Created
August 3, 2016 05:23
-
-
Save SinZ163/ad2fa1ea0b5097afd60bebb054d4dfa7 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oct 23 15:42:07 <Grum> clonejo<barneygale> Gru-m reckons it's drivers but it affects loads of different cards from different manufacturers <-- then it's the subsystem, whatever it is, its not minecraft | |
Oct 23 15:43:17 <barneygale> what subsystem? | |
Oct 23 15:45:12 <nickelpro> Could be a problem in Mesa, unlikely though | |
Oct 23 15:46:23 <nickelpro> More likely LWGL or Minecraft doing something non-obvious that OpenGL doesn't like, don't know the spec well enough to speculate | |
Oct 23 15:48:07 <barneygale> I asked in #lwjgl and they say "we don't do mc support" | |
Oct 23 15:48:23 <dx> unsurprising | |
Oct 23 15:48:32 <barneygale> Yah, it was a long shot | |
Oct 23 15:48:39 <dx> they are probably used to it | |
Oct 23 15:48:47 <barneygale> So as far as I can tell its up to the affected users to figure out the problem, uhg | |
Oct 23 15:49:06 <nickelpro> For a long time MC used an ancient version of LWJGL, that would get frustrating fast | |
Oct 23 15:49:36 <dx> barneygale: if only we could bisect | |
Oct 23 15:50:16 <barneygale> nickelpro: yah. Xorg wouldn't report keyups properly so you would get stuck heading straight for lava haha | |
Oct 23 15:51:17 <nickelpro> barneygale: Ya that was fun, on Arch it wouldn't even launch unless you manually updated the LWJGL jars | |
Oct 23 15:52:07 <dx> ...always worked for me? | |
Oct 23 15:52:28 <dx> with the stuck keys issue | |
Oct 23 15:52:35 <nickelpro> I forget the error it used to throw, I think it was related to not being able to grab a rendering context | |
Oct 23 15:53:29 <dx> barneygale: do we even know if it's a graphics related issue? | |
Oct 23 15:53:45 <dx> or is the pulseaudio theory still around? | |
Oct 23 15:55:37 <nickelpro> dx: Feel like it would be more widespread if it was PA | |
Oct 23 15:56:56 <nickelpro> looks like people running flat alsa still run into it as well | |
Oct 23 15:58:01 <barneygale> dx: pulseaudio theory is a red herring | |
Oct 23 15:58:22 <barneygale> I changed openal to use a null driver, and also to just error out, but it didn't affect it. | |
Oct 23 16:00:05 * mappum has quit (Read error: Connection timed out) | |
Oct 23 16:01:31 <Grum> barneygale: you have the problem? | |
Oct 23 16:01:51 <barneygale> I do. | |
Oct 23 16:02:11 <barneygale> My PC blew up so I'm stuck on this laptop which exhibits it. | |
Oct 23 16:02:46 <barneygale> I'm going to bed soon but I can set up a VNC or something if you promise not to touch anything | |
Oct 23 16:04:39 <barneygale> wait, why are you awake Grum? | |
Oct 23 16:04:45 <Grum> because its morning? | |
Oct 23 16:04:58 <barneygale> Oh yeah, you're ahead not behind. derp. | |
Oct 23 16:04:58 <Grum> do you know how to make a custom version to run? | |
Oct 23 16:05:08 <barneygale> Yeah, I do | |
Oct 23 16:05:12 <barneygale> create folder, edit some json, etc? | |
Oct 23 16:05:22 <Grum> yes, how much time do you have spare now? | |
Oct 23 16:05:32 <barneygale> I can spare an hour or two | |
Oct 23 16:05:53 <Grum> ok gimme a min, i'll see if i can make some custom builds for you to test | |
Oct 23 16:06:10 <barneygale> This is why I love you Grum <3 | |
Oct 23 16:06:37 <SinZ> btw Grum, are you actually going to rewrite EntityBoat? | |
Oct 23 16:07:02 <Grum> no idea SinZ | |
Oct 23 16:07:11 <Grum> barneygale: you do not get me then, i'm going to try to prove its not something we do | |
Oct 23 16:07:24 <Grum> and i'll force-hack out the probing and usage of FBO's | |
Oct 23 16:07:36 <Grum> if that fixes it, its not our problem | |
Oct 23 16:07:39 <Grum> BUT we can make a workaround | |
Oct 23 16:07:52 <Grum> a --linuxsucksdonkeycocksandcannotrunthegameinthefutureunlesstheystop | |
Oct 23 16:08:01 <Grum> so sec=D | |
Oct 23 16:08:03 <barneygale> I'm just happy that progress is being made, even if that progress is eliminating possibilities | |
Oct 23 16:08:19 <SinZ> <Mikeee> told Grum to spend time rewriting that shit :P | |
Oct 23 16:09:00 <Grum> barneygale: building now | |
Oct 23 16:09:13 <Grum> diff: | |
Oct 23 16:09:16 <Grum> - useFbo = separateBlend && GLContext.getCapabilities().GL_ARB_framebuffer_object; | |
Oct 23 16:09:16 <Grum> + useFbo = false && separateBlend && GLContext.getCapabilities().GL_ARB_framebuffer_object; | |
Oct 23 16:09:37 <Grum> i'm sure its some sort of quirky compat mode that now falls on its face | |
Oct 23 16:09:47 <Grum> however we have no intention of not using FBOs in the future | |
Oct 23 16:10:00 <nickelpro> why poll for them then? | |
Oct 23 16:10:04 <Grum> we do want to stop using them and gllists at the same time (we're slowly moving towards VBOs like we should) | |
Oct 23 16:10:19 <Grum> nickelpro: because right now it is optional | |
Oct 23 16:10:33 <Grum> but some videocards/drivers/subsystems lie about support and then die | |
Oct 23 16:10:56 <Grum> meh not awake, building now :P | |
Oct 23 16:11:01 <Grum> its a bit early out here ;) | |
Oct 23 16:11:07 <barneygale> :D | |
Oct 23 16:12:28 <Grum> takes 2-3 mins orso :P | |
Oct 23 16:12:46 <Grum> Total time: 1 mins 41.417 secs | |
Oct 23 16:12:52 <Grum> oh bit shorter, yay | |
Oct 23 16:13:24 <Grum> http://db.grum.nl/minecraft.linux.1.jar | |
Oct 23 16:15:17 <barneygale> Running now | |
Oct 23 16:15:32 <barneygale> err, well, "installing" now. | |
Oct 23 16:16:16 <Grum> ;) | |
Oct 23 16:18:26 <barneygale> Grum: same as before (i.e. not working). Want me to wait a while? | |
Oct 23 16:18:37 <Grum> euuh lol lets see | |
Oct 23 16:19:21 <Grum> building again | |
Oct 23 16:19:28 <dx> the plot thickens | |
Oct 23 16:20:48 <Grum> even less probing to opengl now O.o | |
Oct 23 16:22:49 <Grum> barneygale: http://db.grum.nl/minecraft.linux.2.jar | |
Oct 23 16:24:23 <barneygale> stand by | |
Oct 23 16:24:45 <barneygale> no workie. | |
Oct 23 16:24:55 * mappum (~mappum@c-67-185-176-162.hsd1.wa.comcast.net) has joined #mcdevs | |
Oct 23 16:25:15 <Grum> err right | |
Oct 23 16:27:06 <dx> woo fancy flowers | |
Oct 23 16:27:25 <dx> all over the motherfucking place | |
Oct 23 16:27:51 <Grum> building again | |
Oct 23 16:29:48 <Grum> barneygale: http://db.grum.nl/minecraft.linux.3.jar | |
Oct 23 16:30:01 <Grum> stripping more and more opengl related things | |
Oct 23 16:30:52 <dx> minecraft.linux.512.jar will be a directx version of minecraft | |
Oct 23 16:31:02 <barneygale> hah | |
Oct 23 16:32:23 <barneygale> Grum: failure once again :( | |
Oct 23 16:32:50 <Grum> barneygale: running out of options | |
Oct 23 16:33:09 <SinZ> which version did it start to derp | |
Oct 23 16:33:26 <SinZ> work your way up in commits and figure out what caused it | |
Oct 23 16:33:41 <dx> yeah, git bisect | |
Oct 23 16:34:02 <Grum> i have no idea when it broke | |
Oct 23 16:34:12 <barneygale> I found that out the other day. Let me check again | |
Oct 23 16:35:04 <barneygale> w37 works, w38 doesn't | |
Oct 23 16:35:39 <barneygale> w38 might work after 7 minutes (or whatever), I never bothered to wait that long | |
Oct 23 16:35:53 <Grum> both 'a' ? | |
Oct 23 16:36:07 <Grum> 13w37a / 13w38a ? | |
Oct 23 16:36:27 <barneygale> 13w37b / 13w38c | |
Oct 23 16:36:33 <barneygale> Those are the only ones I see in my launcher | |
Oct 23 16:36:53 <barneygale> Want me to check 13w38a and 13w38b? | |
Oct 23 16:36:54 <Grum> $ git log --oneline snapshot-13w37b..snapshot-13w38a | wc -l | |
Oct 23 16:36:55 <Grum> 106 | |
Oct 23 16:37:06 <Grum> sigh ..... 106 commits o.O | |
Oct 23 16:37:12 <barneygale> ouch | |
Oct 23 16:37:23 <Grum> 12500 lines of diff | |
Oct 23 16:37:27 <dx> that isn't a lot of steps with git bisect! | |
Oct 23 16:37:29 <barneygale> eek1 | |
Oct 23 16:39:09 <SinZ> thats still less than 1.6..1.7 | |
Oct 23 16:39:19 <Grum> yeah starting bisect now | |
Oct 23 16:39:24 <Grum> the thing is, i dont even know if this compiles | |
Oct 23 16:39:45 <Grum> this will be the bestest bisect ever | |
Oct 23 16:39:49 <Grum> remotely tested =) | |
Oct 23 16:40:21 <barneygale> when do I get my first paycheck | |
Oct 23 16:40:32 <Grum> whenever you start using windows | |
Oct 23 16:40:38 <barneygale> >:( | |
Oct 23 16:40:48 <dx> Grum: you can always do "git bisect skip" if it's a broken commit | |
Oct 23 16:40:54 <Grum> i know how to bisect | |
Oct 23 16:41:05 <Grum> i just dont know if these things are buildable :D | |
Oct 23 16:41:17 <Grum> also i think the libraries changed with the release | |
Oct 23 16:41:18 <nickelpro> rebooting, what's the mail time across atlantic? | |
Oct 23 16:41:30 <Grum> barneygale: diff the two snapshot json files please | |
Oct 23 16:41:35 <Grum> see if their libraries changed | |
Oct 23 16:41:43 <Grum> i know we added vecmath | |
Oct 23 16:41:47 <Grum> but see if anything else did | |
Oct 23 16:41:56 <Grum> and then use the 13w38 one for the build i'll be giving you | |
Oct 23 16:42:28 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.1.jar + 13w38's json | |
Oct 23 16:42:49 <barneygale> Grum: https://gist.github.com/barneygale/27e9422e9fc668974ddc | |
Oct 23 16:43:00 <Grum> good | |
Oct 23 16:43:06 <Grum> so use any json for it as base | |
Oct 23 16:43:10 <Grum> we did change libs for 1.7 | |
Oct 23 16:43:18 <Grum> and i'm not compiling against those now | |
Oct 23 16:44:43 <barneygale> Grum: works! | |
Oct 23 16:44:48 <Grum> k | |
Oct 23 16:45:09 <Grum> Bisecting: 26 revisions left to test after this (roughly 5 steps) | |
Oct 23 16:45:11 <Grum> whoo :p | |
Oct 23 16:45:20 <SinZ> 106 down to 26, impressive | |
Oct 23 16:45:30 <Grum> well 52 now | |
Oct 23 16:45:31 <barneygale> is that average or max? hopefully it'll take fewer than that | |
Oct 23 16:45:36 <Grum> should be max | |
Oct 23 16:45:41 <Grum> its a binary walk to the problem | |
Oct 23 16:45:48 <Grum> assumes that one commits causes it | |
Oct 23 16:46:08 <Grum> weird cases let it be more | |
Oct 23 16:46:09 <Grum> but we'll see | |
Oct 23 16:46:36 <Grum> pfft obfuscation taking time :P | |
Oct 23 16:46:54 <barneygale> just give it to me unobfuscated, I totally won't put it on TPB ;) | |
Oct 23 16:47:01 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.2.jar | |
Oct 23 16:48:15 <barneygale> Grum: also works. | |
Oct 23 16:48:27 <Grum> building next | |
Oct 23 16:48:36 <Grum> Bisecting: 13 revisions left to test after this (roughly 4 steps) | |
Oct 23 16:48:49 <Grum> and yeah for 1.7 we have 1100+ commits | |
Oct 23 16:48:59 <Grum> that would take a little bit longer (a least 3-4 bisects) | |
Oct 23 16:49:21 * TomyLobo (~TomyLobo@91-65-113-111-dynip.superkabel.de) has joined #mcdevs | |
Oct 23 16:51:01 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.3.jar | |
Oct 23 16:52:32 <barneygale> Grum: this one seems slightly broken, let me screenshot | |
Oct 23 16:52:48 <Grum> it starts right? | |
Oct 23 16:53:10 <Grum> it's stuck in the bottom-left corner | |
Oct 23 16:53:12 <barneygale> It starts, but the whole menu is squashed into the bottom left hand side the window | |
Oct 23 16:53:15 <barneygale> the rest is black | |
Oct 23 16:53:16 <Grum> yeah | |
Oct 23 16:53:21 <Grum> known issue back then | |
Oct 23 16:53:28 <barneygale> OK, cool | |
Oct 23 16:53:43 <Grum> this does mean your videocard is lying about FBO support in some way | |
Oct 23 16:54:04 <Grum> because you shouldn't have to reset the viewport for that | |
Oct 23 16:54:14 <Grum> building ofc ;) | |
Oct 23 16:55:02 <Grum> buildsystem is getting a workout :P | |
Oct 23 16:55:05 <Grum> happy its stable =) | |
Oct 23 16:55:18 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.4.jar | |
Oct 23 16:55:24 <Grum> oh wait | |
Oct 23 16:55:25 <Grum> not upped yet | |
Oct 23 16:55:28 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.4.jar | |
Oct 23 16:55:30 <Grum> now is! :D | |
Oct 23 16:55:34 <barneygale> Thanks! | |
Oct 23 16:55:48 <Grum> Bisecting: 6 revisions left to test after this (roughly 3 steps) | |
Oct 23 16:56:58 <barneygale> Starts; same issue as previous one. | |
Oct 23 16:57:13 <Grum> building | |
Oct 23 16:59:15 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.5.jar | |
Oct 23 17:00:24 <barneygale> Fuck, I think I made a mistake Grum. Previous one you sent me /doesn't/ work (i.e. bisect.4) | |
Oct 23 17:00:30 <barneygale> I accidentally launched bisect.3 instead | |
Oct 23 17:00:34 <Grum> hmmm | |
Oct 23 17:00:38 <Grum> lol how to rollback! | |
Oct 23 17:00:54 <barneygale> :/ sorry, you're doing all this hard work and I messed it up | |
Oct 23 17:01:27 <barneygale> bisect.5 doesn't work either, as you might expect | |
Oct 23 17:02:38 <Grum> building again | |
Oct 23 17:02:46 <Grum> i just redid the bisecting | |
Oct 23 17:02:48 <Grum> the process is stable | |
Oct 23 17:02:54 <barneygale> yep | |
Oct 23 17:02:55 <Grum> so i just had to make sure i did the right amount of 'good'-s :D | |
Oct 23 17:03:00 <barneygale> :D | |
Oct 23 17:03:10 <Grum> 3xgood yo! | |
Oct 23 17:03:18 <dx> there's git bisect log and git bisect replay | |
Oct 23 17:03:26 <Grum> yeah i failed to notice log | |
Oct 23 17:03:51 <Grum> i did replay, didn't have log ... well ok .. reset ... scroll, copy/paste 3x .. done ;D | |
Oct 23 17:03:58 <dx> lol | |
Oct 23 17:04:10 <Grum> i'll check the log after we finish for the science | |
Oct 23 17:04:16 <dx> yay science | |
Oct 23 17:05:13 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.6.jar | |
Oct 23 17:06:33 <barneygale> Grum: hang | |
Oct 23 17:06:44 <Grum> k | |
Oct 23 17:06:59 <Grum> erikbroes@grumm:~/src/mojang/Minecraft ((a52bd74...)|BISECTING)$ git bisect bad | |
Oct 23 17:07:00 <Grum> Bisecting: 0 revisions left to test after this (roughly 1 step) | |
Oct 23 17:07:00 <Grum> [70f530deeff38044f769b66982526ec0be368213] Don't let players silktouch lamps | |
Oct 23 17:07:06 <dx> wat. | |
Oct 23 17:07:09 <barneygale> wat :/ | |
Oct 23 17:07:13 <Grum> so close! ;D | |
Oct 23 17:07:22 <Grum> the bug was actually the silktouching of lamps that are *ON* | |
Oct 23 17:07:30 <Grum> you can silk-touch them, they just go off ;) | |
Oct 23 17:08:06 <dx> and... linux graphics initialization is a metaphorical lamp that gets silk-touched? | |
Oct 23 17:08:09 * mapppum (~mappum@c-67-185-176-162.hsd1.wa.comcast.net) has joined #mcdevs | |
Oct 23 17:08:12 <barneygale> Surely there's a commit in there somewhere that fixes the shrunken menu? | |
Oct 23 17:08:14 <Grum> no idea | |
Oct 23 17:08:20 <Grum> barneygale: yeah | |
Oct 23 17:08:22 <Grum> but i think its after | |
Oct 23 17:08:32 <Grum> the menu is not the problem | |
Oct 23 17:08:46 <barneygale> Uhg... | |
Oct 23 17:08:49 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.7.jar | |
Oct 23 17:09:29 <dx> so much suspense | |
Oct 23 17:09:38 <Grum> one more to go if this one is bad i think | |
Oct 23 17:11:20 <nickelpro> this is better than football | |
Oct 23 17:11:42 <dx> a lot of things are better than football | |
Oct 23 17:11:47 <barneygale> Grum: bad | |
Oct 23 17:11:51 * mappum has quit (Ping timeout: 272 seconds) | |
Oct 23 17:12:07 <Grum> $ git bisect bad | |
Oct 23 17:12:07 <Grum> Bisecting: 0 revisions left to test after this (roughly 0 steps) | |
Oct 23 17:12:08 <Grum> [0653bbf1f74dc2fe6c727436359309974114ece2] Added a slider for view distance | |
Oct 23 17:12:11 <Grum> cleaning ass again! | |
Oct 23 17:12:48 <barneygale> Is that it? | |
Oct 23 17:12:53 <barneygale> Nothing else to try? | |
Oct 23 17:13:01 <dx> he's building | |
Oct 23 17:13:02 <Grum> building | |
Oct 23 17:13:05 <barneygale> Oh, sorry | |
Oct 23 17:13:06 <Grum> 'clean ass' is building :p | |
Oct 23 17:13:13 <Grum> gradle clean ass(emble) | |
Oct 23 17:13:20 <barneygale> hah! | |
Oct 23 17:13:21 <nickelpro> heh | |
Oct 23 17:13:31 <Grum> yeah i have a little pun every time i make a build for you lot =) | |
Oct 23 17:14:05 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.8.jar | |
Oct 23 17:14:58 <barneygale> Grum: bad | |
Oct 23 17:15:24 <dx> uh oh. | |
Oct 23 17:15:47 <Grum> oh god | |
Oct 23 17:15:51 <Grum> 0653bbf1f74dc2fe6c727436359309974114ece2 is the first bad commit | |
Oct 23 17:15:51 <Grum> commit 0653bbf1f74dc2fe6c727436359309974114ece2 | |
Oct 23 17:15:51 <Grum> Author: Nathan Adams <dinnerbone@dinnerbone.com> | |
Oct 23 17:15:52 <Grum> Date: Thu Sep 19 15:53:28 2013 +0200 | |
Oct 23 17:15:54 <Grum> Added a slider for view distance | |
Oct 23 17:15:56 <Grum> that makes *NO* SENSE | |
Oct 23 17:16:11 <nickelpro> I'm falling out of my seat | |
Oct 23 17:16:55 <Grum> errr | |
Oct 23 17:17:19 <nickelpro> We know for sure the commit prior to it is good? Tested in the bisect? | |
Oct 23 17:17:20 <dx> science time? | |
Oct 23 17:17:27 <barneygale> :/ | |
Oct 23 17:17:47 <dx> nickelpro: actually nope | |
Oct 23 17:17:50 <dx> we got two bads in a row | |
Oct 23 17:18:18 <Grum> 3 | |
Oct 23 17:18:38 <Grum> i dont trust this at all | |
Oct 23 17:18:41 <barneygale> I'm going to double-check everything | |
Oct 23 17:19:01 <barneygale> I may have made a mistake earlier on that I didn't notice (launched the wrong version) | |
Oct 23 17:19:24 <barneygale> Fat lot of fucking help I am. | |
Oct 23 17:19:29 <Grum> nah its ok | |
Oct 23 17:19:32 <Grum> this a fickle process | |
Oct 23 17:20:13 <dx> so.. new bisect starting from the last confirmed bad to the last confirmed good? | |
Oct 23 17:20:32 <Grum> yup | |
Oct 23 17:20:35 <Grum> finding the shas :D | |
Oct 23 17:20:46 <dx> have fun! | |
Oct 23 17:21:05 <nickelpro> I would have really loved if it _was_ the sliders | |
Oct 23 17:21:05 <Grum> if what you did was correct barneygale; then http://db.grum.nl/minecraft.linux.bisect.9.jar -- should be 'good' | |
Oct 23 17:21:20 <Grum> as that is the commit right before the 'one first bad' | |
Oct 23 17:21:25 * barneygale crosses fingers | |
Oct 23 17:21:44 <dx> if .9 is good then it's the sliders? lol | |
Oct 23 17:22:10 <Grum> yes o.O | |
Oct 23 17:22:14 <nickelpro> HAH | |
Oct 23 17:22:26 <barneygale> Grum: good | |
Oct 23 17:22:31 <barneygale> ... | |
Oct 23 17:22:32 <dx> .. | |
Oct 23 17:22:34 <nickelpro> Oh lord | |
Oct 23 17:22:35 <Grum> ok | |
Oct 23 17:22:40 <Grum> looking into that commit | |
Oct 23 17:22:44 <dx> well at least we don't have to bisect again | |
Oct 23 17:22:55 <Grum> you run 64bit java and shit? | |
Oct 23 17:23:09 <barneygale> Java HotSpot(TM) 64-Bit Server VM (build 24.0-b56, mixed mode) | |
Oct 23 17:23:46 <Grum> dafuq | |
Oct 23 17:23:50 <Grum> this has nothing to do with anything | |
Oct 23 17:24:07 <barneygale> Which bisect is the one immediately after that? I'll double-check | |
Oct 23 17:24:46 <Grum> errr | |
Oct 23 17:24:56 <dx> 8 = slider and 9 = slider^? | |
Oct 23 17:25:16 <Grum> pfft hard to figure that out after the fact >.> | |
Oct 23 17:25:23 <barneygale> Yah | |
Oct 23 17:26:22 <barneygale> 8 definitely doesn't work | |
Oct 23 17:27:44 <Grum> 8 hangs, 9 works? | |
Oct 23 17:27:52 <barneygale> Yes | |
Oct 23 17:29:02 <barneygale> Wait, wtf | |
Oct 23 17:29:06 <barneygale> Hold on a minute :/ | |
Oct 23 17:29:26 <Grum> i hope you found something that went wrong | |
Oct 23 17:29:32 <Grum> :p | |
Oct 23 17:29:48 <barneygale> Nope, I was right. 8 hangs, 9 works | |
Oct 23 17:30:02 <barneygale> I just re-downloaded them and double-checked I was launching the right version | |
Oct 23 17:30:06 <Grum> ok | |
Oct 23 17:30:10 <Grum> testing something | |
Oct 23 17:30:13 <barneygale> I've been up for a long time now so don't trust myself with this. | |
Oct 23 17:31:04 <Grum> i found one thing that might be causing an issue | |
Oct 23 17:31:10 <Grum> but .... it has nothing to do with lwjgl :P | |
Oct 23 17:31:13 <Grum> building | |
Oct 23 17:31:24 <Grum> this will be 8 with a tiny change | |
Oct 23 17:33:23 <Grum> - is64bit = is64Bit(); | |
Oct 23 17:33:24 <Grum> + is64bit = false; //is64Bit(); | |
Oct 23 17:34:32 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.10.jar | |
Oct 23 17:35:11 <barneygale> Grum: bad | |
Oct 23 17:35:34 <barneygale> Uhg, that's not good news is it? | |
Oct 23 17:38:27 <md_5> io.netty.handler.codec.DecoderException: java.lang.IndexOutOfBoundsException: readerIndex(3) + length(109) exceeds writerIndex(6): UnpooledHeapByteBuf(ridx: 3, widx: 6, cap: 6) | |
Oct 23 17:38:35 <md_5> I keep getting a too small handshake :c | |
Oct 23 17:38:52 <md_5> at net.md_5.bungee.protocol.DefinedPacket.readString(DefinedPacket.java:23) | |
Oct 23 17:39:09 <md_5> maybe strings are still shorts not varint | |
Oct 23 17:39:19 <Grum> no, port is short | |
Oct 23 17:39:33 <md_5> no, but Strings in general | |
Oct 23 17:39:40 <md_5> hm | |
Oct 23 17:39:43 <md_5> writerIndex is 6 | |
Oct 23 17:39:47 <md_5> how can a handshake be that short | |
Oct 23 17:40:16 <Grum> barneygale: building a new one, found some oddity | |
Oct 23 17:40:21 <Grum> but meh this is feeling rather hopeless :/ | |
Oct 23 17:40:34 <barneygale> I agree. Don't put yourself out too much | |
Oct 23 17:40:58 <barneygale> Are we sure 8 and 9 differ by only that commit? | |
Oct 23 17:41:18 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.11.jar | |
Oct 23 17:42:10 <dx> what's this one? | |
Oct 23 17:42:22 <Grum> fixed some weird oddity in the code | |
Oct 23 17:42:28 <dx> yay oddities | |
Oct 23 17:42:43 <barneygale> Grum: sadly this is bad too | |
Oct 23 17:42:44 <md_5> Grum fix chat on 1.7 | |
Oct 23 17:42:48 <md_5> colours dont carry over the line | |
Oct 23 17:42:55 <Grum> it works | |
Oct 23 17:42:58 <Grum> you just use old shit | |
Oct 23 17:43:00 <Grum> that doesnt work | |
Oct 23 17:43:04 <Grum> TOUGH COOKIE | |
Oct 23 17:43:26 <dx> rip § | |
Oct 23 17:43:29 <dx> will be missed | |
Oct 23 17:43:45 <Grum> goddammit, i have no idea what is causing this | |
Oct 23 17:44:27 <dx> can we have another build of 0653bbf1f74dc2fe6c727436359309974114ece2^ (slider^) just as a sanity check? | |
Oct 23 17:45:21 <Grum> making another build | |
Oct 23 17:46:12 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.12.jar | |
Oct 23 17:46:36 <barneygale> I'll be 2 mins | |
Oct 23 17:48:33 <barneygale> Grum: I'm just checking out jconsole. For non-working builds, when it hangs, "PS Survivor Space" is at 100% while the others are at zero. For the working version, PS Survivor Space is at ~30% and the others (PS Eden and PS Old Gen) are at 1-5%. | |
Oct 23 17:48:48 <barneygale> I don't know if this is relevant, I don't know much about Java memory management. | |
Oct 23 17:48:54 <Grum> neither do i | |
Oct 23 17:49:27 <barneygale> OK. Trying your latest version in a sec | |
Oct 23 17:50:26 * TomyLobo has quit (Quit: Standby mode...) | |
Oct 23 17:50:40 <barneygale> Grum: bad | |
Oct 23 17:50:47 <Grum> good, that was a trick | |
Oct 23 17:50:53 <Grum> that was 8 ;) -- or *REALLY* close to that | |
Oct 23 17:50:56 <Grum> so lets see wtf happened | |
Oct 23 17:51:19 <barneygale> I'll check 8 again | |
Oct 23 17:51:32 <Grum> something is fishy | |
Oct 23 17:51:36 <Grum> meh tripply checking | |
Oct 23 17:51:40 <Grum> its not exactly 8, it should have been | |
Oct 23 17:51:55 <dx> slider^ plz | |
Oct 23 17:52:03 <dx> PLS | |
Oct 23 17:52:09 <Grum> that is what this was | |
Oct 23 17:52:15 <Grum> i had all the changes in the commit stashed | |
Oct 23 17:52:19 <dx> ...what | |
Oct 23 17:52:23 <Grum> yeah | |
Oct 23 17:52:27 <Grum> so something is up | |
Oct 23 17:52:36 <dx> slider^ is what 9 was supposed to be | |
Oct 23 17:53:07 <dx> if slider^ is bad then we need to run another git bisect | |
Oct 23 17:53:27 <Grum> could be the effect isnt stable | |
Oct 23 17:53:42 <dx> hm, true | |
Oct 23 17:54:14 <dx> barneygale: try 9 more | |
Oct 23 17:54:22 <barneygale> repeatedly? | |
Oct 23 17:54:28 <dx> i guess! | |
Oct 23 17:54:33 <barneygale> OK | |
Oct 23 17:58:13 <barneygale> Ran 10 times. Worked every time | |
Oct 23 17:58:23 <barneygale> I can do the same thing for 8? | |
Oct 23 17:58:48 <Grum> i'm just going to make you two fresh builds :P | |
Oct 23 17:59:23 <barneygale> This is one mysterious bug. | |
Oct 23 18:00:00 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.13.jar | |
Oct 23 18:01:22 <barneygale> Grum: bad | |
Oct 23 18:01:52 <barneygale> back in 2-3 mins | |
Oct 23 18:03:28 <Grum> barneygale: http://db.grum.nl/minecraft.linux.bisect.14.jar | |
Oct 23 18:05:55 <barneygale> Grum: good | |
Oct 23 18:06:02 <Grum> ok sigh | |
Oct 23 18:06:05 <Grum> confirmed at the same point | |
Oct 23 18:06:12 <Grum> i need to fragment the commit that causes it | |
Oct 23 18:06:16 <Grum> maybe we can figure out wtf causes it | |
Oct 23 18:07:04 <barneygale> We can certainly take a break and come back to it another day if you'd like? This isn't much fun. | |
Oct 23 18:10:24 <Grum> i dont care | |
Oct 23 18:10:30 <Grum> if you need to sleep, go sleep :D | |
Oct 23 18:10:43 <barneygale> I'll be here for another 20 minutes | |
Oct 23 18:11:26 <dx> Grum: same point? slider? | |
Oct 23 18:11:45 <Grum> yes | |
Oct 23 18:11:52 <Grum> minimizing commit-diffs now | |
Oct 23 18:12:30 <dx> so apparently 14 == 12 == 9 == slider^ | |
Oct 23 18:12:39 <Grum> yup :p | |
Oct 23 18:12:39 <dx> and they are good, bad, good respectively | |
Oct 23 18:12:49 <Grum> mm i think 12 fubarred somehow | |
Oct 23 18:12:58 <barneygale> I'll check 12 again | |
Oct 23 18:14:29 <barneygale> 12 confirmed bad | |
Oct 23 18:15:56 <Grum> ok making a build | |
Oct 23 18:16:07 <Grum> this is the 'bad' commit but completely minimized | |
Oct 23 18:17:50 <Grum> barneygale: http://db.grum.nl/minecraft.linux.try.1.jar | |
Oct 23 18:18:54 <barneygale> Grum: bad | |
Oct 23 18:19:14 <Grum> expected; building new | |
Oct 23 18:19:24 <Grum> each build will have a really tiny little piece extra :/ | |
Oct 23 18:19:51 <barneygale> You have more patience than I | |
Oct 23 18:19:53 <dx> shouldn't it have less? | |
Oct 23 18:20:06 <Grum> nah easier to work towards fail | |
Oct 23 18:20:18 <dx> but we're at fail | |
Oct 23 18:21:11 <Grum> barneygale: http://db.grum.nl/minecraft.linux.try.2.jar | |
Oct 23 18:21:13 <Grum> that one *should* work :P | |
Oct 23 18:21:18 <Grum> i mean, it only removes code =) | |
Oct 23 18:21:52 <barneygale> Grum: good | |
Oct 23 18:21:58 <dx> whoa | |
Oct 23 18:22:15 <SinZ> what changed between 1 and 2 | |
Oct 23 18:22:16 <Grum> oooooh | |
Oct 23 18:22:21 <Grum> no | |
Oct 23 18:22:26 <barneygale> hah! | |
Oct 23 18:22:28 <Grum> this is just like 5% of the commit that caused fail | |
Oct 23 18:22:34 <Grum> but i think i know what it is | |
Oct 23 18:22:52 <dx> apt-get install sliders | |
Oct 23 18:23:13 <dx> those sneaky external dependencies! | |
Oct 23 18:24:24 <Grum> god this is really entangled, sec | |
Oct 23 18:25:29 <Grum> building | |
Oct 23 18:25:44 <dx> how big is that commit diff, btw? | |
Oct 23 18:25:49 <Grum> not big anymore | |
Oct 23 18:25:52 <Grum> but its entangled | |
Oct 23 18:25:54 <dx> lol | |
Oct 23 18:25:56 <Grum> we used the view distance everywhere | |
Oct 23 18:26:01 <Grum> and it was 'inverted' | |
Oct 23 18:26:04 <Grum> higher numbers was smaller range | |
Oct 23 18:26:40 <dx> oh, so it's like a refactor commit | |
Oct 23 18:26:51 <Grum> we sanified the viewdistance | |
Oct 23 18:27:00 <Grum> but! i just noticed it is setting glfog | |
Oct 23 18:27:06 <Grum> so maybe we're going over the range of the min/max | |
Oct 23 18:27:11 <Grum> so i just set it to 0-1 now | |
Oct 23 18:27:15 <barneygale> :o | |
Oct 23 18:27:19 <barneygale> this sounds promising | |
Oct 23 18:27:25 <Grum> barneygale: http://db.grum.nl/minecraft.linux.try.3.jar | |
Oct 23 18:27:28 <Grum> got i hope that works | |
Oct 23 18:27:31 <Grum> *god | |
Oct 23 18:28:12 <barneygale> :( it doesn't | |
Oct 23 18:28:20 <Grum> ok | |
Oct 23 18:29:27 <Grum> got another target | |
Oct 23 18:29:27 <Grum> sec | |
Oct 23 18:31:00 <Grum> hahahahhaa | |
Oct 23 18:31:00 * mapppum has quit (Ping timeout: 245 seconds) | |
Oct 23 18:31:05 <Grum> this MUST be it! | |
Oct 23 18:31:19 * SinZ waits to see if it is | |
Oct 23 18:31:22 <barneygale> I really hope so | |
Oct 23 18:31:23 <Grum> hahahah | |
Oct 23 18:31:31 <Grum> i'm actually rather sure about it | |
Oct 23 18:31:37 <Grum> god glaring error >.> | |
Oct 23 18:31:40 <SinZ> what is it? | |
Oct 23 18:31:42 <barneygale> don't say that, you'll jinx it | |
Oct 23 18:31:55 <Grum> barneygale: thanks so much for this bisect | |
Oct 23 18:32:06 <dx> !! | |
Oct 23 18:32:10 <dexter0> the suspense | |
Oct 23 18:32:11 <barneygale> Grum: thank you so much for doing this! | |
Oct 23 18:32:16 <SinZ> Grum: just add him to credits.txt or whatever | |
Oct 23 18:32:17 <Grum> it *IS* minecraft doing a wrong thing | |
Oct 23 18:32:31 <barneygale> :O | |
Oct 23 18:32:38 <barneygale> Lets not count our chickens before they've hatched | |
Oct 23 18:32:43 <Grum> goddammit finish building already | |
Oct 23 18:32:48 <SinZ> either way, its cleaning the codebase | |
Oct 23 18:32:50 <Grum> so i can checkout master | |
Oct 23 18:32:54 <Grum> and do the change there :D | |
Oct 23 18:32:57 * mappum (~mappum@c-67-185-176-162.hsd1.wa.comcast.net) has joined #mcdevs | |
Oct 23 18:33:12 <SinZ> Grum: so are we getting a 1.7.1 straight away? | |
Oct 23 18:33:18 <Grum> barneygale: http://db.grum.nl/minecraft.linux.try.4.jar <-- works :D | |
Oct 23 18:33:59 <barneygale> Grum: good! | |
Oct 23 18:34:00 <barneygale> :D | |
Oct 23 18:34:03 <dx> :D | |
Oct 23 18:34:06 <SinZ> :D | |
Oct 23 18:34:08 <barneygale> What was it? | |
Oct 23 18:34:09 <Grum> already building 1.7 with that fix | |
Oct 23 18:34:17 <Grum> yeah erm .... no :P | |
Oct 23 18:34:19 <Grum> i'm not saying! | |
Oct 23 18:34:20 <Grum> =D | |
Oct 23 18:34:27 <dexter0> :( | |
Oct 23 18:34:29 <dx> glFuckShitUp(); | |
Oct 23 18:34:29 <Grum> hehe | |
Oct 23 18:34:32 <Grum> nah | |
Oct 23 18:34:33 <SinZ> atleast change the ticket to resolved | |
Oct 23 18:34:33 <barneygale> oh no way! | |
Oct 23 18:34:37 <Grum> MathShitUp | |
Oct 23 18:34:42 <barneygale> It's dinnerbone's commit, we can blame him! | |
Oct 23 18:34:47 <Grum> you can | |
Oct 23 18:34:51 <Grum> though i think i was looking at his screen at the time | |
Oct 23 18:34:54 <Grum> and i also missed it | |
Oct 23 18:35:03 <SinZ> and took awhile to find the derp | |
Oct 23 18:35:07 <dx> it's okay, you can tell us, we promise not to laugh | |
Oct 23 18:35:10 <SinZ> ^ | |
Oct 23 18:35:11 <Grum> i will | |
Oct 23 18:35:13 <Grum> just gimme a sec | |
Oct 23 18:35:19 <Grum> first want to give him a working 1.7 | |
Oct 23 18:35:23 <Grum> and make sure it works | |
Oct 23 18:35:29 <Grum> maybe we do another derptiderp after | |
Oct 23 18:35:40 <barneygale> technical term | |
Oct 23 18:35:40 <SinZ> #mcdevs++ | |
Oct 23 18:35:46 <Grum> thinking about it, its bloody amazing that any people can run this | |
Oct 23 18:35:52 <barneygale> hahahahaha | |
Oct 23 18:36:00 <barneygale> linux: more correct than windows. | |
Oct 23 18:36:08 <dx> *some linux | |
Oct 23 18:36:11 <Grum> barneygale: how much ram do you have? ;D | |
Oct 23 18:36:20 <Grum> barneygale: http://db.grum.nl/minecraft.linux.fixed.1.7.jar | |
Oct 23 18:36:30 <Grum> omg the confidence! | |
Oct 23 18:36:41 <Grum> ok so the diff: | |
Oct 23 18:36:44 <Grum> - int maxChunksWidth = (MAX_VIEW_DISTANCE + 1) * 2; // Why +1? | |
Oct 23 18:36:44 <Grum> + int maxChunksWidth = (Options.RENDER_DISTANCE_EXTREME + 1) * 2; // Why +1? | |
Oct 23 18:37:06 <Grum> and from that you can guess the error ;) | |
Oct 23 18:37:11 <dx> ..the reverse diff, right? | |
Oct 23 18:37:18 <Grum> no this is the one i have now | |
Oct 23 18:37:19 <barneygale> Grum: 6GB | |
Oct 23 18:37:20 <SinZ> please tell me EntityBoat's onUpdate is better than this now http://pastebin.com/c61PBjSq | |
Oct 23 18:37:50 <Grum> barneygale: videocard ram? | |
Oct 23 18:38:10 <barneygale> Give me a minute, I'll just test this first | |
Oct 23 18:38:24 <Grum> someone gimme the ticketnumber :P | |
Oct 23 18:38:38 <Grum> ah found https://mojang.atlassian.net/browse/MC-31714 | |
Oct 23 18:38:38 <barneygale> https://mojang.atlassian.net/browse/MC-31714 | |
Oct 23 18:39:07 * mappum has quit (Ping timeout: 240 seconds) | |
Oct 23 18:39:08 <barneygale> Grum: good! :DDD | |
Oct 23 18:39:25 <Grum> whooooo! :D | |
Oct 23 18:39:29 <dx> \o/ | |
Oct 23 18:39:34 <Grum> \o/ \o/ \o/ \o/ \o/ | |
Oct 23 18:39:45 <SinZ> \o/\o/\o/\o/\o/\o/\o/\o/\o/ | |
Oct 23 18:39:48 <SinZ> Lets hold hands | |
Oct 23 18:39:50 <Grum> anyhow, so what we did | |
Oct 23 18:39:52 <barneygale> Grum: you're an absolute hero! I'll be playing minecraft for many months yet :) | |
Oct 23 18:40:24 <Grum> we accidentally tried to allocate ((512 + 1) * 2) * ((512 + 1) * 2) * 16 glLists O.o | |
Oct 23 18:40:32 <barneygale> hahahahaha | |
Oct 23 18:40:43 <Grum> instead of: ((16 + 1) * 2) * ((16 + 1) * 2) * 16 ;) | |
Oct 23 18:40:50 <dx> dammit barneygale, you broke the promise | |
Oct 23 18:40:59 <AndrewPH> / Why +1? | |
Oct 23 18:41:03 <AndrewPH> //* | |
Oct 23 18:41:12 <barneygale> dx: which promise? | |
Oct 23 18:41:19 <dx> 04:35 < dx> it's okay, you can tell us, we promise not to laugh | |
Oct 23 18:41:29 <barneygale> Oh, foo | |
Oct 23 18:41:35 <Grum> so ... 16842816 instead of 18496 | |
Oct 23 18:41:43 <SinZ> ouch | |
Oct 23 18:41:47 * kcj has quit (Ping timeout: 240 seconds) | |
Oct 23 18:41:51 <barneygale> I wonder why it worked on windows etc | |
Oct 23 18:42:08 <Grum> for that it has to allocate ~64mb of ram to even request this from the videocard | |
Oct 23 18:42:08 <dexter0> I suspect it works on os x b/c os x can page cram. | |
Oct 23 18:42:14 <dexter0> page vram | |
Oct 23 18:42:23 <Grum> nah, i think drivers work differently | |
Oct 23 18:42:28 <Grum> they do not actually have to allocate it | |
Oct 23 18:42:32 <Grum> they just have to block it | |
Oct 23 18:42:40 <Grum> and then give back an array of that amount of unique numbers | |
Oct 23 18:42:52 <Grum> if they do it smart, they just give back the list in a block | |
Oct 23 18:44:16 <Grum> this was nice and hidden in the commit :P |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment