**** BEGIN LOGGING AT Thu Jan 27 02:59:57 2011 Jan 27 03:01:09 philipp64: I'm not sure of the situation with libnl-tiny vs libnl so I'm not touching that patch...Jo should know Jan 27 03:01:56 philipp64: but I suspect if you're going to and -libnl-genl to things with libnl then will need a suitable libnl-genl-tiny as well Jan 27 03:02:01 *add Jan 27 03:02:17 libnl is related to nls_base ? Jan 27 03:02:27 no netlink Jan 27 03:02:53 looking at philipp's packages on -devel Jan 27 03:02:53 oki :) Jan 27 03:03:34 i ve seen a patch which enhance ath5k tx by 2 this morning Jan 27 03:04:05 (20 hours ago) Jan 27 03:04:07 :p Jan 27 03:04:22 4AM here Jan 27 03:13:26 i use ushare (DLNA) on the mounted network shares Jan 27 03:13:38 for my phone and tv :) Jan 27 03:13:48 More spam posts on the forum by the same spammer! Jan 27 03:14:02 yea, its automatic Jan 27 03:14:05 i think Jan 27 03:14:29 the best thing to do is to ignore the spambot posts Jan 27 03:14:30 Just remove the post and deactivate the account. Then, block the subnet. Jan 27 03:14:33 without errors Jan 27 03:14:40 else another account will be created Jan 27 03:15:06 You ignore them, they will come to post more spams on a daily base. Jan 27 03:15:22 mazilo: ignore as in don't actually post Jan 27 03:15:35 This will make the forum unreadable. That's what had happened to FreeWorldDialUp forum. Jan 27 03:15:45 ignore = not displayed Jan 27 03:15:54 i mean Jan 27 03:16:09 i've a protection on my drupal site... Jan 27 03:16:11 I can do that, but it will hurt the forum. Jan 27 03:16:21 some agent thing Jan 27 03:16:53 mazilo: we don't mean you should ignore, we mean once they're seen that account should have it's posts silently ignored Jan 27 03:16:56 When a nice forum starts to be inundated with spams, it will lose is readers and dies out. Jan 27 03:17:22 mazilo: by the forum software Jan 27 03:17:31 yea... same on blogs with comments Jan 27 03:17:33 cshore: That is true. But, it will invite more spammers to post spam posts if not taken care. Jan 27 03:17:55 mazilo: how so, the posts will not be visible Jan 27 03:18:11 mazilo: it's not like they go back and check the old post is still there Jan 27 03:18:31 Like I said, it won't hurt me once I saw even the title of the post and it will disappear from my browser after a few minutes later on once refreshed. Jan 27 03:18:38 module is called "badbehavior" on drupal Jan 27 03:19:44 http://bad-behavior.ioerror.us/documentation/how-it-works/ Jan 27 03:19:48 cshore: The spammers are using the forum to disperse their interests. The more spams they posts, the more points they get on the search engines. Jan 27 03:19:49 mazilo: that is we still need to be alerted about them Jan 27 03:20:20 mazilo: but once we know, instead of deleting and banning, delete, and ignore future posts Jan 27 03:20:43 (i.e. they never show up on the forum) Jan 27 03:21:15 yea :) its what i mean Jan 27 03:21:36 i dont think the bot check result Jan 27 03:21:41 s Jan 27 03:22:12 i dont have any on SMF Jan 27 03:22:23 phpbb is the most know Jan 27 03:22:27 n Jan 27 03:22:29 cshore: I agree with what you said. But, their posts are still there for others to encounter. Jan 27 03:22:37 SMF Small Machines Forum Jan 27 03:22:59 Epsylon3: Thanks for the link. Jan 27 03:23:03 mazilo: not if the old posts are deleted...and remember the new posts are not displayed to the other users Jan 27 03:23:44 mazilo: the bot only *thinks* it is successfully posting Jan 27 03:24:00 they maybe check for error with url thing Jan 27 03:24:03 I think that's a good idea (it doesn't do this currently) Jan 27 03:24:04 post/id Jan 27 03:25:17 cshore: I thought the old posts are still there for posterities to view. Jan 27 03:25:28 a spammer on my smf : http://forum.wdscript.fr/index.php?action=profile;u=1136 Jan 27 03:25:30 posts : 0 Jan 27 03:25:31 :) Jan 27 03:25:32 mazilo: I don't think so Jan 27 03:27:16 now the question is, is there something like this for phpbb Jan 27 03:27:17 ? Jan 27 03:27:22 cshore: If I search on the forum, all (new/old) posts related to the search item will get hit. Jan 27 03:28:15 mazilo: presumably not if the module is coded right Jan 27 03:28:38 cshore: IC. Jan 27 03:29:11 http://phsoftware.de/index.php/content/view/51/8/ Jan 27 03:29:19 PhBad Behave Jan 27 03:29:31 but old Jan 27 03:29:40 its for phpbb2 Jan 27 03:30:07 Various bits of documentation need to be updated. I need to host installation instructions, or links thereto, for any platform to which Bad Behavior has been ported, and several of those (such as phpBB and Movable Type) are missing right now. Jan 27 03:30:25 that means there is support for phpbb in dev 2.1.x Jan 27 03:30:29 http://bad-behavior.ioerror.us/2006/08/05/bad-behavior-21-and-30-roadmap/ Jan 27 03:30:35 ouch 2006 Jan 27 03:32:55 http://bad-behavior.ioerror.us/2011/01/25/bad-behavior-2-1-9/ Jan 27 03:32:59 more recent :) Jan 27 03:33:04 2 days ago Jan 27 03:37:13 Epsylong3: how recent is the build with the nls_base behaviour? Jan 27 03:37:30 It seems it ought to be fixed by AddDepends/nls Jan 27 03:37:39 which current trunk has Jan 27 03:39:33 2.6.37 Jan 27 03:39:40 today Jan 27 03:39:48 er, I meant...okay trunk as of today? Jan 27 03:39:53 yea :) Jan 27 03:40:02 hmmmm.....don't know why that is Jan 27 03:40:05 it should work Jan 27 03:40:23 that is nls_base should be a dependency Jan 27 03:40:43 kmod Jan 27 03:40:51 yeah, that's what I mean Jan 27 03:41:51 DEPENDS:=+kmod-nls-base +kmod-crypto-hmac +kmod-crypto-md5 Jan 27 03:41:58 "package/kernel/modules/fs.mk" Jan 27 03:42:04 to add Jan 27 03:42:07 ? Jan 27 03:42:13 i added that here Jan 27 03:42:13 that's not right Jan 27 03:42:24 no ? Jan 27 03:42:50 nls-base should already be added by $(call AddDepends/nls) Jan 27 03:43:02 kmod-nls-base I mean Jan 27 03:43:12 if required by that platform Jan 27 03:43:18 (not all platforms do) Jan 27 03:43:34 hmm Jan 27 03:43:36 maybe yours has changed from doesn't to does Jan 27 03:43:41 im on orion platform Jan 27 03:43:56 there is no many things in "package/kernel/modules/001-depends.mk" Jan 27 03:44:00 hmac and md5 are suggested rather than dependencies AIUI Jan 27 03:44:05 yea Jan 27 03:44:12 only on use Jan 27 03:44:19 not at boot time Jan 27 03:44:33 right, because not all situations require them Jan 27 03:44:46 dunno, dont have Win95 to check ;) Jan 27 03:44:49 hence making them a dependency would be wrong Jan 27 03:45:09 the problem is the error message is not clear Jan 27 03:45:22 (kernel one i think) Jan 27 03:45:27 VFS CIFS Jan 27 03:45:28 hmmm...well that should be more easily fixed perhaps Jan 27 03:45:45 could you open a ticket suggesting a better error message? Jan 27 03:46:02 hehe a comment is not enough ? Jan 27 03:46:08 in same ticket ? Jan 27 03:46:22 oh, you have one...sorry didn't see it Jan 27 03:46:32 CIFS VFS: could not allocate crypto hmacmd5 Jan 27 03:47:14 maybe a space is missing Jan 27 03:47:20 between hmac and md5 Jan 27 03:47:28 I think you're right Jan 27 03:48:10 but yea, there are cpus with hw crypto Jan 27 03:48:29 so maybe md5 could be provided by another module (not sure about that) Jan 27 03:48:46 hmm....hadn't thought of that....I think you're right hw could supply the crypto instead Jan 27 03:49:11 crypto-ocf or something like that Jan 27 03:49:36 mv-cesa for AES Jan 27 03:49:48 yeah Jan 27 03:50:12 marvell, c? aes (mv-caes) you mean? Jan 27 03:50:22 euh Jan 27 03:50:28 i check the name Jan 27 03:50:49 CONFIG_DEFAULT_kmod-crypto-mv-cesa Jan 27 03:51:01 name of the module :p Jan 27 03:51:03 hmmm....interesting. wonder what it stands for Jan 27 03:51:15 its for AP Jan 27 03:51:21 the AES Jan 27 03:51:34 I mean the acronymn cesa Jan 27 03:51:48 it probably means something to somebody Jan 27 03:52:14 http://www.mail-archive.com/linux-crypto@vger.kernel.org/msg03697.html Jan 27 03:53:25 iDMA support and wires up the CESA engine Jan 27 03:53:42 maybe a marvell name of the unit Jan 27 03:54:14 could be...it's not on wikipedia anyway Jan 27 03:54:27 i check the datasheet... Jan 27 03:55:56 http://tpruvot.free.fr/files/MV88F5182-opensource-manual.pdf Jan 27 03:56:05 nothing Jan 27 03:56:24 hmmmm....oh well, life little mysteries I guess Jan 27 03:56:29 *life's Jan 27 03:56:43 http://tpruvot.free.fr/files/MV88F5182-opensource-manual.pdf Jan 27 03:56:47 arf Jan 27 03:57:02 Security Cryptographic Engine Jan 27 03:57:21 http://tpruvot.free.fr/files/MV88F5182-usermanual.pdf Jan 27 03:57:29 there is details here Jan 27 03:57:31 heh, knew it meant something Jan 27 03:57:34 page 5 Jan 27 03:58:02 AES, DES MD5 3DES SHA1 Jan 27 03:58:14 so look like its not linked to cifs :p Jan 27 03:58:42 heh, right Jan 27 03:59:05 A = accelerator Jan 27 03:59:12 Security Accelerator Jan 27 03:59:32 Crypto Engine and Security Accelerator i think Jan 27 03:59:38 makes sense Jan 27 03:59:45 yea Jan 27 03:59:52 its title of chapter 11 Jan 27 03:59:54 maybe even no and Jan 27 04:00:00 heh Jan 27 04:00:04 or with :) Jan 27 04:00:27 Cryptographic Engines and Security Accelerator Jan 27 04:00:32 exactly Jan 27 04:00:34 who ever knew such choice could make life difficult ;-) Jan 27 04:00:47 4 engines, independant Jan 27 04:00:54 interesting Jan 27 04:01:01 so you could have four threads at once? Jan 27 04:01:52 not really Jan 27 04:01:59 yea Jan 27 04:02:06 but they cant do all the same things Jan 27 04:02:33 ah, so you if you had four different types of encryption you could do them at the same time, but no two the same? Jan 27 04:02:49 its like a FPU, you send and wait result Jan 27 04:02:53 right Jan 27 04:03:09 or a DSP probably Jan 27 04:03:48 though I think a DSP needs more guidance from supporting software Jan 27 04:03:57 hmmm Jan 27 04:04:02 maybe they can do 4 threads Jan 27 04:04:37 well, I've got to go....my dog is getting anxious to get outside Jan 27 04:05:09 HMAC is supported too Jan 27 04:05:33 ok.. :) good walk Jan 27 04:06:38 Yea and HMAC-MD5 directly Jan 27 04:07:25 sysupgrade... Jan 27 04:33:05 blogic: ping Jan 27 04:34:32 * The 5082/5181l/5182/6082/6082l/6183 have crypto Jan 27 04:34:32 * while 5180n/5181/5281 don't have crypto. Jan 27 04:35:13 good question, my uboot tells 5181l Jan 27 04:36:39 Orion ID: MV88F5181-Rev-B1. TCLK=166666667. Jan 27 04:38:07 jow_laptop: ping? Jan 27 04:47:30 no... ok mv_cesa is unused on our routers Jan 27 04:47:53 cpu is 5181 Jan 27 05:27:09 Hi anyone have some idea fix this error: http://tksite.gotdns.org:8010/broken_packages/xburst/fetchmail/compile.txt Jan 27 06:03:30 a missing include i think Jan 27 06:03:42 i dont have sources of fetchmail here to check Jan 27 06:03:51 md5c.c Jan 27 06:08:09 MD5Context Jan 27 06:08:22 struct (declared in fm_md5.h) Jan 27 06:08:33 http://gitorious.org/fetchmail/fetchmail/blobs/master/fm_md5.h Jan 27 06:13:05 Epsylon3: yes. I just check the "config.h" in fetchmail. it's have "#define HAVE_MD5_H 1" Jan 27 06:13:48 Epsylon3: but in STAGING_DIR/usr/include, there is no "md5.h" , that is why the md5c.c don't have the right head file Jan 27 06:14:10 hmmm check you have fm_md5 in fetchmail dir Jan 27 06:14:13 .h Jan 27 06:14:47 http://gitorious.org/fetchmail/fetchmail/blobs/master/md5c.c#line18 Jan 27 06:14:49 like here Jan 27 06:14:52 (trunk) Jan 27 06:15:29 Epsylon3: my version fm_md5.h is different with trunk version. Jan 27 06:16:03 MD5Context is in ? Jan 27 06:16:03 Epsylon3:the openwrt version of fetchmail fm_md5.h have such lines: Jan 27 06:16:04 #ifdef HAVE_MD5_H Jan 27 06:16:05 #include Jan 27 06:16:05 #else Jan 27 06:16:20 Epsylon3: it's have. but here is a define ^ Jan 27 06:16:33 oh Jan 27 06:16:41 ok i see Jan 27 06:17:00 so check your crypto module crypto-md5 Jan 27 06:17:05 its needed Jan 27 06:17:13 in kernel modules Jan 27 06:18:51 checking for library containing MD5Init... no Jan 27 06:20:01 http://manpages.unixforum.co.uk/man-pages/unix/freebsd-6.2/9/MD5Init-man-page.html Jan 27 06:20:05 #include Jan 27 06:20:20 dunno if only is ok Jan 27 06:25:26 Epsylon3: just tried , not working. Jan 27 06:25:30 Epsylon3: check this : http://gitorious.org/fetchmail/fetchmail/commit/738dcc102db67a741893e2979c034df9904e2925 Jan 27 06:25:49 Epsylon3: we need update to fetchmail to trunk version :) Jan 27 06:26:20 yea Jan 27 06:26:29 or disable HAVE_MD5 Jan 27 06:26:46 undef HAVE_MD5_H Jan 27 06:27:07 or... just include fm_md5 :) Jan 27 06:30:53 Epsylon3: yes, compare to create a patch , I would like update the fetchmail package's Makefile Jan 27 06:31:56 hmm wait next tag :) Jan 27 06:34:13 Epsylon3: how to check tag in gitorious.org? Jan 27 06:34:32 on the right... but gitorious is buggy Jan 27 06:34:36 git clone :) Jan 27 06:35:40 hmm Wednesday December 15 2010 Jan 27 06:35:56 http://gitorious.org/fetchmail/fetchmail/commits/next Jan 27 06:38:35 did you check you have md5 in kernel ? Jan 27 06:38:52 selected Jan 27 06:39:22 git clone http://git.gitorious.org/fetchmail/fetchmail.git; git tag Jan 27 06:41:55 sysupgrade... Jan 27 06:42:41 Epsylon3: md5 disabled in kernel. Jan 27 07:21:55 ^^ Jan 27 07:31:59 xiangfu : how to become a slave ? ^^ Jan 27 07:32:15 in build system... Jan 27 07:32:24 http://tksite.gotdns.org:8010/builders/orion Jan 27 07:34:25 howto include adm8668 in build system? :D Jan 27 07:34:47 i've a box with Gigabit connection Jan 27 07:36:34 feed install adm8668 -d m ? Jan 27 07:36:40 scripts/feed install adm8668 -d m ? Jan 27 07:36:42 Epsylon3: sorry for my bad English. don't understand "how to become a slave ?" what buildslave do? what should I do ? Jan 27 07:37:02 on the page : http://tksite.gotdns.org:8010/builders/orion Jan 27 07:37:11 we can see build "slaves" Jan 27 07:38:02 because, i compile the orion trunk 5 times per days ;) Jan 27 07:38:56 http://epsy.ath.cx:88/orion/ Jan 27 07:39:07 Epsylon3: hmm... I build out openwrt every two days. because the build need ~ 30 hours. include ~2400 packages Jan 27 07:39:14 s/out/our Jan 27 07:39:40 ouch Jan 27 07:39:54 :) yea in fact i dont compile every package Jan 27 07:40:21 the build slaves are users or computers? Jan 27 07:40:38 in fact i dunno Jan 27 07:40:48 its why i ask the question Jan 27 07:41:13 Epsylon3: I also don't know this buildbot. :( Jan 27 07:41:51 Epsylon3: we have two computer for develop openwrt, one is buildhost , build every two days. Jan 27 07:42:03 Epsylon3: http://fidelio.qi-hardware.com/~xiangfu/compile-log/ :) Jan 27 07:42:20 Epsylon3: those are compile script files : http://fidelio.qi-hardware.com/~xiangfu/bin/ Jan 27 07:42:24 30 hours sounds like a long time. I used to be able to build all of Astlinux (about 200 packages) in 2 hours on a Quad core Phenom-II w/ 8GB and a RAID array. Jan 27 07:42:45 ouch Jan 27 07:42:52 and how many Watts ? :p Jan 27 07:42:55 of course, you take the hit running "configure" over and over... since it can be parallelized. Jan 27 07:43:01 500W Jan 27 07:43:04 my box is 20W :p Jan 27 07:43:11 philipp64|laptop: my computer is not that good. 2GB mem, AMD 3.0G cpu Jan 27 07:43:14 Atom 330 Jan 27 07:43:17 dual core HT Jan 27 07:43:50 If I had 3 SSD's in a RAID array, I think I could cut it down by 30%. Jan 27 07:43:54 with ccache Jan 27 07:44:15 yea... Jan 27 07:44:19 i've one here Jan 27 07:44:21 Vertex 2 Jan 27 07:44:24 but after a while, everything basically fits into about 5GB of cache anyway... Jan 27 07:44:31 on my main computer, for web projects Jan 27 07:44:52 to search in sources Jan 27 07:45:03 anyway, time for bed. Jan 27 07:45:05 i think about buying one for my little box :) Jan 27 07:45:16 a bed? Jan 27 07:45:23 hehe :)) Jan 27 07:45:35 what's a vertex 2? Jan 27 07:45:40 OCZ Vertex 2 Jan 27 07:45:44 350 MB/s Jan 27 07:46:07 hmm Jan 27 07:46:19 no Jan 27 07:46:21 285MB Jan 27 07:46:31 in read or write Jan 27 07:46:43 its cheap Jan 27 07:46:46 yeah... having the 0 seek time is nice. :-) Jan 27 07:46:57 yea Jan 27 07:47:01 what's "cheap"? Jan 27 07:47:10 90 euros for the 60GB Jan 27 07:47:23 120 USD Jan 27 07:47:44 but price is bigger now... weird :)) Jan 27 07:47:48 a 200GB Intel SSD is $300 here. Jan 27 07:47:49 200 packages 2 hours : 2400 packages 24 hours. :) Jan 27 07:49:17 maybe 2x100Mb is better than 1x200 :) Jan 27 07:50:37 now we have many sata ports Jan 27 07:51:25 http://www.guruht.com/2010/01/ocz-vertex-2-ssd-pro-sandforce-vs-intel.html Jan 27 07:51:31 ^^ Jan 27 07:53:22 Critical Feature Comparison of OCZ and Intel SSD's: Jan 27 07:53:24 OCZ Vertex 2 60 GB Jan 27 07:53:26 Max Read: up to 285 MB/s, Jan 27 07:53:28 Max Write: up to 275 MB/s, Jan 27 07:53:30 Sustained Write: up to 250 MB/s, 4k Random Write (Aligned): 50,000 IOPS, Seek Time: 0.1ms Jan 27 07:53:32 INTEL X25M 80 GB SSD Jan 27 07:53:34 Native Command Queuing, which enables up to 32 simultaneous operations. Jan 27 07:53:36 sustained sequential read speeds of up to 250 MBps, Jan 27 07:53:38 sustained sequential write speeds of up to 70 MBps, Jan 27 07:53:40 and a read latency of just 85 microseconds. Jan 27 07:55:24 a good article : http://www.storagereview.com/ocz_vertex_2_review_120gb Jan 27 07:57:08 1.5W max :) Jan 27 07:59:46 just got my 2x2GB memory. rebooting :) Jan 27 08:02:02 :) Jan 27 08:02:15 xiangfu : what is "Qi" ? Jan 27 08:02:36 i see that everywhere in "chinese" projects Jan 27 08:05:02 the bootloader? Jan 27 08:05:56 In traditional Chinese culture, qi (; pronounced in English; in Standard Mandarin; Korean: gi; Japanese: ki; Vietnamese: khí, ) is an active principle forming part of any living thing. Jan 27 08:08:48 gmorning Jan 27 08:09:48 Hi. Jan 27 08:09:53 yea Jan 27 08:09:58 Chi Jan 27 08:10:08 in french Jan 27 08:10:18 Mana in gamer :) Jan 27 08:10:37 Energy Jan 27 08:11:10 The Force in Starwars :) Jan 27 08:11:31 i really need to sleep... Jan 27 08:11:41 but i wait my new netgear :p Jan 27 08:12:02 neutronscott: interesting Jan 27 08:12:35 Epsylon3: do you see it in living products? Jan 27 08:13:32 hmm Jan 27 08:13:51 static discharges ? :) Jan 27 08:13:55 EMI ? Jan 27 08:15:04 dynamic ones Jan 27 08:17:11 oh "living products" = "produits de beauté" ? Jan 27 08:18:29 cosmetics ? Jan 27 08:19:05 hmm no Jan 27 08:19:27 yoghurt and cheese Jan 27 08:19:33 :P Jan 27 08:19:35 ohh Jan 27 08:19:59 no i dont ;p Jan 27 08:20:05 maybe in eggs Jan 27 08:20:06 :p Jan 27 08:20:27 and salad Jan 27 08:21:04 what are you going on about! Jan 27 08:21:14 about Tsi Jan 27 08:21:22 Chi? Jan 27 08:22:12 yea Qi at first :) Jan 27 08:22:24 i know the boot loader yea Jan 27 08:22:33 tres interesont Jan 27 08:23:19 well lets use Qi on the Apple TV 2 and port openwrt Jan 27 08:23:28 ^^ Jan 27 08:23:36 it'd make an interesting AP Jan 27 08:23:38 or Samsung ones Jan 27 08:23:49 Serial port is on VGA pins 4 and 11 Jan 27 08:23:53 :) Jan 27 08:24:00 really... Jan 27 08:24:02 mandarin apple Jan 27 08:24:13 samsung what? Jan 27 08:24:36 Samsung TVs (AllShare / Internet@TV) Jan 27 08:24:57 oh actual TV. that'd be silly Jan 27 08:25:11 http://tobias.schroepf.de/doku/doku.php?id=garage:samsung_tv_hacking Jan 27 08:25:16 apple tv is $99 and about the likes of a SheevaPlug Jan 27 08:25:52 1.5inch screen? Jan 27 08:26:19 !p i use a 32" TV for my main LCD screen Jan 27 08:26:29 1920x1080 :) Jan 27 08:26:37 apple tv? no screen. is set-top box version of ipod really Jan 27 08:26:52 a tv withous screen? cool Jan 27 08:26:56 a 23" on the right, and a 4:3 22" on the left Jan 27 08:27:19 you're rich. Jan 27 08:27:35 a 23" cost 250 USD Jan 27 08:28:02 the tv... hmm 375 EUR Jan 27 08:28:09 oh, i forget, all my money goes to a wife Jan 27 08:28:13 :D Jan 27 08:28:23 haha, me no... i ve made the choice Jan 27 08:28:37 i live like in "Friends" Jan 27 08:28:41 else i would have more gadgets Jan 27 08:28:41 we are 5 here Jan 27 08:29:02 yes. share rent. Jan 27 08:29:06 yea Jan 27 08:29:13 dunno the english word :) Jan 27 08:29:22 share rent. Jan 27 08:29:23 roommate? Jan 27 08:29:31 you can buy cheap gadgets from chinese e-shops Jan 27 08:29:53 yea :p i do that Jan 27 08:29:58 i ve some tablets Jan 27 08:29:59 :p Jan 27 08:30:06 SmartQ Jan 27 08:30:07 i buy toys from dealextreme Jan 27 08:30:20 I've bought an HTC phone clone Jan 27 08:30:22 a SmartQ5 and a SmartQV3 Jan 27 08:30:28 yeah deal extreme Jan 27 08:30:39 just got a Motorola Defy Jan 27 08:30:43 very cheap Jan 27 08:30:47 I see that they have brought android phones now Jan 27 08:30:51 and very not chinese Jan 27 08:30:52 :p Jan 27 08:30:53 1hr to work. hope laundry is dry in time. :o Jan 27 08:31:03 solid like a rock Jan 27 08:31:21 dealextreme doesnt have much phone for USA band Jan 27 08:32:05 that's a pity Jan 27 08:32:05 also waterproof :) Jan 27 08:32:08 why i got it Jan 27 08:32:11 and no one has phone for T-Mobile USA's perverse 3G band Jan 27 08:32:24 the first water proof phone Jan 27 08:32:25 :) Jan 27 08:32:28 USA always uses it's own standards, different from the rest of the world Jan 27 08:33:02 Defy is also Quad band no ? Jan 27 08:33:26 yeah well. when you invent it, and everyone else makes revision 1 after seeing room for improvement ;) Jan 27 08:33:30 I don't know Jan 27 08:33:34 jk Jan 27 08:33:45 http://www.talkandroid.com/14220-t-mobile-usa-getting-the-rugged-motorola-defy/ Jan 27 08:34:09 erf, the TMobile one is white Jan 27 08:34:22 CC arch/mips/adm5120/zyxel/p-334wt.o Jan 27 08:34:23 arch/mips/adm5120/zyxel/p-334wt.c:34:19: error: macro "MIPS_MACHINE" requires 4 arguments, but only 3 given Jan 27 08:34:53 hmm Jan 27 08:34:58 i want device that is like blackberry formfactor. Motorola Charm made low-end device. Screen resolution was deal breaker tho :p Jan 27 08:35:34 http://tomoyo.sourceforge.jp/cgi-bin/lxr/source/arch/mips/include/asm/mips_machine.h?a=mips#L25 Jan 27 08:36:03 mach_type mach_id mach_name *mach_setup() Jan 27 08:36:35 jQuery UI 1.8.9 Jan 27 08:36:38 rmm Jan 27 08:37:48 Algerian Arabic Australian and New Zealand Datepicker localizations Jan 27 08:37:56 mmm nothing new Jan 27 08:38:49 russell--: there is a missing comma between the arguments Jan 27 08:39:14 i'ts "P-334WT" "ZyXEL Prestige 334WT" Jan 27 08:39:25 and should be "P-334WT", "ZyXEL Prestige 334WT" Jan 27 08:41:20 Epsylon3: http://patchwork.cyranjo.org/patch/562/ Jan 27 08:41:26 err Jan 27 08:41:35 russell--: ^ Jan 27 08:42:27 what is the site :) Jan 27 08:42:34 another place to post patches ? Jan 27 08:43:03 -devel, trac, and this one... Jan 27 08:43:48 lol, i just found that patch too, thanks! Jan 27 08:44:32 Incoming patches Jan 27 08:44:34 from here ? Jan 27 08:44:40 from where ? Jan 27 08:44:48 erg.. hand tired Jan 27 08:45:15 seems be from mailing list Jan 27 08:46:01 maybe not Jan 27 08:46:20 no... i dont think so, i ve sent some last days Jan 27 08:47:26 Patches that have been sent to a mailing list are 'caught' by the system, and appear on a web page. Any comments posted that reference the patch are appended to the patch page too. Jan 27 08:47:41 acinonyx * r25134 /trunk/package/iw/patches/401-antenna.patch: [package] iw: Extend antenna setting 'all' to apply for either rx or tx antenna Jan 27 08:47:50 gah, where do kernel patches go? Jan 27 08:47:55 guess it didnt catch mine either Jan 27 08:49:04 ah, found it Jan 27 08:49:25 target/linux/... Jan 27 08:50:11 Openwrt-devel Jan 27 08:50:31 maybe my patches were not in good format Jan 27 08:50:58 "mime" instead of text Jan 27 08:51:22 no philipp64|laptop 's are in there Jan 27 08:51:28 they were attachments Jan 27 08:51:48 yea, right Jan 27 08:51:56 i've received them Jan 27 08:51:57 might be your filter. i see mine after i removed the filter "action required" Jan 27 08:52:03 i guess florian accepted them. Jan 27 08:53:02 or maybe its only the florian's branch Jan 27 08:53:26 its Kaloz for orion patches Jan 27 08:54:39 Acinonyx, KanjiMonster: thanks! Jan 27 08:55:01 nevermind... Kaloz has commited (locally) some of my patches Jan 27 08:56:24 acinonyx * r25135 /trunk/package/iw/patches/ (4 files): [package] iw: Refresh patches Jan 27 09:14:03 acinonyx * r25136 /trunk/package/mac80211/files/lib/wifi/mac80211.sh: [package] mac80211: Fix antenna setting when only one of {rx,tx}antenna options is specified Jan 27 09:16:52 build #71 of cobalt is complete: Failure [failed compile_6] Build details are at http://tksite.gotdns.org:8010/builders/cobalt/builds/71 Jan 27 09:22:48 build #61 of x86 is complete: Failure [failed compile_1] Build details are at http://tksite.gotdns.org:8010/builders/x86/builds/61 Jan 27 09:28:07 philipp64|laptop: yo Jan 27 09:54:31 http://tksite.gotdns.org:8010/builders/orion/builds/69/steps/compile_1/logs/stdio Jan 27 09:54:37 Checking 'zip'... failed. Jan 27 09:54:37 : Please install zip. Jan 27 09:54:42 weird :p Jan 27 10:13:53 <_trine> whats the correct format for an openwrt email address is it name@openwrt.net Jan 27 10:14:20 acinonyx * r25137 /trunk/package/dnsmasq/files/dnsmasq.init: [package] dnsmasq: add support for option "--dhcp-script" (thanks Bastian Bittorf) Jan 27 10:16:06 _trine: ? Jan 27 10:16:14 do oyu go to openwrt.org or openwrt.net ? Jan 27 10:17:59 <_trine> blogic, I think I can do both Jan 27 10:19:39 <_trine> I sent it to net so I shall see if it gets sent or returned to me Jan 27 10:19:44 _trine: the answer is @openwrt.org Jan 27 10:19:53 <_trine> damn Jan 27 10:19:56 <_trine> thanks Jan 27 10:20:33 <_trine> its sent to both now Jan 27 10:22:36 _trine: think ? :) is the browser so hard to start and verify ? :) Jan 27 10:36:36 florian accepted mine on the patchwork page but didn't commit them. :o Jan 27 10:38:22 best part of being in antarctica. the massive amounts of chocolate from 2003 are given away gladly for free. :D Jan 27 10:38:36 <_trine> blogic, which is easier for you,, to remonstrate with me which took many more characters than the answer would have taken or just to have said its 'org' ? Jan 27 10:38:50 <_trine> I dunno Jan 27 10:40:32 to explain to you that the internet is for looking up things and not for having them looked up Jan 27 10:40:50 give a man a fish and he has food Jan 27 10:40:57 show him how to catch a fish .... Jan 27 10:45:01 <_trine> blogic, my last work on this is bollocks Jan 27 10:48:10 hmm weird system : http://pastebin.com/H1D8vgHd Jan 27 10:50:08 UserServiceNonVolSettings::GetSingletonInstance: WARNING - the singleton instance is NULL, and someone is accessing it! Jan 27 11:26:35 build #60 of uml is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/uml/builds/60 Jan 27 11:47:40 who the fuck distributes a program with -Werror Jan 27 11:47:58 liboping does not build because of 2 unused variables :-/ Jan 27 11:49:11 the oping author does ^^ Jan 27 11:49:24 he is nuts Jan 27 11:49:41 it's good for development, but surely not distribution Jan 27 11:54:42 is there some special reason why openwrt firewall seems to get sip connections so wrong that they won't work at all ? my old endian firewall did this job much better Jan 27 11:57:32 it doesn't "get" anything, its just a bunch of input and forward accept/reject rules Jan 27 11:57:51 plus an accept of established,related Jan 27 11:58:36 maybe you're missing the sip conntrack helpers Jan 27 11:58:55 mmhh, maybe Jan 27 11:59:23 havn't seen them at least Jan 27 11:59:46 they should be part of iptables-mod-conntrack-extra (or mod-nat-extra, not sure) Jan 27 12:00:37 i will look Jan 27 12:00:57 xMff: where have you been on the congress ? Jan 27 12:01:17 I was on vacation, didn't get tickets anyway Jan 27 12:01:27 :-( Jan 27 12:01:36 i hope you had a nice vacation Jan 27 12:01:39 yes Jan 27 12:01:53 you seen my current project ? Jan 27 12:02:25 iptables-mod-nat-extra btw. Jan 27 12:02:31 no I didn't see your currewnt project Jan 27 12:02:45 https://github.com/poelzi/ulatencyd/ Jan 27 12:03:42 ah nice Jan 27 12:03:58 it even protects you from swap of death Jan 27 12:04:21 poelzi: interesting Jan 27 12:05:16 i can make a make -j 60 on my 2 core machine on the linux kernel, start the memleak program and still look a video :-) Jan 27 12:07:16 interesting Jan 27 12:07:32 sorry, i am repeating myself, but that is interesting :D Jan 27 12:07:37 and it uses lua i see Jan 27 12:08:02 i will look at that in detail later Jan 27 12:08:05 yes. most of the rules are written in lua. as heuristics are best scripted. the core is c for performance Jan 27 12:08:13 * blogic makes a note on his todo list Jan 27 12:08:16 yes Jan 27 12:08:27 in fact, parsing the proc filesystem is the slowest part of all :-) Jan 27 12:08:28 i agree that at first sight it makes all the sense Jan 27 12:08:29 and thats c Jan 27 12:08:37 hehe Jan 27 12:08:49 https://github.com/poelzi/ulatencyd/blob/master/rules/scheduler_desktop.lua Jan 27 12:08:50 do you calculate mem usage of processes aswell ? Jan 27 12:09:19 i only care about mem usage when memory is low Jan 27 12:09:50 i recently wanted to know the exact amount of alloctaed memory and it is really not that easy Jan 27 12:10:02 but ok, different subject Jan 27 12:10:05 it is rss in proc Jan 27 12:10:08 yes Jan 27 12:10:22 took me a moment to find Jan 27 12:10:51 it is much harder to find other stuff like what pages are shared between which processes etc Jan 27 12:14:20 yes Jan 27 12:15:54 bbl Jan 27 12:16:35 poelzi: why do you call mount via a system call ? Jan 27 12:16:39 i damaged my notebook display some days ago and bought a new display. thought: hey, the same model with other submodel number has a higher resolution, so why not buy a better one. now i found out that i need another flex cable for that :-( Jan 27 12:16:43 sorry i am reading the code :) Jan 27 12:17:03 i call "mount ...." now Jan 27 12:17:08 from lua Jan 27 12:17:13 ok Jan 27 12:17:31 thats the old code. same as the broken libcgroups stuff Jan 27 12:17:40 it's not used anymore Jan 27 12:18:24 kk Jan 27 12:19:07 libcgroups is large and totally buggy. instead of 120kb lib i use now 300 lines lua which do the job much better ;-) Jan 27 12:19:29 yes, been there done that Jan 27 12:19:41 most libs are bloattastic Jan 27 12:21:21 kaloz * r25138 /trunk/include/kernel-defaults.mk: fixup board specific KERNELNAME overrides Jan 27 12:21:24 it is more flexible with mount points. so you can mount theme where you want, etc. but you will run into problems then as not all cgroup subsystems support nasted hierarchies Jan 27 12:24:03 bored. wonder if i can drive a vga monitor with gpio... Jan 27 12:26:23 if you got only like 100 pixels ;-) Jan 27 12:27:04 hi guys , is there a way to override LINUX_VERSION in target/profile (to have different LINUX_VERSION in two profiles under same subtarget) ? Jan 27 12:27:42 build #65 of ramips is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/ramips/builds/65 Jan 27 12:29:44 poelzi: too slow? of course it'd use all my cpu time... heh Jan 27 12:29:57 * neutronscott looking at vga timing now Jan 27 12:32:13 the monitor will sync-on-green. only need 1 signal pin 1 ground, 1 resistor..! Jan 27 12:32:56 blogic: wanna apply https://gist.github.com/798448 ? Jan 27 12:33:28 fixes the oping does not build because of 'i find warnings should be errors stupididty' Jan 27 12:36:15 i detest those. Jan 27 12:36:43 i use -werror :) Jan 27 12:37:04 for development yes, for deployment no Jan 27 12:37:05 but yeah i'll look into it in a sec Jan 27 12:39:17 build #65 of rb532 is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/rb532/builds/65 Jan 27 12:45:49 build #75 of atheros is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/atheros/builds/75 Jan 27 12:51:39 kaloz * r25139 /trunk/target/linux/leon/ (32 files in 3 dirs): [target/leon]: add preliminary LEON support Jan 27 12:52:46 kaloz * r25140 /trunk/target/linux/leon/patches/002-sparc_uimage_target.patch: [target/leon]: fix LD name Jan 27 12:54:21 kaloz * r25141 /trunk/toolchain/uClibc/ (3 files in 3 dirs): [toolchain/uClibc]: LEON is SPARC v8 Jan 27 13:04:01 build #66 of ppc44x is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/ppc44x/builds/66 Jan 27 13:18:02 kaloz * r25142 /trunk/toolchain/gcc/ (Config.in Config.version): [toolchain]: use the linaro gcc by default (right now this only affects x86 targets) Jan 27 13:19:16 mb * r25143 /trunk/target/linux/omap24xx/patches-2.6.37/ (5 files): omap24xx: Refresh 2.6.37 patches Jan 27 13:22:51 voyo: no Jan 27 13:23:55 jow_laptop: hey, I know answer from you already ;) but its always worth a try ... Jan 27 13:24:50 kaloz * r25144 /trunk/ (3 files in 2 dirs): [toolchain]: get rid of gcc 4.1.2 and 4.2.4 support Jan 27 13:25:17 voyo: why would you want to do that ? Jan 27 13:27:33 blogic: I defined subtarget like x86/voyo'smachines , where I want to put definitions for things im using - ie. Xen, vmware, some geos , some atom boards etc. for xen and vmware I can't for some reason use same kernel like for other... Jan 27 13:28:43 kaloz * r25145 /trunk/toolchain/gcc/patches/ (4.1.2/ 4.2.4/): [toolchain]: nuke unneeded patchsets as well Jan 27 13:29:23 its just more flexible, to redefine more things in deeper sub-tree Jan 27 13:29:54 build #62 of ar7 is complete: Exception [exception failed slave lost shell_13 compile_12] Build details are at http://tksite.gotdns.org:8010/builders/ar7/builds/62 Jan 27 13:32:07 kaloz * r25146 /trunk/package/busybox/Makefile: [package/busybox]: we don't support gcc 4.5.0 Jan 27 13:34:05 kaloz * r25147 /trunk/target/linux/generic/ (8 files in 8 dirs): [target/generic/mips]: we don't support gcc 4.1.x Jan 27 13:37:03 jow_laptop: are you still willing (and got little time ) to help me with my single xen target at least ? Jan 27 13:38:16 voyo: later today I can help you. Right now I'm a bit busy with other stuff Jan 27 13:38:42 jow_laptop: ok, appreciate Jan 27 13:55:12 kaloz * r25148 /trunk/ (5 files in 3 dirs): cleanup toolchain version handling Jan 27 13:57:00 <_trine> the lack of connectivity has been resolved in KAMIKAZE (bleeding edge, r25136) Jan 27 13:57:13 <_trine> on the LAN Jan 27 14:12:14 * mb__ is fighting the evil force: libtool Jan 27 14:12:56 the dark side is strong in this one Jan 27 14:13:11 boah i felt a disturbacne Jan 27 14:13:32 devlish words in your mouth Jan 27 14:13:36 :) Jan 27 14:25:15 http://freshmeat.net/articles/stop-the-autoconf-insanity-why-we-need-a-new-build-system Jan 27 14:25:47 build #79 of brcm63xx is complete: Success [build successful] Build details are at http://tksite.gotdns.org:8010/builders/brcm63xx/builds/79 Jan 27 14:26:52 mb * r25149 /packages/net/gpsd/ (Makefile patches/100-cross-compile.patch): gpsd: Fix crosscompile and libtool issues Jan 27 14:36:42 mb * r25150 /packages/net/gpsd/patches/100-cross-compile.patch: gpsd: Add missing define Jan 27 14:44:37 jow_laptop: thats 8 years old news, and still autoconf is used widely ;) Jan 27 14:54:02 I am aware of that Jan 27 14:55:20 i take autoconf based packages over custom hacked up build systems any time Jan 27 14:55:47 at least autoconf based stuff has a good chance of cross compiling properly with generic code to handle it Jan 27 14:55:56 but I have to agree , at least partialy. currently Im struggling with compiling open-vmware-tools with owrt, makefile & configure & crosscompilation nightmare... Jan 27 14:56:19 i have yet to see a viable alternative to the autofools Jan 27 14:57:28 nbd: yeah, good chance. but not 100% of chance ;) Jan 27 14:57:54 correct, but most of the nastiness can be handled with generic code that covers almost all packages Jan 27 14:58:38 and that's better than having to dig through and fix hundreds of custom build systems Jan 27 14:59:22 package authors typically don't give any thought to cross compiling at all Jan 27 14:59:36 and when they use autoconf then most of the time they don't have to because it'll just work Jan 27 14:59:49 unless they're doing something stupid such as using AC_TEST_RUN or whatever that's called Jan 27 15:00:05 where they need to run the binaries that they compiled (just for testing) Jan 27 15:00:32 nbd: try to compile mono for arm on owrt ;) Jan 27 15:00:44 (on x86 based host) Jan 27 15:00:45 or the embedded libtool is broken, or it uses macros that don't exist anywhere, or it uses feature test macros that are only satisfied if theres "GNU" in the version string, or the macros are outdated, or the autoconf api changed recently Jan 27 15:01:32 often it helps, but sometimes fixing the autofoo is more work than compiling manually Jan 27 15:01:46 yup... Jan 27 15:02:49 actually autoconf can be fixed by a human while libtool cannot :) Jan 27 15:03:08 and they're so tightly interleaved, its not fun Jan 27 15:04:58 libtool definitely is the worst of the bunch Jan 27 15:04:59 they should kill libtool and start a decent m4/autoconf macro libary, that'd be a start Jan 27 15:05:34 right now, finding appropriate macros is like searching toolz for windows Jan 27 15:35:55 iperf package is broken. It fails on some pthread symbols. Jan 27 15:36:05 details? Jan 27 15:36:22 undefined reference to pthread_create and pthread_detach Jan 27 15:36:28 It uses g++-uc Jan 27 15:36:47 are you compiling plain iperf or iperf-mt? Jan 27 15:36:48 Oh, it has "-pthread" on the commandline. Typo? Jan 27 15:37:07 iperf-single Jan 27 15:37:10 ok Jan 27 15:37:12 oh yeah, I have same issue with iperf Jan 27 15:37:16 -pthread normally isn't a typi Jan 27 15:37:22 both iperf and iperf-mt are broken Jan 27 15:37:46 ah, typo in the package Jan 27 15:38:13 testing the fix now Jan 27 15:38:50 actually maybe not same.. Jan 27 15:39:04 nbd * r25151 /packages/net/iperf/Makefile: iperf: fix a typo that broke the single-threaded variant Jan 27 15:39:05 there Jan 27 15:39:18 here is mine http://paste.openwrt.org/d5dcd7565 Jan 27 15:47:30 nbd: Thanks, that fixes it Jan 27 16:00:17 jow * r25152 /trunk/package/mac80211/files/lib/wifi/mac80211.sh: [package] mac80211: fix order of iw parameters, fix broken antenna config call Jan 27 16:24:38 mb * r25153 /packages/Xorg/lib/pango/Makefile: pango is not parallel build safe w.r.t. its internal libs Jan 27 17:18:44 mb * r25154 /packages/Xorg/lib/vte/patches/ (. 100-compile-fixes.patch): vte: Compile fix Jan 27 17:31:05 mb * r25155 /packages/Xorg/xorg/lib/ (libSM/Makefile libXpm/Makefile): libSM, libXpm: dependency fixes Jan 27 18:05:48 hm, xorg-server (or dri?) fails in a strange way: http://pastebin.com/yv8v3sWK Jan 27 18:06:39 missing macros or .pc files? Jan 27 18:12:27 is someone using imq here ? Jan 27 18:25:32 ah it's a bug in the mesa patch. Jan 27 18:32:49 blogic: how's that testing coming? :-) Jan 27 18:38:30 mb * r25156 /packages/Xorg/lib/mesa/patches/010-fix-pkgconfig-files.patch: mesa: Fix version information in .pc files Jan 27 18:42:39 mb * r25157 /packages/Xorg/xorg/xserver/xorg-server/Makefile: xorg-server: Fix linkage Jan 27 18:52:38 * Epsylon3 is now auto-away after 2h idle Jan 27 19:04:54 mb * r25158 /feeds/lxde/apps/lxtask/Makefile: lxtask: Force include of libintl.h to avoid implicit linkage to intl functions. Jan 27 19:05:14 neutronscott: did I hear my name? Jan 27 19:28:35 build #70 of orion is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/orion/builds/70 Jan 27 19:52:52 mb * r25159 /trunk/tools/missing-macros/src/ (README m4/xaw.m4): Add missing XAW macros Jan 27 19:58:21 sooo there we go. Tree compiles :) Jan 27 20:20:42 mb * r25160 /trunk/target/linux/omap24xx/Makefile: omap24xx: Update to .37 Jan 27 20:30:42 mb * r25161 /trunk/target/linux/omap24xx/base-files/etc/inittab: omap24xx: Fix serial console for .37 Jan 27 20:55:09 jow * r25162 /trunk/target/linux/at91/patches/ (600-usb_vbus_active_high.patch 700-tqma9263-support.patch): [at91] prepare TQ-Components TQMa9263 kernel support Jan 27 20:55:40 jow * r25163 /trunk/target/linux/at91/files/arch/arm/mach-at91/board-tqma9263.c: [at91] add TQMa9263 board setup Jan 27 20:56:07 jow * r25164 /trunk/target/linux/at91/tqma9263/ (. config-default target.mk): [at91] prepare TQMa9263 subtarget Jan 27 20:57:15 jow * r25165 /trunk/target/linux/at91/ (5 files in 2 dirs): [at91] move Netus G20 into a subtarget and rework toplevel config and makefile Jan 27 21:00:21 mb * r25166 /trunk/target/linux/omap24xx/config-2.6.37: omap24xx: Support ext4 as rootfs Jan 27 21:34:03 g'evening Jan 27 21:42:55 build #61 of avr32 is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/avr32/builds/61 Jan 27 21:49:41 florian * r25167 /trunk/package/acx-mac80211/Makefile: Jan 27 21:49:41 [package] acx-mac80211: support loading different firmwares Jan 27 21:49:41 Signed-off-by: Alexey Torkhov Jan 27 21:49:44 florian * r25168 /trunk/package/acx/files/lib/wifi/acx.sh: Jan 27 21:49:44 [package] acx: don't fail if 'wifi' led is missing Jan 27 21:49:44 Signed-off-by: Alexey Torkhov Jan 27 21:49:54 florian * r25170 /trunk/target/linux/adm5120/router_be/profiles/200-ZyXEL.mk: Jan 27 21:49:54 [adm5120] zyxel: use acx-mac80211 driver by default Jan 27 21:49:54 Signed-off-by: Alexey Torkhov Jan 27 21:49:56 florian * r25169 /trunk/target/linux/adm5120/files/arch/mips/adm5120/zyxel/p-334wt.c: Jan 27 21:49:56 [adm5120] fix p-334wt build Jan 27 21:49:56 Signed-off-by: Alexey Torkhov Jan 27 21:49:59 florian * r25171 /trunk/target/linux/adm5120/base-files/etc/config/ (. network system): Jan 27 21:49:59 [adm5120] add default network/leds config (#6759) Jan 27 21:49:59 Signed-off-by: Alexey Torkhov Jan 27 21:50:01 florian * r25172 /trunk/package/base-files/files/etc/init.d/led: Jan 27 21:50:01 [package] base-files: support port_state in led config (#6763) Jan 27 21:50:01 Signed-off-by: Alexey Torkhov Jan 27 21:50:05 florian * r25173 /trunk/package/admswconfig/files/admswswitch.sh: Jan 27 21:50:05 [package] admswconfig: reset interface after applying config Jan 27 21:50:05 Signed-off-by: Alexey Torkhov Jan 27 21:50:08 wooooo Jan 27 21:50:08 florian * r25174 /trunk/package/kernel/modules/usb.mk: Jan 27 21:50:08 [package] enable MOS7720 usb-serial devices Jan 27 21:50:08 Enable build of kernel module for MOS7720 usb-serial devices. Jan 27 21:50:08 Tested on WGT634U and WL500g running trunk @r24238. Jan 27 21:50:08 Signed-off-by: Marc Zyngier Jan 27 21:50:14 florian * r25175 /trunk/target/linux/adm8668/ (7 files in 3 dirs): Jan 27 21:50:14 [adm8668] clean up mach-adm8668 includes Jan 27 21:50:14 This patch cleans up the include directory, as they were from vendors Jan 27 21:50:14 2.4 GPL source. Now only what's used is there. Jan 27 21:50:15 Signed-off-by: Scott Nicholas Jan 27 21:50:18 florian * r25176 /trunk/target/linux/adm8668/ (3 files in 2 dirs): (log message trimmed) Jan 27 21:50:18 [adm8668] fix firstboot with new MTD map driver Jan 27 21:50:18 my-mkimage previously did not include the fs_mark of deadc0de because Jan 27 21:50:18 upon formatting of jffs2 partition, U-Boot's CRC validation would not Jan 27 21:50:18 allow the image to boot. This new MTD map will shrink the Jan 27 21:50:18 kernel+rootfs+fs_mark image and recalculate the CRC, so that only the Jan 27 21:50:18 kernel is part of the image validation. This also improves boot time, Jan 27 21:50:37 theres my patch :D Jan 27 21:50:38 someones on a commit run i see... Jan 27 21:50:45 florian's alive! Jan 27 21:50:52 indeed. Jan 27 21:51:06 <[florian]> ah ah Jan 27 21:51:08 <[florian]> yes sir Jan 27 21:51:32 <[florian]> neutronscott: had to slightly change your cleanup patch, tell me if that runs ok Jan 27 21:51:44 <[florian]> neutronscott: basically removing the BIT_* and IRQ_MASK stuff Jan 27 21:51:46 okay, how so? Jan 27 21:51:49 <[florian]> which need to be added back Jan 27 21:52:15 oh.. ? Jan 27 21:52:36 guess i had another patch to send.. i got rid of BIT_ Jan 27 21:52:37 <[florian]> no problem, you deleted just a little bit more Jan 27 21:52:49 <[florian]> or maybe that was patchwork which did not like your patches, I guess Jan 27 21:53:54 i changed my irq.c to be adm5120's. they're same, so rather use known code than what i came up with reading the MIPS porting guides last year Jan 27 21:54:05 build #57 of etrax is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/etrax/builds/57 Jan 27 21:54:41 <[florian]> neutronscott: when we get both merged mainline, it will make sense to create a common driver for these Jan 27 21:55:17 adm8668 could be together with 5120 i suppose. Jan 27 21:55:29 <[florian]> I do not know how much they differ Jan 27 21:55:35 <[florian]> I would need to have a closer look Jan 27 21:55:42 <[florian]> for sure the ethernet core is different Jan 27 21:56:01 i didn't look at much besides this irq.c Jan 27 21:56:10 same. jtag = same too. Jan 27 21:58:20 tonight i'll look. i need to put the wifi away and do the switch and gpio. Jan 27 21:58:37 <[florian]> did you fix your wifi performance issues with 2.6 kernel? Jan 27 21:58:39 i saw a 6996m patch on mailing list. should help :D Jan 27 21:59:06 Helmut's patches in rt2x00.git should help as much as possible. Jan 27 22:00:32 not sure if a patch against compat-wireless would be helpful. it's bumped so often... Jan 27 22:00:35 mb * r25177 /packages/Xorg/driver/xf86-input-tslib/ (10 files in 2 dirs): xf86-input-tslib: Update to 0.0.6. This fixes touchscreen on linux-2.6.37. Jan 27 22:12:46 <[florian]> neutronscott: I am afk now, see you later Jan 27 22:13:12 ok thanks. Jan 27 22:14:06 i'm ready for bed. waiting for post office to open to send postcards. Jan 27 22:14:29 anyone want a postcard from antarctica? i've 3 extras :D Jan 27 22:15:57 mb * r25178 /trunk/target/linux/omap24xx/base-files/etc/config/fstab: omap24xx: Don't automatically mount maemo fs Jan 27 22:20:08 jow * r25179 /trunk/package/firewall/ (Makefile files/lib/core_rule.sh): [package] firewall: fix wrong rule order if multiple protocols are used Jan 27 22:20:48 jow * r25180 /branches/backfire/package/firewall/ (Makefile files/uci_firewall.sh): [backfire] backport r25179 Jan 27 22:29:16 jow * r25181 /packages/libs/mysql/Makefile: [packages] mysql: ensure linking against libuClibc++, also install system table data sql file Jan 27 22:38:15 neutronscott: you've been /msgd :) Jan 27 22:42:02 jow * r25182 /packages/libs/mysql/ (Makefile patches/502-fix-hostname.patch): [packages] mysql: fix /bin/hostname invocation in table install script, openwrt does not have this command Jan 27 22:43:32 jow * r25183 /packages/libs/mysql/patches/502-fix-hostname.patch: [packages] mysql: use proper proc file in hostname patch Jan 27 23:46:09 build #62 of sibyte is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/sibyte/builds/62 Jan 28 00:12:01 mb * r25184 /packages/Xorg/lib/imlib2/Makefile: imlib2: Bump to 1.4.4 Jan 28 00:17:27 build #59 of kirkwood is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/kirkwood/builds/59 Jan 28 00:34:49 mb * r25185 /feeds/desktop/apps/tint2/ (. Makefile patches/ patches/100-remove-xinerama.patch): Add tint2 panel Jan 28 00:35:33 build #55 of rdc is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/rdc/builds/55 Jan 28 00:55:39 build #57 of xburst is complete: Failure [failed compile_13] Build details are at http://tksite.gotdns.org:8010/builders/xburst/builds/57 Jan 28 01:10:56 nbd * r25186 /trunk/toolchain/uClibc/patches-0.9.32/180-pthread_cleanup_fix.patch: uClibc 0.9.32: fix infinite recursion on application exit triggered by linking libc before libpthread Jan 28 01:10:59 nbd * r25187 /trunk/toolchain/uClibc/ (Config.in Config.version): uClibc: make 0.9.32-git default, it fixes lots of bugs compared to earlier versions (especially threading issues), and the known regressions have been fixed Jan 28 01:53:46 build #54 of octeon is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/octeon/builds/54 Jan 28 02:23:40 build #80 of brcm63xx is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/brcm63xx/builds/80 Jan 28 02:29:16 nbd * r25188 /trunk/toolchain/binutils/patches/2.20.1/310-backport_fPIE_mips_fix.patch: binutils: backport a fix for broken relocation entries on mips with -fPIE, fixes dbus-server crash with sstrip Jan 28 02:34:51 ping nbd Jan 28 02:35:02 pong Jan 28 02:35:06 about to go to sleep Jan 28 02:36:02 nbd ucibc 0.9.32 doesnt compile Jan 28 02:36:21 it trys to download the tar.gz instead of the git co Jan 28 02:37:06 ah, somebody updated it again without uploading a tarball Jan 28 02:37:06 sec Jan 28 02:37:46 but shouldnt it use the git clone? Jan 28 02:37:48 ifeq ($(CONFIG_UCLIBC_VERSION_0_9_32),y) Jan 28 02:37:48 PKG_SOURCE_URL:=git://git.busybox.net/uClibc Jan 28 02:37:51 @nbd Jan 28 02:38:13 well, it's normal if it tries to download the tarball Jan 28 02:38:15 and i just checked Jan 28 02:38:17 the tarball is on mirror2 Jan 28 02:38:24 so why doesn't it compile for you? Jan 28 02:38:33 i doesnt find the source Jan 28 02:38:52 show me Jan 28 02:39:01 i try it now again Jan 28 02:39:35 http://nopaste.info/e194cde61c.html Jan 28 02:39:38 @mbd Jan 28 02:39:40 nbd Jan 28 02:39:58 ah Jan 28 02:40:46 bug in the makefile Jan 28 02:40:52 which messes up the tarball name Jan 28 02:41:07 didn't show up in my tests because i enabled advanced toolchain settings Jan 28 02:42:11 nbd * r25189 /trunk/toolchain/uClibc/Makefile: uClibc: fix version check for 0.9.32 Jan 28 02:42:44 that should fix it Jan 28 02:43:15 okay i will try it when its in the git tree - i only use git Jan 28 02:44:26 triggered a resync Jan 28 02:44:28 you can pull now Jan 28 02:45:08 okay Jan 28 02:46:08 nbd: it finds the file now and downloads it Jan 28 02:46:35 thanks Jan 28 02:49:49 nbd i'll go now have a nice night Jan 28 02:54:45 build #75 of s3c24xx is complete: Failure [failed compile_3] Build details are at http://tksite.gotdns.org:8010/builders/s3c24xx/builds/75 **** ENDING LOGGING AT Fri Jan 28 02:59:56 2011