**** BEGIN LOGGING AT Mon Nov 16 02:59:56 2009
Nov 16 08:05:28 <mvzaanen> Morning
Nov 16 08:06:02 <u4ia> hi
Nov 16 08:35:12 <serdar__> jancona, there?
Nov 16 09:11:53 <panicking> hi jancona
Nov 16 09:12:02 <panicking> I'm working on camera support on the freerunner
Nov 16 09:12:07 <panicking> and I'm facing a problem
Nov 16 09:12:14 <panicking> it can be a global one problem
Nov 16 09:12:16 <panicking> are you there
Nov 16 09:12:17 <panicking> ?
Nov 16 09:13:20 <mvzaanen> the freerunner doesn't have a camera?
Nov 16 09:17:01 <panicking> yes I know
Nov 16 09:17:02 <panicking> :)
Nov 16 09:17:07 <panicking> but we can use in android
Nov 16 09:17:21 <panicking> and it's possible to find an design a micro camera
Nov 16 09:17:29 <panicking> for it and plug on the usbhost
Nov 16 09:18:41 <panicking> Ok, the problem is in the short-enums
Nov 16 09:18:51 <panicking> so a find a bug in the code
Nov 16 09:19:30 <panicking> because the sizeof is different
Nov 16 09:19:34 <panicking> and the ioctl fail
Nov 16 09:20:50 <larlin> panicking, would that be in the compiler?
Nov 16 09:21:41 <larlin> of isi t in the c library?
Nov 16 09:23:31 <panicking> the ioctl
Nov 16 09:23:40 <panicking> use the sizeof(struct) for
Nov 16 09:23:43 <panicking> create cmd
Nov 16 09:23:49 <panicking> so if the sizeof is different
Nov 16 09:23:56 <panicking> from the kernel point of view
Nov 16 09:24:05 <panicking> and the userspace point of view
Nov 16 09:24:10 <panicking> the cmd is different
Nov 16 09:24:38 <larlin> ahh that is a problem
Nov 16 09:25:14 <panicking> it can be affected a  lot of subsytem
Nov 16 09:29:16 <mvzaanen> so where does this size difference come from?
Nov 16 09:29:27 <mvzaanen> aren't both user and kernel space using the same include?
Nov 16 09:29:27 <mvzaanen> s
Nov 16 09:29:45 <panicking> compilation issue
Nov 16 09:29:48 <panicking> of enum type
Nov 16 09:29:53 <panicking> sizeof(enum) = 1
Nov 16 09:30:01 <panicking> -fno-short-enum
Nov 16 09:30:05 <panicking> sizeof(enum) = 4
Nov 16 09:30:09 <panicking> so a struct can be
Nov 16 09:30:14 <panicking> of different size
Nov 16 09:30:16 <panicking> is it clear
Nov 16 09:30:24 <panicking> gcc -fno-short-enum pippo.c
Nov 16 09:30:28 <panicking> gcc pippo.c
Nov 16 09:30:59 <mvzaanen> ah, I see
Nov 16 09:31:08 <panicking> so for example
Nov 16 09:31:12 <panicking> in the audio makefile
Nov 16 09:31:29 <panicking> LOCAL_CFLAGS := -fno-short-enums
Nov 16 09:31:32 <panicking> we have this one
Nov 16 09:31:35 <mvzaanen> yeah
Nov 16 09:31:39 <mvzaanen> I see the problem...
Nov 16 09:31:48 <panicking> ok how many external application
Nov 16 09:31:50 <panicking> use the ioctl
Nov 16 09:31:50 <panicking> ?
Nov 16 09:31:53 <mvzaanen> where does this difference come from? (I mean the difference in -f flags)?
Nov 16 09:32:04 <mvzaanen> is this from the "original" Android?
Nov 16 09:32:30 <panicking> and some of them can be fail because
Nov 16 09:32:43 <panicking> the ioctl use a struct as param
Nov 16 09:32:47 <panicking> with some enum type
Nov 16 09:32:53 <panicking> so it's very tricky
Nov 16 09:52:46 <mvzaanen> but if all are compiled the same way, it shouldn't be a problem, should it?
Nov 16 09:56:56 <panicking> yes, if you compile in the same way
Nov 16 09:57:00 <panicking> it's not a problem
Nov 16 09:57:13 <panicking> but are we sure that all the application are compiled in the same way
Nov 16 09:57:47 <serdar__> hi
Nov 16 09:59:42 <panicking> hi
Nov 16 10:03:31 <mvzaanen> well... no
Nov 16 10:03:40 <mvzaanen> but then they will simply crash on an ioctl call
Nov 16 10:05:30 <panicking> or return an EINVAL
Nov 16 10:05:39 <panicking> because they check the value
Nov 16 10:05:46 <panicking> of the size
Nov 16 10:05:53 <panicking> so if the size is different they fail
Nov 16 10:05:56 <panicking> and nothing crash
Nov 16 10:06:04 <panicking> just it answer that is not supported
Nov 16 10:06:37 <serdar__> there is a syncing problem with android.kernel.org.
Nov 16 10:07:28 <panicking> serdar__, do you try to compile eclair
Nov 16 10:07:28 <panicking> ?
Nov 16 10:08:41 <serdar__> i can't sync
Nov 16 10:09:21 <mvzaanen> I see, panicking
Nov 16 10:11:31 <serdar__> serdar@serdar-laptop:~/eclair$ repo init -u git://android.git.kernel.org/platform/manifest.git
Nov 16 10:11:31 <serdar__> Getting repo ...
Nov 16 10:11:31 <serdar__>    from git://android.git.kernel.org/tools/repo.git
Nov 16 10:11:31 <serdar__> fatal: The remote end hung up unexpectedly
Nov 16 10:32:07 <serdar__> have someone tried golang?
Nov 16 10:53:33 <u4ia> golang?
Nov 16 10:55:33 <serdar__> the google go language
Nov 16 10:55:48 <serdar__> but I guess it will be renamed to issue 9
Nov 16 10:57:36 <u4ia> haven't tried it yet
Nov 16 10:58:27 <u4ia> ah, "I thought of the name first" issue
Nov 16 10:59:06 <serdar__> look at issue 9 http://code.google.com/p/go/issues/detail?id=9
Nov 16 10:59:09 <serdar__> 909 comments
Nov 16 10:59:17 <u4ia> hold on, not yet done scrolling
Nov 16 10:59:19 <u4ia> pfff
Nov 16 10:59:27 <u4ia> :)
Nov 16 10:59:49 <serdar__> but renaming google go to issue9 will be nice.
Nov 16 10:59:55 <u4ia> yeah
Nov 16 11:00:02 <u4ia> or follow comment 79
Nov 16 11:00:48 <serdar__> look at #135
Nov 16 11:01:41 <u4ia> hehehe
Nov 16 11:02:02 <u4ia> way too much free time
Nov 16 11:02:24 <u4ia> but 'Issue9' has a nice ring to it
Nov 16 11:02:55 <serdar__> #go-nuts
Nov 16 11:04:18 <u4ia> hmm
Nov 16 11:04:36 <u4ia> apparently weechat doesn't like this many people in a channel..
Nov 16 11:06:28 <u4ia> the nicklist is filling 80% of the screen :)
Nov 16 11:11:42 <larlin> u4ia, and this not a particularly big channel =)
Nov 16 11:12:49 <u4ia> i'll just stay in channels with under 50 users, and I'll be fine :)
Nov 16 13:11:32 <Serdar> oh my god, they killed chanserv.
Nov 16 13:12:44 <th0br0> umm, is the new weekly release up already?
Nov 16 13:13:52 <th0br0> apparently
Nov 16 13:15:27 <larlin> th0br0, it's automagic
Nov 16 13:18:36 <th0br0> Ah ok larlin...
Nov 16 13:18:51 <th0br0> Well, given the commits I doubt that the SMS issues and so on have been fixed yet :(
Nov 16 13:28:19 <Serdar> I can't sync!
Nov 16 13:50:12 <pbaxter> hi all
Nov 16 14:09:46 <Serdar> hi jancona
Nov 16 14:33:29 * GovCon acts like a kernel and panics
Nov 16 14:33:54 * larlin thinks GovCon is a brokend kernel
Nov 16 20:00:18 <serdar_> hi panicking
Nov 16 20:02:35 <panicking> Hi
Nov 16 20:32:24 <pbaxter> hi all
Nov 16 20:32:34 <pbaxter> someone's running eclair?
Nov 16 20:32:43 <pbaxter> and someone's running w3eekly47?
Nov 16 20:32:46 <pbaxter> weekly47*
Nov 16 21:32:37 <u4ia> still can't sync :|
**** ENDING LOGGING AT Tue Nov 17 02:59:56 2009