#qi-hardware IRC log for Tuesday, 2013-11-19

apeleteGood morning everyone08:44
apeletelarsc: thinking about submitting the fix for hnp upstream, so that review can start asap09:21
apeletedo you think that's a good idea or not ?09:22
larscprobably a good idea09:27
apeletelarsc: any advice on how to proceed ?09:28
larscgit format-patch; git send-email ;)09:29
apeletelarsc: nice, thanks :-)09:31
apeletewill let you know before doing it, probably tonight after work09:32
apeletelarsc: what kernel version and tree should the fix be rebased on ?10:04
apeletelatest 3.13-rcX ? linus' tree or someone else's (felipe balbi maybe) ?10:04
larscthe later10:18
larsccheck the MAINTAINERS file10:18
larscthere is probably a git tree for USB10:18
apeleteyeah, felipe is the maintainer for musb, and he's got a tree for usb10:20
apeletelarsc: will mirror his tree and rebase the fix on it, thanks10:21
DocScrutinizer05what's the problem with HNP?14:39
DocScrutinizer05(not that in real life HNP is *ever* used. But I seen it working when we experimented with H-E-N prototypes)14:41
DocScrutinizer05larsc: ^^^14:42
larscthat the board doesn't have HNP support14:44
larscbut the linux kernel assumes it has14:44
DocScrutinizer05oh well14:44
larsceasy to fix14:44
DocScrutinizer05makes sense to kick it out then14:44
DocScrutinizer05nobody needs HNP14:44
DocScrutinizer05at least I hope no silly printer is using HNP to switch the camera from USB-A to mass-storage to download the pictures to print14:46
DocScrutinizer05it's a brainfart concept14:46
DocScrutinizer05you'll need it when you want a USB-OTG stamp from USBcert14:48
DocScrutinizer05but OTG is overengineered anyway14:50
wpwrakHNP is what happens when you let the intern design your OTG:14:50
wpwrakthe intern thinks about it for a moment, realizes that host/device role could be communicated simply through the ID pin, and drafts that up. hands the draft to the experts on the same day.14:51
DocScrutinizer05HNP is about overriding the ID assignment14:52
wpwrakthey look at it and are HORRIFIED. where's all the complexity ? what is there to make their jobs look important ? what to make the certificate a difficult and therefore worthwhile goal ?14:52
wpwrakthen they go in to design HNP. and the world is whole again :)14:53
DocScrutinizer05HNP is defined as swapping A/B roles as defined by ID pin14:53
wpwrakyes, i know :)14:53
wpwrakat least that's what it says14:54
DocScrutinizer05without ID there's no HNP really14:54
DocScrutinizer05when you use a USB hardware that doesn't care about ID then it's not OTG and thus doesn't need to bother about HNP14:54
wpwrakit's true purpose is to add the complexity that was forgotten in that silly ID pin concept. you have to uphold the brand reputation, don't you ? USB - certified complexity14:55
wpwraks/it's/its/ # damn14:55
apeletelarsc: hi19:40
apeletelarsc: I just rebased the hnp patch on top of felipe balbi's master branch19:43
apeletebeen wondering, how to test it since there is no qi_lb60_defconfig file upstream ?19:44
apeletewhat config file should I use to do a test build before submitting ?19:45
larscI think you can just copy the qi_lb60_defconfig19:46
apeleteok, that's what I did actually, and it builds, but how is the kernel community going to test it ?19:47
larscthey'll have to trust you19:48
larscYou can test the patch without the hardware anyway19:48
apeleteok, fine :-)19:49
larscthat makes more sense19:50
apeleteI will try it on the nanonote on my side before submitting it19:50
apeleteyeah, just thought about asking you before doing something stupid ;-)19:52
wpwraks/it's/its/ # damn20:45
apeletelarsc: what do you think about it ? -> http://paste.debian.net/66683/ http://paste.debian.net/66684/21:43
apeleteshouldn't you be in Cc: ?21:43
DocScrutinizer05roh: ping23:24
DocScrutinizer05roh: CCC?23:24
--- Wed Nov 20 201300:00

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