**** BEGIN LOGGING AT Sat Jul 05 02:59:57 2008 Jul 05 05:08:26 thoughts on http://www.myopenrouter.com/ ? Jul 05 05:09:59 dwb: it prompts me to install Flash Jul 05 05:10:13 = I don't like it Jul 05 05:12:11 http://www.techworld.com/mobility/news/index.cfm?newsID=102078&pagtype=all Jul 05 05:13:01 Bummed they didn't mention OpenWrt (or even CoovaAP), but they link to DD-WRT, Tomato, and Sveasoft... ouch Jul 05 05:13:28 dwb: it might help if OpenWrt had a stable release :þ Jul 05 05:13:58 yeah, I wonder... Jul 05 05:14:18 btw, that article *does* mention OpenWrt Jul 05 05:14:42 Perhaps they would have been happier if kamikaze kept bit of whiterussian like nvram -- at least for broadcom Jul 05 05:14:54 Oh... :) didn't see it when scanning :) Jul 05 05:15:05 shrug, just get decent docs and mark it stable Jul 05 05:15:21 The most popular open source firmware, Tomato and DD-WRT, are available on WGR614L, and support will soon be added for OpenWRT. Jul 05 05:15:48 Is Tomato and DD-WRT really the most popular? I guess Jul 05 05:16:24 they have WUIs? Jul 05 05:20:11 Ok, they talk about OpenWrt in the myrouter site: http://www.myopenrouter.com/blog/entry/10062/WGR614L---Router-for-developers/ Jul 05 05:22:01 dunno what's so special about that router Jul 05 05:22:05 it doesn't even have 8 MB flash Jul 05 05:23:30 I don't know, but guess it is just a response to the Linksys GL unofficial "open-source router" -- they just made it more official with a hardware warranty that will stay intact Jul 05 05:30:56 I guess kamikaze will have a webif built in for the next release? wondering how 'big' it will be (with all dependencies)... Jul 05 05:32:05 I assume building without the webif will be optional in the buildroot Jul 05 05:32:44 I was wondering a little about the webif too Jul 05 05:34:34 I'm of two minds: one, it's great for openwrt to have a real webif for kamikaze. but two, I'm not convinced lua is right for me. another (more obscure) language to learn. hmm. Jul 05 05:37:41 there is some info if you google it... Jul 05 05:37:52 and I just hate mixing HTML with functionality.... Jul 05 05:38:05 I now... it seems to be 'up and coming'... Jul 05 05:38:43 I guess I'm wondering what's wrong with 'sh' ... it's a bit more universal Jul 05 05:38:55 I haven't gotten to look at the syntax yet Jul 05 05:39:17 and more useful to more people... Jul 05 05:41:04 it gives the noobs a warm fuzzy feeling, makes it easier to get people to try it Jul 05 05:42:14 True... just hoped it would have more for the non-noob... but, I wouldn't say using lua makes it more accessible for noob-devs to work on... Jul 05 05:43:52 I guess I've spent too much time with my own ideas for a new webif :) Jul 05 05:44:20 dwb: I know how you feel Jul 05 05:46:05 I have a feeling its not going to suck too bad Jul 05 05:46:28 Just hoping that everything doesn't change to lua outside of the webif... I don't believe the webif technology should be tied to more than GUI Jul 05 05:47:01 nothing OpenWrt sucks... :) Jul 05 05:52:16 I have also been thinking: since the GUI is configuring apps not specific to OpenWrt, it is interesting to have a GUI platform that also isn't tied to OpenWrt. Perhaps then better GUIs can be built by a larger community that extends beyond openwrt... I know.. wishful thinking. Jul 05 05:54:47 http://www.vimeo.com/1068368 <- I dont understand Jul 05 05:58:23 me neither Jul 05 06:01:24 http://ap.coova.org/cgi-bin/ewt-cgi/com.coova.ewt.Home/Home.html <-- example GUI .. I have a atheros firmware built with it, if anyone wants to play with it Jul 05 06:02:22 The idea is simple: The GUI is a self-contained "application" that uses Ajax "services" for all data communication... Jul 05 06:03:05 The GUI itself can be served up by the router, or even completely removed and put into a desktop app --leaving just the small 'services' scripts on the router Jul 05 06:03:52 The GUI 'logic' (what dialogs appear under what config conditions) is all coded in XML and is not limited to Openwrt.. Jul 05 06:04:07 http://coova.org/wordpress/index.php/2007/05/03/embedded-web-20/ a bit old now, but the idea remains the same... Jul 05 06:43:55 * Nix wanders in Jul 05 06:44:25 sup Nix Jul 05 06:44:35 hey there Jul 05 06:44:42 Nothing much.. just woke up Jul 05 06:44:43 :-) Jul 05 06:45:49 Hey, I have a new snapshot of the FreeRADIUS/JRadius GUI I've been working on ... can be used to edit mod_sql tables as of now (still working on it writing FR configs) Jul 05 06:46:32 nice Jul 05 06:46:52 the new 2.1 freeradius is going to be NIIICE.. Jul 05 06:47:16 I really need to find the time to finish off my mod_python UI for it... Jul 05 06:47:33 I am interested to see your efforts too of course ;-) Jul 05 06:47:46 hehehe... so many UIs out there :) Jul 05 06:48:16 http://ap.coova.org/coova-jradius-server-1.0.0-SNAPSHOT-release.zip Jul 05 06:48:36 uses slightly modified SQL: http://dev.coova.org/svn/cjradius/files/freeradius.sql Jul 05 06:49:10 Configure configuration.properties to taste, then run "start.sh" and goto http://localhost:1800/ewt/com.coova.ewt.Home/JRadius.html Jul 05 06:50:07 ok.. adding lat and long to the nas table is a good idea Jul 05 06:50:36 Working on a simple JRadius handler that does the functions of mod_sql - but using Java Jul 05 06:50:42 have you tested with postgresql or only mysql? Jul 05 06:51:04 mysql... but in Java, it's just a matter of getting the right driver jar and changing configuration.properties Jul 05 06:51:12 yeah. I did something similar with python.. seemed easier to handle the DB myself.. Jul 05 06:51:39 but they is only because of some limits that existed.. I think that may no longer be the case Jul 05 06:52:10 I should release my session killer.. its so much better then simultaneous use checks. Jul 05 06:52:41 instead of rejecting the second sesssion, you kill the first one with PoD... which is actually what most people want.. Jul 05 06:52:51 because it is usually stale.. Jul 05 06:53:06 yeah... I plan on adding more functionality like that into the JRadius SQLHandler too... Jul 05 06:53:25 and in the case of abuse.. you cant tell if the "valid" session is the first of second anyway.. so the behaviour is no more broken than the original.. Jul 05 06:53:29 to add more features 'around' the standard mod_sql tables... for better integration. Jul 05 06:54:08 how is the coova firmware coming along? I havent had a chance to look at it for a while.. Jul 05 06:54:31 have been concentrating on voip products.. not wireless Jul 05 06:54:39 with the snapshot above, the most usuable bit is the SQL table editors... which do name completion for you on the attribute name (no more misspellings!) ;) Jul 05 06:54:47 :-) Jul 05 06:55:14 I have a snapshot of a kamikaze version... complete with CoovaEWT GUI -- but, it's not fully functional. Jul 05 06:55:26 ok Jul 05 06:55:32 I have it built for atheros, I can give you links if you want Jul 05 06:55:47 Hmm.. no its ok.. I can build it myself ;-) Jul 05 06:56:02 I must find the time to throw it on some of our danube boards.. Jul 05 06:56:06 ... except for the GUI bit.. Jul 05 06:56:14 they are really powerfull.. 333MHz MIPs.. Jul 05 06:56:28 with onboard atheros wifi Jul 05 06:56:45 nice.. Jul 05 06:57:49 http://ap.coova.org/kamikaze/mips/openwrt-atheros-root.squashfs & http://ap.coova.org/kamikaze/mips/openwrt-atheros-vmlinux.lzma Jul 05 06:58:21 It's standard kamikaze (svn, about a month old) + GUI Jul 05 06:58:26 and ram problems are a thing of the past.. we have 64MB İN THESE BABİES :-) Jul 05 06:58:41 very nice... Jul 05 06:58:49 is the voip bit all opensourced too? Jul 05 07:04:51 not yet.. Jul 05 07:04:55 ;-) Jul 05 07:05:24 we use an open source stack.. but we rely on Infineon's TAPI and DSP driver set which is closed :-( Jul 05 07:12:50 b44.c Jul 05 07:13:48 I don't really know how to interpret the table at the start of http://wiki.openwrt.org/OpenWrtDocs/Hardware/Asus/WL500GP Jul 05 07:14:47 If you are using the Broadcom wireless card, does the open source driver (b44.c) work, allowing one to use a new enough 2.6 kernel? Jul 05 07:16:20 Or were the random sefaults fix in SVN 9285 required for running 2.6 even if one has a transplanted Atheros wireless? Jul 05 07:16:21 b44 is the switch driver Jul 05 07:17:11 oh, I thought I'd seen b44.c on a PC for the wireless. I must have been mistaken. Jul 05 07:20:24 http://www.turkeytravelplanner.com/fun/BackwardsMermaid.html Jul 05 07:20:26 rofl Jul 05 07:23:59 b43 is the name of the wireless card driver and 42 is the meaning of the universe. Jul 05 07:24:54 I have the impression that the b43 driver works on PCs but not in OpenWRT. Is this correct? Jul 05 07:33:52 DHR: it somewhat works in sta mode Jul 05 08:06:13 loswillios: can you check if the change I just made is accurate? It is at the start, before the first table: http://wiki.openwrt.org/OpenWrtDocs/Hardware/Asus/WL500GP Jul 05 08:06:43 I don't actually know what STA mode is. If I did, I would have put a link there. Jul 05 08:17:12 DHR: sta is client mode Jul 05 08:18:29 DHR: leave the ticketlinks (2677,9285) out. they are only confusing and don't matter Jul 05 10:05:50 florian * r11632 /packages/ipv6/babel/Makefile: Upgrade babel to 0.14 Jul 05 10:18:45 florian * r11633 /packages/utils/zile/Makefile: Upgrade zile to 2.2.59 (#3700) Jul 05 10:21:51 florian * r11634 /packages/utils/tar/Makefile: Upgrade tar to 1.20 (#3698) Jul 05 10:24:01 florian * r11635 /packages/utils/stress/Makefile: Update stress to 1.0.0 (#3697) Jul 05 10:26:01 florian * r11636 /packages/utils/smartmontools/ (Makefile patches/001-compile-fix.patch): Upgrade smartmontools to 5.38 (#3696) Jul 05 10:29:33 florian * r11637 /packages/utils/setpwc/Makefile: Upgrade setpwc to 1.2 (#3695) Jul 05 10:33:32 florian * r11638 /packages/utils/sdparm/Makefile: Upgrade sdparm to 1.03 (#3694) Jul 05 10:49:29 florian * r11639 /packages/utils/rdiff-backup/ (Makefile patches/001-python_path.patch): Upgrade rdiff-backup to 1.0.5 (#3691) Jul 05 12:15:14 florian * r11640 /packages/utils/pwgen/Makefile: Upgrade pwgen to 2.06 (#3690) Jul 05 12:18:56 florian * r11641 /packages/utils/pciutils/Makefile: Upgrade pciutils to 3.0.0 (#3689) Jul 05 12:25:02 florian * r11642 /packages/utils/mgetty/ (Makefile patches/001-mgetty.patch): Upgrade mgetty to 1.1.36 (#3687) Jul 05 12:25:35 florian * r11643 /packages/utils/mksh/ (Makefile patches/100-dot_mkshrc): Upgrade mksh to R33d (#3688) Jul 05 12:26:41 florian * r11644 /packages/utils/loop-aes/Makefile: Upgrade loop-aes to 3.2c (#3686) Jul 05 12:29:35 florian * r11645 /packages/utils/i2c-tools/Makefile: Upgrade i2c-tools to 3.0.1 (#3684) Jul 05 12:31:47 florian * r11646 /packages/utils/heyu/ (Makefile patches/001-avoid_configure.patch): Upgrade heyu to 2.3.2 (#3683) Jul 05 12:33:16 florian * r11647 /packages/utils/hdparm/ (Makefile patches/100-compile-fix.patch): Upgrade hdparm to 8.9 (#3682) Jul 05 12:34:38 florian * r11648 /packages/utils/empty/Makefile: Upgrade empty to 0.6.15b (#3681) Jul 05 12:35:53 florian * r11649 /packages/utils/dtach/Makefile: Upgrade dtach to 0.8 (#3680) Jul 05 12:42:25 florian * r11650 /packages/utils/bemused/ (Makefile patches/100-iconv.patch): Upgrade bemused to r062 (#3679) **** ENDING LOGGING AT Sat Jul 05 12:43:11 2008 **** BEGIN LOGGING AT Sat Jul 05 12:43:32 2008 Jul 05 12:45:03 florian * r11651 /packages/multimedia/motion/ (3 files in 2 dirs): Upgrade motion to 3.2.10.1 (#3678) Jul 05 12:47:08 florian * r11652 /packages/mail/fetchmail/ (3 files in 2 dirs): Upgrade fetchmail to 6.3.9-rc2 (#3677) Jul 05 12:59:14 florian * r11653 /packages/net/apache/ (6 files in 3 dirs): Upgrade apache to 2.2.9 (#3671) Jul 05 13:02:14 florian * r11654 /packages/net/xl2tpd/ (Makefile patches/001-opt_flags.patch): Upgrade xl2tpd to 1.1.12 (#3670) Jul 05 13:05:04 florian * r11655 /packages/net/xinetd/ (4 files in 2 dirs): Upgrade xinetd to 2.3.14 (#3669) Jul 05 13:07:16 florian * r11656 /packages/net/wifidog/ (Makefile files/wifidog.conf): Upgrade wifidog to 1.1.5 (#3667) Jul 05 13:15:00 florian * r11657 /packages/net/transmission/Makefile: Upgrade transmission to 1.22 (#3665) Jul 05 13:18:39 florian * r11658 /packages/net/wdiag/Makefile: Upgrade wdiag to 0.9 (#3664) Jul 05 13:20:50 florian * r11659 /packages/net/vpnc/ (3 files in 2 dirs): Upgrade vpnc to 0.5.1 (#3663) Jul 05 13:24:12 florian * r11660 /packages/net/unfs3/ (Makefile patches/100-no_lfl.patch): Upgrade unfs3 to 0.9.20 (#3662) Jul 05 13:32:37 florian * r11661 /packages/net/vpnc/patches/003-no_makeman.patch: Fix vpnc installation on man pages Jul 05 13:38:24 florian * r11662 /trunk/target/linux/rdc/image/Makefile: Jul 05 13:38:24 If the *.img file is larger than 0x1e8000 loading it with Jul 05 13:38:24 flash -s 0x100000 -d 0xffe08000 -l Jul 05 13:38:24 on the bootloader prompt will destroy the bootloader and brick the device. The attached patch tries to implement a size check and renames the image into *.img.too_big. It applies to svn r11401. Jul 05 13:38:26 Signed-off-by: Joerg Albert Jul 05 13:43:57 florian * r11663 /trunk/package/dnsmasq/files/dnsmasq.init: Jul 05 13:43:57 This adds dhcp-boot support to dnsmasq-init: Jul 05 13:43:57 option name "virt0" Jul 05 13:43:57 option filename "pxelinux.0" Jul 05 13:43:57 option servername "mainframe0.dev" Jul 05 13:43:57 option serveraddress "9.3.2.1" Jul 05 13:43:59 closes #3529 Jul 05 13:59:15 florian * r11664 /packages/ipv6/ahcpd/files/ahcpd.config: Add a note to ahcpd to explain that the init script is slightly broken with the no_dns option (#3550) Jul 05 14:49:12 matteo * r11665 /trunk/target/linux/avr32/ (4 files in 2 dirs): avr32: update to latest atmel sources Jul 05 16:49:11 kaloz * r11666 /trunk/target/linux/ixp4xx/patches-2.6.26/110-pronghorn_series_support.patch: update pronghorn patch Jul 05 16:52:17 kaloz * r11667 /trunk/ (11 files in 11 dirs): update targets to 2.6.25.10 Jul 05 18:44:28 matteo * r11668 /packages/lang/perl/patches/100-static_uudmap.patch: perl: finally fix the uudmap.h bug (closes #3709) Jul 05 19:29:56 matteo * r11669 /packages/lang/perl/patches/100-static_uudmap.patch: perl: refresh patch and fix generate_uudmap permissions Jul 05 19:30:51 stupid stupid stupid perl Jul 05 20:57:39 hi Jul 05 20:57:45 aroscha: huh? Jul 05 20:58:02 nico, florian - sugar is running on the classmate 2 Jul 05 22:11:25 nbd * r11670 /trunk/include/kernel-defaults.mk: spurious initramfs breakage fix Jul 05 22:22:22 blogic * r11671 /packages/lang/perl/ (files/config.sh-i386-linux-uclibc.in perlbase.mk): adds a dependency and fixes compile for x86 Jul 05 22:24:09 I'm interested in going to 2.6 on a system with the broadcom wireless. Unfortunately the open source b43 only works in STA mode, I understand. Does anyone know if this is likely to change soon? Jul 05 22:24:42 i sucessfully used b43 in ad-hoc mode some time ago Jul 05 22:25:08 support in recent kernel version seems to have matured a bit Jul 05 22:25:13 great news! Jul 05 22:25:38 matteo * r11672 /packages/lang/perl/ (6 files in 2 dirs): more perl fixes Jul 05 22:26:08 but i don't know what's the current state in recent openwrt trunk version Jul 05 22:27:09 xMff: Is this written up anywhere? I've looked a bit in the bcm43xx-dev mailing list archives and on their web site. And on OpenWRT's. But I certainly could have missed things. Jul 05 22:27:18 there where some issues in openwrt when you configured b43 to use ad-hoc mode Jul 05 22:27:53 the system always came up in manged mode, you had to do a ifconfig down, iwconfig, ifconfig up to get it to work Jul 05 22:28:23 hmm, i just follow recent kernel developments, i own a broadcom 4306 pcmcia card Jul 05 22:29:05 bcm43xx seems to be basically dead Jul 05 22:29:23 I want to use 2.6 on an Asus WL-500gP to run Openswan. Openswan has little support for 2.4 Jul 05 22:29:34 i get my informations from testing kernels on my laptop and by following commitlogs in wireless-testing.git Jul 05 22:30:18 what kind of w-lan mode do you intend to use? Jul 05 22:30:21 master? Jul 05 22:30:25 The bcm43xx-dev list is now really for the b43 and b43-legacy Jul 05 22:30:34 ah okay Jul 05 22:31:10 i didn't bother to check the lists back then, i just made my hands dirty by compiling stuff and searching for errors afterward ;) Jul 05 22:31:27 I don't know enough about wireless modes. Jul 05 22:31:53 Hacking on notebooks is a little more "comfortable" than hacking on routers. Jul 05 22:32:02 ok, I assume you want to use your wl500 as access point? Jul 05 22:32:22 not in client mode or so Jul 05 22:32:26 although the serial console is disappearing from notebooks and is still in many routers. Jul 05 22:32:30 Right. Jul 05 22:33:13 DHR: Johannes Berg is working on mac80211 AP mode using b43 as his driver of choice. Jul 05 22:33:35 ah, good news :) Jul 05 22:33:45 yes! Jul 05 22:33:52 * xMff is waiting for proper ad-hoc support Jul 05 22:34:32 http://vger.kernel.org/vger-lists.html#linux-wireless is the list to track. Jul 05 22:34:49 My impression is that OpenWRT would be improved if the various systems with Broadcom wireless (eg. the iconic WRT54gl) could move to 2.6 Jul 05 22:35:47 not to mention the flexibility gains if you can use an up-to-date 2.6 kernel Jul 05 22:37:07 farnz: is there an overlap between the kernel wireless list and https://lists.berlios.de/mailman/listinfo/bcm43xx-dev ? Jul 05 22:37:44 Looks like there's similar people on both. Jul 05 22:40:56 DHR: I'm working on bcm33xx :þ Jul 05 22:41:16 Surely you'd be working on b43 by now? :) Jul 05 22:41:23 what do I care about b43? Jul 05 22:41:53 Sorry, misread; thought you said bcm43xx (which was what people were on about). Jul 05 22:42:09 isn't b43 the successor to b43xx and hence where the majority of progress is being made? Jul 05 22:42:26 s/progress/effort/ maybe Jul 05 22:42:43 DHR: isn't b43 just wireless? Jul 05 22:42:58 jep Jul 05 22:43:10 hum Jul 05 22:43:12 I think so. And b44 is the/a switch, apparently Jul 05 22:43:17 bcm33xx doesn't even have wireless :þ Jul 05 22:43:36 bcm43xx is what b43 used to be called. bcm33xx is only related in as much as they're both from Broadcom Jul 05 22:43:57 bcm43xx wasn't the whole platform? Jul 05 22:44:01 DHR: b44 is an Ethernet chip (the BCM43xx series). Jul 05 22:44:04 luke-jr: No. Jul 05 22:44:08 i c Jul 05 22:44:12 Oh, I misread. I assumed you mean b43xx when you typed bcm33xx. Jul 05 22:45:07 b44 is a driver for the bcm44xx and bcm47xx series of Ethernet chips. Jul 05 22:45:46 from last night: b44 is the switch driver Jul 05 22:46:43 DHR: I'm looking at Linux git as of Friday. Jul 05 22:48:05 I was confused by numbers then too. I was looking at https://dev.openwrt.org/changeset/9285 and thought it had something to do with wireless (from an ambiguous reference in the wl800gP wiki page). Jul 05 22:48:31 Ah, the b44 series ethernet chips are often paired with Broadcom's roboswitch. Jul 05 22:49:11 so anyone want to assist with bcm33xx? Jul 05 23:06:11 blogic * r11673 /trunk/target/linux/ifxmips/files/ (18 files in 5 dirs): lots of ifxmips fixes and features Jul 05 23:22:22 blogic * r11674 /trunk/target/linux/ifxmips/files/arch/mips/ifxmips/board.c: use boards mac on boards with brn loader, on ifxmips Jul 06 00:15:20 luke-jr: I don't even know if I have a bcm33xx. What is it in? Jul 06 00:15:36 DHR: about 50% of cable modems Jul 06 00:17:36 I have a cable modem: Motorola SB85100. I don't know what I could gain by hacking it. Jul 06 00:19:20 5100? Jul 06 00:19:26 that should have a bcm3349 inside Jul 06 00:20:07 yeah, 5100. I seem to be having trouble with numbers. Jul 06 00:20:29 what do you gain by hacking routers at all? Jul 06 00:21:05 I'm just wanting to get rid of proprietary software Jul 06 00:21:49 but potential gains are things like increasing your download speeds, unlimited IP addresses, unblocking blocked ports, etc Jul 06 00:22:03 I'm not using the 5100 as a router, more like a modem. I have a PC as a router. I'd like to replace it with something lighter weight like the Asus box. Jul 06 00:22:07 at least some of those are illegal if your ISP doesn't like them tho Jul 06 00:22:44 I try to follow rules, but some I find unconscionable. Like "no servers". Jul 06 00:22:50 yeah Jul 06 00:23:00 if we get Linux on the modem, we can bypass that Jul 06 00:23:05 ;) Jul 06 00:23:36 or even run your own private LAN on the cable network :þ Jul 06 00:24:03 Really? Are modems able to act like head-ends on cable? Jul 06 00:24:17 cable networks are more or less hubs ☺ Jul 06 00:24:28 the modems just download config files to limit speeds and such when they boot Jul 06 00:24:53 run tcpdump on your WAN and watch the ARP flooding ;) Jul 06 00:25:05 my DPC2100 (bcm3349) runs eCos, so I'm hoping to get some usable source from the GPL violation Jul 06 00:25:15 I know that you see broadcast stuff, but that's about it. Jul 06 00:25:37 DHR: the only reason you don't see everything is because the modems use proxy ARP instead of real bridging Jul 06 00:25:51 if they bridged, you'd see all the traffic on the cable network segment Jul 06 00:26:22 (unless the ISPs enable BPI+, in which case you'd see encrypted data) Jul 06 00:26:31 I see bootp crap from far and wide. I think larger than the segment. Jul 06 00:26:47 segments are pretty big ;) Jul 06 00:27:30 hehe, i get dhcp requests via wan on my cable connection Jul 06 00:27:37 the only reason your ISP can pinpoint you at all is because they have your modem's MAC in a database with your address Jul 06 00:27:46 otherwise cable would be anonymous Jul 06 00:28:02 just don't serve any of those dhcp requests! Jul 06 00:28:08 lol Jul 06 00:28:56 they seem to have my ethernet interface's mac address too. Jul 06 00:28:58 DHR: also, with Linux we could potentially get IPv6 support Jul 06 00:29:24 but the ISP would not offer IPv6 transit, I expect. Jul 06 00:29:31 shrug Jul 06 00:29:32 dunno Jul 06 00:29:45 cable IPv6 usually needs DOCSIS 3.0 Jul 06 00:29:53 which isn't supported by 99% of modems Jul 06 00:30:06 I have a portable class C that I get routed in through ADSL. The cable company would not do that for many reasons. Jul 06 00:30:29 they are control freaks Jul 06 00:30:38 DHR: your modem also limtis the # of DHCP clients on your network Jul 06 00:30:54 otherwise, your ISP will hand out as many IPs as you want Jul 06 00:30:56 ;) Jul 06 00:31:14 Very interesting. Jul 06 00:31:40 So: if you have a hacked modem, and cloned the MAC address of a registered modem, you could have fun? Jul 06 00:31:57 eh, if you have that you can steal service too Jul 06 00:32:15 but I wouldn't blame the cable co for going after you on that one ;) Jul 06 00:33:04 I think that they offered 2 IP addresses for extra money. Perhaps even more. Jul 06 00:33:20 Time Warner includes 4 with the basic residential service Jul 06 00:33:28 and allows servers Jul 06 00:33:53 I'm with Rogers (in Toronto). They act like the monopolists that they are. Jul 06 00:34:09 I'm with Cox. they suck too Jul 06 00:34:10 :þ Jul 06 00:34:56 anyhow, bcm33xx seem to do some weird stuff at least to boot Jul 06 00:35:09 qemu gets stuck looping an exception if I try booting the firmware in it Jul 06 00:35:38 Since you are not the copyright holder, what is your strategy to get the modem maker on eCos GPL issues? Jul 06 00:35:47 I emailed the copyright holder. Jul 06 00:35:51 Hopefully they will act. Jul 06 00:35:53 redhat? Jul 06 00:36:08 ecos-maintainers, gpl-violations, and FSF Jul 06 00:36:52 I've tried to poke Linksys about GPL violations (as a customer and a copyright holder) and have not gotten very far. Jul 06 00:37:06 ask a lawyer to write them a letter? ;) Jul 06 00:37:29 I did start the busybox vs. Hava war. Jul 06 00:37:36 (get them to surrender the b43 wifi code while you're at it) Jul 06 00:38:53 DHR: I used JTAG to dump the flash, so in theory it can be reverse engineered too Jul 06 00:39:22 many folks think that binary drivers are kosher if they are modules. I think that it is an untested theory. Jul 06 00:39:36 I think it's unquestionably illegal. ☺ Jul 06 00:40:43 I disassembled firmware in the days when 16k was a lot. I hate to think what it is like now with megabytes of code. Jul 06 00:41:49 nbd * r11675 /trunk/package/madwifi/patches/200-no_debug.patch: be a bit more verbose about madwifi errors Jul 06 00:43:02 DHR: 770 KB for eCos Jul 06 00:43:17 DHR: and I have symbol tables for the VxWorks images Jul 06 00:43:22 help a bit? ;) Jul 06 00:44:33 symbol tables include addresses of cours Jul 06 00:44:35 symbols help, of course. But big code is tough even if you have the source. Especially without hardware documentation. Jul 06 00:45:27 If I had space cycles for disassembly, I think that the Broadcom wireless would be top of my list. Jul 06 00:45:33 s/space/spare Jul 06 00:47:13 that's already been reveng'd Jul 06 00:47:15 I spent much of yesterday trying to understand someone elses lex and yacc code. Mostly because it was crap. That's higher level than MIPS or ARM binaries Jul 06 00:48:47 the broadcom wireless isn't reverse engineered enough to work well in OpenWRT for an AP. As far as I can tell. Jul 06 00:50:56 did you read the docs? Jul 06 00:51:19 the reveng can be good without the reeng being complete ;) Jul 06 00:51:38 On the cable side, I'm annoyed that digital TV will preclude third part PVRs (like my Myth box). Here, all channels are encrypted (in US, FCC requires local channels to be in the clear, as I understand it) Jul 06 00:52:11 might be able to add TV support to the cable modems too i guess Jul 06 00:52:17 but TV isn't worth a crap IMO Jul 06 00:52:17 :þ Jul 06 00:53:00 We don't have DMCA here. But we might soon. So hacking cable TV box is more interesting than hacking cable modem. Jul 06 00:53:49 I think my cable box has a SPARC chip in it. Jul 06 00:54:02 cable TV box Jul 06 00:59:50 blogic * r11676 /trunk/ (6 files in 5 dirs): make atheros wifi cards used on arcaydian 4519 ifxmips based boards work Jul 06 01:33:55 o **** ENDING LOGGING AT Sun Jul 06 02:59:56 2008