Time Nick Message 03:20 spillz thoughts? https://forum.minetest.net/viewtopic.php?pid=136794#p136794 (coding that GUI was surprisingly excruciating) 14:01 sfan5 any more comments to https://github.com/minetest/minetest_game/pull/250 ? 17:59 sapier wow 24 hours after merging a patch most ppl didn't check and no blocker issue? is everyone sleeping? ;-) 18:00 sapier ok almost 24 hours 18:01 sapier I'm gonna merge #1208 #1205 and #1199 soon so please comment if anyone recognizes issues in there 18:03 ShadowNinja sapier: Well, I figure commiting something with a bug or two is better than leaving it in limbo for a year. 18:04 sapier You're right ShadowNinja, at least as long as we're not about to release and the one commiting is prepared to fix upcoming bugs within reasonable time 18:05 ShadowNinja Yes, that's true. 18:06 ShadowNinja 1208 seems fine with s/}\n\telse {/} else {/ 18:07 ShadowNinja I suppose 1205 doesn't cause any harm. 18:08 sapier sorry what do you want to be changed at 1208? 18:09 ShadowNinja https://github.com/minetest/minetest/pull/1199/files#diff-57ef810d2cc823e1e1b7fe73af93cf8aR276 this should be removed thater than commented out. 18:09 sapier well I'm still hoping someone knowing bsd better then me finds a way to fix it for bsd 18:10 Taoki Hey everyone. Latest GIt master is broken and won't compile with gcc 18:10 sapier what? 18:10 Taoki Last commit seems to cause this 18:10 ShadowNinja sapier: http://pastebin.ubuntu.com/7227402/ 18:10 Taoki Hold on 18:10 Taoki http://pastebin.com/raw.php?i=AbMaGxiy 18:10 sapier else one line up? 18:10 Taoki And yes I have gettext disabled from cmake 18:11 Taoki sapier: Last commit has your name, so please see if you can find what's causing that 18:11 ShadowNinja sapier: Yep. And spaces after commas please. :-) 18:12 * ShadowNinja compiles 18:12 sapier come on spaces ... empty room ... do we really have to waste that much ? ;-) *just kidding* 18:13 sapier taoki debug or release build? 18:13 Taoki I always compile release 18:13 ShadowNinja sapier: There's a #error after a #warning there. And should there be a warning at all? 18:14 sapier you're right that error is relic of a test I'm gonna remove it 18:16 sapier taoki can you do a clean remove CMakeCache.txt and do a new cmake? 18:16 Taoki Sure, will after new commit 18:17 sapier what do you supply to cmake? I can't reproduce this bug with my default config 18:18 ShadowNinja Minetest git compiles without so much as a warning with GCC 4.8.2. 18:19 sapier I recently fixed all clang warnings too but as noone dared to agree to them they're already outdated again 18:20 Taoki sapier: cmake-gui with default settings apart from enabling freetype support (linux) 18:23 sapier I suggest trying a fresh clone and trying command line 18:23 Taoki Nothing is changed in the default files, so that shouldn't be needed 18:24 Taoki Otherwise GIt revert would show it 18:24 sapier ShadowNinja: I think so yes as that warning is shown on unsupported architectures, for all those we know we shouldn't run in this case 18:24 sapier the cmake config has changed 18:24 Taoki I cleared the old cmake cache. I'll see if anything else remains 18:25 sapier if you've generated files in there which aren't up to date that would explain this error 18:25 sapier but if this happes we might have a dependency issue 18:25 Taoki I deleted the build folder and made sure cmakelists.txt is the new one like everything else on GIT 18:26 sapier build is not a clean 18:26 sapier there are generated headers in src folder 18:26 sapier come one taoki instead of discussing cloning and checking a clean build would already be completed ;-) 18:27 sapier ShadowNinja: #1208 fine now? 18:27 Taoki I'm looking at the files changed in GIT, nothing strange. I'll try a new GIT clone if you insist though 18:29 Taoki sapier: What cmake command should I run exactly? Never used cmake in CLI before. How do I specify build folder location also 18:29 sapier I use "cmake -DRUN_IN_PLACE=1 -DCMAKE_RELEASE_TYPE=Debug/Release" ... of course you should only specify debug OR release 18:30 sapier wait 18:30 sapier CMAKE_BUILD_TYPE 18:30 Taoki ok. How do i specify build folder 18:30 sapier get into that folder? 18:30 Taoki Don't want it to try installing it system-wide 18:30 Taoki Ah, so it depends where you run cmake from? 18:31 sapier yes 18:31 Taoki ok 18:31 Taoki I'll try root GIT folder for this test since its temporary 18:31 sapier well if this is a gui reason I wonder why it's there in new versions but not in old ones 18:31 Taoki I'm trying CLI too to be sure 18:34 Taoki http://pastebin.com/raw.php?i=sq402N4t Says errors ocurred at the end but there aren't any listed 18:36 sapier CMAKE_BUILD_TYPE 18:36 sapier sorry 18:37 Taoki Ah 18:37 sapier and don't forget the -D in front of it 18:39 Taoki Seems like it might be building now. Which makes little sense since I did new cmake config and "make clean" for existing GIT. I'll re-check the whole scr folder to clear it of anything that might be left there 18:40 Taoki I've never heard of anything persisting a new cmake config and a "make clean" before though x.x 18:40 Taoki So this is new... didn't suspect new checkout might matter 18:41 sapier did you try it with gui and new clone too? 18:41 Taoki will now 18:42 sapier if not it could still be gui related 18:42 sapier maybe we miss something required by gui 18:43 Taoki I only know last commit breaks it, since I compiled it just fine with the one before it 18:43 Taoki Trying now with GUI cmake and all settings default 18:44 Taoki Strange, now this works too. But I'm not seeing any old files left in the old checkout by GIT. Hmmm 18:45 Taoki So new checkout always works 18:45 sapier if you find out what happened tell us so we can help others who stumble upon same issue 18:45 Taoki Only difference is that I alwo enabled Freetype. I'll try with that too and see 18:46 sapier pushing 1208 now 18:49 Taoki Yeah, it's working with that too... really weird 18:52 sapier merging 1205 next 18:55 Taoki Wow. Apparently if I change the name of the path the error re-appears. 18:56 Taoki Even if I use the new git clone 18:56 sapier so cmake somewhere adds the path in generated file 18:56 sapier s 18:57 Taoki Weit, might have been a wrong impression, it appears as if it's still going 18:58 Taoki An error deceived me 18:58 Taoki Erm, warning, sorry >.< 18:58 Taoki [ 5%] Building C object src/lua/build/CMakeFiles/lua.dir/loadlib.o 18:58 Taoki /windows/D/Small games/Minetest/Minetest_GIT/src/lua/src/loadlib.c: In function ‘ll_sym’: 18:58 Taoki /windows/D/Small games/Minetest/Minetest_GIT/src/lua/src/loadlib.c:76:21: warning: ISO C forbids conversion of object pointer to function pointer type [-Wpedantic] 18:58 Taoki lua_CFunction f = (lua_CFunction)dlsym(lib, sym); 18:58 sapier no problem I've seen to many strange errors to not believe strange things 18:59 Taoki So far though, with the new GIT checkout it is working. Although the old GIt checkout contained the EXACT same thing, which I checked multiple times. My mind is blown 18:59 Taoki Guess I'll put the new checkout in old path and copy any extra files back again 18:59 Taoki See that warning I posted above though 18:59 Taoki Maybe it needs fixing just in case 19:00 sapier possibly but it's not that easy to fix as in c function and variable pointers arent same thing 19:00 sapier most time it's gonna work as they in fact are same thing, but that's not guaranteed to be on all cpu architectures 19:01 Taoki ok. Yeah warnings aren't bad since it's all ok as long as it compiles 19:09 sapier did anyone check #1170 yet? 19:10 ShadowNinja It seems fine, I was going to suggest merging it before but was going to wait for kahrl to show up. 19:11 ShadowNinja Hmmm, he's been around recently. 19:12 sapier I've 1199 ready to merge is it fine the way it's now? 19:12 sapier removed error and added a request to fix the bsd issue to code :-) hope someone reads it :) 19:15 ShadowNinja sapier: Remove the broken code and change the english to "BSD doesn't seem to support thread names" or so. 19:15 sapier it's not exactly broken it just doesn't do what it's supposed to do 19:16 ShadowNinja Also: "Unknown platform for setThreadName, you won't have custom thread names" 19:16 sapier it's meant for others to see it's not gonna work this way :-) 19:17 ShadowNinja You could mention that setproctitle sets the proc name rather than thread name or so. 19:17 ShadowNinja But that seems fairly obvious from the function name. 19:17 sapier prctl is processcontrol too ;-) 19:33 sapier ShadowNinja: now ok? 19:37 Taoki sapier: GIT didn't show some folders which were there. I deleted them and it compiled again 19:37 sapier ok 20:12 sapier ok mergeing 1199 now 20:47 sapier wtf ... does anyone else have issues to clone something from github? 21:06 VanessaE works fine here, sapier 21:06 sapier hmm works for me again now too ... wonder what this was 21:07 VanessaE probably another DDoS :P 21:07 VanessaE "21:02 UTC We continue to investigate problems with HTTPS based Git operations. Things are working again, but moving very slowly." 21:08 VanessaE https://status.github.com/messages 21:08 sapier ok :-) that's a good explanation ... https ddos wonder if this is related to the openssl bug 21:09 iqualfragile sapier: my threads do not seem to be named 21:09 sapier what os? 21:09 iqualfragile debian 21:09 sapier strange 21:09 sapier do you see a warning on compile? 21:09 iqualfragile dempends, maybee i missunderstand 21:09 iqualfragile they would have names in htop? 21:10 sapier don't know about htop but they should be named if you do top -H -p 21:10 iqualfragile well, postgress shows up in htop 21:10 iqualfragile (named) 21:11 iqualfragile or was it squid 21:11 sapier does it work with top -H? 21:13 iqualfragile sapier: yes, it does, but the naming might be a bit… ambigious 21:13 iqualfragile minetest: would be better 21:13 sapier it's limited to 12 chars ;-) 21:15 iqualfragile sapier: well, maybee stop naming them like we would be running java? 21:15 sapier maybe start commenting prior merge ;-) ... you're free to change the names and create a pull request ;-) 21:16 sapier I don't care about names as long as I understand what they are supposed to do 21:23 iqualfragile sapier: postgress seems to be using something else, as the titles are definitly longer 21:23 sapier possible ... maybe pthread names 21:24 sapier are posgress names shown in top -H too? 22:54 ShadowNinja This *should* add support for BSD and Windows thread names: http://ix.io/bBn 23:02 ShadowNinja OSX too: http://ix.io/bBp 23:02 ShadowNinja http://stackoverflow.com/questions/2369738/can-i-set-the-name-of-a-thread-in-pthreads-linux 23:27 ShadowNinja hmmmm, PilzAdam, sfan5: Can you check if that will work on BSD and Windows?