#qi-hardware IRC log for Sunday, 2013-03-10

viricwpwrak: referendum in the malvines today, no?09:35
wpwrakviric: yeah. i'm sure it'll be a close call ;-)12:16
viric'close call'?12:16
wpwrakthe outcome uncertain until the last moment12:18
viricaha ok12:42
viricha, these british let territories do referendums12:42
viricunbelivable under the Spanish crown :)12:43
wpwrakwell, if the underlings were quick enough, they may have had their little secessionist cabal before the armada arrived to crush it :)12:55
viricbritish armada?13:22
wpwraknaw, the spanish armada. the one that helped to make such things unthinkable13:29
wpwrakof course, the british navy had a similar role13:30
viricyes13:58
wpwrakof course, the whole falklands/malvinas thing is mere theater. and argentina's posturing isn't helping its supposed case.14:11
larscis there a referendum whether they want to join argentina or stay with the empire?14:12
larscwikipedia already referes it in the past tense: "The Referendum was held on 10 March 2013."14:14
wpwraki think for now it's just whether they want to stay with the empire14:16
wpwrakregarding the date, i've read that it's today and tomorrow. lemme check ...14:17
rohwpwrak: i never understood some conflicts... or is there oil on falkland?14:18
biotthere has been some oil found, but that's not the issue.14:19
biotthe current (and past) noise made by the argentine government is intended to distract from severe domestic problems14:19
wpwrakyeah. today and tomorrow. results are expected tomorrow night (local time)14:19
rohi see.. so its prehistroric brainproblems of politicians14:19
wpwrakbiot: what's new, eh ? :)14:19
biotnah, not even nationalism14:19
biotwpwrak: sad but true14:19
biotargentina's economy is in a bad state, and the country will again default on its debt14:20
biotthe noise over the malvinas is just that, they won't do anything about it14:21
wpwrakroh: there's supposedly quite a bit of oil, but not too easy to retrieve14:21
wpwrakbiot: we'll see about the default. at the moment, it seems that there won't be an overly harsh ruling in new york.14:22
biotwpwrak: meh, that's just one settlement over past bad debt, not a big deal in the grand scheme of things.14:22
wpwrakbiot: not that i'd expect a default to make much of a difference. the government is already doing all it can to discourage investment.14:23
biotyeah, it's seriously horrible mismanagement14:23
biotthere's a sometime contributor I know in argentina, and it's nearly impossible for him to import electronics now, even really cheap things from china14:24
biotno foreign currency, you know14:24
biotthe only way they can get gear at his university is through the *military*14:24
wpwrakthat, and also simply scary things. like restrictions on acquiring foreign currencies.14:24
biotwrap your head around that :-(14:24
wpwrakyeah, that issue :)14:24
wpwrakimport should still be fine if you avoid a few traps14:25
wpwrakbut you either need a credit card that works abroad (in theory most do) or directly have money outside the country14:26
wpwrak(traps) like connecting to mains. a big no-no.14:27
bioteh?14:27
wpwrakalso, declared value should be below USD 100014:27
biotwhat do you mean, connecting to mains?14:28
wpwrakanything that connects to mains needs to be certified. argentina has its own certificate, not shared with any other country. only very few products are certified for use in argentina. so if they spot that customs, you drop out of the normal customs process.14:28
wpwrakconnecting to 220 V14:28
biotoh damn, sounds terrible14:28
wpwrakso no power cable, no wall wart, etc.14:28
biotright14:29
biotit's stuff like this...14:29
wpwrakwhat you can do it ask the sender to remove any power supplies before shipping. they usually can do this.14:29
wpwrakyou can also try to ship the power supply separately. sometimes that passes. and if it doesn't, you just scrap the shipment.14:30
wpwrakbecause, if it gets red-flagged, it stops at customs. for courier, that's at ezeiza (the international airport of buenos aires). then you have to go there, no matter where in the country you live. there, they will tell you to get a customs broker to resolve the issue.14:31
wpwrakthings you can't do (according to the customs officials) are: 1) just remove the offending bit and accept the rest of the shipment. 2) return the whole package to the sender.14:32
biothaha14:32
rohi just was told from somebody working at a us chemical company that they also opened up a factory in china because of the shipping issues to europe14:32
rohits easier to recieve stuff from china in europe than from the us.14:33
rohso its not only argentinia doing more than stupid things14:33
wpwraknow you have three options: 1) decide it's now worth it, pay the storage fees and abandon the shipment. 2) get your device certified, which involves at least three additional parties, should cost a handsome amount of money, will take time, etc. 3) have the customs broker negotiate a bribe.14:34
wpwrakroh: nice paradigm shift :)14:34
biotwpwrak: I'm guessing 3) is the commonly used option14:34
wpwrakyup. that's the most reliable and efficient way to resolve such issues.14:35
rohwpwrak: ack. actually its funny.. i only ever recieved one package from china declared correctly.. and that was from wolfgang *g*14:35
wpwraksupposedly, just not having a wall plug in the box is sufficient to avoid trouble. e.g., if you have a device that connects directly to mains via a cable, remove the cable. but i haven't tested that hypothesis yet.14:36
rohall others were marked gift on the CN22 form14:36
rohthere are even packagings which have the CN22 printed on, with the tick in the 'gift' field already marked14:36
biotwpwrak: the implication is that they open EVERY package to check, which is in itself pretty surreal14:36
whitequarkchina is a very charitable country14:36
whitequarkdidn't you know that?14:36
wpwrakroh: this may be a good thing. if everything from china has bad paperwork, customs simply can't make a fuss about it14:36
rohwpwrak: i think so ;)14:37
wpwrakbiot: oh, they certainly do14:38
wpwrakat argenting customs, you can also run into "soft" problems. things you can resolve with the only trouble being that you have to go to the airport. e.g., if the paperwork is incomplete or if they suspect you're trying to sneak something illegal past them. e.g. they once suspected smartphone parts to be some contraption to "blank" stolen phones.14:42
wpwrak(as if i really had to import anything if i wanted to do that ;-)14:42
biothaha14:42
biotthey fancy themselves pretty smart huh14:42
wpwrakwell, there's probably a few thousand phone "recyclers" for each hardware developer in the country, so i guess a false positive every now and then shouldn't come as a surprise14:45
lkclhi folks has anyone tracked down a reference design set of schematics and PCB files for the jz4760?19:06
larscdoesn't ingenic have them on their website?20:13
lkcllarsc: no.  they have a PDF.  that's not a set of schematics in ORCAD format, and it's not a 6-layer PCB file in e.g. PADS format.20:18
lkclfreescale commissioned boundarydevices.com to produce a reference design with the iMX6 - full schematics including PADS files were released.20:19
lkclTI always release full schematics and PADS files20:19
larscwrite them a mail20:19
lkclsamsung not so much20:19
lkcli did.  i was just wondering if anyone else had already done so20:19
wpwrakwow. ingenic are releasing data sheets again ? wow. and there's even the 4740. alas, no PM.20:38
wpwrak(unless it's hiding in a tar)20:39
larscdid they depublish them again?20:42
rohwpwrak: well.. its over for interresting anyhow20:51
rohwpwrak: A10 has won that market already.20:51
wpwraklarsc: in march 2011, they had everything on ftp for a while. when wolfgang congratulated them, they hurriedly pulled them. don't know when they put things back.20:52
wpwrakroh: even though there seems to be even less material on the A10. well, maybe in chinese, but it's kinda suspicious that even sites that claim to have all there is to know about the A10 have only a list of registers definitions fished out of the kernel source.20:53
lkclPM... as in PMIC?  i just found the ACT860020:53
lkclwpwrak: not even most employees in allwinner know everything there is to know about the A10 :)20:54
rohwpwrak: i prefer registermaps and workin driver over broken documentation every day.20:54
lkclit even has VGA in for goodness sake!20:54
whitequark#arm-netbook might know more about A10...20:54
wpwraklkcl: PM as in Programming Manual. the book that describes the registers, the fileds in them, the values, and with a little luck also the semantics20:54
rohwpwrak: and its atleast not 'prehistoric' when it comes to performance.20:55
lkclroh: likewise.  you've uh... seen the size of the iMX6 reference docs?  fiive thouuusand pages...20:55
wpwraklkcl: you mean if you don't know how it works, you're more confident that you'll get it to work ? (-:C20:55
whitequarkexynos 4210 is 11k or so...20:55
whitequarkand that isn't even a complete doc. and that doesn't include the ARM reference20:56
rohlkcl: hehe... freescale documentation is usually nice. also their drivers were (dunno about now) well written.. some by i think t. gleixner?20:56
rohthe problem with freescale is price. their chips are much too expensive for a not-extremely-high-volume consumer electronics product.20:56
whitequarklkcl: I recall that samsung's partner/subsidiary (?) hardkernel has had great troubles prying documentation on the devkits from samsung20:57
rohwithin china there were usually copied documentation and even develboards from samsung20:58
whitequarkand they're very obviously linked. one of odroids is almost sgs2 (it even looks like one). hardkernel's website is referenced by samsung under "devkits" or so20:58
larscpengutronix did imx, tlgx is linutronix20:58
rohlarsc: ah. thanks for clearing that up. both do good work from by pov20:58
wpwrakroh: nice drivers are boring. i kinda liked the "let's roll up the sleeves, put on the boots, take out the pitch forks and get to work" feeling you got with, say, some samsung stuff20:58
lkclwhitequark: that sounds about right.  i have one of the original odroid hand-helds - yyep, kernel source: nnope on docs :)20:59
wpwraks/pitch forks/pitchforks/20:59
hellekinwpwrak: so you prefer broken stuff?21:12
wpwrakstuff that can be salvaged :)21:13
wpwrakwell, i have to admit that the biggest part of the joy i felt was that the driver didn't look like a total disaster. just something that needed a bit of cleanup and it could actually go upstream.21:14
hellekinthat was not a driver, wpwrak, but a salmon :)21:15
wpwrak*scratch,scratch*21:17
whitequarkthat's a surprisingly good analogy for the hoops you need to jump through (sic) in order to get your code accepted21:18
Action: larsc never jumped through any hoops21:21
Action: whitequark has had three attempts to merge trivial patches in LLVM. no replies.21:22
larscmaybe llvm doesn't have an andrew morten21:36
lindi-hmm, all of my patches have been merged I think21:37
wpwrakhoops ? i once even had a patch torn from my hands by linus himself :)21:41
wpwrak(okay, that was admittedly a good while ago :)21:41
rohhm. i also never have seen proper patches dropped that way. if they were addressed correctly21:42
whitequarkI've pinged them and asked on IRC if they were written/submitted correctly21:43
whitequarkno idea :/21:43
whitequarkone of them merely replaces bool with LLVMBool21:44
whitequark(was a typo)21:44
rohwell.. all my 3-liners were accepted by the subsys maintainers back then21:44
larscwpwrak: Back in your days the only prerequisite to get a patch accepted was that the code worked most of the time ;)21:54
wpwrakor that it looked at least as if it might :)21:55
larscthese days patches don't get accept if there is a newline too much or one is missing21:59
larscs/accept/accepted/21:59
whitequarkok, a ping on IRC succeeded this time!22:00
wpwrakyeah, the level of pickiness is quite high. well above what's necessary for most purposes.22:00
viricsometimes pickiness is not on the proper topics, either23:02
wpwrakkewl. my keyboard has sun-like "option" buttons that - according to xev - translate to "Muhenkan" and "Henkan_Mode", whatever these may be. life is full of surprises :)23:23
--- Mon Mar 11 201300:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!