=== zyga is now known as zyga-afk [09:26] I have some boards that runs debian and some of them runs ubuntu. How can I understand which one is ubuntu and which one is debian? uname -a shows "Linux omap 3.2.0-psp7 #1 Fri Apr 13 04:55:05 UTC 2012 armv7l armv7l armv7l GNU/Linux" [09:27] angs: on Ubuntu, "lsb_release -a" will tell you. I don't know if there's a better way. [09:27] thanks a lot, it shows it is ubuntu:) === zyga-afk is now known as zyga === doko_ is now known as doko [11:15] ogra_: hey [11:16] ogra_: gpio library is not available for Ubuntu arm ? :o [11:18] perhaps it is the sysfs that is advertised for usage nowadays. [13:13] lilstevie, do you know if there are git trees of tf101 kernels with better history and changes rationale than what ASUS' tarballs provide? [13:13] janimo, yes/no [13:13] lilstevie, please expand on the yes branch :) [13:14] * janimo figures this is an interactive adventure [13:14] janimo, we can match to where they most likely branched from nvidias tree [13:14] with their history [13:14] but asus themselves strip *ALL* history before releasing it [13:15] lilstevie, that is what I was going to check, seeing only it is a 2.6.39.4 [13:15] but compared to upstream pristine linux-stable 2.6.39.4 there [13:15] is a 15Mb patch :) [13:15] yeah, well matches with nvidias android tree [13:15] look at androids 2.6.39.4 [13:15] it is a pain though [13:15] cause there are a few tags [13:16] RaYmAn has a script which diffs each tag against the asus release and suggests the closest match [13:17] we have done it to the tf201, tf101 is next on the list to be moved over to nvtegra [14:17] lilstevie, by moved over to nvtegra and 'we' what do you mean? Kernels used by the mod community? [14:44] androidroot team, me bumble-bee RaYmAn kmdm and IEF [14:44] janimo, ^^ [14:47] lilstevie, ok [16:34] I cant believe theres no tutorials for running ubuntu on nookcolor besides that silly chroot nonsense. [16:39] hehe === zyga is now known as zyga-afk [17:20] xase_: I was going to work on that for Oneiric, but was swamped with the arm server bringup. Sadly, I have since moved on to a differnt job (and subsequentially lost time/interest). Sorry. [17:51] GrueMaster: no problem do you have an archive of your work available? [17:53] No. I didn't get too far into it. I barely spent a week on it, mainly going through the Nook source and seeing what I needed to implement. [17:56] damn [19:17] marvin24, are you working more on nvec until it gets out of staging? [19:21] janimo: yep [19:22] marvin24, do you have the docs from nvidia? [19:22] yes [19:22] the README in the tree suggests there are no docs at all [19:22] oh, that deserves a patch ;-) [19:23] in fact, I "appeared" some month ago [19:23] ok, just making sure you're not missing out :) I just discovered they're public last week, and downloaded them today [19:23] together with the tegra2/3 documentation [19:23] yes, Nov 11 from what I saw [19:23] janimo: public? [19:23] ah tegra3 as well? I thought that's still WIP [19:23] you need to register first ... [19:24] well, public with annoyance, but not private/NDA etc :) [19:24] mmh, I think there is a NDA at least, but I didn't read it to the end [19:25] marvin24, so this means the EC doc that was floating around of the chip used in the ac100 is not necessary, as you only need the tegra2 side/protocol [19:25] ah did not read either :) [19:26] mmh, I don't know of any EC doc floating around ... wait [19:27] ah, found it [19:31] of course, for this has no NDA ;-) [19:31] normally, you would have to register here http://developer.nvidia.com/tegra-2-technical-reference-manual [19:42] yes, I registered and downloaded this and the TRM today [19:42] there was an EC doc for an embedded keyboard controller in the ac100 last year [19:42] which I was told was the only doc that helped the nvec initial development [19:42] maybe that was the original source [19:43] it had some documentation inside [19:46] janimo: can you try to get linux-backports-modules-cw-3.3-precise-generic building on ARM ? [19:46] is that a package? [19:46] eh, no, the binary [19:46] wait ... [19:46] found it [19:47] but it's for the 3.2.0 precise kernel [19:47] linux-meta is the source :-( [19:47] should be buildable on any kernel [19:47] not even amd64 is enabled [19:48] I have it here on my amd64 ... [19:48] https://launchpad.net/ubuntu/precise/amd64/linux-backports-modules-cw-3.3-precise-generic [19:48] ok, so Architecture: is misleading [19:49] linux-backports-modules-3.2.0 the package [19:49] marvin24, which drivers are needed on arm? [19:50] I don't know, but paz00 needs rt2xx [19:50] I think they can also be useful for panda whatever [19:52] the package contains all wifi drivers [19:54] btw, I still can only get a 3.1 kernel booting with "console=ttyS0,115200n8 quiet" enabled [19:54] this ensures nothing it printed to the console on boot [19:55] so not yet ready for uploading I guess? Any chance of newer kernels being better? [19:55] how much of what's working now is not yet ported to 3.5? [20:13] everything gpu and power save related is missing [20:13] (which is too important to be skipped) [20:13] marvin24, are the changes in 3.5 too big so that code need significant rewrite? [20:14] I see nvidia devs steadily sending patches upstream [20:14] yeah, still unportable I think [20:14] there are lots of nv engineers already working on it [20:14] but it takes time [20:15] somehow nv decided to rewrite everything at least three times [20:15] this is the price you have to pay to get a product quickly to market [20:16] you just cannot do it in a sane way [20:16] and mainlining is only low priority compared to that [20:17] I don't want to know how many developers worked on porting win8 to tegra3 [20:18] and even less I want to see that code [20:18] :) [20:19] in that case they may not rewrite at all as there's no mainlining required ever, and win8 will be around for many years [20:19] maybe it's only the FOSS model that gives grief to developers. Or at least this type of grief [20:21] I would be interested in some kind of statistics how much developer month it takes to port some arch to linux and windows [20:21] and how much time it consumes to maintain it [20:21] during product lifetime === cmagina-lunch is now known as cmagina === heathkid|2 is now known as heathkid [23:31] marvin24, tbh because of UEFI, it would be more the UEFI code that you would want to see less :p