**** BEGIN LOGGING AT Fri Dec 02 03:00:09 2005 Dec 02 03:07:53 Hi folks, quick question. I'm adding new native package. I had to write custom do_stage which install couple of binaries and corresponding libraries in staging and I'm using devshell to setup all PATHs, The problem that binaries from staging/bin could not run, because they could not find libraries. Some ideas what I'm doing wrong ? Dec 02 03:09:55 morning morning morning Dec 02 03:21:02 heh.. sometimes its good to have apache2+mysql locally.. Dec 02 03:35:41 morning Dec 02 03:36:25 hi greentux Dec 02 03:36:35 hi hrw|work Dec 02 03:36:43 ~seen lrg Dec 02 03:36:48 lrg is currently on #oe Dec 02 03:36:49 ~seen lrg|home Dec 02 03:36:50 lrg|home is currently on #oe (52m 11s) Dec 02 03:36:55 I'm here Dec 02 03:37:00 :) Dec 02 03:37:09 greentux: got my borzoi yesterday Dec 02 03:37:12 ~ibot Dec 02 03:37:13 extra, extra, read all about it, ibot is a blootbot written in perl run by TimRiker on his server. logs on http://ibot.rikers.org// , ibot, jbot, apt are all the same process. It uses sqlite, but mysql or other SQL storage is also supported. Dec 02 03:37:25 lrg|home: ooohh very fast Dec 02 03:37:28 thanks Dec 02 03:37:55 greentux: I'll fax back the receipt on Monday when I'm back at the office Dec 02 03:38:39 lrg|home: ok, np. hope you can improve the "sound" ... which company produces the soundchip? Dec 02 03:39:04 greentux: that would be my employer Wolfson Dec 02 03:39:41 aaaahhh Dec 02 03:40:46 lrg|home: so that's why wmXXXX are something familiar for you ;) Dec 02 03:40:59 hrw|work, yes. Dec 02 03:42:53 ok, have to reboot the pc... cu Dec 02 03:51:20 ~kill ms ie Dec 02 03:51:22 * ibot shoots a fluxneutron gun at ms ie Dec 02 04:04:38 hi Dec 02 04:09:45 hi gints Dec 02 04:11:18 gints: ping Dec 02 04:12:10 does exist a page where I can post a report about a broken package? Dec 02 04:12:20 http://bugs.openembedded.org/ Dec 02 04:15:32 hi Dec 02 04:17:17 hi mickeyl Dec 02 04:17:38 mickeyl: how things? Dec 02 04:19:08 hey mickeyl Dec 02 04:19:53 hrw|work: thanks, fairly well. still struggling a bit with my health, but nothing serious. Dec 02 04:20:05 ph.d. progressing slowly, but progressing at least :) Dec 02 04:20:15 how are things at your end? Dec 02 04:20:17 mickeyl: we might have some pcmcia breakage in the branch: http://handhelds.org/hypermail/familiar/295/29505.html Dec 02 04:20:32 I'll try to investigate after finishing my assignments Dec 02 04:20:55 mickeyl: have to buy new bed but otherwise good Dec 02 04:21:06 koen: please try downgrading module-init-tools - i have the slight idea that this could be the case Dec 02 04:21:34 dropbear is also broken Dec 02 04:21:45 define broken Dec 02 04:22:22 3 weeks until christmas... shock, horror. i have no idea what to give Dec 02 04:22:38 the homepage of the developer seems to be deleted from the server Dec 02 04:22:40 mickeyl: same here Dec 02 04:23:08 Cobelius: http://matt.ucc.asn.au/dropbear/dropbear.html works Dec 02 04:23:10 Cobelius: http://matt.ucc.asn.au/dropbear/ works here Dec 02 04:23:23 * koen bought his own christmas present Dec 02 04:23:35 now I need to figure out what my girlfriend wants Dec 02 04:23:46 * mickeyl can't buy something for himself because everything would just distract him Dec 02 04:24:03 heh Dec 02 04:24:08 * mickeyl whistles 'diamonds are a girls' best friend' Dec 02 04:24:21 this year we go to Ania parents for Christmas and I dont know what to buy Dec 02 04:25:07 hmmm some minuts ago I got: NOTE: Task failed: Fetch failed: http://matt.ucc.asn.au/dropbear/releases/dropbear-0.46.tar.bz2 Dec 02 04:25:07 NOTE: package dropbear-0.46-r3: task do_fetch: failed Dec 02 04:25:13 and the site doesnt' work Dec 02 04:25:49 sounds like a problem on your site Dec 02 04:25:59 try to ping/traceroute it Dec 02 04:26:10 maybe someone cut a cable Dec 02 04:26:11 heh Dec 02 04:26:34 12:26:54 (42.62 KB/s) - `dropbear-0.46.tar.bz2' saved [1413918/1413918] Dec 02 04:27:33 mickeyl, koen: will we give users christmas gift (new release) or skip it to valentines? Dec 02 04:28:12 christmas gift. I really want to start with the EABI thing next year Dec 02 04:28:14 okay Dec 02 04:28:15 koen? Dec 02 04:28:20 same here Dec 02 04:28:46 mickeyl: our best bet is to put the 2005q3 csl toolchain into OE Dec 02 04:29:07 gcc 4.1 HEAD has eabi support too Dec 02 04:29:09 koen: *nod* + the kernel patches on rmk's devel site Dec 02 04:29:16 ah, that's good to know Dec 02 04:30:19 dropbear home page works for me too Dec 02 04:35:21 Why is it called dropbear? Dec 02 04:35:44 ask author Dec 02 04:36:53 iirc dropbear is a mythical animal in australia Dec 02 04:37:17 hmm or rather a real one Dec 02 04:37:26 see http://www.cfr.com.au/dropbears/ Dec 02 04:41:09 if kernel has to be patched for EABI then what with 2.4-crapix? Dec 02 04:41:57 morning all Dec 02 04:43:27 RP: Morning Richard Dec 02 04:44:51 thanks mickeyl :) Dec 02 04:45:03 hrw|work: that's the major question Dec 02 04:45:06 morning Richard Dec 02 04:45:14 The answer is 2.4 dies Dec 02 04:45:26 :-/ Dec 02 04:45:48 RP: my collie and 2.6.... Dec 02 04:45:56 RP: no SD no fun Dec 02 04:46:01 unless someone ports the new EABI into the old 2.4 kernels... Dec 02 04:46:10 and don't look at me ;-) Dec 02 04:46:29 hrw|work: just write some userspace wrappers for 2.4 kernel modules we dont have source for :-) Dec 02 04:46:55 Reverse engineer them? Dec 02 04:47:17 bigal: Perfectly possible but we need someone with the immense amount of time it would take Dec 02 04:47:48 RP: I imagine it'd be bloody hard. Dec 02 04:48:13 bigal: I know I could do it. It just takes a lot of time Dec 02 04:48:45 Its why atilib has never happened for the w100 :-/ Dec 02 04:48:55 someone buy RP a chinese slave farm for reverse engineering Dec 02 04:49:17 * bigal grins. Dec 02 04:50:35 RP: Time always seems to be in short supply, too. Dec 02 04:51:08 bigal: Yes, I don't have nearly enough of it. My todo list is depressing Dec 02 04:52:34 todolist... yesterday I added opie-today tweaking into it.. Dec 02 04:53:56 it need 6s to load todolist (and 4-5s for rest) Dec 02 04:59:40 hey Dec 02 05:01:33 * zecke screwed up his exam... Dec 02 05:03:21 hi zecke, lardman|work Dec 02 05:03:28 hi hrw|work Dec 02 05:20:57 zecke will you be a guest at 22c3 or will you hold an lecture? Dec 02 05:21:03 good morning Dec 02 05:26:20 guest Dec 02 05:26:35 so I can understand this bettter, is it the native packages that get installed in STAGEDIR? Dec 02 05:26:43 %s/bettter/better Dec 02 05:27:08 both nativa and cross packages go into staging Dec 02 05:27:21 native and llibs for target go in staging_dir Dec 02 05:27:37 they are seperated by a machine thingy Dec 02 05:29:04 ok.. so if I do a -c clean on one of the libraries, it'll get yanked out of staging? Dec 02 05:29:15 Tygger-Bob: no Dec 02 05:29:19 grr Dec 02 05:30:12 how does one do that, then? Dec 02 05:30:14 rf -rf tmp before you go home and run reuild :) Dec 02 05:30:24 :( Dec 02 05:30:26 that's one of our x-mas wishes: staging under packagamanagement control Dec 02 05:30:45 ahhh.. ok. Dec 02 05:30:56 that would also mean you can save your toolchain as a bunch of ipkgs Dec 02 05:31:07 cool Dec 02 05:32:03 so.. if I DID wipe my tmp dir, and put a PREFERRED-VERSION command in for libnet to only use 1.02a in my local.conf, that might solve this problem for me? Dec 02 05:33:40 or just remove the offending libnet stuff out of staging Dec 02 05:34:40 and just rebuild libnet using -b <1.02a bb file>? Dec 02 05:35:07 that should do the trick Dec 02 05:35:15 ok.. Dec 02 05:35:40 because right now, none of the apps I'm using work with 1.1, so they all break. Dec 02 05:35:52 dsniff, thcrut, and one other one too Dec 02 05:37:59 Tygger-Bob: fix their depends then Dec 02 05:38:40 that still breaks is a newer libnet was built before iirc Dec 02 05:38:55 hrw|work, doesn't help Dec 02 05:39:01 koen, exactly.. Dec 02 05:39:16 I'd consider renaming then to libnet-1.0_1.02a.bb and libnet-1.1_1.1.4324523.bb Dec 02 05:39:25 koen: good idea Dec 02 05:39:30 just like with gtk-1.2 and gtk-2.0 Dec 02 05:39:53 that solves the selectability Dec 02 05:40:12 a few tweaks to do_stage should solve the build issues Dec 02 05:40:20 * Tygger-Bob 's confused.. what does that do? Dec 02 05:40:35 install the includes and headers into staging Dec 02 05:40:46 s/headers/libs/ Dec 02 05:43:14 so that "upgrades" libnet to 1.02a? Dec 02 05:47:07 any people use oe for upgrade packages of openzaurus(collie)? Dec 02 05:47:50 Dormi|OFF: oe is used to build oz so yes Dec 02 05:50:38 i have some problems to build oz reading http://oe.handhelds.org/cgi-bin/moin.cgi/GettingStarted Dec 02 05:51:18 my problem is when to change of use monotone Dec 02 05:53:12 monotone work fine but a have some config files and environment variables in org.openembedded.dev produce bitbake segmentacion fault or problem with python scripts Dec 02 05:53:59 please, any people whant to send me any example to use collie configuration Dec 02 05:54:10 does only I have the problem or is the monotone server down? Dec 02 05:54:35 no, monotone rule fine Dec 02 05:54:46 monotone: network error: failed to connect: Connection refused Dec 02 05:55:12 Cobelius, I upgrade yesterday all fine Dec 02 05:55:25 Cobelius: use one of the mirrors Dec 02 05:55:46 mickeyl: could you remove your viewmtn? since the db isn't there anymore it's pretty useless Dec 02 05:56:20 is somewhere a list of the mirrors? Dec 02 05:56:39 Cobelius: http://openembedded.org/ Dec 02 05:57:01 Dormi|OFF: MACHINE="collie" DISTRO="openzaurus-3.5.4" Dec 02 05:57:31 Cobelius, * Dec 02 05:57:31 monotone.nslu2-linux.org Dec 02 05:57:31 * Dec 02 05:57:31 ewi546.ewi.utwente.nl Dec 02 05:57:31 * Dec 02 05:57:32 dominion.kabel.utwente.nl Dec 02 05:57:39 ahh okay thank you Dec 02 05:57:46 ~pastebin Dec 02 05:57:47 [pastebin] a place to paste your stuff without flooding the channel - try http://pastebin.ca/ Dec 02 05:57:49 ~flood Dec 02 05:57:51 flood is probably spewing loads of output into a channel; *very* rude in most channels and often grounds for banning. If you want to show a lot of output to someone, ask them to join you in #flood and paste the output there. Dec 02 06:12:55 RP: I now have PM working on touch for tosa. I'll email it to you later today for Dirk to try Dec 02 06:13:46 lrg|home: Ok, thanks Dec 02 06:53:19 http://www.windowsfordevices.com/news/NS3737841146.html Dec 02 06:54:16 looks like modified HTC Wizard Dec 02 06:54:46 looks like embeddix accidentally compiled wince Dec 02 06:54:50 I have got a problem trying to complile gpe-mini-browser, In the populate stage package osb-jscore-0.5.0+cvs20050430-r1: task do_populate_staging: startedERROR: function do_stage failed Dec 02 06:55:12 what does staging involved? can I reset it somehow Dec 02 06:57:14 I need to get it to refetch that package, its empty Dec 02 07:08:01 ok well that package is part of gtk-webcore Dec 02 07:08:20 I'll see if I can download it manually and see if I get it going that way Dec 02 07:10:34 ahh thats not going to work cause the patches won't be applied Dec 02 07:10:41 darmou: that depend on the error you get Dec 02 07:15:19 hi lardman Dec 02 07:15:31 welcome back from lunch:) Dec 02 07:16:47 thanks Dec 02 07:34:58 oo I got a bit further FATAL: oe_libinstall: unable to locate shared library Dec 02 07:37:59 darmou: okay then there is no libNAME.so* file? Dec 02 07:38:06 darmou: look at the log Dec 02 07:43:37 hmm the log does not name the missing lib Dec 02 07:50:22 ping liam Dec 02 07:52:33 incinerator, hi Dec 02 07:53:40 * france is back (gone 22:04:51) Dec 02 07:55:11 does anyone know how I could find out what the missing lib that oe is looking for? Dec 02 07:56:53 hiho Dec 02 07:57:18 liam, i know someone how could get you a borzoi for 400 quid.... Dec 02 07:57:27 if you're interested... Dec 02 07:57:58 alternative: I buy the borzoi and you can have my 860 for 150 Dec 02 07:58:48 incinerator: thanks, I'll probably end up buying one next year. (after I sell my old car) Dec 02 07:59:02 fair enough... Dec 02 07:59:49 incinerator: I'll keep you informed when the time comes, thanks Dec 02 07:59:59 okies Dec 02 08:02:29 guys - remind me one thing please.. in c++ I can do switch(var){ case 1: do_sth();break; default: case 2: do_sth2();break;} or not? Dec 02 08:03:12 sure you can fall through Dec 02 08:03:33 but I prefer to do case 2:default: do_sth2(); break;} Dec 02 08:03:45 first I have to get it working.. Dec 02 08:03:59 or will add default value to function declaration Dec 02 08:04:38 again 3-4 minutes to get libopiepim. Dec 02 08:10:19 has anyone been able to get mplayer compiled for collie? Dec 02 08:10:30 is oe? Dec 02 08:10:36 in oe sorry? Dec 02 08:13:26 koen: have you tried to build packages natively? Dec 02 08:13:46 darmou: it is in OZ repo so it should be buildable Dec 02 08:15:30 i installed bitbake but it fails to run (missing .py imports files) Dec 02 08:15:39 *import Dec 02 08:16:33 CSMan: wrong PYTHONPATH Dec 02 08:18:20 CSMan: `which bitbake` Dec 02 08:30:48 zecke: the one from ipkgfind Dec 02 08:30:55 for 0.8.2 Dec 02 08:31:18 CSMan: ah that is what you mean with native Dec 02 08:31:19 the one that has your home path =P Dec 02 08:31:38 CSMan: I have never looked at the bb-file for bitbake, it is probably outdated :( Dec 02 08:31:41 ah, no, that's koen's path Dec 02 08:31:52 anyway sorry ;) Dec 02 08:33:59 hrw|work: the path seems to be ok, but some .py files imported in bitbake are not in the ipkg Dec 02 08:35:06 holly fsck! Dec 02 08:35:14 * CSMan hasn't traveled in a while Dec 02 08:35:26 $$$$ Dec 02 08:35:44 hrw|work I get armv4l/jrevdct_arm.S: Assembler messages: Dec 02 08:36:03 with a whole lot of assembler errors Dec 02 08:36:28 which arm is collie? Dec 02 08:36:52 the left one ;) Dec 02 08:37:03 ~collie Dec 02 08:37:05 extra, extra, read all about it, collie is sharp sl-5500 and sl-5000d, or a dog Dec 02 08:38:17 ~collie arm version Dec 02 08:38:38 everything you need should be in conf/machine/collie.conf Dec 02 08:41:41 well I did get the machine arch right in my local.conf I specified it as arm, thought so Dec 02 08:53:09 hi all Dec 02 08:54:05 I just compiled libSDL with fbcon support, and tested supertux with it: it works great :-) Dec 02 08:54:39 but I had to export SDL_NOMOUSE=1, or it complained about the missing mouse Dec 02 08:54:58 is there any way to use the touchscreen as a mouse, in fbcon? Dec 02 08:55:50 someone could write gpm subdriver for it Dec 02 09:08:10 goodnight all Dec 02 09:47:40 cu Dec 02 10:25:05 re Dec 02 10:30:21 does anyone know the syntax of the /etc/pointercal file? Dec 02 10:37:08 use the source luke Dec 02 10:40:33 :-( cal->a[0] = (int)((a*z + b*zx + c*zy)*(scaling)); Dec 02 10:40:33 cal->a[1] = (int)((b*z + e*zx + f*zy)*(scaling)); Dec 02 10:40:33 cal->a[2] = (int)((c*z + f*zx + i*zy)*(scaling)); Dec 02 10:41:05 am I supposed to understand anything :-) Dec 02 10:41:34 what device you use? Dec 02 10:44:13 C860 Dec 02 10:44:32 I've hacked gpm, Dec 02 10:44:46 ah.. gpm hacker.. I forgot.. Dec 02 10:44:49 and now I can read the absolute position of the pen, as returned by evdev Dec 02 10:45:12 but they are raw values, they need to be converted in some way Dec 02 10:45:18 You might as well learn how pointercal works, because you'll want to use those filters anyway Dec 02 10:45:20 they're good Dec 02 10:45:37 tslib? Dec 02 10:45:45 does not tslib already filter values? Dec 02 10:46:22 I guess so, but I don't think gpm's author would appraise a patch that adds a dependency on tslib Dec 02 10:46:25 Next on a very special, as the OSS turns - GPM mets tslib Dec 02 10:46:44 Mardy: tslib is the best way iirc Dec 02 10:47:04 Of course, if GPM wants to work with touchscreens, its going to need a filter - and if its going to introduce a dependency, tslib would be the right dependency to introduce Dec 02 10:47:11 Mardy: do you want to support 270° rotated ts too? etc.. Dec 02 10:48:05 weird fuzzy ts from hacked devkit etc? Dec 02 10:50:01 mvista prerotated fbs......... Dec 02 10:51:13 hrw: it's still early to say that; we'll see, if anything good comes out :-) Dec 02 11:06:56 funny Dec 02 11:07:11 there's no way to exit from the ts_test program Dec 02 11:07:21 even the power button doesn't work Dec 02 11:07:29 ctrl-c? Dec 02 11:07:40 nope :-( Dec 02 11:08:15 what's the key combination to switch to another virtual console? Dec 02 11:08:27 fn+left/right Dec 02 11:08:57 no, ts_test is really evil :-) Dec 02 11:09:16 let's see if it survives a reboot Dec 02 11:12:00 Mardy: not supporting tslib but directly reading from the input device would be braindead. tslib is the de-facto standard atm. Dec 02 11:13:19 mickeyl: agreed :-) Dec 02 11:14:16 since 20051106 I got 1352 unique spam mails Dec 02 11:14:23 if you don't need to handle calibration on your own, using tslib is a three liner :) Dec 02 11:14:31 hrw: congrats :D Dec 02 11:14:49 someone want some? Dec 02 11:14:57 thanks, I have enough on my own Dec 02 11:14:57 heh Dec 02 11:15:13 my spam archive contain just ~160M bzipped spam Dec 02 11:15:31 ~2 years collection Dec 02 11:16:40 ~lart mickeyl for sqlite off in libopie Dec 02 11:16:40 * ibot turns mickeyl into a lifesized tux doll for sqlite off in libopie Dec 02 11:18:36 heh, wie talked about that for collie Dec 02 11:18:41 s/wie/we/ Dec 02 11:18:46 I know I know Dec 02 11:20:42 heh.. my home wifi still sluggish... desktop->ap->husky Dec 02 11:21:49 guess how much time 10M file need Dec 02 11:22:01 hehe. what's the limiting factor? Dec 02 11:22:44 husky probably Dec 02 11:23:02 hmm... i need run netbench one of these days Dec 02 11:23:18 netperf? Dec 02 11:23:27 yeah Dec 02 11:26:36 10M file fetched using wget in 4m56s Dec 02 11:27:49 http://pastebin.ca/32128 - iwconfigs from husky and router Dec 02 11:28:36 19dBm? Dec 02 11:28:45 hmm that doesn't look good Dec 02 11:28:58 the bitrate is down to lowest Dec 02 11:29:01 1 Mb/s Dec 02 11:29:07 it's falling back to 1mb/s Dec 02 11:29:53 router is set to any value Dec 02 11:31:13 the cf-card antenna might be damaged Dec 02 11:32:52 too bad that I dont know does dlink dcf660w can be open without destroying Dec 02 11:33:30 my second hand buffalo card came apart by itself Dec 02 11:35:25 I have to get friend(s) with wifi equipment to go here so we can test does other cards will work (but not other in husky) Dec 02 11:37:58 ok. time to hack opiepim more Dec 02 11:44:22 mickeyl: what do you think of http://bugs.openembedded.org/attachment.cgi?id=404&action=view ? Dec 02 11:46:08 opie pim databases looks weird... Dec 02 11:47:20 koen: that looks correct at first glance. Dec 02 11:47:25 who did that? Dec 02 11:47:44 ah, guillermo Dec 02 11:47:48 no Dec 02 11:47:53 what's the bug # ? Dec 02 11:49:16 http://bugs.openembedded.org/show_bug.cgi?id=505 Dec 02 11:49:36 that should give up wpa for the h5xxx internal wlan Dec 02 11:50:21 Cool. Dec 02 11:50:35 Do we have someone with a h5xxx to check that? Dec 02 11:50:46 I have a h5xxx but no wpa Dec 02 11:51:55 the h5xxx will use the berlios driver by default, so this change shouldn't break anything Dec 02 11:52:12 it only accomodates people wanting to test the usb version of the atmelwlandriver Dec 02 11:52:15 publish it on familiar ML and someone will find to test Dec 02 11:54:35 like this: http://handhelds.org/hypermail/familiar/295/29528.html ? Dec 02 11:55:26 no link to the oz mailinglists on oz.org? Dec 02 11:55:49 hi Dec 02 11:56:31 koen: looks like Dec 02 11:56:39 hi Bernardo Dec 02 11:56:51 koen: ipaq hx2750 is not supported by familiar? Dec 02 11:59:08 hrw: I have no idea on how to build an image for it Dec 02 11:59:48 and no-one has asked for inclusion Dec 02 12:00:12 http://handhelds.org/hypermail/familiar/295/29508.html Dec 02 12:00:48 koen: I saw that - thats why I was asking. Dec 02 12:00:59 ipaq-pxa270 machine with linux-openzaurus kernel? Dec 02 12:01:39 * koen really doesn't know Dec 02 12:03:19 probably the only one is RP Dec 02 12:03:36 * koen is off to Harry potter Dec 02 12:03:42 later all Dec 02 12:03:46 mmmh harry potter Dec 02 12:03:48 cu koen|cinema Dec 02 12:03:55 have fun koen Dec 02 12:03:57 * hrw watched it in Wednesday Dec 02 12:04:05 i had no chance yet Dec 02 12:04:05 that movie is trailer to book ;) Dec 02 12:04:07 hopefully next week Dec 02 12:04:12 heh, same here Dec 02 12:04:55 heh.. today one guy ended work here.. he had 6th book in english.. Dec 02 13:13:40 hi kergoth Dec 02 13:17:13 is there some way to get the screen resolution out of tslib? Dec 02 13:17:28 Hi pb_ Dec 02 13:17:55 hi ue Dec 02 13:18:03 Mardy: no; tslib doesn't know the resolution Dec 02 13:18:03 tslib gives me coordinates in the rectangle 640x320 (for my zaurus), Dec 02 13:18:25 and I need to convert these coordinates to tty characters (80x25) Dec 02 13:19:24 or maybe some way to get the console character size? Dec 02 13:19:37 I might hard encode that to 8x16, maybe Dec 02 13:20:39 does FBIOGET_VSCREENINFO work, if I have the Z. in text mode? Dec 02 13:24:15 well, let's try Dec 02 13:52:58 CosmicPenguin: did I just email you ? Dec 02 13:53:07 yep - hey.. :) Dec 02 13:53:19 CosmicPenguin, cool Dec 02 13:53:25 I've been reading the patch - looks very cool Dec 02 13:53:31 small world Dec 02 13:53:52 People who are interested in this sort of stuff tend to clump together Dec 02 13:54:02 CosmicPenguin, Have you thought about audio yet on the 1200 ? Dec 02 13:54:42 CosmicPenguin, I had noticed that the ac97 stuff was different from the older alchemys Dec 02 13:54:44 We have a validated ALSA driver sitting in my queue to get integrated and pushed upstream Dec 02 13:54:55 Excellent. Dec 02 13:55:21 It would be cool if we could add it into asoc Dec 02 13:55:37 i would like to do that Dec 02 13:56:58 On the Au1200 we actually have both AC97 and I2S Dec 02 13:57:05 have you done anything with I2S yet? Dec 02 13:57:10 CosmicPenguin: Richard and I almost have a stable API, Dec 02 13:57:25 CosmicPenguin, Yes, it supports I2s, pcm and ac97 Dec 02 13:57:57 all that would need done for au1200 is a dma driver and interface driver for i2s and ac97 Dec 02 13:58:06 excellent Dec 02 13:58:21 the DMA driver is the easy part Dec 02 13:58:25 * CosmicPenguin hugs the Au1200 DDMA Dec 02 13:58:37 it's best to wait until nearer xmas as things are moving quite quick atm Dec 02 13:58:56 We would also need to add a codec driver for the AC97 codec, which is a SigmaTel STAC9752T Dec 02 13:59:12 But the I2S codec is a WM8731, so yay for code reuse Dec 02 13:59:52 np, it should probably already work for STAC9752T as long as it's already supported under ac97_codec.c Dec 02 13:59:58 cool Dec 02 14:00:11 WM8731 even has dynamic power managment :) Dec 02 14:00:18 thats double plus good Dec 02 14:00:27 When do you think you'll be ready to go upstream? Dec 02 14:00:33 Richard is using the 31 driver in the corgi iirc Dec 02 14:01:31 CosmicPenguin: we are looking at submitting the core, dpm, pxa platform and some codec driver to alsa-kernel by xmas Dec 02 14:01:51 so it's possible it could be in mainline by Feb Dec 02 14:02:23 I'd be able to start hacking a pb1500 platform layer after xmas Dec 02 14:03:03 I'll start playing with a platform layer for the Au1200 (which would work on the 1550 too, since they share dma engines) Dec 02 14:03:26 And if that all works out, I'll dump our current driver and move to this one Dec 02 14:04:14 the platform stuff is mostly stable, pls have a look at pxa2xx-pcm.c (dma) and pxa2xx-ac97.c and pxa2xx-i2s.c (interface) for a templates Dec 02 14:05:24 you should be able to cut and paste your existing driver quite easily into a platform layer, if I've done my job ok ;) Dec 02 14:06:39 cool Dec 02 14:07:24 CosmicPenguin: I'll catch up tomorrow and next week, my wife is now calling me away to clean some dishes :( Dec 02 14:07:31 excellent Dec 02 14:07:33 thanks for the heads up Dec 02 14:33:52 NOTE: package module-init-tools-3.2-pre9-r0: task do_fetch: failed Dec 02 14:34:15 this is from org.openembedded.oz354fam083 Dec 02 14:34:49 its having difficulting finding "module-init-tools-3.2-pre9.tar.bz2" Dec 02 14:40:18 jt_dnb: this is now fixed, you should update your repository Dec 02 14:49:22 Will updating my repository require me to recompile everything from scratch :S? Dec 02 14:53:42 It shouldn't Dec 02 14:53:49 but even if it does, at least this bug will be fixed Dec 02 14:53:56 jt_dnb: It shouldn't require you to have to rebuild anything that isn't included in the update and isn't required by your configuration. Dec 02 14:58:12 ok thanks Dec 02 15:17:59 ~seen lrg|home Dec 02 15:18:09 lrg|home