01:20 < ws-client> **<ChillerDragon>** how is NOT FINISHED related to that
01:20 < ws-client> **<ChillerDragon>** the screenshot i showed shows non 0 scores for all players
01:20 < ws-client> **<ChillerDragon>** nothing related to finishes
01:21 < ws-client> **<ChillerDragon>** if there is no timescore it should not break sorting even if not finished is sent @Assa 
01:21 < ws-client> **<ChillerDragon>** --
01:21 < ws-client> **<ChillerDragon>** yo wtf is https://github.com/NVIDIA/open-gpu-kernel-modules
01:21 < ws-client> **<ChillerDragon>** nvidia W?
01:21 < ws-client> **<ChillerDragon>** no more noveueueuu needed?
01:21 < ws-client> **<ChillerDragon>** nvidia open source company
01:35 < bridge> <.dartmonkey> It's not really full open source but it's really great
01:35 < bridge> <.dartmonkey> This is the future of NVIDIA drivers on Linux
01:36 < bridge> <.dartmonkey> The only problem is that it only supports recent cards
01:42 < bridge> <.dartmonkey> Not to mention that it's a kernel module, so it's a bit more of a hassle to install and takes up more space on the EFI partition
02:46 < bridge> <kínverskursamkynhneigður> rare chiller latency
02:48 < bridge> <kínverskursamkynhneigður> but ye, they did it cuz they can now. since newer cards have  proprietary core logic in the firmware now, and features stuff are in their user-space driver which is still proprietary.
02:48 < bridge> <kínverskursamkynhneigður> but ye, they did it cuz they can now. since newer cards have proprietary core logic in the firmware now, and features stuff are in their user-space driver which is still proprietary.
02:49 < bridge> <kínverskursamkynhneigður> also probably why they don't support older cards? not sure about  it tho
02:51 < bridge> <kínverskursamkynhneigður> but they do seem to help NVK (the open source user space driver) now too.
04:26 < bridge> <Assa> If the server sends finish times it should.
04:29 < bridge> <Assa> You are just not following protocol of this netmessage
04:32 < bridge> <Assa> Please use UNSET, which is literally what this is for
04:44 < bridge> <Assa> We now have defacto 2 scores we can sort by. The server is sending both - for legacy reasons. And sorting by the new Message should be prefered due to #11363 and the -9999 timescore bug. Old servers don't send this message making it sort by score. You are sending it and thus it sorts by it.
04:44 < bridge> <DDNet> https://github.com/ddnet/ddnet/issues/11363
04:47 < bridge> <Assa> Yes it should Display this - and if the display PR is merged, ddnet insta will show no scores, because nobody finished. Fix your server.
05:06 < bridge> <Assa> I opened an upstream issue in ddnet insta for you - with my phone since I am not home
07:17 < chillerbot12> Elo @Assa I just wanted to remind you that: check if the server saved your time on run_Jellyfish_Fields
07:18 < bridge> <louis> good map
08:42 < bridge> <kínverskursamkynhneigður> @0xdeen duplicated entries. probably
08:42 < bridge> <kínverskursamkynhneigður> https://cdn.discordapp.com/attachments/293493549758939136/1452204651817599037/image.png?ex=6948f679&is=6947a4f9&hm=f1064f3bca014386298eb82818ed35cd33f931399fdc648ed442226e4860c12f&
08:43 < bridge> <kínverskursamkynhneigður> from here: `https://ddnet.org/players/?json2=Cor`
08:47 < bridge> <byfox> In fact, there are actually two map.
08:47 < bridge> <byfox> 
08:47 < bridge> <byfox> I also noticed that the site can't display maps with the same name.
08:47 < bridge> <kínverskursamkynhneigður> oh these are two different maps?
08:48 < bridge> <kínverskursamkynhneigður> or do you mean there is another map that have multiples
08:48 < bridge> <byfox> I completely understood the problem, the card was moved to event and they forgot to remove it from moderate
08:48 < bridge> <byfox> https://cdn.discordapp.com/attachments/293493549758939136/1452206155362205806/image.png?ex=6948f7e0&is=6947a660&hm=a2697614ffe6bbcc7502c72a0afdac621db59c6ee0811376d7c1bfd640891968&
08:49 < bridge> <kínverskursamkynhneigður> either way, probably just a database error deen should be able to fix it
08:49 < bridge> <kínverskursamkynhneigður> it broke some of my tools tho lol
08:49 < bridge> <kínverskursamkynhneigður> and probably made point calculation a bit of wack for a while. people would lose 9 points over this, rip.
09:59 < bridge> <melon> oh nurr D:
09:59 < bridge> <melon> i even triplechecked after we missed a few, when the new category was introduced
09:59 < bridge> <melon> sadge i missed that
10:02 < bridge> <Assa> should have quadruple checked :justatest:
12:09 < bridge> <learath2> @0xdeen || @davide55 do we have some sort of rule on ger10 that breaks long running ssh connections? It keeps killing my gdb session 🙁
12:12 < bridge> <yabarmori> @heinrich5991
12:19 < bridge> <Assa> as a workaround use a screen session? :justatest:
12:20 < bridge> <learath2> I have a feeling it’s a veeery low level thing just keeping an eye on traffic
12:20 < bridge> <learath2> Every time it happens I have to think about glibc malloc internals again to navigate to where I was 😛
12:29 < bridge> <Assa> 4-in-the-morning-Assa is so unfriendly, missing sleep no good
12:29 < bridge> <learath2> 280 bytes of random garbage written into the heap, I wonder what this is
12:30 < bridge> <louis> My fault was playing around in there
12:30 < bridge> <learath2> Do not be doing that in the future please
12:30 < bridge> <learath2> I have to keep fiddling with the arrow key so the great firewall of ger10 does not shoot me down
12:32 < bridge> <learath2> I see what did that 😄
12:32 < bridge> <learath2> That did look a little too ascii to be random bs
12:48 < bridge> <Assa> ger10 still exists?
12:49 < bridge> <Assa> I an sure I only get Ger1 in the serverbrowser lately
12:49 < bridge> <learath2> ta da
12:49 < bridge> <learath2> https://cdn.discordapp.com/attachments/293493549758939136/1452266638798880920/image.png?ex=69493034&is=6947deb4&hm=97340b7d91bf4c663edcc0d7874f69f1521b258171f0305e191e020a15dc6a26&
12:49 < bridge> <learath2> I really enjoy debugging weird stuff like this, is there a job for this?
12:50 < bridge> <aegisub> so true
12:50 < bridge> <aegisub> I need one too
12:50 < bridge> <aegisub> or well my favorite kind of debugging right now is "why is this video file broken"
12:51 < bridge> <learath2> Only god himself knows what noby was thinking when using a fixed size heap allocation for a string that did not even need to be on the heap
12:58 < bridge> <learath2> Also which one of you yoinked our `clamp` function out of existence?
13:01 < bridge> <learath2> #10306
13:01 < bridge> <DDNet> https://github.com/ddnet/ddnet/pull/10306
13:03 < bridge> <Assa> why do you need your own one?
13:04 < bridge> <learath2> I didn't but the antibot was linking to it 😄
13:05 < bridge> <learath2> I didn't but the antibot was getting it there too 😄
13:15 < bridge> <melon> :monkaS:  so it wasnt an issue on my end
13:15 < bridge> <melon> :monkaS: :monkaS: :monkaS:
13:20 < bridge> <learath2> Can you be here in the next 15 minutes? I'll need a quick review for the antibot
13:21 < bridge> <melon> if you give me 20, sure
13:22 < bridge> <learath2> I just need it merged within like 30 minutes or so
13:22 < bridge> <melon> i'll slurp up that pizza faster then :kek: - ping me when you're ready
13:28 < bridge> <learath2> This is like a masters thesis on how not to do C
13:28 < bridge> <learath2> https://cdn.discordapp.com/attachments/293493549758939136/1452276574144692388/image.png?ex=69493975&is=6947e7f5&hm=5bee98db0aa335f858b303b726ad36d29e15442b2be29a1bcb734925bda6d54a&
13:31 < bridge> <louis> never doubt noby C
13:32 < bridge> <melon> noby broke so many rules i didnt dare to touch :kekw:
13:32 < bridge> <melon> one of my favourites was seeing `struct something *psomething = (struct something *)calloc(...);`
13:33 < bridge> <learath2> If you ever see `strcat` or `sprintf` in a C program, it is 99.99999% of the time insane.
13:33 < bridge> <Assa> textbook exploit
13:34 < bridge> <Assa> imagine sending the right inputs to RCE the antibot :kek:
13:34 < bridge> <learath2> I hadn't seen the one I am fixing until today. He was just allocating some char arrays on the heap, for no reason really
13:35 < bridge> <Assa> The serverbrowser was doing this as well, you can solve this with a unique ptr or a vector
13:41 < bridge> <learath2> No it wasn't, you have never in your life seen this pattern
13:42 < bridge> <learath2> I don't think anyone has ever done this before. This is probably the first time this construction has been used in the almost 3 decades C existed
13:44 < bridge> <learath2> @blaiszephyr actually no rush I want to do this last one properly, enough jank
13:44 < bridge> <melon> xd fair enough
13:45 < bridge> <melon> ill enjoy the rest of my pizza then
13:52 < ws-client> **<ChillerDragon>** @learath2 you link your antibot against the server binaries clamp???
14:06 < bridge> <Assa> :pepeW:
14:22 < bridge> <learath2> @blaiszephyr reviewpls
14:23 < bridge> <learath2> I kept the commits tiny, so it should be very enjoyable 😛
14:25 < bridge> <melon> on it :)
14:27 < bridge> <learath2> We will be closing a 5 year issue with this :Wow:
14:28 < bridge> <learath2> We will be closing a 5 year issue with this :NoelleWOW_Erizebett:
14:29 < bridge> <melon> reviewed
14:30 < bridge> <melon> i hereby request everyone to do their commits like you do
14:30 < bridge> <melon> that was the fastest i've ever managed to skim through changes in order
14:32 < bridge> <learath2> It does take a bit more thinking upfront to plan out your changes, not always possible 🙂
14:34 < bridge> <learath2> @blaiszephyr that array there can at best be a VLA, but I don't want to use VLAs, I like VLAs but most people hate VLAs
14:36 < bridge> <melon> i agree, fuck VLAs
14:36 < bridge> <melon> well, lgtm - my fault if all servers go into mental shutdown after
14:36 < bridge> <learath2> Fixed the `free` though, re-review?
14:36 < bridge> <melon> damn this change looks dangerous, better write a unit test for this
14:36 < bridge> <melon> https://cdn.discordapp.com/attachments/293493549758939136/1452293738847141991/image.png?ex=69494971&is=6947f7f1&hm=7af17232becb1ae66001b720c5252ccda2f0d7c1dd796cfb11d60fe2321495e9&
14:37 < bridge> <melon> /s - i'll compile it locally one last time, fiddle for 5 minutes and merge if i dont find anything
14:37 < bridge> <learath2> This looks terrifying
14:37 < bridge> <learath2> https://cdn.discordapp.com/attachments/293493549758939136/1452293843549552734/image.png?ex=6949498a&is=6947f80a&hm=64f43cc66c44ea56b02a0142a076ef512debd0109a3c358967e4711b694927ab&
14:37 < bridge> <vena6080> @essigautomat hey, i just got an idea, we are using tune for a message in the chat, but for the specific utilization i thought about a tune that make a broadcast (only for the player, not everyone), could be cool to add.
14:39 < bridge> <learath2> if you don't mind, can you also do a quick dump to see if I broke any of the strings just in case? You are more familiar with its output
14:39 < bridge> <learath2> It looked correct to my eye
14:42 < bridge> <melon> ye did that, they looked just fine to me - and our parser also doesnt go up in flames
14:42 < bridge> <melon> ye did that, they looked just fine to me - and our parser also didnt go up in flames
14:42 < bridge> <melon> ye did that, they looked just fine to me - and our parser didnt go up in flames
14:43 < bridge> <learath2> ping me when you merge, I'll compile and deploy it
14:43 < bridge> <melon> @learath2 merged
14:50 < bridge> <learath2> We have some amazing connections to some of our game servers
14:50 < bridge> <learath2> https://cdn.discordapp.com/attachments/293493549758939136/1452297258757521529/image.png?ex=69494cb8&is=6947fb38&hm=d321142f7995c164ac2bcc23b7e6b95a756239f92c25eabd6bf89b667004e6de&
14:51 < bridge> <melon> ah, feels like home
14:51 < bridge> <melon> "MOM DONT SHUT IT OFF I'VE BEEN INSTALLING THE DIABLO 2 PATCH FOR 5 HOURS NOW!"
14:52 < bridge> <melon> (the patch was like.. 3MB? xd)
14:52 < bridge> <learath2> Is there something wrong with the internet today btw? I have been having issues but I thought it was just my turkishinternet
14:52 < bridge> <learath2> but it seems the main ddnet server isn't having a better time 😄
14:53 < bridge> <melon> either AWS, Cloudflare or the third world war °-°, my internet is just fine at the moment xd
14:54 < ws-client> **<ChillerDragon>** @learath2 do you know `git add -p` ?
14:54 < ws-client> **<ChillerDragon>** u can write code first and split commits later
14:54 < bridge> <Assa> `tune_zone_enter_broadcast` :justatest: 
14:54 < bridge> <Assa> 
14:54 < bridge> <Assa> With envelope trigger tiles you could do even fancier things
14:55 < bridge> <learath2> Yes, but I use it through `git add -i` usually. Either way, you still do need to plan it out as it can only split lines, I sometimes want to refactor and fix something
14:55 < bridge> <vena6080> omg, when the trigger tiles will be finished ?
14:55 < bridge> <vena6080> omg, when will the trigger tiles will be finished ?
14:55 < bridge> <Assa> They ARE finished, I am waiting for Review
14:55 < bridge> <learath2> I will first do the refactor than the fix in that case. E.g. in this case that buffer both needed to become larger and it needed to become a stack buffer instead. So I first stackified it, then in the next commit I resized it
14:56 < bridge> <vena6080> but can we test it or map with it before the rls?
14:56 < bridge> <Assa> @vena6080 #11398
14:56 < bridge> <DDNet> https://github.com/ddnet/ddnet/pull/11398
14:57 < bridge> <Assa> yes of course and I even MUST do it due to heinrichs RFC
14:57 < bridge> <pipou_tw> Oh well, our map is already 9.8mb, can't wait for it :justatest:
14:58 < bridge> <Assa> I have a 37 MB map I'd submit with this tiles :justatest:
14:59 < bridge> <pipou_tw> Are you making a movie or what XD
15:00 < bridge> <Assa> <https://youtu.be/7WPTr4meLIY?si=IN0sO6uOYNmbCoYS> ... maybe :justatest:
15:01 < bridge> <pipou_tw> It's mad
15:10 < bridge> <Assa> this is a funmap tbf xD no discussion
15:10 < bridge> <learath2> @blaiszephyr can you do me a quick favor and check on an empty server if everything looks fine?
15:10 < bridge> <melon> already did, they look fine
15:11 < bridge> <melon> i'll recheck in ~30 mins after a restarted one gets a few players
15:11 < bridge> <louis> https://discord.com/channels/252358080522747904/342454474117218334/1452301364842594347 :justatest:
15:11 < bridge> <Assa> @vena6080 @pipou_tw do you want to test the tiles? They are entirely client sided and you might be able to Download it from the actions
15:12 < bridge> <vena6080> yes pls
15:13 < bridge> <Assa> @learath2 can you help this awesome gentletees? I am to skillchecked with phone :justatest:
15:15 < bridge> <Assa> does this link contain Downloads at the bottom for you? https://github.com/ddnet/ddnet/actions/runs/20007707142
15:19 < bridge> <melon> it only does when you're logged in @essigautomat
15:19 < bridge> <melon> people without a github account cant download artifacts
15:19 < bridge> <pipou_tw> It says download for ubuntu only
15:20 < bridge> <melon> 2 weeks ago, maybe they were hit with a cleanup - Assa rerun the CI
15:21 < bridge> <Assa> I am logged in, but the gh apps seems not to show it
15:21 < bridge> <Assa> I restarted the build workflow, it takes ~ 30 minutes ._.
15:22 < bridge> <Assa> I find the artifacts more trustworthy then anyone sending client zips
15:22 < bridge> <melon> absolutely xd
15:26 < bridge> <remakepower> Map magic
15:47 < bridge> <Assa> @pipou_tw @vena6080 Download should be available now under the same link
15:55 < bridge> <pipou_tw> Alright thanks, will try it
15:55 < bridge> <Assa> Please take a look at the description or testmap I provided, the env_zone command is not trivial
17:41 < bridge> <milkeeycat> :lol:
17:41 < bridge> <milkeeycat> https://cdn.discordapp.com/attachments/293493549758939136/1452340096085917818/aqyOL4p_460swp.png?ex=6949749e&is=6948231e&hm=a6e59a5ec0cacc1aaf7229bd24e454ef0f63fc8fbf1c0bbc11e1fe36e285fe35&
17:42 < bridge> <alw5> einstein
19:39 < bridge> <sonic (Solly)> @chillerdragon i do the write code then split quite often my ide helps me ^~^
23:47 < bridge> <learath2> Confirming this one was much harder, crashes due to heap overflows are soo annoying to diagnose
23:49 < bridge> <robyt3> Is #6843 also the same? :justatest:  No core though
23:49 < bridge> <DDNet> https://github.com/ddnet/ddnet/issues/6843
23:50 < bridge> <robyt3> Also
23:50 < bridge> <robyt3> Also #5668
23:50 < bridge> <robyt3> #5668
23:50 < bridge> <DDNet> https://github.com/ddnet/ddnet/issues/5668
23:52 < bridge> <learath2> Doesn't really look like it, but without a core impossible to say, I have been just tracking down what malloc is trying to do, then looking at the chunk it's messing with to see what's broken about it
23:53 < bridge> <learath2> I did fix 3-5 potential heap overflows within the antibot though, so some things may get better
23:55 < bridge> <learath2> With heap overflows backtraces are really not all that useful, unless we get lucky and the parameter to a good call isn't optimized out
23:56 < bridge> <learath2> With heap overflows backtraces are really not all that useful, unless we get lucky and the parameter to a useful call isn't optimized out (like p to `_int_free`, or `_int_free_merge_chunk`)