device 1 but don't see them arriving on device 2 Mar 01 14:55:09 florian * r19933 /trunk/toolchain/kernel-headers/Makefile: [toolchain] fix compilation w/ 2.6.33 Mar 01 14:55:17 florian * r19934 /trunk/target/linux/generic-2.6/ (2 files in 2 dirs): [kernel] prevent UML build failure against decompress_lzo.c Mar 01 14:55:36 florian * r19935 /trunk/target/linux/brcm63xx/ (2 files in 2 dirs): [brcm63xx] remove mail headers from patches Mar 01 14:56:08 blogic really cheap Mar 01 14:56:19 KanjiMonster: I found your post on the ubnt forum, did you announce your work in openwrt forum as well ? or created some ticket ? Mar 01 14:56:39 tripolar_: indeed Mar 01 14:56:44 tripolar_: i ordered 3 units Mar 01 14:57:16 stintel: only indirecly, I stated that i am working on making it to work, but I did write an update Mar 01 14:57:17 <[florian]> tripolar_: please try now Mar 01 14:57:19 blogic nice - tell me how it works when u get them Mar 01 14:57:25 +not Mar 01 14:57:31 [florian]: okay Mar 01 14:57:47 tripolar_: k Mar 01 14:57:57 [florian]: i have to wait until its in the git co Mar 01 14:58:15 KanjiMonster: ok Mar 01 14:58:28 KanjiMonster: if you want me to test anything, feel free to ask Mar 01 14:58:44 blogic how much does a logic sniffer noramlly coast - i used ony running on windows at the universitiy which was really nice Mar 01 14:58:45 KanjiMonster: currently have two rspro I am experimenting with ;-) Mar 01 14:59:47 ping Nico Mar 01 14:59:56 tripolar_: good ones cost thousands Mar 01 15:01:18 ping nbd Mar 01 15:01:20 ping xMff Mar 01 15:01:44 hmmm, my rspro with ath9k in client mode seems stuck in HT20 Mar 01 15:02:02 there were some patches on linux-wireless@ Mar 01 15:02:05 my rspro with ath9k ap is running in HT40+, Mar 01 15:02:06 i'll merge those soon Mar 01 15:02:08 nbd: ah Mar 01 15:02:16 nbd: fyi, MFP works Mar 01 15:02:20 cool Mar 01 15:02:29 but sometimes it doesn't :P Mar 01 15:02:30 i will commit my minstrel_ht implementation soon Mar 01 15:02:38 interesting Mar 01 15:03:08 nbd: do you want me to do up a patch for the switch to /overlay for jffs? Mar 01 15:03:18 yeah, that would be nice Mar 01 15:03:42 nbd: pay the way tomorrow I'm gone till late Friday Mar 01 15:03:46 pay=by Mar 01 15:04:14 ok Mar 01 15:04:28 nbd: but I'll do the overlay today...should be a pretty easy change Mar 01 15:07:36 someone could quickly look at the wpa_supplicant.sh patch I sent to the ML yesterday? or can I resend it as part of series to enable 802.11w? Mar 01 15:14:18 stintel: you can resend it Mar 01 15:14:30 just make a note that you did Mar 01 15:18:29 ok Mar 01 15:23:04 [florian]: im just trying to compile - but dont i need binutils 20 to get it working or should it work with 18? Mar 01 15:23:34 <[florian]> tripolar_: you do not, I just tried with the default settings here and they work just fine Mar 01 15:24:55 okay Mar 01 15:25:18 because i use 20 here in my desktop sys and when i downgraded it to 18 the kernel didnt compile Mar 01 15:25:38 maybe i have to recompile gcc with 18 to get it working Mar 01 15:29:25 [florian]: your patch seems to work here Mar 01 15:30:39 <[florian]> tripolar_: good :) Mar 01 15:31:55 [florian]: build isnt ready yet but i comes to point never reached befor ;) Mar 01 15:32:02 with 2.6.33 Mar 01 15:33:33 <[florian]> tripolar_: I could successfully build and run 2.6.33 on brcm63xx yesterday so do not worry ;) Mar 01 15:34:52 :) Mar 01 15:35:18 [florian] there are still some config symbols missing in generic 2.6.33 config Mar 01 15:35:38 <[florian]> tripolar_: I though I added them already, but that was only for brcm63xx so, I certainly missed some Mar 01 15:35:56 i'll send them later to the mailing list Mar 01 16:10:29 hi all, i'm trying to develop an application on Openwrt. My application is multi-threaded(there is 3 thread). When i check with pgrep i see 6 pieces of my app. Mar 01 16:10:34 What's going on ?? Mar 01 16:12:15 its your app, how are we supposed to know? Mar 01 16:12:46 maybe you forked before spawning the threads or something Mar 01 16:18:12 why all the threads has a pid Mar 01 16:18:28 is it just for the processes ?? Mar 01 16:20:29 same pid or different? Mar 01 16:21:04 and do you actually use threads (pthread) or child processes created with fork() ? Mar 01 16:22:56 there are 6 pid Mar 01 16:23:10 and i'm using pthreads Mar 01 16:24:20 augurel: which kernel and openwrt version? Mar 01 16:24:37 Linux OpenWrt 2.6.32.7 #5 Wed Feb 24 17:52:50 EET 2010 mips GNU/Linux Mar 01 16:24:41 is openwrt bug tracker supposed to send e-mails on replies to tickets? Mar 01 16:24:56 greenlion: only if you add yourself to the CC: field Mar 01 16:25:18 jow_laptop, even if I'm reporter? Mar 01 16:25:26 greenlion: yes Mar 01 16:25:31 think so Mar 01 16:25:57 augurel: well, most likely you did something really odd Mar 01 16:26:22 augurel: maybe you used some really old pthread version or something Mar 01 16:26:49 augurel: iirc OpenWrt's uClibc uses linuxthreads_old which uses a different pid for each thread Mar 01 16:26:55 jow_laptop, oh well. thanks. It sent me email on creation of ticket, but not on reply https://dev.openwrt.org/ticket/6768 Mar 01 16:27:16 also this implementation spawns at least one additional thread for management purposes Mar 01 16:27:21 i think i'm using stdcpp Mar 01 16:27:34 oh sorry, Mar 01 16:28:33 i think i'm using /lib/libpthread-0.9.30.1.so Mar 01 16:28:37 hmm... but it sent me mail on other mine ticket, where I did comment myself... Mar 01 16:28:49 so the number of processes you see in `ps` and whether you get different pids depends on the actual pthread implementation in use, linuxthreads-old, linuxthreads-new or ntpl Mar 01 16:29:04 ntpl is broken in uClibc therfore linuxthreads is used Mar 01 16:29:10 but I'm not sure whether old or new Mar 01 16:29:43 jow_laptop: the fork approach is from pre 2.2 times afaik ... Mar 01 16:29:51 this would be REALLY old Mar 01 16:30:14 how could i know, which library that i use? Mar 01 16:30:17 rtz2: doesn't matter if it is old, the fact is that anything newer is broken in uclibc Mar 01 16:30:32 #include : this is my include Mar 01 16:30:44 augurel: -0.9.30.1.so sounds like it is a part of uclibc Mar 01 16:31:05 should i update libpthread ?? Mar 01 16:31:25 why? Because you're scared by the entries in "ps" ? Mar 01 16:31:38 as I said, anything newer is most likely broken Mar 01 16:31:50 broken means bad ?? Mar 01 16:31:59 broken means it does not work Mar 01 16:32:02 or not stable?? Mar 01 16:32:32 augurel: you could always fix it yourself ;) Mar 01 16:32:38 ok, thank u. Mar 01 16:34:17 k, just checked, OpenWrt uses linuxthreads-new Mar 01 16:34:32 by the way, i'm using a mikrotik board(rb450), when i plugged out the power cable(or reboot) the openwrt, the clock is resets and set itself to 1970-Jan-01 Mar 01 16:34:42 do u have any experience about this ? Mar 01 16:35:10 does it even have a RTC? If not it is normal Mar 01 16:35:27 how can i check it? Mar 01 16:35:32 datasheet Mar 01 16:35:40 oh, it is a hardware Mar 01 16:35:47 yes Mar 01 16:35:47 real time clock ?? Mar 01 16:35:52 correct Mar 01 16:36:00 most router platforms don't have one Mar 01 16:36:13 so how could we adjust time? Mar 01 16:36:14 and time needs to be sinced via ntp after each powerloss Mar 01 16:36:22 *synced Mar 01 16:36:53 actually i'm trying to use my board as a special purpose mini computer Mar 01 16:37:17 i may not have a connection every time Mar 01 16:37:47 if the board has no battery backed rtc, you can't do anything to preserve the time over a powerloss Mar 01 16:40:33 i think it doesn't has a RTC Mar 01 16:41:19 i couldn't find any battery on the board Mar 01 16:41:24 you can store the current time on shutdown and restore the time on boot but then your system time will lag behind a few seconds more after each powercycle Mar 01 16:42:41 or you could install a gps dongle and use that as time source Mar 01 16:43:07 when i connect to board via serial cable, i reach to root console autmatically Mar 01 16:43:19 is there a setting or sth ?? Mar 01 16:43:41 ? Mar 01 16:44:40 when i plugged my serial cable and i ran a serial tool(minicom), i automatically reach to console, that root logged in Mar 01 16:44:48 could i explained? Mar 01 16:45:46 i think, couldn't :( Mar 01 16:46:59 augurel: I don't completle understand your question, but it's normall to reach the root console over serial Mar 01 16:47:45 i want to disable it, it is better for me if it asks for login Mar 01 16:47:57 like ssh Mar 01 16:48:04 is it possible? Mar 01 16:48:48 augurel: you want to disable it completly or only ask for a password? Mar 01 16:48:57 only ask for a password? Mar 01 16:49:51 augurel: /bin/login is only a dummy afaik, you could try to replace it with the busybox login applet Mar 01 16:50:01 augurel: but there will still be failsafe Mar 01 16:50:20 augurel: you would have to compile your own image to disable that Mar 01 16:50:34 i have compiled already Mar 01 16:50:50 i should add busybox?? Mar 01 16:51:07 augurel: I think the applet is already included Mar 01 16:51:18 yes it is Mar 01 16:51:22 i checked Mar 01 17:10:05 augurel: there will also always remain bootloader access over serial Mar 01 17:11:20 any way to disable it ?? Mar 01 17:11:59 hmm, some boot loaders have a boot_wait environment variable that you could set to 0 Mar 01 17:12:23 augurel: i'm curious: why do you require all this? Mar 01 17:13:42 i'm trying to use this router boards as a small computers Mar 01 17:14:09 so i don't want to connect anyone to these boxes Mar 01 17:14:26 ? Mar 01 17:14:51 augurel: you mean for industrial control or something? Mar 01 17:15:13 routers _are_ small computers, embedded ones Mar 01 17:15:31 do you mean you want to make them pcs? Mar 01 17:16:05 yes Mar 01 17:16:38 augurel: what do you use for video? Mar 01 17:17:19 no, not a general purpose one Mar 01 17:17:41 just make some network operations, controls, management issues Mar 01 17:18:37 augurels: ah; you want it to replace expensive network devices like Cisco? Mar 01 17:21:30 augurel: disabeling the bootloader will be hard, short of compiling your own, you won't be able to prevent access to it Mar 01 18:01:36 jow * r19936 /trunk/scripts/ipkg: Mar 01 18:01:36 [scripts] ipkg: Mar 01 18:01:37 strip directory entries from package file list when offline installing .ipk archives, Mar 01 18:01:37 this prevents opkg from removing directories like /usr later on when trying to Mar 01 18:01:37 uninstall preinstalled packages (partial fix for #6772) Mar 01 18:20:42 nbd * r19937 /trunk/package/base-files/files/lib/preinit/30_failsafe_wait: base-files: fix enter failsafe message Mar 01 18:31:29 tripolar * r19938 /packages/net/ctorrent/Makefile: ctorrent depends on openssl Mar 01 18:37:23 ping nbd Mar 01 18:37:38 pong Mar 01 18:38:19 nbd could u please apply my openssl 0.9.8m patch Mar 01 18:38:27 sure Mar 01 18:38:27 just send it to the mailing list Mar 01 18:38:39 i had to modify some ocf parts Mar 01 18:38:50 they already applied some stuff Mar 01 18:39:37 and all of the stuff that you removed was applied upstream? Mar 01 18:40:39 nbd yeah Mar 01 18:40:46 aok Mar 01 18:40:54 most of the stuff where upstream patches Mar 01 18:41:05 all 9* files Mar 01 18:41:15 maybe the files should be renamed Mar 01 18:41:29 the remaining Mar 01 18:41:42 i just compiled m here and it works like a charm :) Mar 01 18:41:52 m=openssl-*m Mar 01 18:44:13 nbd if i did it right brcm47xx is also ready for .33 but i have to wait until the image is ready :) Mar 01 18:46:45 nbd * r19939 /trunk/package/openssl/ (14 files in 2 dirs): openssl: upgrade to 0.9.8m (patch by Peter Wagner) Mar 01 19:28:51 tripolar * r19940 /packages/net/irssi/Makefile: irssi Makefile style ++ :) Mar 01 19:31:30 tripolar * r19941 /packages/net/ntpd/Makefile: Mar 01 19:31:30 switch ntpd to the build variants infrastructure Mar 01 19:31:30 -- d die folgenden Zeilen werden ignoriert -- Mar 01 19:31:30 M net/ntpd/Makefile Mar 01 19:40:27 ping nbd Mar 01 19:40:37 it works Mar 01 19:40:38 Linux server 2.6.33 #1 Mon Mar 1 20:31:05 CET 2010 mips GNU/Linux Mar 01 19:40:52 could u add the files to the brcm47xx profile? Mar 01 19:42:09 tripolar_ thnx :) will test it to tomorrow then :) Mar 01 19:43:19 crow you are welcome :) Mar 01 19:48:51 nbd the patch has an error i will subbmit a new new in a few minutes to the ml Mar 01 20:00:28 802.11w patches sent to ml Mar 01 20:05:44 heh. 0/0 should've been 0/4 Mar 01 20:05:48 * stintel puts on a noob hat Mar 01 20:58:28 ping blogic Mar 01 21:17:00 build #19 of adm5120 is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/adm5120/builds/19 Mar 01 21:19:08 acinonyx * r19942 /trunk/package/dnsmasq/Makefile: [package] dnsmasq: Remove deprecated HAVE_ISC_READER=1 flag (#6757) Mar 01 21:27:12 1-705-728-1148 Mar 01 21:27:39 I think that's his home number Mar 01 21:28:17 wrong channel? Mar 01 21:28:24 doh! Mar 01 21:28:26 yep Mar 01 21:32:27 ping Acinonyx Mar 01 21:32:33 pong Mar 01 21:32:51 Acinonyx could u please commit my brcm47xx 2.6.33 patches Mar 01 21:33:13 and the missing config symbols for 2.6.33 Mar 01 21:33:19 I wish I could. I don't have svn write access to targets Mar 01 21:33:25 ohh okay Mar 01 21:33:31 I also have patches for 2.6.32 which i cannot commit Mar 01 21:33:37 *had Mar 01 21:33:41 :) Mar 01 21:44:01 anyone having trouble with wireless on .33. My boss says he can configure the wlan but his devices can't see it since up updated his image Mar 01 21:45:44 cshore: madwifi devices? Mar 01 21:46:05 b43 Mar 01 21:46:24 no clue Mar 01 21:46:27 :( Mar 01 21:48:56 nbd * r19943 /trunk/package/mac80211/ (7 files in 2 dirs): mac80211: add the new 802.11n minstrel rate control implementation (optional, not used by default) Mar 01 21:49:50 share and enjoy! Mar 01 21:49:52 ;) Mar 01 21:59:03 nbd: my jffs image doesn't seem to want pick up the overlay directory from build_dir/target.../root... . It's there but not making it into the image Mar 01 22:00:02 I've replace all the hard coded references to jffs with overlay Mar 01 22:00:31 nbd: does a normal jffs2 image include a /jffs directory? Mar 01 22:05:25 I see the problem. /jffs / overlay was rm'd from any jffs2 image in image.mk Mar 01 22:11:52 nbd can i commit my brcm47xx patches? Mar 01 22:12:32 haven't looked at them yet Mar 01 22:12:34 busy with other stuff Mar 01 22:15:28 nbd u dont have to look at them - just commit them ;) Mar 01 22:24:39 nbd: I've sent the patch with the /overlay change for jffs (i.e. no more /jffs, now named /overlay whether squashfs/jffs or external root) Mar 01 22:25:14 ok, thx Mar 01 22:25:18 i'll take a look at it later Mar 01 22:28:52 cshore: why not just stay with /jffs? Mar 01 22:34:36 tripolar_: because overlay_root in opkg.conf needs to point to the overlay and we don't want to use /jffs for external root (external root uses an overlay too) Mar 01 22:37:19 cshore why does the external root use an overlay? Mar 01 22:38:04 because it can be used transparently then Mar 01 22:38:12 for opkg packages etc. Mar 01 22:38:30 without altering PATH, LD_LIBRARY_PATH, install destination and a couple of other things Mar 01 22:42:17 xMff okay Mar 01 22:42:53 so the root is still the normal squashfs and the usb disk for example is the overlay? Mar 01 22:43:13 yes, instead of what jffs2 is usually is Mar 01 22:43:55 but is there some way to tell the rootoverlay to not use the external overlay altough the disk is attached? Mar 01 22:44:31 well in this case you just don't install the extroot packages Mar 01 22:44:40 and rely on the standard mass storage mounting Mar 01 22:44:41 tripolar_: yes, don't enable is_rootfs for the partition Mar 01 22:44:46 or that :P Mar 01 22:45:23 is it possible to change the is_rootfs variable when im allready on the usb disk overlay? Mar 01 22:45:51 you would mount the old jffs and edit /etc/config/fstab Mar 01 22:45:57 xMff: I'm having no luck with glibc .. it's crazy - was compiling fine b4, but kernel wouldn't boot. now kernel boots (up to init) but glibc fails to build *sigh* Mar 01 22:47:26 tripolar_: or go to failsafe mode and mount and edit jffs Mar 01 22:47:42 cshore i still like my dirt pivot_root hack more :) Mar 01 22:47:47 dirty Mar 01 22:48:14 tripolar_: you're entitled to your opinion :) Mar 01 22:48:55 i know - but your solution is better - if someone likes to change the external root partion - this isnt possible with my hack Mar 01 22:49:46 tripolar_: genericity was what I was aiming for Mar 01 22:50:09 tripolar_: let people do what they want Mar 01 22:50:31 good idea Mar 01 22:51:29 cshore if the external root device isnt attached although it is configured - would the sys still come up? Mar 01 22:51:53 yes...using the jffs Mar 01 22:52:11 tripolar: it's falls back Mar 01 22:52:16 okay Mar 01 22:53:01 tripolar_: that was a popularly requested feature Mar 01 22:53:15 frogonwheels: had my share of fun with uclibc as well, they all can be a pita Mar 01 23:02:08 is there some way when i add an option in uclibc to get the include files into stating dir without a make toolchain/{clean,compile}? Mar 01 23:02:37 nbd do u know this? Mar 01 23:08:08 gnight all Mar 01 23:15:40 has someone expirience with package makesfiles? a compile attr and installed its libs into staging_dir but i still get this error when compiling libcap Mar 01 23:15:44 /mnt/data/Openwrt/trunk_git/staging_dir/toolchain-mipsel_gcc-4.3.3+cs_uClibc-0.9.30.2/usr/lib/gcc/mipsel-openwrt-linux-uclibc/4.3.3/../../../../mipsel-openwrt-linux-uclibc/bin/ld: cannot find -lattr Mar 01 23:16:03 although i have Mar 01 23:16:03 staging_dir/target-mipsel_uClibc-0.9.30.2/usr/lib/libattr.la Mar 01 23:16:04 staging_dir/target-mipsel_uClibc-0.9.30.2/usr/lib/libattr.a Mar 01 23:16:04 staging_dir/target-mipsel_uClibc-0.9.30.2/usr/lib/libattr.so Mar 01 23:43:01 cshore: the brcm63xx b43 issue is because the mac address is not detected, /sys/class/ieee80211/phy0/macaddress is all zeroes Mar 01 23:49:15 xMff: what can I do about it?....wait for patch? Mar 01 23:49:31 xMff: or what? Mar 01 23:49:50 marc said there was a patch for this already, maybe it got lost during the .30 -> .32 transition Mar 01 23:50:01 xMff: ah, I see, ok Mar 01 23:50:17 xMff: I'll look Mar 01 23:50:55 <{Nico}> cshore: back Mar 01 23:51:10 {Nico}: hi, you were trying to contact me? Mar 01 23:51:46 <{Nico}> yeah, some potential bugs in firstboot Mar 01 23:51:48 <{Nico}> http://openwrt.pastebin.com/GcFJtZSQ Mar 01 23:52:30 <{Nico}> and Mar 01 23:52:31 <{Nico}> http://openwrt.pastebin.com/kFpzYRBD Mar 01 23:52:35 yes, that was reported, but I don't have access to fix it. Your fix is right Mar 01 23:54:12 {Nico}, yes that one too Mar 01 23:54:48 nico * r19944 /trunk/package/base-files/files/lib/firstboot/99_10_with_fo_cleanup: firstboot: fix typo (closes: #6738) Mar 01 23:55:05 cshore: I think [florian] just forgot to rename files-2.6.30 to files-2.6.32 (or files) when bumping brcm63xx to .32 Mar 01 23:55:39 xMff: probably...I don't know when he bumped, but last I checked there was only 2.6.30 Mar 01 23:56:01 xMff: if 2.6.32 is missing should I rename it? Mar 01 23:56:03 <{Nico}> last update to 2.6.33 was a mess Mar 01 23:56:04 at least the sprom fixups are in files-2.6.30 Mar 01 23:56:20 files/ is just a bunch of empty directories (git...) Mar 01 23:56:26 <{Nico}> they should have used 'svn cp' to keep history Mar 01 23:56:45 xMff: oh, wait one of the upcoming revisions has the stuff in the kernel tree Mar 01 23:57:00 mainline I mean Mar 01 23:57:03 <{Nico}> i'm wondering if it's worth reverting everything and starting again Mar 01 23:57:05 all of it? even the sprom fixes? Mar 01 23:57:15 xMff: I'm not sure Mar 01 23:57:52 xMff: I just know Florian said that brcm63xx had been merged into mainline for .32 or .33 (not sure which) Mar 01 23:58:52 build #28 of ppc44x is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/ppc44x/builds/28 Mar 02 00:02:00 <{Nico}> for the 2nd fix, i would say package/base-files/files/lib/firstboot/10_determine_parts is not needed anymore Mar 02 00:02:16 <{Nico}> except the check for squashfs/jffs2 Mar 02 00:04:34 cshore: http://pastebin.ca/1818122 - this are the diffs between 2.6.32.9 and the old stuff for 2.6.30 Mar 02 00:05:14 I think it's needed for the squashfs vs other fs check Mar 02 00:07:42 xMff: I think most of those differences are new or things we need for OpenWRT Mar 02 00:08:10 green is stuff that was removed upstream, red is stuff that is upstream but not in openwrt Mar 02 00:08:18 should have reversed it Mar 02 00:08:36 xMff: I figured it out Mar 02 00:08:44 anyway - this is above my paygrade Mar 02 00:08:54 heheh Mar 02 00:09:34 I think the problem for WLAN is the CONFIG_SSB_PCIHOST Mar 02 00:09:44 ifdef Mar 02 00:10:20 yes I think that too but Marc just showed me a patch where some srom content was just patched into the kernel instead of loading it from somewhere Mar 02 00:10:36 hmm.... Mar 02 00:20:38 oh boy, the wrt54gl I got back since yesterday is running dd-wrt Mar 02 00:20:40 yuk Mar 02 00:21:53 xMff: Marc's patch is very old...I don't think we need it anymore Mar 02 00:24:58 cshore: allright Mar 02 00:25:24 I think I need to figure what files we need from files Mar 02 00:27:37 xMff: I think they need to be converted to patches since some of the files are included in mainline now Mar 02 00:28:12 cshore: right Mar 02 01:08:03 build #19 of gemini is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/gemini/builds/19 **** ENDING LOGGING AT Tue Mar 02 02:59:58 2010