| 00:19.19 | CIA-8 | 03jp30 * 10unslung/make/ (php-apache.mk php-thttpd.mk): build fixes to php-apache; stop php-thttpd downloading multiple copies of the php sources |
| 00:38.25 | CIA-8 | 03jp30 * 10unslung/make/php-apache.mk: put back PHP_APACHE_VERSION; PHP_VERSION cannot occur in targets in php-apache.mk |
| 00:47.14 | CIA-8 | 03jp30 * 10unslung/make/byrequest.mk: byrequest: consistent ipk version number |
| 00:57.04 | CIA-8 | 03jp30 * 10unslung/Makefile: promote getmail, procmail, mod-python, py-mx-base, py-mysql, py-psycopg |
| 01:22.53 | *** join/#nslu2-linux linoxx (~linoxx@adsl-065-012-178-027.sip.asm.bellsouth.net) |
| 01:23.30 | linoxx | whenever i boot up my nslu2, the ready/status light is orange and it doesn't have any ip address |
| 01:23.48 | linoxx | i had flashed it with the unslug firmware and had rebooted it multiple times and it worked fine then |
| 01:27.55 | *** join/#nslu2-linux dyoung-web (www-data@netblock-66-159-209-60.dslextreme.com) |
| 01:28.04 | dyoung-web | morning. |
| 01:28.16 | linoxx | lo. |
| 01:28.57 | dyoung-web | rwhitby-away, My mule doesnt have space. I'll send it conventional Intl Priority. |
| 01:31.07 | dyoung-web | [g2] |
| 01:33.10 | linoxx | dyoung-web, would you happen to know what it means when the orange status light is continually glowing on the slug |
| 01:33.18 | linoxx | and not to mention, when it doesn't get an ip |
| 01:34.37 | [g2] | dyoung-web, hello |
| 01:46.05 | dyoung-web | hi |
| 01:46.14 | dyoung-web | orange light constant on is redboot. |
| 01:46.46 | linoxx | dyoung-web, so redboot has a problem? :X |
| 01:47.07 | dyoung-web | what did you do before the problem popped up |
| 01:47.14 | dyoung-web | its usually means a bad flash |
| 01:47.21 | linoxx | well i had rebooted a bunch of times |
| 01:47.37 | linoxx | i had taken it over to a friend's house |
| 01:47.40 | linoxx | after it had been working |
| 01:47.47 | linoxx | booted it up - i get orange light |
| 01:48.47 | dyoung-web | fw version? |
| 01:49.23 | linoxx | it was unslug |
| 01:50.03 | linoxx | i think 3.17 beta |
| 01:50.56 | linoxx | checked the wiki, but didn't find anyting about the orange glow |
| 01:51.00 | linoxx | that's why i came |
| 01:56.03 | jacques | I don't suppose you have a serial port on it? |
| 01:56.17 | [g2] | or JTAG hooked up :) |
| 01:56.30 | linoxx | jacques, no |
| 01:56.37 | linoxx | [g2], ohh boy this doesn't sound good |
| 01:56.53 | [g2] | linoxx, I'm just joking |
| 01:56.58 | *** join/#nslu2-linux Ksilebo (russ@d149-67-147-112.try.wideopenwest.com) |
| 01:57.03 | [g2] | have you tried the redboot telnet in ? |
| 01:57.18 | jacques | d`oh I always forget about the redboot telnet |
| 01:57.27 | jacques | that's definitely the first thing to try |
| 01:57.27 | linoxx | [g2], i ran this: nmap -sP -T5 192.168.0.* to see if the ip was on the network |
| 01:57.29 | linoxx | i didn't see it |
| 01:57.42 | jacques | 192.168.1.77 isn't it ? |
| 01:57.54 | dyoung-web | linoxx: you had the misfortune of finding 3 out of 5 guys that have actually used JTAG on the thing. |
| 01:58.10 | linoxx | :XXXXXXXX |
| 01:58.12 | dyoung-web | jacques: redboot GDB Debug address is 192.168.0.1 |
| 01:58.28 | jacques | ah ok sorry for spreading misinformation |
| 01:58.33 | [g2] | and setup the MAC addr for the box so no need to ARP |
| 01:58.34 | linoxx | so i have to use route-add or whatever it was to get to the thing? |
| 01:58.57 | dyoung-web | easier IMHO to setup a alias for the interface. |
| 01:59.00 | jacques | dyoung-web, lol I just read "linoxx: you had the misfortune of finding 3 out of 5 guys that have actually used JTAG on the thing." |
| 01:59.29 | [g2] | I think any ethernet cable will do |
| 01:59.44 | dyoung-web | :-) |
| 01:59.44 | [g2] | the slug is auto sensing right ? |
| 02:00.02 | [g2] | or all my hubs are :) |
| 02:00.15 | linoxx | dyoung-web, how do you setup the alias to it? |
| 02:00.34 | dyoung-web | ifconfig eth0:1 192.168.0.10 |
| 02:00.36 | dyoung-web | for instance. |
| 02:00.41 | dyoung-web | (assuming you re running linux) |
| 02:01.04 | dyoung-web | under windows; just press the "add" button for IP address in the TCP/IP tab in the network setup for the interface |
| 02:01.08 | Tiersten | [g2]: Yeah. It seems to be auto MDI/MDIX |
| 02:01.18 | linoxx | dyoung-web, ah |
| 02:01.27 | dyoung-web | hehe and now another JTAG guy shows up. ;-) |
| 02:01.35 | linoxx | dyoung-web, is it safe to do under a wireless connection? |
| 02:01.59 | dyoung-web | I would use a wired connection myself.... but I suppose it ought to work.... |
| 02:03.27 | linoxx | dyoung-web, i hate to ask too many questions, but is 192.168.0.10 in your example the ip of the slug? |
| 02:03.51 | Tiersten | No guarantees about what happens if the wireless connection craps out half way through though |
| 02:03.53 | dyoung-web | the slug is SET at 192.168.0.1; the example above is to make your linux box appear on the same network. |
| 02:04.00 | dyoung-web | (as .10) |
| 02:04.25 | dyoung-web | (but thats a shot in the dark without knowing anything about your network setup) |
| 02:06.38 | linoxx | one sec, i have to disconnect |
| 02:06.40 | *** part/#nslu2-linux linoxx (~linoxx@adsl-065-012-178-027.sip.asm.bellsouth.net) |
| 02:09.42 | *** join/#nslu2-linux linoxx_ (~linoxx@adsl-065-012-178-027.sip.asm.bellsouth.net) |
| 02:09.47 | linoxx_ | allright back |
| 02:12.35 | linoxx_ | i set up the alias, but wonder how good it will do due to the fact my router is 192.168.0.1 |
| 02:13.30 | jacques | :-\ |
| 02:13.37 | linoxx_ | luckily i have 2 routers |
| 02:14.16 | linoxx_ | so, guess i need to move this over.... |
| 02:14.17 | linoxx_ | heh |
| 02:14.23 | linoxx_ | be right back |
| 02:15.36 | dyoung-web | that may be a problem eh. |
| 02:15.40 | jacques | yeah |
| 02:16.47 | *** join/#nslu2-linux linoxx_ (~linoxx@adsl-065-012-178-027.sip.asm.bellsouth.net) |
| 02:16.59 | linoxx_ | ookay |
| 02:17.29 | linoxx_ | woot now no more seting aliases up |
| 02:18.31 | linoxx_ | now its just flashing green |
| 02:20.21 | dyoung-web | have you tried reflashing it yet? |
| 02:21.19 | jacques | flashing green is better than solid orange |
| 02:21.47 | dyoung-web | oh yea, sorry I didnt read that properly. |
| 02:21.50 | linoxx_ | i'm about to try that right now |
| 02:22.05 | dyoung-web | flashing green means its probably booting. |
| 02:22.27 | dyoung-web | are there disks attached right now? |
| 02:22.37 | linoxx_ | its been doing that for a while now, and no there aren't any |
| 02:24.10 | dyoung-web | I see. I suppose you already found http://www.nslu2-linux.org/wiki/HowTo/RecoverFromABadFlash ? |
| 02:29.55 | dyoung-web | so guys., on that page someone added something on the bottom today. |
| 02:30.10 | dyoung-web | How does one "accidentally" overwrite the MAC address like that? |
| 02:31.05 | Tiersten | Yep. I was wondering that |
| 02:31.34 | Tiersten | I guess you could have written something which flipped all those bits... Bit of an odd thing to do |
| 02:31.54 | dyoung-web | unless he was messing around with assign mode. |
| 02:33.11 | Tiersten | you'd end up with garbage though |
| 02:33.50 | Tiersten | you can't flip a bit back to 1 so you'd end up with some odd address |
| 02:35.16 | dyoung-web | hmm |
| 02:35.24 | dyoung-web | Very Odd. |
| 02:43.49 | *** join/#nslu2-linux linoxx__ (~linoxx@adsl-065-012-178-027.sip.asm.bellsouth.net) |
| 03:09.15 | Christopher | anyone remember why freeradius is in the needs to be worked on list? |
| 03:09.32 | Christopher | It just built for me |
| 03:14.42 | *** join/#nslu2-linux jp30 (~josh@jp30.nslu2-linux) |
| 03:14.58 | dyoung-web | Christopher, ask jp30. |
| 03:15.15 | dyoung-web | <Christopher> anyone remember why freeradius is in the needs to be worked on list? |
| 03:15.33 | jp30 | nope, before my time, i think |
| 03:15.38 | dyoung-web | Hmm |
| 03:15.54 | dyoung-web | In that case, No idea. |
| 03:16.07 | jp30 | i'll take a look and see if anything obvious is wrong with it |
| 03:20.50 | Christopher | k, thanks |
| 03:21.23 | jp30 | Christopher, i've looked and have some build errors |
| 03:22.24 | jp30 | In file included from sql_mysql.c:44: |
| 03:22.24 | jp30 | ../../rlm_sql.h:15:25: ltdl.h: No such file or directory |
| 03:22.25 | jp30 | In file included from sql_mysql.c:44: |
| 03:22.27 | jp30 | ../../rlm_sql.h:68: error: parse error before "lt_dlhandle" |
| 03:22.29 | jp30 | . |
| 03:23.19 | jp30 | ltdl.h and "lt_dlhandle" refer to bits of libtool... libtool is being staged ok, so it must be a matter of missing CPPFLAGS |
| 03:23.24 | Christopher | hmm, maybe i need to compile mysql and then freeradius |
| 03:23.41 | Christopher | i probably need to make it so it doesn't enable the mysql support |
| 03:24.01 | jp30 | if that's the case, then freeradius is broken, since it would need to stage mysql |
| 03:24.16 | jp30 | it dates from before there was a mysql package in unslung anyway |
| 03:24.25 | Christopher | right |
| 03:24.30 | jp30 | my guess is that it is getting confused by a host mysql install |
| 03:24.45 | jp30 | are you edmondsc by any chance? |
| 03:24.48 | Christopher | yeah, i'm thinking about telling it not to compile in mysql support |
| 03:24.50 | Christopher | yep |
| 03:24.57 | jp30 | ah |
| 03:25.05 | Christopher | sorry |
| 03:25.25 | Christopher | I use christopher here because that's who i've been in the channel |
| 03:25.42 | jp30 | s'ok. i don't know anything about radius - you can have it depend on mysql if that makes sense |
| 03:26.03 | Christopher | yeah, it can use it, but it doesn't have to |
| 03:26.11 | Christopher | I'll mess with it and see what's best |
| 03:26.21 | Christopher | thanks for the information |
| 03:26.45 | jp30 | good; move it to READY_FOR_TESTING when you want me to try building it on the official build system |
| 03:26.57 | Christopher-bbl | alright |
| 03:43.34 | *** join/#nslu2-linux rwhitby-web (www-data@netblock-66-159-209-60.dslextreme.com) |
| 03:44.31 | rwhitby-web | dyoung-zzzz: yeah, intl priority is probably easier than finding someone in EU to buy and send it |
| 03:46.33 | rwhitby-web | linoxx__: did you get your slug working again? |
| 03:47.12 | rwhitby-web | nettie: re bk: we'll move to whatever openembedded moves too. I hope they move to cogito :-) |
| 03:48.17 | rwhitby-web | jp30: did you sort out the slug purchase with ka6sox? |
| 03:50.38 | jp30 | nope, haven't talked to him yet |
| 03:54.31 | rwhitby-web | ok, just ask him directly, I'm sure he'll be happy to help out :-) |
| 03:54.46 | rwhitby-web | and he'll just send me the bill :-) |
| 03:57.09 | jp30 | ok, i'll send him an email |
| 03:57.51 | jp30 | jf tells me thttpd doesn't like thread-safe php - he seems to be right |
| 03:58.06 | rwhitby-web | yeah, what do you think we should do about this? |
| 03:58.28 | rwhitby-web | I'd like to be able to have thttpd and apache both running on the slug at the same time, and both sharing the php.mk package |
| 03:58.49 | rwhitby-web | which means "dumbing down" apache to not use the thread-safe stuff |
| 03:58.53 | jp30 | well, we can do that - we just can't have apache worker MPM |
| 03:59.20 | rwhitby-web | that's apache threads, right? which will dramatically save memory usage? |
| 03:59.51 | rwhitby-web | the trouble is that I think I want that too :-) |
| 03:59.52 | jp30 | that's right. experiments with mediawiki suggest that real php apps will be swap-bound w/ prefork |
| 04:00.43 | jp30 | by paring down apache and using worker, i was just about able to run mediawiki with no swap |
| 04:00.50 | *** join/#nslu2-linux ka6sox-web (www-data@netblock-66-159-209-60.dslextreme.com) |
| 04:00.52 | rwhitby-web | are we destined to emit two ipks (one ts, and one not) for all php-related packages? |
| 04:01.03 | rwhitby-web | ka6sox-web: speak of the angel |
| 04:01.31 | ka6sox-web | uh oh...I stepped in it now! |
| 04:01.55 | jp30 | well, i wanted to look at how well mediawiki and phpmyadmin work with thttpd... perhaps we could give up on apache |
| 04:01.59 | ka6sox-web | how many slugs/wiley's? |
| 04:02.17 | jp30 | wileys? |
| 04:02.35 | rwhitby-web | one wiley deluxe for dyoung, one slug for dyoung (to replace jf's slug) and one slug for jp30. |
| 04:02.39 | jp30 | ah |
| 04:02.59 | rwhitby-web | ka6sox-web: that name's gonna stick. |
| 04:03.22 | ka6sox-web | I like my Hard Wiley |
| 04:03.24 | rwhitby-web | I'll see if Oleg wants to call his firmware "Wiley" :-) |
| 04:03.39 | rwhitby-web | Unslung/Wiley |
| 04:03.46 | ka6sox-web | Wiley Coyote |
| 04:04.07 | ka6sox-web | okay so 2 slugs and 1 Wiley Delux |
| 04:04.35 | rwhitby-web | yep, and jp30 will contact you direct for shipping details. Send the bill for the slugs and shipping to me. |
| 04:05.08 | jp30 | ka6sox, shall i email you my postal address to tking at users.sf? |
| 04:05.27 | ka6sox-web | ka6sox@gmail.com |
| 04:07.20 | rwhitby-web | ka6sox-web: so the hard wiley is working well? |
| 04:07.50 | rwhitby-web | My mini wiley should arrive in the next couple of days |
| 04:08.09 | ka6sox-web | yeah. working fine at 6kms...soon to test at 17kms |
| 04:08.35 | rwhitby-web | nice. did it recognise the usb audio out of the box? |
| 04:08.59 | *** join/#nslu2-linux beewoolie-away (~beewoolie@florence.buici.com) |
| 04:09.00 | ka6sox-web | no...I used the stuff from jf/rpedde |
| 04:09.03 | peteru-home | rwhitby-web: Did you end up making an .bb for the toppy specific library/ftpd? |
| 04:09.22 | ka6sox-web | beewoolie-away: bad addy for you. |
| 04:09.24 | jp30 | ka6sox-web, email on its way. thanks! |
| 04:09.27 | ka6sox-web | came back. |
| 04:09.50 | beewoolie-away | ka6sox-away: Oh really? What did you use? |
| 04:09.51 | rwhitby-web | jp30: I couldn't get OpenShowCenter to work on thttpd, so I went back to Apache. Will test again tonight. We need to do the alternatives thing for the htpasswd executable. Do you know how to do that? |
| 04:10.02 | rwhitby-web | peteru-home: nope, just Unslung. |
| 04:10.16 | ka6sox-web | let me look at my gmail account again. |
| 04:10.28 | peteru-home | rwhitby-web: What's the unslung package called? |
| 04:10.35 | ka6sox-web | I have a new Slug here... |
| 04:10.38 | rwhitby-web | ftpd-topfield and libtopfield |
| 04:10.48 | peteru-home | thank you |
| 04:10.52 | ka6sox-web | Sound Slug...only one problem...Slugs cant' hear! |
| 04:11.11 | beewoolie-away | They can feel the vibe. |
| 04:11.15 | rwhitby-web | if you give them a brain transplant they can ... |
| 04:11.26 | jacques | if a slug falls in the forrest... |
| 04:11.35 | jacques | forest |
| 04:11.37 | jp30 | rwhitby, i noticed the problem with htpasswd; what do you mean by "the alternatives thing"? |
| 04:12.27 | rwhitby-web | jp30: you have multiple htpasswd exes (in different spots), and then use the alternatives support to automatically do the symlinks based on package priorities |
| 04:13.27 | jp30 | this sounds like a feature of ipkg i wasn't aware of. |
| 04:14.14 | rwhitby-web | http://www.handhelds.org/hypermail/familiar/4/0441.html |
| 04:16.46 | jp30 | thanks, i'll study that, looks like it is the solution to htpasswd issue. are we using it for any other unslung packages yet? |
| 04:17.26 | rwhitby-web | no, we were considering using it for busybox vs coreutils and stuff like that, but decided that those really should be incompatible |
| 04:18.27 | jp30 | how about coreutils vs. procps? |
| 04:18.31 | rwhitby-web | whereas having both thttpd and apache installed does make sense, and they just both happen to supply htpasswd |
| 04:19.15 | rwhitby-web | could probably use alternatives for procps vs coreutils |
| 04:20.55 | jp30 | ok, i'll look into it. i want to solve the threads problem first, as it is preventing me pushing other stuff to the feeds at the mo. |
| 04:22.05 | rwhitby-web | jp30: cool |
| 04:22.12 | rwhitby-web | ka6sox-web: we need this: http://ipkgfind.handhelds.org/ |
| 04:22.37 | rwhitby-web | jp30: so do we need to go for two ipks emitted from php packages? |
| 04:25.23 | *** join/#nslu2-linux jbot (ibot@apt.bot.TimRiker.active.supporter.pdpc) |
| 04:25.23 | *** topic/#nslu2-linux is Developer (not end-user) discussion about modifying the Linux operating system on the Linksys NSLU2. Use #nslu2-general for end-user questions. See www.nslu2-linux.org for info, and be sure to read www.nslu2-linux.org/wiki/Main/FAQ before asking questions. |
| 04:26.18 | jp30 | the php extensions we have at the moment are all parts of the php distribution. if we didn't have a standalone php, they would end up built into the web servers too (or in the case of apache, built into mod_php) |
| 04:31.15 | rwhitby-web | but still as dynamic linked libraries, right> |
| 04:31.17 | rwhitby-web | ? |
| 04:32.34 | jp30 | they could be, if we wanted. have to have two sets though |
| 04:33.04 | rwhitby-web | I think we always want them to be, so people can decide not to load some extensions to same RAM |
| 04:33.26 | rwhitby-web | can we get a single php.mk to emit two sets of ipks? |
| 04:34.59 | jp30 | that sounds like the php-nothreads / php-threads proposal... ? |
| 04:35.21 | jp30 | maybe it would be worth enumerating the options |
| 04:35.56 | jp30 | i can think of about 4 |
| 04:37.22 | jp30 | 1) go back to apache prefork, shared php extensions, no threads; 2) hack thttpd until it supports threadsafe php; 3) fork php into php and php-nothreads; 4) give up on php.mk, and build php independently to support apache, thttpd, and cli/cgi |
| 04:39.17 | jp30 | seems to me that (1) is a good holding position; (4) is strictly better than (3); (2) means hard work and likely portability issues with wl500g |
| 04:39.19 | rwhitby-web | jp30: what about new non-core php modules? most of them are not thread-safe yet. |
| 04:39.52 | rwhitby-web | yeah, I think 2) is a non-starter |
| 04:40.47 | jp30 | non-core module issues: that's why i didn't initially think that apache / worker would work. but when i found that mysql was thread-safe... |
| 04:41.17 | rwhitby-web | not sure that 4) flies when you've got possible non-thread-safe third-party modules. You would need to allow the user to choose a non-thread-safe php for apache if they needed that one third-party module which has not been made thread safe yet ... |
| 04:41.57 | jp30 | if we want to support non-thread safe modules, the only thing to do is apache/prefork, and that means (1) |
| 04:42.07 | rwhitby-web | that's why I agree that 1) is a good holding position, but doesn't allow us to go to MPM for core and thread-safe php modules. |
| 04:43.58 | rwhitby-web | and 4) is the best of all worlds, as you can choose to use apache prefork with php-non-threaded (and all your non-threaded modules) or you can choose apache MPM and php-thread-safe and a restricted set of third-party modules (which are made to conflict with the non-thread-safe php ipks) |
| 04:44.19 | rwhitby-web | s/non-threaded/non-thread-safe/ |
| 04:44.45 | rwhitby-web | oops - I meant 3) for that last statement |
| 04:45.22 | jp30 | (3) would also require us to fork every threadsafe non-core php extension |
| 04:45.43 | rwhitby-web | yeah, I know. |
| 04:46.56 | jp30 | it sounds as though you are also thinking of offering the user a choice of apache worker and prefork. that is hard in itself, as the apache include files in staging are mpm specific |
| 04:47.50 | rwhitby-web | hmm. |
| 04:48.09 | rwhitby-web | BTW, seems that others have to contend with forking php modules: see the bottom of http://www.sourceguardian.com/ixeds/how_to_install_ixed.html |
| 04:49.06 | jp30 | i am trying phpmyadmin under php-thttpd now - it is too slow :( |
| 04:49.20 | jp30 | i wonder if eaccelerator can work with thttpd? |
| 04:49.31 | rwhitby-web | dunno |
| 04:50.20 | jp30 | i will try... |
| 04:51.48 | rwhitby-web | jp30: what was the memory footprint of the apache/worker running mediawiki? |
| 04:52.54 | jp30 | about 17MB if i remember correctly |
| 04:53.28 | rwhitby-web | anyway, we have thrashed out all the issues here. I have full trust that you will make the right decision re 3) vs 4) |
| 04:54.06 | rwhitby-web | if apache/worker running mediawiki can fit in 17MB, then that's reason enough to not stick with 1) |
| 04:56.03 | rwhitby-web | (BTW, it would have looked like I was pushing 3 much more than 4, but if 3 doesn't really give us anything over 4 then I'm quite happy to go with 4) |
| 04:57.33 | rwhitby-web | i.e. I'm confident that the close working model that you have jean-fabrice have in place is just as good as a common php.mk to ensure we don't diverge them any more than is absolutely required. |
| 04:57.50 | rwhitby-web | s/you have/you and/ |
| 04:58.32 | rwhitby-web | cause the only real downside to 4) is the risk of people adding modules to one but not the other, and we can easily fix that later if it happens. |
| 04:59.03 | jp30 | ok, i'll think about this some more, and see how i get on playing with thttpd. most likely outcome for now is falling back to (1) so we can get some .ipks shipped to users, and get php-thttpd tested on slugs |
| 04:59.52 | rwhitby-web | ok, but I really want to try the apache/worker myself :-) |
| 05:03.11 | jp30 | you can build it from today's cvs by adding --with-mpm=worker to apache's configuration flags |
| 05:05.19 | *** join/#nslu2-linux rwhitby-web1 (www-data@netblock-66-159-209-60.dslextreme.com) |
| 05:07.43 | rwhitby-web1 | jp30: is that the only change required? |
| 05:08.31 | jp30 | should be - you will also need to dirclean and rebuild php-apache afterwards |
| 05:14.34 | ka6sox-web | rats |
| 05:18.09 | jp30 | rwhitby, i'll check in my latest tuned httpd.conf - that will help with worker if you want to try it |
| 05:18.57 | CIA-8 | 03jp30 * 10unslung/sources/apache/httpd-std.conf.in: apache: turn worker mpm settings |
| 05:37.50 | ka6sox-away | jp30 got the info just fine |
| 05:39.52 | jp30 | good... i trust you will be reimbursed for postage? |
| 05:40.23 | ka6sox-web | its all arrainged |
| 05:42.26 | jp30 | fab. i will soon have a little slug farm here |
| 05:43.32 | ka6sox-web | yep... |
| 05:43.39 | ka6sox-web | jacques: ping |
| 05:56.11 | jacques | ka6sox-web, hi |
| 06:01.16 | ka6sox-web | howdy, how are you doing tonight? |
| 06:04.53 | *** join/#nslu2-linux ruDelph (~delph@194.154.83.193) |
| 06:06.16 | jacques | ka6sox-web, kinda drowsy |
| 06:06.28 | jacques | but it's way too early to go to sleep |
| 06:33.57 | *** join/#nslu2-linux Helius (~jeff@ip68-102-109-95.ks.ok.cox.net) |
| 07:05.04 | jf-work | moringn |
| 07:05.13 | jf-work | hum... morning ! |
| 07:05.33 | jacques | :-) |
| 07:06.24 | jp30 | morning, jf-work |
| 07:08.58 | ka6sox-web | ~slugtime |
| 07:08.59 | jbot | i heard slugtime is 40 hour days 10 hours in 4 timezones with overlap |
| 07:16.15 | *** join/#nslu2-linux nslu2-log (dyoung@netblock-66-159-209-60.dslextreme.com) |
| 07:17.54 | CIA-8 | 03jp30 * 10unslung/ (7 files in 2 dirs): back to non-threadsafe php by default |
| 07:29.25 | CIA-8 | 03jp30 * 10unslung/Makefile: promote php-thttpd |
| 07:57.46 | jf-work | jp30: do you know why php-thttpd.mk and php-apache.mk link against libxml2 and do not make libxml2 staged ? |
| 08:07.51 | jp30 | jf-work, they shouldn't - will fix |
| 08:09.08 | jf-work | I remember rwhitby-away make'em link against instead of using php extensions. Maybe there's an (obscur for me) reason |
| 08:10.13 | jp30 | sorry, i should have been more explicit. i mean they should link against libxml2, they shouldn't fail to stage it |
| 08:10.41 | jp30 | php's libxml2 support doesn't build as a DSO extension |
| 08:10.49 | jf-work | hum.. ok |
| 08:10.59 | jf-work | SDo I changed php-thttpd to stage libxml2 |
| 08:11.06 | jf-work | s/SDo/so |
| 08:11.25 | jf-work | I'm now using your fix to php-thttpd for download target. have cleanup php-thttpd.mk (not using .downloaded flag anymore). Seems to build |
| 08:13.15 | CIA-8 | 03jp30 * 10unslung/make/php-apache.mk: php-apache: stage libxml2 |
| 08:13.20 | jp30 | good. i've just committed a fix to php-apache to stage libxml2 |
| 08:13.31 | jp30 | ...now i must sleep. 'night! |
| 08:14.06 | jf-work | 'night jp30 |
| 08:24.23 | CIA-8 | 03jeanfabrice * 10unslung/make/php-thttpd.mk: stage libxml2 and cleanup source download stuff |
| 10:18.21 | dyoung-zzzz | wow lots happening. |
| 10:20.09 | rwhitby | I'm keen to test the worker thread apache/php to see if it can do the openshowcenter media server without swapping .... |
| 10:23.15 | rwhitby | Very nice howto from jeanfabrice at http://forum.chupa.nl/showpost.php?p=14309&postcount=4 |
| 10:25.23 | *** join/#nslu2-linux jf-work (~none@home.bobo-rousselin.com) |
| 10:32.12 | dyoung-zzzz | I have another mule going to France next week. |
| 10:32.35 | dyoung-zzzz | But its going to be faster for me to just intl priority it I think... |
| 10:33.06 | dyoung-zzzz | nice howto. |
| 10:34.02 | jf-work | dyoung-zzzz: do it as easy and painless it is for you... I still have to buy the adapter |
| 10:35.02 | dyoung-zzzz | I'll send it tommrow. I dont want to put too much burden on my french friends. |
| 10:35.12 | jf-work | ok |
| 10:35.34 | dyoung-zzzz | I had forgotten they are all going for a wedding, so are carrying a LOT of stuff back. |
| 10:37.53 | jf-work | :) Where does the wedding take place ? |
| 10:38.15 | dyoung-zzzz | abut 200km south of you I believe. |
| 10:38.48 | jf-work | x fr en pense |
| 10:38.56 | dyoung-zzzz | if I read this map correctly at least. Saint Clement de Reviere. |
| 10:39.40 | dyoung-zzzz | (they gave me the invitation in french which I havnt quite translated yet....) |
| 10:40.06 | jf-work | humm. nice & sunny place |
| 10:40.06 | dyoung-zzzz | I think thats north of Montpellier |
| 10:41.01 | dyoung-zzzz | Kai Steinbach is such a polite guy. |
| 10:52.54 | rwhitby | yep |
| 11:18.23 | *** join/#nslu2-linux mrblack51 (mrblack51@wa-pullman-cuda1k-78.losaca.adelphia.net) |
| 12:36.01 | dyoung-zzzz | ~seen voodooz |
| 12:36.06 | jbot | voodooz <~Stephane6@CPE00c0f021712c-CM014110003785.cpe.net.cable.rogers.com> was last seen on IRC in channel #nslu2-linux, 97d 10h 17m 45s ago, saying: 'Anybody ever heard if any of those PCMCIA-to-USB cards work under linux? (I'm too broke to upgrade my portable :( )'. |
| 12:36.18 | dyoung-zzzz | ~seen voodooz-work |
| 12:36.19 | jbot | i haven't seen 'voodooz-work', dyoung-zzzz |
| 12:36.28 | *** join/#nslu2-linux jf-work1 (~none@home.bobo-rousselin.com) |
| 12:36.31 | dyoung-zzzz | ~seen voodooz_work |
| 12:36.32 | jbot | voodooz_work is currently on #openjtag #nslu2-linux |
| 12:44.28 | VoodooZ_Work | hey! |
| 12:44.56 | Tiersten | Morning |
| 12:45.03 | VoodooZ_Work | morning guys |
| 12:45.25 | VoodooZ_Work | hey dyoung. |
| 12:45.29 | VoodooZ_Work | Did you get my email? |
| 12:47.04 | *** join/#nslu2-linux Machiavelli (~Machiavel@0x535b9779.arcnxe1.adsl-dhcp.tele.dk) |
| 12:47.25 | macsat | hi guys |
| 12:47.31 | dyoung-zzzz | VoodooZ_Work, Hi! |
| 12:47.36 | dyoung-zzzz | I got your mail |
| 12:47.41 | VoodooZ_Work | long time no see! |
| 12:47.42 | dyoung-zzzz | Theres a funny story about that.... |
| 12:47.44 | VoodooZ_Work | cool. |
| 12:47.47 | VoodooZ_Work | what? |
| 12:47.56 | dyoung-zzzz | I dont actually *have* a PWC yet. |
| 12:48.06 | VoodooZ_Work | really? That's funny alright! |
| 12:48.24 | dyoung-zzzz | However, the person you'll want to talk to is odoc (on channel) |
| 12:48.36 | VoodooZ_Work | I did a lot of research online but couldn't find any performance figures. |
| 12:48.54 | dyoung-zzzz | He has one and tested it and provided the endian patch that made it actually work. |
| 12:49.00 | VoodooZ_Work | odoc? Never seen him around. |
| 12:49.05 | dyoung-zzzz | if you want long enough I'll buy one eventually. |
| 12:49.05 | VoodooZ_Work | cool. |
| 12:49.11 | dyoung-zzzz | s/want/wait |
| 12:49.21 | dyoung-zzzz | and can give you the numbers then... |
| 12:49.21 | VoodooZ_Work | Does he come here? |
| 12:49.35 | dyoung-zzzz | Yeah, he pops in periodically. |
| 12:49.52 | VoodooZ_Work | ok. I'll try to locate him. |
| 12:50.21 | VoodooZ_Work | meanwhile I'm finding ways to optimize my computer vision code further. |
| 12:50.41 | VoodooZ_Work | The bottleneck is still the USB bus as it takes ~160ms per frame using double-buffering. |
| 12:50.54 | VoodooZ_Work | That's why I'm wondering if USB2.0 would help. |
| 12:51.24 | dyoung-zzzz | I think odoc is usnig it to grab a image "every now and then"; but I think he listed as the "official" conatct for nslu2/pwc on the pwc page now. |
| 12:51.39 | dyoung-zzzz | so he ought to be willing to run some tests for you. |
| 12:52.20 | Tiersten | Wasn't there a big thing on the kernel list about the PWC driver recently? It had a hook for a binary module which gave you access to the "higher" resolution images |
| 12:52.37 | VoodooZ_Work | Tiersten: yep. A real drama. |
| 12:52.39 | Tiersten | Turns out the camera and/or binary module just did loads of interpolation to give you a higher resolution but fuzzier image |
| 12:52.48 | dyoung-zzzz | Okay not quite yet, buy http://www.saillard.org/linux/pwc/projects.php might help. |
| 12:53.00 | VoodooZ_Work | yeah, compression is required for higher frame rate and VGA |
| 12:53.13 | VoodooZ_Work | yeah, I looked it up. |
| 12:53.35 | dyoung-zzzz | Right, the saillard driver doesnt have the binary hook. |
| 12:53.43 | VoodooZ_Work | I hope the decompressor is not too cpu intensive though. |
| 12:53.48 | Tiersten | I've got a EyeToy actually which is VGA |
| 12:53.56 | VoodooZ_Work | really? |
| 12:53.59 | Tiersten | That is apparently supported by one of the OVxxx drivers |
| 12:54.00 | Tiersten | yeah |
| 12:54.09 | Tiersten | No clue if it's okay with the endianess |
| 12:54.11 | VoodooZ_Work | Yep. That's what I'm using too. |
| 12:54.22 | VoodooZ_Work | Just add CONFIG_OV511 to your defconfig and voila! |
| 12:54.25 | dyoung-zzzz | Tiersten, The endianness should be resolved. |
| 12:54.27 | Tiersten | Ah. At least they say it's VGA :) |
| 12:54.39 | dyoung-zzzz | *should*..... |
| 12:54.39 | VoodooZ_Work | Is it usb 2.0? |
| 12:54.45 | Tiersten | No. USB1 |
| 12:54.53 | VoodooZ_Work | Probably not without compression though. |
| 12:55.45 | dyoung-zzzz | I can do futhur testing when I actually have one. For now I need to defer to odoc's expertise. |
| 12:56.05 | VoodooZ_Work | dyoung-zzzz, No problems, Thanks for the tip. |
| 12:56.14 | Tiersten | http://www.planetanalog.com/printableArticle.jhtml?articleID=16600457 |
| 12:56.29 | Tiersten | Need to see if the driver supports VGA |
| 12:56.35 | Tiersten | It might be using it in QVGA or something |
| 12:57.24 | VoodooZ_Work | dyoung-zzzz: Do you know ODOC's real name? Where is he listed on the PWC page/wiki? |
| 12:57.35 | rwhitby | peteru-home: just cross-compiled libdvb for unslung ... |
| 12:57.54 | VoodooZ_Work | VGA would be very slow even with compression. |
| 12:58.11 | VoodooZ_Work | I get around 5-6fps at 352x288. |
| 12:58.15 | Tiersten | yeah |
| 12:58.27 | VoodooZ_Work | And that's without computer vision code! |
| 12:58.37 | dyoung-zzzz | Hes not listed yet; but he will be evetually. He just asked recently if I wanted to be the nslu2/pwc contact, and told him he should feel free to take that on since he actually has one. |
| 12:58.56 | VoodooZ_Work | So I need to cut down that transfering time to a minimum. say 10-15fps! |
| 12:59.12 | VoodooZ_Work | ok. |
| 12:59.22 | dyoung-zzzz | VoodooZ_Work, do you mind if I pass your email address to him? |
| 12:59.30 | VoodooZ_Work | not at all. please do so. |
| 12:59.32 | dyoung-zzzz | rather forward your message to him. |
| 12:59.34 | dyoung-zzzz | Okay done. |
| 12:59.36 | VoodooZ_Work | yep |
| 12:59.39 | VoodooZ_Work | thanks mate |
| 12:59.55 | rwhitby | talk like an Australian day still! |
| 13:00.03 | VoodooZ_Work | hehehe |
| 13:00.32 | VoodooZ_Work | Or is it pirate day? |
| 13:00.39 | VoodooZ_Work | matey! |
| 13:00.54 | Tiersten | arrr... |
| 13:02.03 | Tiersten | Hmm. The EyeToy uses a OV519 which uses JPEG compression it seems |
| 13:02.41 | dyoung-zzzz | Sent. |
| 13:04.00 | dyoung-zzzz | For your refernece, hes in CST. |
| 13:04.05 | dyoung-zzzz | CET I mean |
| 13:04.43 | dyoung-zzzz | So he may pop in in a few hours after work is finished. |
| 13:11.48 | rwhitby | sweet - topf2ps runs on the slug. |
| 13:12.27 | rwhitby | (converts a Topfield transport stream to mpeg2 program stream, which the Neuston MC-500 can display) |
| 13:12.34 | Tiersten | Nice |
| 13:13.02 | dyoung-zzzz | Woo |
| 13:13.08 | rwhitby | so I can pull the .rec file, remultiplex, and serve to the Neuston, all on the slug. |
| 13:13.18 | dyoung-zzzz | so if I mail you a bunch of blank DVD's................ |
| 13:13.27 | jf-work1 | Had anyone ever give a look at sane |
| 13:13.31 | Tiersten | How many streams can you pull from the toppy at one time? |
| 13:13.41 | Tiersten | pingu? |
| 13:13.55 | dyoung-zzzz | Pingu is my hero. |
| 13:14.01 | Tiersten | The penguin? |
| 13:14.02 | rwhitby | Tiersten: Topfield can record two transport streams at a time. |
| 13:14.21 | rwhitby | I'm not streaming - it's batch mode |
| 13:14.31 | VoodooZ_Work | dyoung-zzzz: ok. |
| 13:14.34 | Tiersten | rwhitby: How many clients can be downloading from the topfield? |
| 13:15.27 | rwhitby | Tiersten: one - it's a usb device interface |
| 13:15.33 | Tiersten | ah okay |
| 13:16.01 | Tiersten | Was wondering if you could multiplex the requests and have several people downloading :) |
| 13:16.30 | dyoung-zzzz | www.rwhitby-toppy.org/pingu |
| 13:16.43 | Tiersten | heh |
| 13:17.27 | rwhitby | dyoung-zzzz: how about you just send me a ssh key, and then you can do it yourself :-) |
| 13:18.27 | dyoung-zzzz | You have adequate UL bandwidth? |
| 13:18.43 | Tiersten | You not get Pingu on TV where you live? |
| 13:18.58 | dyoung-zzzz | or are you gonna be iptabling me down to 5k or something? ;-) |
| 13:19.03 | rwhitby | nope. 256K upload, with total 20Gb per month shaped to 64k |
| 13:19.20 | dyoung-zzzz | Pingu is impossible to find/get here. |
| 13:20.38 | CIA-8 | 03rwhitby * 10unslung/ (make/libdvb.mk sources/libdvb/topf2ps.patch): Added libdvb |
| 13:21.32 | CIA-8 | 03rwhitby * 10unslung/Makefile: libdvb ready for testing cross. |
| 13:21.46 | rwhitby | k |
| 13:22.02 | Tiersten | Pingu is on nearly every day |
| 13:22.16 | Tiersten | Been on for years actually |
| 13:22.19 | Tiersten | heh |
| 13:24.18 | dyoung-zzzz | AFAIK, pingu has never aired here. |
| 13:24.41 | dyoung-zzzz | I only know about it because hes popular in asia, and a lot of character goods are found in the store. |
| 13:24.46 | dyoung-zzzz | (s) |
| 13:24.48 | Tiersten | Only 5 minutes an episode so they're not that large |
| 13:24.49 | Tiersten | Night |
| 13:24.58 | rwhitby | night |
| 13:25.09 | rwhitby | dyoung-zzzz: each .rec file is about 260MB |
| 13:25.58 | dyoung-zzzz | heh I was hoping to get the already transcoded files. |
| 13:26.23 | rwhitby | I'm remuxing, not transcoding. So the mpeg2 file is still 260MB. |
| 13:26.34 | dyoung-zzzz | Ooh, okay gotcha. |
| 13:26.45 | rwhitby | 15 MB/sec - high quality Pingu |
| 13:27.02 | dyoung-zzzz | drat, so I guess we wont be using that key I just sent you. ;-) |
| 13:27.33 | rwhitby | well, it's almost the end of the month |
| 13:27.44 | rwhitby | let's see how much I've got left .... |
| 13:28.01 | dyoung-zzzz | I like the seal dde. |
| 13:28.03 | dyoung-zzzz | dude. |
| 13:28.12 | rwhitby | Robby |
| 13:28.22 | dyoung-zzzz | Hes so cute it makes nauseated. |
| 13:28.38 | dyoung-zzzz | but so endearing. |
| 13:29.28 | dyoung-zzzz | night all |
| 13:35.32 | jf-work1 | have anyone tried to port sane ? |
| 13:35.43 | jf-work1 | s/have/has |
| 13:38.07 | dyoung-zzzz | I looked at it |
| 13:38.28 | dyoung-zzzz | because it has some bits to grab stuff from usb camera too |
| 13:38.35 | dyoung-zzzz | but it was more than I wanted to chew at the time |
| 13:38.37 | jf-work | it compiles fine for me (not in the unslung because need some trick like openldap (build call for self generated binary). |
| 13:39.08 | jf-work | but at launch time, it is crappy like mysql is on my wiley (!) |
| 14:17.40 | *** join/#nslu2-linux _chris_ (~chris@pcp04638208pcs.gambrl01.md.comcast.net) |
| 14:33.20 | *** join/#nslu2-linux mrb51 (mrblack51@wa-pullman-cuda1k-78.losaca.adelphia.net) |
| 14:46.40 | peteru-home | Pingu was very popular in Japan. At one stage they had a promotion where if you bought 4 doughnuts and a large softdrink, they'd throw in a Pingu mini-disc player carry pouch :) |
| 14:47.23 | peteru-home | When Pingu started being so popular, Sanrio decided to compete and started pushing "Daniel" - Hello Kitty's boyfriend. |
| 14:49.21 | Tiersten | I thought Doraemon was better personally :) |
| 14:49.54 | peteru-home | ffmpeg should let you transcode the MPEG-2 streams to MPEG4/DivX to make them more suitable for internet - not sure how well the slug is likely to cope with that task though. |
| 14:50.12 | Tiersten | The NSLU2 isn't powerful enough for that |
| 14:50.18 | Tiersten | By a long way |
| 14:50.47 | Tiersten | Would be nice to be able to do that however |
| 14:51.57 | peteru-home | I wasn't even contemplating streaming. |
| 14:52.18 | Tiersten | yeah but even batch conversion would take ages |
| 14:52.32 | Tiersten | Like a couple of days sort of thing |
| 14:52.40 | peteru-home | Our DVB-T is about 6Mbit/s which comes to 2GB per hour of TV. |
| 14:52.52 | peteru-home | that's pretty slow to transfer |
| 14:54.00 | Tiersten | All of the portable MPEG2/4 devices which have an input do it using a hardware encoder chip |
| 14:56.30 | peteru-home | that involves an analogue stage, although in theory it should be possible to decode MPEG2 to YUV frames and then get the MPEG4 encoder to encode the raw frames. |
| 14:56.52 | Tiersten | yeah but the hard part isn't the analog -> digital part. it's the compression stage |
| 14:57.12 | Tiersten | a few years back, people ripping DVDs and compressing it to DiVX would have to leave it running for hours and hours |
| 14:57.30 | Tiersten | P3 era |
| 14:57.45 | Tiersten | That was MPEG2 -> MPEG4 but at a higher bitrate on average |
| 14:59.14 | peteru-home | in my last job, we had a solution that used a hardware encoder to transcode between various digital formats at faster than real-time. We could do IMX to DV50 at 3x, all it needed was a Linux box with a PCI slot. |
| 14:59.39 | peteru-home | then again, this was aimed at broadcasters |
| 15:00.15 | Tiersten | Yeah |
| 15:00.39 | Tiersten | You'd lose the USB 2 however if you wanted to add a encoder to the NSLU2 |
| 15:01.46 | peteru-home | there are just some tasks that the slug shouldn't try to do - leave it to hardware designed for the task. |
| 15:02.56 | Tiersten | Yeah |
| 15:03.25 | macsat | is anyone running mailserver (pop, imap, smtp) on the NSLU2 ? |
| 15:12.33 | *** join/#nslu2-linux pTweety (~pTweety@dialin-212-144-093-028.arcor-ip.net) |
| 15:27.53 | CIA-8 | 03ptweety * 10unslung/ (make/cyrus-imapd.mk sources/cyrus-imapd/cyrus.cross.patch): |
| 15:27.54 | CIA-8 | Now ipk can be created when doing cross compile. |
| 15:27.54 | CIA-8 | But perl-stuff is still missing when compiling cross. |
| 15:32.16 | *** join/#nslu2-linux beewoolie-away (~beewoolie@florence.buici.com) |
| 15:47.03 | *** join/#nslu2-linux ByronT-Away (byrontodd@adsl-212-195-69.bhm.bellsouth.net) |
| 16:09.11 | *** join/#nslu2-linux ByronT (byrontodd@ByronT.nslu2-linux) |
| 16:19.23 | *** join/#nslu2-linux jipi (~jipi@cm16.gamma183.maxonline.com.sg) |
| 18:06.11 | *** join/#nslu2-linux [cc]smart (~smart@gw.ptr-62-65-149-158.customer.ch.netstream.com) |
| 18:42.53 | *** join/#nslu2-linux clueless (~Clue@p54AE8495.dip0.t-ipconnect.de) |
| 18:43.02 | *** join/#nslu2-linux ByronT (byrontodd@ByronT.nslu2-linux) |
| 18:46.00 | *** join/#nslu2-linux ucl (~Clue@p54AE8495.dip0.t-ipconnect.de) |
| 18:47.21 | *** join/#nslu2-linux ucl (~Clue@p54AE8495.dip0.t-ipconnect.de) |
| 18:53.01 | *** join/#nslu2-linux caplink811 (~Joerg@dsl-213-023-145-111.arcor-ip.net) |
| 19:02.41 | CIA-8 | 03jp30 * 10unslung/Makefile: demote libdvb |
| 19:40.15 | *** join/#nslu2-linux Tiersten (~tman@Tiersten.nslu2-linux) |
| 19:41.54 | *** join/#nslu2-linux jf-asleep (~jeanfabri@home.bobo-rousselin.com) |
| 20:23.28 | jeanfabrice | out of channel subject but is there a way to be pushed with the mailing list without creating a yahoo account ? |
| 20:32.56 | ka6sox-office | I have a yahoo account but the list goes to my email account (not the yahoo one) |
| 20:34.16 | NAiL | Send a mail to nslu2-general-subscribe@yahoogroups.com I guess |
| 20:44.15 | *** join/#nslu2-linux copperbeech (~copperbee@82-40-186-72.cable.ubr04.newm.blueyonder.co.uk) |
| 20:44.58 | jeanfabrice | ty. I'll try it |
| 20:47.52 | NAiL | np, gl |
| 21:01.48 | jp30-work | jeanfabrice, what's the status of thttpdphp.mk? is there a reason to keep it in the package system, or is it obsoleted by php-thttpd? |
| 21:08.32 | jeanfabrice | jp30-work: thttpdphp is now obslolete. It can be destroyed. |
| 21:09.08 | jeanfabrice | It doesn't build using the wl500g branch. Don't know about the slug branch |
| 21:09.09 | jp30-work | ok, i'll destroy it then :) |
| 21:09.15 | jeanfabrice | ok |
| 21:11.13 | jeanfabrice | I was wondering how both branch maintainer/developper could be inform of package created by the other branch. As I update only my part of the makefile and so you may do, I will probably not see package you create |
| 21:14.31 | jp30-work | i tend not to pay much attention to which packages get created - i wait until they show up in READY_FOR_TESTING. i suppose the trouble is, how do we trigger (e.g.) wl500g testing of a package submitted for slug |
| 21:14.42 | jp30-work | (or vice versa) |
| 21:14.51 | jp30-work | ...is that what you were thinking about? |
| 21:15.15 | jeanfabrice | could you put your package in the READY_FOR_TESTING wl500g section ? |
| 21:15.32 | CIA-8 | 03jp30 * 10unslung/ (11 files in 3 dirs): remove obsolete thttpdphp package |
| 21:15.42 | jp30-work | would you want me to if i didn't know whether it would work on wl500g? |
| 21:15.52 | jeanfabrice | yep |
| 21:17.26 | jeanfabrice | That's give me a visibility on package you create and so I can test if they build fine. I ask for package contributors on wl500g forum but did not get a lot of answer right now. So you may create more package than I do. |
| 21:17.34 | jp30-work | the trouble is, i don't take responsibility for runtime QA testing on slug - are you sure you want to take responsibility to test every package at runtime on wl500g? |
| 21:19.15 | jp30-work | ...i mean, i'm expecting that package developers won't submit something until they have tested that it works at runtime on the slug |
| 21:19.32 | jp30-work | ...but they can't be expected to test on both slug and wl500g |
| 21:20.18 | jp30-work | ...of course, if you like, I could pop each package i promote on slug into WL500G_PACKAGES_READY_FOR_TESTING as i promote them |
| 21:25.46 | jeanfabrice | that is what I would have said. Package you promote are package which built correctly on nudi for slug. If I see them in WL500G_PACKAGES_READY_FOR_TESTING, I can take a few minutes trying to build them on my machine just to see if they build. |
| 21:26.56 | jeanfabrice | I don't think I will take responsability fo runtime QA testing on wiley too... but hope that users installing packages will give feedback |
| 21:32.48 | jp30-work | ok, i'll do that then. |
| 21:34.10 | jeanfabrice | As an example, I see your message on the mailing list today and saw that you've promoted esmtp and libesmtp. That could be interesant package for wl500 package user |
| 21:35.25 | jp30-work | a further ramble on QA - the only thing i would be worried about if i were you is that there will in some cases be nobody reponsible for ensuring that any given package actually works at runtime on the wiley. |
| 21:35.54 | ka6sox-office | omg...it stuck. |
| 21:36.39 | jp30-work | :) |
| 21:37.01 | jp30-work | ka6sox, what is the story with libao? is someone actually working on fixing it? |
| 21:37.18 | ka6sox-office | not to my knowledge. |
| 21:37.21 | jeanfabrice | agreed jp30-work |
| 21:37.25 | ka6sox-office | it needs to be "fixed" |
| 21:38.36 | jp30-work | is it slugbugged? |
| 21:39.50 | jeanfabrice | jp30-work: the problem I have is that actually, you create 10 times more package than me for asus community. I was thinking a way to grow fast the ipkg available for the wiley |
| 21:41.09 | ka6sox-office | um....I'll kick the poor sod who forgot to Slugbug it... |
| 21:41.44 | jp30-work | jf, in that case, perhaps the benefit of rapidly creating packages offsets the cost of unreliability |
| 21:43.03 | jp30-work | maybe we could create a shared CROSS_PACKAGES line in the root Makefile; packages in there are known to work on both targets |
| 21:43.06 | jeanfabrice | agree again :-/ |
| 21:43.07 | ka6sox-office | jeanfabrice, is there a mechanism to get bugs reported back? |
| 21:43.44 | jeanfabrice | ka6sox-office: nothing well organised. |
| 21:44.10 | jp30-work | ...then you could more easily scan a SLUG_ONLY_CROSS_PACKAGES for additions |
| 21:46.59 | *** join/#nslu2-linux ByronT (byrontodd@ByronT.nslu2-linux) |
| 21:47.28 | jeanfabrice | jp30-work: and we would synchronized the shared CROSS_PACKAGES on irc ? |
| 21:51.34 | jp30-work | my idea is: i would promote packages from WL500G_ONLY to CROSS, you could promote packages from SLUG_ONLY to CROSS |
| 21:56.55 | jeanfabrice | jp30-work: that sounds fine to test. |
| 21:57.42 | jp30-work | jf, let's think about it; i would like feedback from rwhitby before making a change to the organisation of the root Makefile anyway |
| 21:58.27 | jp30-work | meanwhile i'll pop any newly promoted slug packages in WL500G_READY_FOR_TESTING. how's that? |
| 21:58.28 | jeanfabrice | yes, np |
| 22:14.14 | jeanfabrice | jp30-work: does esmtp shouldn't depend on libesmtp ? |
| 22:19.01 | CIA-8 | 03jeanfabrice * 10unslung/Makefile: Promote esmtp and libesmtp to wiley |
| 22:20.57 | *** join/#nslu2-linux glc (~glc@glc.nslu2-linux) |
| 22:33.06 | eno-away | jeanfabrice: i think you're right, esmtp should depend on libesmtp |
| 22:33.42 | jp30-work | eno, jf, yeah, it should |
| 22:33.59 | jeanfabrice | want me to fix ? |
| 22:35.09 | jp30-work | go ahead |
| 22:35.14 | CIA-8 | 03bzhou * 10unslung/make/esmtp.mk: added libesmtp to _DEPENDS; bumped _IPK_VERSION |
| 22:35.33 | eno-away | sorry, just went across |
| 22:35.49 | jeanfabrice | too fast..pfiouu. |
| 22:38.27 | ka6sox-office | rwhitby-away ping |
| 22:47.50 | *** join/#nslu2-linux rwhitby-nslu2 (~rwhitby@rwhitby.nslu2-linux) |
| 22:52.53 | rwhitby-away | jp30-work, jeanfabrice: just read about the shared cross variable. great idea. |
| 22:53.44 | rwhitby-away | BTW, you two guys own the build system, so whilst I appreciate you asking my opinion on things, don't let me hold up progress if you don't get a timely response - just go with what you two decide (after discussing it here in the open forum of course). |
| 23:00.32 | *** join/#nslu2-linux _chris_ (~chris@pcp09943963pcs.hyatsv01.md.comcast.net) |
| 23:56.22 | *** join/#nslu2-linux rwhitby-web (www-data@netblock-66-159-209-60.dslextreme.com) |