00:01 <wolfspraul> pcercuei: what are your thoughts on wayland?

00:02 <wolfspraul> (thanks for the heads up here btw, I think wayland is an exciting project...)

00:02 <wolfspraul> I just don't know that well yet where it will go and what it will mean

00:02 <wolfspraul> how about vnc/remote?

00:02 <wolfspraul> how about sdl/framebuffer stuff or devices with small/very small screens...

01:04 <pcercuei> wolfspraul: I don't have thoughts

01:04 <pcercuei> I'm nowhere near an Xorg expert

01:05 <pcercuei> but what I know for sure is that having more projects is always better than just one

01:18 <wolfspraul> he :-)

01:18 <wolfspraul> that's a start, yes

01:18 <wolfspraul> do you plan to try it out?

01:18 <wolfspraul> I mean wayland

01:25 <pcercuei> maybe

01:25 <pcercuei> if it doesn't fuck up my system

01:25 <pcercuei> I'm running Debian wheezy, which is going to be marked as "stable" soon

01:29 <pabs3> FSVO "soon"

04:31 <qi-bot> [commit] Paul Cercueil: Save changes made on the link of an OPK on the user directory. (packages) http://qi-hw.com/p/gmenu2x/a04b9e3

04:31 <qi-bot> [commit] Paul Cercueil: Memorize the selector's directory as the default one for later use (packages) http://qi-hw.com/p/gmenu2x/c94968f

04:31 <qi-bot> [commit] Paul Cercueil: Assign an icon for a given link only if it hasn't been done before (packages) http://qi-hw.com/p/gmenu2x/dec1a89

04:31 <qi-bot> [commit] Paul Cercueil: Hide parameters not suitable for OPKs on the context menu (packages) http://qi-hw.com/p/gmenu2x/4c4f452

04:38 <qi-bot> [commit] Paul Cercueil: Default the 'browser' option to 'true'. (packages) http://qi-hw.com/p/gmenu2x/64eb86f

04:48 <qi-bot> [commit] Paul Cercueil: Drop the 'remove link' option on the context menu of OPKs (packages) http://qi-hw.com/p/gmenu2x/df981eb

08:35 <qi-bot> The build was successful: http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.minimal-20121021-2325

08:43 <e2580> Anyone interested in a sneak peek at a hardware based security device we are about ready to release?

10:46 <qi-bot> The build was successful: http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.full_system-20121023-1042

11:47 <kyak> xiangfu: hi! the build looks good. Wonder why qball failed to build (builds fine here).. Do you have a build log or could do make package/qball/{clean,compile} V=s (but you have to remove qball_src.zip from dl/)

11:48 <kyak> some people re-release a tarball without notifying, others don't even care about md5sum and don't even bother to change a tarball name upon release :)

11:59 <kyak> hm.. looks like openwrt doesn't test checksums if it doesn't have to download the source (already in dl/)

12:00 <kyak> which makes it really difficult to update packages whose source name doesn't change

12:01 <kyak> i just made some changes locally in dl/qball_src.zip - and openwrt didn't even notice

12:01 <kyak> jow_laptop: do you have an idea?

12:03 <jow_laptop> yeah, see if debian has an own tarball of it (the blah.orig.tar.gz ones) and use that

12:03 <kyak> nope, it's not packaged for debian

12:04 <jow_laptop> then you have to do it yourself

12:04 <kyak> do what?

12:04 <jow_laptop> or try to find a repo somewhere which you can reference

12:04 <jow_laptop> making sane, versioned tarballs

12:04 <kyak> ok.. is it possible to download to a versioned tarball?

12:05 <jow_laptop> ???

12:05 <kyak> like, grab qball_src.zip, but save it as qball_src_VER.zip

12:05 <kyak> in dl

12:05 <jow_laptop> probably not

12:05 <kyak> then when version changes, download will not find it try to redownload

12:05 <kyak> ok. Another option - can we check md5sum even if package is in dl/ ?

12:06 <jow_laptop> that would extremely slow down the build process

12:07 <jow_laptop> right now the md5sum calculation happens directly during download

12:07 <kyak> you know what.. forget about it. Unversioned tarballs really piss me off - so it's the author's fault

12:07 <jow_laptop> yes

12:07 <jow_laptop> does he have a repo?

12:08 <kyak> no, he has a web page and updates the source from time to time

12:08 <jow_laptop> link?

12:08 <kyak> sec

12:08 <kyak> http://homepage.ntlworld.com/mark.harman/comp_qball.html

12:08 <jow_laptop> bbl

12:09 <kyak> well, i even approached the author once asking him to version his source

12:09 <kyak> he said: "Unfortunately I've got some places that link to my website (e.g., http://qt-apps.org/ you can advertise Qt applications, with links directly to the archives), so it would be difficult to keep these all updated. And I don't have web space to store two copies I'm afraid. I'm not sure how OpenWrt packaging works - can you simply rename the files, or could it be hosted elsewhere?"

12:10 <kyak> that was almost two years ago.. probably i should approach him again and tell about wonders of github

12:11 <larsc> we can put a versioned mirror up on downloads.openwrt.org I think we do that for some other packages as well

12:12 <kyak> sounds like a good idea

12:12 <kyak> hosting versioned tarball for the author :)

12:36 <xiangfu> kyak, after re-download the dl/qball_src.zip, the qball compile just fine. I start build the new image again (with qball=y)

12:58 <jow_laptop> kyak: "There is also a Git repository on the Sourceforge page."+

12:58 <jow_laptop> so just reference that in the openwrt makefile and it will create beautifully versioned tarballs

12:58 <jow_laptop> which you can then mirror

13:11 <kyak> xiangfu: great, thanks :) hope this one is going to be final!

13:11 <kyak> jow_laptop: let me check...

13:12 <xiangfu> kyak, yes.

13:12 <jow_laptop> git clone git://git.code.sf.net/p/q-ball/code q-ball-code

13:13 <kyak> jow_laptop: thanks for your eyes :) This phrase hadn't always been there (for SF repo)

13:13 <kyak> the SF is down for me at the moment. .some error 500

13:49 <qi-bot> The build was successful: http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.full_system-20121023-1432

14:08 <kyak> xiangfu: i see there are no patches in http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.full_system-20121023-1432/patches/ for qt4. Did it build fine as it is?

14:11 <xiangfu> kyak, just checked. I forget update the patches. the qt4 patch should be http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.full_system-20121023-1432/patches/feeds.xorg.patch

14:12 <kyak> ok

14:17 <qi-bot> [commit] Paul Cercueil: Fix a bug where having multiple MIME types would crash GMenu2X (packages) http://qi-hw.com/p/gmenu2x/b8112d1

14:26 <qi-bot> [commit] Xiangfu: cgmienr: fix depends (master) http://qi-hw.com/p/openwrt-packages/a511011

14:31 <qi-bot> [commit] Paul Cercueil: For non-OPK programs, chdir() to the app's directory as well (packages) http://qi-hw.com/p/gmenu2x/0b922e9

14:33 <qi-bot> [commit] Xiangfu: ascii-paint: fix depends (master) http://qi-hw.com/p/openwrt-packages/2e44a43

14:35 <pcercuei> commit fight!

14:40 <kyak> --)

15:08 <qi-bot> [commit] Xiangfu: guile2: fix depends (master) http://qi-hw.com/p/openwrt-packages/45cd9dd

15:08 <qi-bot> [commit] Xiangfu: fpgatools: rename mini-jtag to fpgatools, include fp2bit/bit2fp (master) http://qi-hw.com/p/openwrt-packages/a295fd4

15:09 <pcercuei> ok, you win :(

15:12 <wpwrak> pcercuei: a foolish general thinks relentless heroic sacrifices will turn the tide even if the battle seems lost. a wise general falls back upon seeing the battle is lost and saves his troops for the counterattack.

15:13 <qi-bot> [commit] kyak: qball: use git repo (master) http://qi-hw.com/p/openwrt-packages/01e3c1f

15:17 <mth> wpwrak: saves them for the next (git) push?

15:22 <qi-bot> [commit] Xiangfu: dopewars: fix depends (master) http://qi-hw.com/p/openwrt-packages/b30b596

15:26 <kyak> ok, xiangfu is on update spree :)

15:26 <kyak> i think this calls for a final-final image :)

15:33 <xiangfu> kyak, yes. I will update the pacakge/index. for those packages.

15:34 <xiangfu> kyak, now I am waiting the 'guile2' compile the result. It is really taking too much time on the guile2-host. :-)

15:40 <wpwrak> mth: yeah. always keep a few unpushed commits as reserve units :)

15:40 <xiangfu> :)

15:43 <xiangfu> wpwrak, that is not good for your ally. :-D

16:09 <kyak> xiangfu: hm.. so strange. fw_printenv returns bootargs=mem=32M console=tty0 console=ttyS0,57600n8 ubi.mtd=2 rootfstype=ubifs root=ubi0:rootfs rw rootwait consoleblank=30 quiet g_ether.host_addr=72:8d:1f:c4:e8:ca, but cat /proc/cmdline is "mem=32M console=tty0 console=ttyS0,57600n8 ubi.mtd=2 rootfstype=ubifs root=ubi0:rootfs rw rootwait"

16:10 <xiangfu> kyak, let me check the u-boot. ENV ADDRESS. there is a lot of update in U-Boot.

16:11 <kyak> i remember that previous uboot didn't have such a problem.. i used this method to modify cmdline

16:25 <qi-bot> [commit] Xiangfu: nanonote-files: update fw_env.config with new bootloader configure (master) http://qi-hw.com/p/openwrt-packages/91b5bd4

16:25 <xiangfu> kyak, fixed in this commit.

16:25 <xiangfu> copy the new fw_env.config to /etc/

16:26 <xiangfu> here: http://projects.qi-hardware.com/index.php/p/openwrt-packages/source/tree/91b5bd4/nanonote-files/files/etc/fw_env.config

16:27 <xiangfu> since this file is under files/. I have to re-generate the image.

16:36 <xiangfu> kyak, have to sleep. good night. I will check the result tomorrow. thanks for feedback. the silent-boot working now. :)

17:05 <kyak> yep, works well now

19:35 <qi-bot> The build was successful: http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.full_system-20121023-1826