**** BEGIN LOGGING AT Wed Nov 14 03:00:00 2012
Nov 14 03:38:29 <Herrie|ZzzZZzzzz> Hi all
Nov 14 03:38:41 <Herrie|ZzzZZzzzz> Ping morphis, JaMa
Nov 14 03:46:06 <Herrie|ZzzZZzzzz> Looks like my Tuna build worked OK
Nov 14 03:46:22 <Herrie|ZzzZZzzzz> NOTE: Tasks Summary: Attempted 4523 tasks of which 4269 didn't need to be rerun and all succeeded.  Summary: There were 5 WARNING messages shown.
Nov 14 03:46:41 <Herrie|ZzzZZzzzz> So not sure what's causing the issue for hpveer and palmpre2 ?
Nov 14 03:49:23 <Herrie|ZzzZZzzzz> retrying palmpre2 now to see if it starts working now all of the sudden after tuna was OK
Nov 14 03:55:22 <Herrie|ZzzZZzzzz> No luck, still "EGL/egl.h: No such file or directory" and "cp: cannot stat `include/generated/bounds.h': No such file or directory"
Nov 14 03:59:25 <Herrie|ZzzZZzzzz> Any help is welcome, since it's my first time toying around with OE :) Just need some help getting it properly setup and it should be OK I guess
Nov 14 04:03:02 <Roy__> Herrie|ZzzZZzzzz: I haven't seen that before but I think ka6sox-away said that others had.  You might try checking out a clean environment and following the steps again.
Nov 14 05:39:06 <morphis> Herrie|ZzzZZzzzz: as I already sad: you have no implementation for virtual/egl or virtual/gles for palmpre/palmpre2/hpveer and as qt is building with opengl support as default the error message is valid
Nov 14 05:53:14 <Herrie> morphis: How to solve this, I'm a n00b at OE ;)
Nov 14 05:53:20 <Herrie> So any help would be appreciated :)
Nov 14 05:54:06 <morphis> Herrie: you need a gles/egl implementation for this
Nov 14 05:54:22 <morphis> for example inside the hpveer is an adreno graphics chip
Nov 14 05:54:29 <morphis> so you need the binaries + headers for it
Nov 14 05:55:21 <morphis> Herrie: should be easier for the palmpre/palmpre2 as there is the SGX sdk available
Nov 14 05:55:56 <Herrie> OK, if you can give me instructions on what to do for the Pre2, I'll work out he Veer myself I guess
Nov 14 05:56:01 <Herrie> Just need some help to get started
Nov 14 05:56:03 <morphis> puuh
Nov 14 05:56:16 <morphis> I can give you only hints
Nov 14 05:56:41 <morphis> search for meta-ti, it has a recipe for omap3-sgx-drivers and libgles-omap3
Nov 14 05:59:07 <morphis> Herrie: it's not that easy to stick just a and b together, you have to do real engineering work here
Nov 14 06:00:09 <Herrie> morphis I know, but need to start somewhere right ;)
Nov 14 06:00:28 <Herrie> OK found meta-ti in Angstrom, would that do?
Nov 14 06:01:06 <morphis> you can try if the drivers compiles with the palmpre/palmpre2 kernel
Nov 14 06:01:50 <morphis> I have to leave
Nov 14 06:01:51 <morphis> bye
Nov 14 06:03:26 <Herrie> OK will try that
Nov 14 06:03:29 <Herrie> Thnx for the hinst
Nov 14 06:18:23 <Herrie> OK seems like I'm getting somewhere
Nov 14 06:18:46 <Herrie> Managed to get OMAP3 libgles and sgx
Nov 14 06:18:52 <Herrie> It's cooking again now :)
Nov 14 06:33:59 <Herrie> Looks like OMAP3 drivers work OK, but still got the problem with "cp: cannot stat `include/generated/bounds.h': No such file or directory" ?
Nov 14 13:03:34 <swisstomcat> ping
Nov 14 14:44:35 <ka6sox-farfarawa> swisstomcat, pong
Nov 14 15:39:47 <morphis> morning
Nov 14 15:40:13 <morphis> ShiftyAxel|Away: ping
Nov 14 16:10:27 <ShiftyAxel> morphis: ping
Nov 14 16:10:29 <ShiftyAxel> er, pong
Nov 14 16:10:31 <ShiftyAxel> you get the idea
Nov 14 16:10:35 <morphis> :)
Nov 14 16:11:05 <morphis> ShiftyAxel: I updated luna-sysmgr with some fixes related to security/kernel-support + merged recent upstream changes in
Nov 14 16:12:09 <ShiftyAxel> excellent :D
Nov 14 16:12:54 <ShiftyAxel> that reminds me- what needs to be done for us to get the rounded corner shaders working again?
Nov 14 16:13:16 <ShiftyAxel> I remember seeing your commit about the Qt GL headers
Nov 14 16:14:49 <morphis> they need some private Qt header files
Nov 14 16:16:02 <ShiftyAxel> proprietary?
Nov 14 16:18:18 <morphis> no
Nov 14 16:18:25 <morphis> just header files qt doesn't install
Nov 14 16:20:09 <ShiftyAxel> hm
Nov 14 16:20:28 <ShiftyAxel> could we stage them? or do we not have access to them at all?
Nov 14 16:20:46 <morphis> we have access, they are part of the qt source
Nov 14 16:21:05 <morphis> yes, we just have to stage them or work around using them (which would be imho the better solution)
Nov 14 16:21:20 <ShiftyAxel> which is where I come in, I suspect :)
Nov 14 16:21:53 <morphis> yes :)
Nov 14 16:36:08 <ShiftyAxel> ah, so this was palm being... somewhat cheeky
Nov 14 16:36:23 <ShiftyAxel> there's a big 'THIS IS NOT PART OF THE API' message at the top of the header...
Nov 14 16:39:34 <GodGinrai> lol
Nov 14 16:40:51 <ShiftyAxel> hum
Nov 14 16:41:19 <ShiftyAxel> looks like i'm going to have to reimplement the required bits inside luna
Nov 14 16:42:32 <ShiftyAxel> ooor modify it to use standard Qt API bits
Nov 14 16:42:34 <ShiftyAxel> that seems more sensible
Nov 14 17:01:23 <ShiftyAxel> morphis: looks like this may well be difficult
Nov 14 17:03:34 <EricBlade> morphis: this is relevant to my interests
Nov 14 17:05:12 <ShiftyAxel> EricBlade: palm used a non-api Qt header that has a license on it
Nov 14 17:05:15 <ShiftyAxel> for the rounded corners
Nov 14 17:05:44 <ShiftyAxel> the right way to do this would be to remove that dependency and reimplement it in a more standard fashion
Nov 14 17:05:58 <ShiftyAxel> however, it looks like that class implements shader effects on painters
Nov 14 17:06:04 <stefan_schmidt_w> wow, because rounded corners are so novel. Who would think about that!! :)
Nov 14 17:06:22 <ShiftyAxel> which isn't in the qt API until 5.0 I believe
Nov 14 17:06:23 <stefan_schmidt_w> SCNR
Nov 14 17:06:52 <ShiftyAxel> YKTSNR
Nov 14 17:07:36 <stefan_schmidt_w> Oh, that one is new to me. What does it stand for?
Nov 14 17:11:45 <ShiftyAxel> you killed the signal to noise ratio :P
Nov 14 17:25:57 <EricBlade> ShiftyAxel: i was thinking more the other thigns he was talking about for "security related"
Nov 14 17:26:11 <ShiftyAxel> ah, fair enough
Nov 14 17:26:16 <EricBlade> "fixes related to security/kernel-support"
Nov 14 17:26:57 <ShiftyAxel> morphis: I say we stage it for now... HP may rewrite it using the new 5.0 bits when they switch over, and I could probably do that if they don't
Nov 14 17:27:09 <EricBlade> although, if you give me a little more of a clue as to what specifically you're talking about in regards to this rounded shader thing, I can have a look into it
Nov 14 17:27:24 <ShiftyAxel> that would be most awesome
Nov 14 17:27:47 <EricBlade> I remember seeing something recently about the rounded shader somewhere, but i'm not sure what i'm looking for
Nov 14 17:28:10 <ShiftyAxel> webOS-ports/luna-sysmgr/morphis/device-support has a commit relating to it
Nov 14 17:28:26 <ShiftyAxel> basically, we're not installing a certain header file that HP is using
Nov 14 17:28:42 <ShiftyAxel> because it's not part of Qt, so doesn't get installed with the OE build
Nov 14 17:29:02 <ShiftyAxel> or rather, it's part of Qt, but not part of the Qt API, it's an implementation reference that happens to work
Nov 14 17:29:18 <EricBlade> ok, yes, this also sounds rather relevant to my interests
Nov 14 17:29:22 <ShiftyAxel> QtOpenGL/qglcustomshaderstage_p.h
Nov 14 17:30:06 <ShiftyAxel> so our options are stage it, reimplement it with Qt4.8, or wait for Qt5
Nov 14 17:33:43 <EricBlade> ok, so, I see that that include file is used in three different spots ...
Nov 14 17:38:26 <EricBlade> Interesting.  Is there a licensing issue involved here?
Nov 14 17:38:52 <ShiftyAxel> possibly, it depends on whether or not the licenses are compatible
Nov 14 17:38:56 <ShiftyAxel> but it's more the non-standardness
Nov 14 17:40:30 <EricBlade> can you point me to the commit that we're talking about?
Nov 14 17:42:04 <ShiftyAxel> https://github.com/webOS-ports/luna-sysmgr/commit/55102ee155059d4c9e8d59e039eed90171de99dd
Nov 14 17:43:17 <EricBlade> hmm. I thought that you had this working at one point?
Nov 14 17:44:11 <ShiftyAxel> afaik it's been disabled on GNex all along
Nov 14 17:44:36 <EricBlade> hm. I coulda swore you demoed it working.  I could also be losing my mind.
Nov 14 17:44:39 <scoutcamper> just for future reference, i tried to git format-patch and git am commits 1 at a time from LunaCE to OWO Luna, there are more errors than i would care to remember. In other words, moving commits from LunaCE to OWO is not going to be as easy as i had hoped....
Nov 14 17:45:05 <ShiftyAxel> EricBlade: you're going loopy :)
Nov 14 17:45:21 <EricBlade> understandable.  I'm not dealing well with old age.
Nov 14 17:45:23 <EricBlade> :D
Nov 14 17:45:29 <ShiftyAxel> there is a rounded corner 'shader' of sorts functioning... but I can't for the life of me figure out how, or from where...
Nov 14 17:46:16 <ShiftyAxel> it takes a few seconds to kick in, if you look carefully you can see that cards are square for a while before suddenly getting a little bit of rounding
Nov 14 17:46:57 <scoutcamper> and a simple git rebase also fails to work....
Nov 14 17:47:17 <EricBlade> the rest of Qt is LGPL isn't it?
Nov 14 17:48:09 * ShiftyAxel knows not
Nov 14 17:48:20 <ShiftyAxel> haaaaang on a second... looks like there's a method to make the painter round corners
Nov 14 17:48:31 <EricBlade> hmm. ok looks like it follows the same licensing as the rest of Qt
Nov 14 17:48:41 <ShiftyAxel> also screw you apple trackpad
Nov 14 17:48:43 <EricBlade> interesting that i'm seeing very different versions of this file online vs what's in my local archive
Nov 14 17:48:55 <EricBlade> different as far as the license block goes
Nov 14 17:49:56 <ShiftyAxel> hm...
Nov 14 17:50:21 <ShiftyAxel> if I can get it to render with QPainterPath, I think the shader stuff can be discarded entirely
Nov 14 17:52:31 <EricBlade> if you'd like to rebuild it, I guess feel free, but if it will work just with adding an include file to the Qt recipe or something, that sounds like the better option, since i'd guess it would work differently in Qt5 anyway
Nov 14 17:53:33 <ShiftyAxel> myself I'm up for staging the file for now, and either waiting on a do-over w/Qt5 or doing it over ourselves when the time comes
Nov 14 17:53:42 <ShiftyAxel> I'll let morphis weigh in on this though :)
Nov 14 17:55:42 <EricBlade> i don't see that there should be a licensing issue, at least the version of the file that's here has the same licensing as the rest of it from what i see
Nov 14 17:57:12 <EricBlade> and worrying over whether the interface changes since it's not officially part of the release seems like a non-issue, unless there's changes in the 4.8.2 to 4.8.3, as otherwise moving Qt versions is quite an undertaking to begin with
Nov 14 18:17:19 <ShiftyAxel> just testing big rounded corners now
Nov 14 18:17:31 <ShiftyAxel> it looks promising, since this can be done easily in Qt
Nov 14 18:17:37 <ShiftyAxel> and have antialiasing also
Nov 14 18:26:25 <ShiftyAxel> woot, we have working corners without the header
Nov 14 18:26:38 <ShiftyAxel> finally, webOS can look webOS-ey again :)
Nov 14 18:33:39 <ShiftyAxel> good lord these looks nice
Nov 14 18:34:08 <ShiftyAxel> https://twitter.com/shiftyaxel/status/268783811888828416
Nov 14 18:36:40 <ShiftyAxel> I'm quite tempted to try moving webOS toward using squircles for rounded corners
Nov 14 18:36:55 <ShiftyAxel> my Pre3's a squircle
Nov 14 18:58:22 <swisstomcat> awesome!
Nov 14 19:00:02 <ShiftyAxel> it looks like these are the 'backup corners' so to speak
Nov 14 19:00:30 <ShiftyAxel> antialiasing isn't working on this qpainter either
Nov 14 19:00:44 <ShiftyAxel> so they're a bit laddery, but it's barely noticeable on the gnex
Nov 14 19:58:40 <Garfonso> love the rounded corners. :)
Nov 14 20:04:46 <ShiftyAxel> they're rather swish, aren't they? :D
Nov 14 20:05:12 <ShiftyAxel> yet another step in the battle for webOS looking like webOS again
Nov 14 21:31:01 <GodGinrai> ShiftyAxel: I'm happy you got the corners working.  No rounded corners was glaring me in the face every time I looked at your screenshots. XD
Nov 14 22:09:51 <bhuey> folks
Nov 14 23:44:02 <PatrickC|Sick> http://www.androidpit.com/apple-patent
Nov 14 23:44:04 <PatrickC|Sick> wow Apple...
Nov 14 23:44:05 <PatrickC|Sick> wow
Nov 15 00:07:27 <linuxjacques> yeah :-\
**** ENDING LOGGING AT Thu Nov 15 02:59:58 2012