#qi-hardware IRC log for Sunday, 2011-10-23

DocScrutinizerwhat's labsw01:51
kristianpaulan usb controlled swich01:54
kristianpaulmade by werner :)01:55
kristianpaulDocScrutinizer: http://downloads.qi-hardware.com/people/werner/labsw/web/01:55
kristianpaulActually i had been thinking what else i can do with it besides powercycle a M101:56
DocScrutinizerpowercycle M1 - what's the use of such project?01:56
kristianpaulhunting for nor corruption in M101:57
DocScrutinizerfor sure a joyful thing to do01:57
DocScrutinizersounds a bit like "what's square root of 2?" though01:59
DocScrutinizer1 * 1 ? no01:59
DocScrutinizer1.000001 * 1.000001? no01:59
kristianpaul"It would be desirable to be able to also switch AC" indeed :)02:02
DocScrutinizerwell, depends on the type of relay you use, no?02:02
kristianpauli would prefer a triac for ac anyway02:04
wpwrakthe relays are quite decent :) just a little big02:09
kristianpaullittle noisy!!02:09
kristianpauli can stand that for a minute02:09
DocScrutinizeruse SolidState relays then02:09
DocScrutinizerI bet there are some available with even same footprint02:10
wpwrakthe noise isn't *that* bad :)02:10
kristianpaulwell, i was making that for 8 in/out that time02:12
kristianpaulfirst vesion was rele based, next triac :-)02:12
kristianpaulme happy ;)02:12
kristianpaulwas a very basic board i did for some EE were doign theit thesis,02:13
kristianpaulpic16F, sms send/receive via AT commands :)02:14
kristianpauland zero noise :)02:14
wpwrak(AT commands) ;-))02:15
kristianpaulSo if understand correcly that 20000+ cycles  was with no pull up mod?02:17
kristianpaulor you still finding the  2 square root, as DocScrutinizer pointed :)02:18
wpwrak20000+ was with WE# pull-up ... and something corrupting the lock bit of block 102:18
wpwrakwell, 1st block, so block 0 :)02:19
wolfspraulgood morning :-)03:24
rohwolfspraul: morning03:53
qi-bot[commit] Maarten ter Huurne: PowerSaver: re-layouted code. (master) http://qi-hw.com/p/gmenu2x/ae685a908:43
qi-bot[commit] Maarten ter Huurne: InputManager: SDL_NumJoysticks is a function, try calling it. (master) http://qi-hw.com/p/gmenu2x/0cb164508:43
qi-bot[commit] Maarten ter Huurne: InputManager: re-layouted code. (master) http://qi-hw.com/p/gmenu2x/f96969708:43
qi-bot[commit] Maarten ter Huurne: Include SDL header as <SDL.h> instead of <SDL/SDL.h>. (master) http://qi-hw.com/p/gmenu2x/ba6003608:43
qi-bot[commit] Maarten ter Huurne: InputManager: reduce namespace pollution. (master) http://qi-hw.com/p/gmenu2x/c6db9ff08:43
qi-bot[commit] Maarten ter Huurne: InputManager: C++-style type definitions. (master) http://qi-hw.com/p/gmenu2x/315bf0a08:43
qi-bot[commit] Maarten ter Huurne: BrowseDialog: simplify code using InputManager::waitForPressedButton(). (master) http://qi-hw.com/p/gmenu2x/ccf55e508:43
qi-bot[commit] Maarten ter Huurne: MessageBox: minor cleanup. (master) http://qi-hw.com/p/gmenu2x/40fd35d08:43
qi-bot[commit] Maarten ter Huurne: SettingsDialog: simplify code using InputManager::waitForPressedButton(). (master) http://qi-hw.com/p/gmenu2x/a38c28308:43
qi-bot[commit] Maarten ter Huurne: MenuSettingBool: Support toggling of boolean settings using left/right. (master) http://qi-hw.com/p/gmenu2x/641c98908:43
qi-bot[commit] Maarten ter Huurne: InputManager: have "joystick" as a member instead of a global. (master) http://qi-hw.com/p/gmenu2x/433be5d08:43
qi-bot[commit] Maarten ter Huurne: InputManager: Don't break if joystick support is disabled in SDL. (master) http://qi-hw.com/p/gmenu2x/9548aa008:43
qi-bot[commit] Maarten ter Huurne: InputManager: improved config file parsing. (master) http://qi-hw.com/p/gmenu2x/a685b4308:43
qi-botThe build was successfull, see images here: http://fidelio.qi-hardware.com/~xiangfu/compile-log/openwrt-xburst.full_system-10222011-0809/12:57
qi-bot[commit] Maarten ter Huurne: InputManager: removed "clear" and "manual" logical buttons. (master) http://qi-hw.com/p/gmenu2x/28f6cd215:45
qi-bot[commit] Maarten ter Huurne: MessageBox: Use arrays instead of vectors for buttons. (master) http://qi-hw.com/p/gmenu2x/3252c7715:45
qi-bot[commit] Maarten ter Huurne: MessageBox: header cleanups. (master) http://qi-hw.com/p/gmenu2x/9e876a615:45
qi-bot[commit] Maarten ter Huurne: SettingsDialog: removed unused action variable and macros. (master) http://qi-hw.com/p/gmenu2x/0b878c115:45
qi-bot[commit] Maarten ter Huurne: InputDialog: re-layouted code. (master) http://qi-hw.com/p/gmenu2x/fa5a98115:45
qi-bot[commit] Maarten ter Huurne: InputDialog: do not return anything from drawVirtualKeyboard(). (master) http://qi-hw.com/p/gmenu2x/55cabce15:45
qi-bot[commit] Maarten ter Huurne: IconButton: removed unused local variable. (master) http://qi-hw.com/p/gmenu2x/76613d115:45
qi-bot[commit] Maarten ter Huurne: IconButton: fixed up-to-date check in setPosition(). (master) http://qi-hw.com/p/gmenu2x/abe913015:45
qi-bot[commit] Maarten ter Huurne: IconButton: removed unused functionality. (master) http://qi-hw.com/p/gmenu2x/62facf415:45
qi-bot[commit] Maarten ter Huurne: IconButton: header cleanup. (master) http://qi-hw.com/p/gmenu2x/cfd294b15:45
qi-bot[commit] Maarten ter Huurne: InputDialog: header cleanup. (master) http://qi-hw.com/p/gmenu2x/40fcd5115:45
qi-bot[commit] Maarten ter Huurne: LinkApp: various cleanups. (master) http://qi-hw.com/p/gmenu2x/14bae0415:45
qi-bot[commit] Maarten ter Huurne: cpu: Include header instead of repeating function declaration. (master) http://qi-hw.com/p/gmenu2x/33e7f1e15:45
qi-bot[commit] Maarten ter Huurne: Massive header cleanup. (master) http://qi-hw.com/p/gmenu2x/e0e7e8715:45
qi-bot[commit] Maarten ter Huurne: More header cleanups. (master) http://qi-hw.com/p/gmenu2x/a01d89215:45
qi-bot[commit] Maarten ter Huurne: utilities: Removed unused functions. (master) http://qi-hw.com/p/gmenu2x/40c510a15:45
qi-bot[commit] Maarten ter Huurne: utilities: Replace min/max functions by the versions from STL. (master) http://qi-hw.com/p/gmenu2x/aea1c4415:45
qi-bot[commit] Maarten ter Huurne: Minimized gmenu2x.h. (master) http://qi-hw.com/p/gmenu2x/f28e4f615:45
qi-bot[commit] Maarten ter Huurne: gmenu2x.cpp: Re-ordered includes. (master) http://qi-hw.com/p/gmenu2x/cf8ebbc15:45
Aylamth: great work15:58
mthyeah, but as usual with gmenu2x I still haven't fixed the thing I wanted to fix today ;)16:01
Aylamth: what was the correct command to compile the python scripts?16:11
jow_laptopkyak: if you find a few minutes, please check whether the menuconfig situation improved now. The odd behaviour was most likely caused by recursive depends injected by the luci suite16:15
kyakjow_laptop: sure, i can do that. What is the commit that has most likely solved the problem?16:55
jow_laptopif you do not include the luci feed in your builds its probably another package causing the problems16:57
jow_laptopin this case, run  make defconfig 2>/dev/null | grep recursive16:57
kyakno, i don't have luci in there16:59
kyakthere are several packages with recursive dependencies17:00
jow_laptopthose are most likely causing it17:00
kyakqt4 is among them17:02
kyaklubpurple, too17:02
kyakit's a little bit strange17:02
kyakfor example, it mentiones alex417:03
kyakalex4 depends on alex4-data, while alex4-data depends on alex4. Is is the kind of thing causing problems for kconfig?17:03
jow_laptopI had  stuff like package A with DEPENDS:=+B and package B with DEPENDS:=A  ... that made menuconfig freak out17:03
kyakstrange.. it has been like this for while17:04
kyaki'll try changing it now17:04
jow_laptopyes, somehow menuconf got stricter, but it has always been a problem17:04
kyakinteresting.. libpurple is mentioned as recursive, but it actually isn't: http://projects.qi-hardware.com/index.php/p/openwrt-packages/source/tree/master/pidgin/Makefile17:07
kyakDEPENDS:=+libxml2 +libgnutls +libncursesw +glib2 $(ICONV_DEPENDS) $(INTL_DEPENDS)17:07
jow_laptopcan you paste the full recursive lines somewhere?17:10
jow_laptopits a bit hard to understand as select-depends also inject reverse depends into the selected package17:10
jow_laptopyou can inspect the generated Kconfig source with "./scripts/metadata.pl package_config tmp/.packageinfo"17:11
kyakjow_laptop: i'll be back in a couple of hours17:12
mthAyla: python -OO -E -c 'import compileall ; compileall.compile_dir("path")'17:20
Aylamth: ah thanks17:21
AylaI'm having an issue with a makefile right now, maybe one of you have the answer17:22
AylaI have this rule:17:22
Ayla$(PYGAME_LIB_DIR)/%.so: $(PYGAME_SRC_DIR)/%.c17:22
Ayla    $(TARGET_CC) $(CFLAGS) $^ $(LDFLAGS) -o $@17:23
Aylait does work fine, as long as the .c files are present when the makefile is executed17:23
Aylahowever, the sources are extracted from a TGZ archive;17:24
Aylaif I launch the makefile, it will fail to build any of the .so lib17:24
Aylaif I launch it again, it works fine17:24
Aylahow can I bypass that?17:24
jow_laptopwhat triggers the .tar.gz unpack?17:27
AylaI untar it manually17:31
Aylainside a rule17:31
Aylalet me upload that Makefile, it'll be easier17:32
Aylajow_laptop: http://pastebin.com/h8YMQ2Q917:34
jow_laptopcan't you make the %.so target simply depend on $(PYGAME_DIR)/.downloaded as well?17:35
Aylathe current fix is line 122: I call an external "make" process17:35
jow_laptop$(PYGAME_LIB_DIR)/%.so: $(PYGAME_DIR)/.downloaded $(PYGAME_SRC_DIR)/%.c17:36
Aylaprobably, what would that change?17:36
jow_laptopit would force make to complete ".downloaded" which in turn would trigger wget and tar -xzf17:36
Aylathe sources are correctly extracted, that's not the problem17:37
jow_laptopbefore diving into the %.so rule17:37
mthit would be cleaner to depend the .so files on the .c files17:42
mthand use $^ to refer to them inside the rule itself, to avoid duplication17:43
mthand then make the .c files depend on .downloaded17:43
Aylasome of the .so are constitued with two .c files17:45
Aylaor maybe $^ extends to all the C files?17:48
Aylathat are supplied at the right of the ':' delimiter17:48
mthiirc $^ is all dependencies and $< is the first dependency, but maybe I mixed them up17:48
mthdoesn't pygame itself already include a Makefile? or a setup.py?17:49
Aylait has setup.py17:50
mthit's possible setup.py already has the functionality to build the libs17:54
mthI don't know if it can handle cross compilation though17:54
mthbut maybe it's easier to patch setup.py than the replicate its functionality17:54
Aylait's a bit of a mess17:54
Aylathat's why I went with a Makefile17:54
AylaSiENcE did the same and bypassed setup.py with another python script17:55
Aylaok, here is a new makefile: http://pastebin.com/yyGhZ8kS17:57
Aylawhich does not work at all17:57
AylaMake says that there's no rule to build output/build/pygame-1.9.1release/lib/base.so17:58
kyakjow_laptop: what i should be looking for in the generated Kconfig source?18:32
jow_laptopselect, dpends18:34
kyakin case of libpurple, it doesn't lok recursive18:35
jow_laptopsorry I got distracted19:10
jow_laptopkyak: can you upload the full kconfig output somewhere? so that I can follow the dependency chains manually19:11
mthwell, there is indeed no rule to build the "base" module21:56
mth(for Ayla)21:57
Aylamth: you should highlight me22:08
Aylamth: the rule to build the "base" module should be the one line 36 on http://pastebin.com/yyGhZ8kS22:08
mthI did highlight you, one line afterwards22:16
mthis there a src/base.c?22:18
mthbedtime for me, I'll check back tomorrow22:22
Aylairssi will highlight the message only if my nick is the very first word of the message22:26
Aylaand there is a src/base.c as soon as the tarball is extracted22:26
mthwell, that's not what most IRC clients do afaik22:27
mthI don't know if it makes a difference, but I usually use the 3-part target line22:28
mthso "targets: dest_pattern: src_pattern"22:28
Action: mth really gone now22:29
--- Mon Oct 24 201100:00

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