**** BEGIN LOGGING AT Wed Aug 16 03:00:01 2017 Aug 16 03:04:29 meta-darwin is sort of incomplete Aug 16 03:04:40 so practical dont have high hopes there Aug 16 03:05:04 although I would say we have docker now a days so dont care Aug 16 03:41:13 bluelightning: I think the last time I did OE stuff was ~2005 :-P Aug 16 03:50:40 mithro: indeed... I remember your name from the handhelds.org times though I'm not sure if our paths crossed Aug 16 03:52:21 A couple of years ago (early 2015), RP poked me about some copy on write code in bitbake that I wrote back then. Aug 16 08:00:48 mithro: hi, we're still using it :) Aug 16 08:01:48 RP: you have an email from me :-P Aug 16 08:14:52 mithro: I see it, I will reply later on Aug 16 08:51:21 mithro: heya! long time no see. how are things? Aug 16 09:14:51 mickeyl: Hey! Yeah long time! Things are super busy, I've been working at Google almost 10 years now - I'm currently a Software Engineer working on Chrome, in my spare time I create FPGA FOSS hardware to record and live stream conferences Aug 16 09:15:36 mickeyl: What have you been up to? Aug 16 09:18:10 mithro: sounds good (especially the FPGA stuff :-). After OpenMoko went down, I've been dragged into iOS contract work – which I mostly like – and the occasional server-side stuff here and there. Been looking to do something with OE for fun again, but couldn't decide on a board to play with yet. Aug 16 09:18:48 mickeyl: Come play with FPGAs with me :-P - https://hdmi2usb.tv Aug 16 09:20:08 mithro: wow, that sounds pretty interesting. You're using Python anywhere in the stack? Aug 16 09:20:32 mickeyl: Anywhere -- everywhere is the goal :-P Aug 16 09:20:36 hehe Aug 16 09:20:38 excellent Aug 16 09:20:39 mickeyl: upy-fpga.github.io Aug 16 09:21:07 mickeyl: We use a Python based HDL for creating the FPGA gateware Aug 16 09:21:29 mickeyl: https://m-labs.hk/gateware.html Aug 16 09:22:30 One exciting thing is that I spoke at 33C3 - https://www.youtube.com/watch?v=yqYBcZzMPGQ Aug 16 09:23:24 cool stuff indeed. Last time I was to at a congress was 6 years ago. Been meaning to go again, but it always conflicts with my birthday (which happens to be on the 26th of December...) Aug 16 09:23:57 It's terrible to get to from Australia Aug 16 09:29:21 i can imagine. 20h flight or what? Aug 16 09:31:18 Between ~24-28 hours in the air depending on the route Aug 16 09:35:43 mickeyl: I gave a talk at PyCon AU on how we are using Python with our FPGAs -> https://www.youtube.com/watch?v=zdLiI-XLTcI -- it was recorded using the very hardware I'm talking about :-P Aug 16 09:39:10 mithro: thanks, will bookmark adn take a look at that. I still love Python, in particular since they introduced the 1st class async/await support. Something I have wanted in Python ever since I've seen it introduced in Vala. Aug 16 09:39:35 I love Python so much I started PyCon AU :-P Aug 16 09:39:42 :D Aug 16 09:41:51 mithro: neat, maybe one day I will be able to attend now that I am just over the pond again Aug 16 09:42:03 bluelightning: Your in NZ? Aug 16 09:42:12 mithro: I am indeed Aug 16 09:42:24 Ahh cool, Kiwi PyCon is pretty cool Aug 16 09:42:39 * bluelightning did not even know there was such a thing Aug 16 09:43:09 https://python.nz/ - Location: Auckland, New Zealand Date: 2/3 December Aug 16 09:43:30 well, it's in my home city so really no excuse not to attend :D Aug 16 09:54:57 bluelightning: Fair enough Aug 16 09:56:05 I'm almost tempted to submit a talk proposal Aug 16 09:56:17 would be on bitbake (or OE) of course Aug 16 11:20:19 RP: btw - I assume there was probably a better place to send that email, happy for you to forward it to other people if you think they can give good advice... Aug 16 11:43:31 mithro: the yocto mailing list would perhaps be a good place but I will reply. I might add Paul in on that as he knows this area well too Aug 16 11:43:52 mithro: just struggling to fight fires this morning, I'll get there! Aug 16 11:43:54 Thanks! No huge hurry Aug 16 11:44:42 Does anyone remember what to add to a Beagle C3 image in order to use the DSP? Aug 16 12:35:40 Hi! I'd like to use xmlrpc-c library in Yocto, but this recipe is missing from meta-openembedded and is available in openembedded repo only Aug 16 12:36:13 http://cgit.openembedded.org/openembedded/tree/recipes/xmlrpc-c?h=master Aug 16 12:36:35 are there any plans to move it to meta-openembedded? Aug 16 12:45:01 yegorich1: once somebody forward ports and submits it, there are plans of including it :) Aug 16 12:45:20 less than a month ago even, so ... :) Aug 16 12:45:41 err. someone sent a patch* less than a month ago, even :) Aug 16 12:45:43 LetoThe2nd: OK. I'll take a look at this Aug 16 12:47:30 bencoh: http://lists.openembedded.org/pipermail/openembedded-devel/2017-July/113760.html Aug 16 12:50:36 yegorich1: bencoh: http://git.openembedded.org/meta-openembedded/tree/meta-oe/recipes-devtools/xmlrpc-c Aug 16 12:54:54 is there a way to select two recipes and have printed out / displayed how they are linked? I sometimes want this when I notice that recipe A inexplicably has a dependency on recipe B Aug 16 12:55:34 depexp only allows for one recipe name Aug 16 13:22:22 dv_: not that I know of. Usually for that I grep one of the recipe names in the .dot files Aug 16 13:42:15 FAIL. co-worker got eclipse glasses for free from the Library Aug 16 13:44:48 dv_: openembedded-core/scripts/contrib/graph-tool find-paths package-depends.dot Aug 16 13:47:59 georgem: neat Aug 16 13:59:48 anyone run into an sdk not asking for install path when run on ubuntu with dash Aug 16 13:59:58 and this is with jethro era stuff Aug 16 14:00:23 georgem, any bets on his eyes working Monday afternoon Aug 16 15:46:07 why is patchwork complaing about patches I sent yesterday? Aug 16 15:46:13 did something get re-run or unstuck? Aug 16 16:19:29 grr, we have a version sync issue with pyqt5 and sip in pyro, need version bump in pyro Aug 16 18:17:07 kergoth: I sent a patch to oe-core and included you on Cc Aug 16 18:17:46 kergoth: we use same update-alternatives replacement used by Mentor, here at OS and we found it is broken with this change Aug 16 18:17:55 kergoth: did you notice it? Aug 16 18:26:11 if I build meta-toolchain with MACHINE=A, then again with MACHINE=B (with all other config the same), should the nativesdk-* recipes be rebuilt the second time? Aug 16 18:26:30 I kind of expect them not to, but they do seem to Aug 16 18:58:52 it depends Aug 16 18:59:15 depends on the differences of the machines Aug 16 19:07:41 otavio: yeah, there must be some config difference that I'm not seeing that's affecting nativesdk. I'll tinker with bitbake-diffsigs a bit Aug 16 19:17:56 smurray: IIRC it includes the kernel headesr too Aug 16 19:18:19 otavio: hrm, not just linux-libc-headers? Aug 16 22:42:00 nativesdk should remain same Aug 16 22:42:11 unless you change SDK host Aug 16 23:02:27 khem: I'm thinking it may be a side-effect of one of the machines having multilib and the other not Aug 16 23:03:01 I need to poke around with bitbake-diffsigs some more Aug 16 23:24:45 hmmm, yes that could be, but I think it should not affect native or nativesdk case. Aug 16 23:25:00 so most probably over depends Aug 16 23:58:24 it does seem to be multilib AFAICT. If I include multilib.conf in the 32-bit platform and set MULTILIBS = "", that makes the signature diffs go away Aug 16 23:59:06 kind of looks like it's the prefixing of MLPREFIX on the *SYSROOT variables that multilib.conf does Aug 17 00:00:16 I'm doing some build tests now to see if it does keep those from being rebuilt Aug 17 01:07:57 khem: hmm, looking like that wasn't all of it, definitely need to diffsig some other recipes **** ENDING LOGGING AT Thu Aug 17 03:00:00 2017