←2021-03-08 2021-03-09 2021-03-10→ ↑2021 ↑all
00:17:12 -!- Guest56 has quit (Quit: My MacBook has gone to sleep. ZZZzzz…).
00:29:33 -!- Guest56 has joined.
00:31:34 -!- dcristofani has joined.
00:53:44 -!- really21 has joined.
00:55:28 -!- Guest56_ has joined.
00:56:46 -!- Guest56 has quit (Ping timeout: 276 seconds).
01:05:00 <Guest56_> HELLO zzo38
01:05:11 <zzo38> Hello, what question today?
01:06:42 <Guest56_> Nice to see you around here, I didn't know you were on esoteric as well. I ocassionally read your posts on the Usenet. Is Sqlnetnews still being maintained?
01:07:49 -!- arseniiv has quit (Ping timeout: 260 seconds).
01:07:56 <zzo38> Mostly I am working on a different project (Free Hero Mesh) at this time, although later I may work on sqlnetnews more. If you have any contributions or suggeations, then you may write where I may read them.
01:08:32 <shachaf> What is Free Hero Mesh?
01:08:52 <zzo38> It is a puzzle game engine, for grid-based puzzle games such as sokoban and Hero Hearts.
01:09:20 <Guest56_> Sounds intresting - is it avaliable for download or something?
01:10:04 <zzo38> It is also meant to be compatible with puzzles from Everett Kaser's Hero Mesh (once they are converted), although it also has many new features too (including dynamic types, higher limits, etc).
01:10:27 <zzo38> Guest56_: There is a Fossil repository at: http://zzo38computer.org/fossil/heromesh.ui
01:11:15 <zzo38> It is in a state ready for testing, although it isn't close to complete yet.
01:12:33 -!- dcristofani has quit (Ping timeout: 264 seconds).
01:16:45 <zzo38> For now, I am trying to figure out what is causing some levels of the FALLING puzzle set with the $SeekerCloser class to not work properly.
01:22:24 <Guest56_> FYI: I am maintaning my own logs of this channel, alongside the ones above
01:22:26 <zzo38> Please tell me if you have any comments, suggestions, bug reports, etc.
01:23:26 <Guest56_> I will check your think out zzo
01:23:28 <Guest56_> thing
02:03:29 -!- FreeFull has quit.
02:06:53 <esowiki> [[User talk:Ais523]] https://esolangs.org/w/index.php?diff=81236&oldid=81123 * Trump Bot * (+790)
02:07:21 <esowiki> [[User:Sertdfyguhi]] https://esolangs.org/w/index.php?diff=81237&oldid=79175 * Sertdfyguhi * (+16) /* esolangs I made */
02:08:10 <esowiki> [[Plts]] https://esolangs.org/w/index.php?diff=81238&oldid=79172 * Sertdfyguhi * (-6) update github name
02:08:40 <esowiki> [[Ewpl]] https://esolangs.org/w/index.php?diff=81239&oldid=79181 * Sertdfyguhi * (-6) update github name
02:09:06 <esowiki> [[Dotter]] https://esolangs.org/w/index.php?diff=81240&oldid=79323 * Sertdfyguhi * (-6) update github name
02:09:25 <esowiki> [[User:Sertdfyguhi]] M https://esolangs.org/w/index.php?diff=81241&oldid=81237 * Sertdfyguhi * (-6) update github name
02:20:09 <nakilon> IRL pixel art language: https://en.wikipedia.org/wiki/Kufic
02:26:56 -!- craigo has joined.
02:40:54 -!- dcristofani has joined.
02:45:00 -!- Lord_of_Life_ has joined.
02:46:59 -!- Lord_of_Life has quit (Ping timeout: 260 seconds).
02:46:59 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
03:36:07 -!- hendursaga has quit (Ping timeout: 268 seconds).
03:38:57 -!- dcristofani has quit (Ping timeout: 264 seconds).
03:39:20 -!- hendursaga has joined.
03:42:14 -!- dcristofani has joined.
03:44:15 -!- sprock has joined.
03:50:29 -!- craigo_ has joined.
03:54:20 -!- int-e_ has joined.
03:55:00 -!- scoofy_ has joined.
03:55:06 -!- sprocklem has joined.
03:55:39 -!- ^_ has joined.
03:57:40 -!- lifthrasiir_ has joined.
03:58:25 -!- Bowserinator_ has joined.
03:59:24 -!- sprock has quit (*.net *.split).
03:59:24 -!- dcristofani has quit (*.net *.split).
03:59:25 -!- zzo38 has quit (*.net *.split).
03:59:25 -!- scoofy has quit (*.net *.split).
03:59:25 -!- int-e has quit (*.net *.split).
03:59:25 -!- V has quit (*.net *.split).
03:59:25 -!- craigo has quit (*.net *.split).
03:59:25 -!- really21 has quit (*.net *.split).
03:59:26 -!- mniip has quit (*.net *.split).
03:59:26 -!- Bowserinator has quit (*.net *.split).
03:59:26 -!- lifthrasiir has quit (*.net *.split).
03:59:55 -!- zzo38 has joined.
04:01:26 -!- mniip has joined.
04:01:33 -!- dcristofani has joined.
04:05:23 -!- Bowserinator_ has changed nick to Bowserinator.
04:11:35 -!- really2 has joined.
04:15:36 -!- delta23 has quit (Quit: Zzz).
04:44:29 -!- sprocklem has quit (Quit: ...).
05:05:58 -!- craigo__ has joined.
05:06:33 -!- tromp has quit (Remote host closed the connection).
05:08:44 -!- craigo_ has quit (Ping timeout: 260 seconds).
05:11:02 -!- dcristofani has quit (Ping timeout: 260 seconds).
05:11:46 <esowiki> [[Wenyan]] https://esolangs.org/w/index.php?diff=81242&oldid=81218 * ColorfulGalaxy * (+96)
05:16:18 -!- dcristofani has joined.
05:43:32 -!- tromp has joined.
05:43:50 -!- dcristofani has quit (Ping timeout: 264 seconds).
05:47:58 -!- tromp has quit (Ping timeout: 245 seconds).
05:50:43 -!- scoofy_ has quit (Read error: Connection reset by peer).
06:22:46 -!- Lord_of_Life has quit (Quit: Laa shay'a waqi'un moutlaq bale kouloun moumkine).
06:25:08 -!- Lord_of_Life has joined.
06:37:38 -!- tromp has joined.
06:42:08 -!- tromp has quit (Ping timeout: 245 seconds).
06:52:54 -!- craigo__ has quit (Ping timeout: 246 seconds).
06:55:52 -!- Guest56_ has quit (Quit: My MacBook has gone to sleep. ZZZzzz…).
07:00:07 -!- really2 has quit (Ping timeout: 260 seconds).
07:02:45 -!- Sgeo has quit (Read error: Connection reset by peer).
07:08:21 -!- delta23 has joined.
07:13:52 -!- craigo__ has joined.
07:21:58 <esowiki> [[User talk:ColorfulGalaxy]] https://esolangs.org/w/index.php?diff=81243&oldid=81200 * Trump Bot * (+256) I sneezed by accident while posting and I can't delete it
07:22:00 <nakilon> this one reminds me the game on ipad https://www.youtube.com/watch?v=WqNockPqPBg
07:22:42 <nakilon> there were wires and signals going through them by similar rules
07:45:56 -!- tromp has joined.
08:03:47 -!- delta23 has quit (Quit: Leaving).
08:23:26 -!- dcristofani has joined.
08:39:18 -!- dcristofani has quit (Ping timeout: 246 seconds).
08:39:48 -!- LKoen has joined.
08:44:30 -!- int-e_ has changed nick to int-e.
08:45:31 <zzo38> O, and there is a mirror of the Fossil repository for Free Hero Mesh at http://chiselapp.com/user/zzo38/repository/freeheromesh in case you need it
09:03:29 <nakilon> $ cc -std=c11 temp.c
09:03:30 <nakilon> temp.c:8:5: warning: implicit declaration of function 'timespec_get' is invalid in C99
09:04:02 <nakilon> why doesn't cc agree to use the C11? it seems to understand the flag but the warning says...
09:09:44 -!- hendursa1 has joined.
09:10:32 <nakilon> the same error with the snippet from https://en.cppreference.com/w/c/chrono/timespec_get while there on page their snippet runner compiles it fine with clang c11
09:11:35 -!- hendursaga has quit (Ping timeout: 268 seconds).
09:12:35 <nakilon> and when I press the "Share" button to get this link https://coliru.stacked-crooked.com/view?id=e84b765e6e9eb6d3and copy the command line from there I get the same error
09:12:48 <nakilon> I guess it's smth with my compiler but I didn't touch it...
09:14:19 <nakilon> they have clang 5.0 while mine is Apple LLVM version 10.0.1 (clang-1001.0.46.4) though and I don't know a shit about how do they correlate
09:49:20 <shachaf> MADV_FREE is a kind of bizarre API.
09:49:42 <shachaf> It should probably be set up to that either reads or writes give you the page, not just writes.
09:50:10 <shachaf> As it is you can have the odd behavior that you read a value from memory that's nominally available, and then you wait a while, and the value is different.
09:57:39 -!- dcristofani has joined.
11:01:23 <fizzie> nakilon: The "implicit declaration -- invalid in C99" warning (in Clang) is badly worded, and really just means "invalid *since* C99". But I don't know the actual cause of your problem; with my Clang, -std=c11 successfully adds timespec_get to the declarations from <time.h>, it just still gives the same warning for actually unknown names.
12:07:52 -!- ^_ has changed nick to V.
12:10:55 -!- dcristofani has quit (Ping timeout: 256 seconds).
12:11:38 -!- dcristofani has joined.
13:11:56 -!- jj97 has joined.
13:13:27 -!- jj97 has quit (Client Quit).
13:21:28 -!- arseniiv has joined.
13:24:40 <esowiki> [[PASM]] M https://esolangs.org/w/index.php?diff=81244&oldid=79641 * PythonshellDebugwindow * (-1) Fix name
13:25:34 <esowiki> [[PASM]] M https://esolangs.org/w/index.php?diff=81245&oldid=81244 * PythonshellDebugwindow * (+1) /* Registers */ l->n
13:44:34 -!- Bowserinator has quit (Quit: Blame iczero something happened).
13:46:22 -!- Bowserinator has joined.
13:48:50 -!- LKoen has quit (Quit: “It’s only logical. First you learn to talk, then you learn to think. Too bad it’s not the other way round.”).
13:51:57 -!- arcsor5 has joined.
14:18:51 -!- LKoen has joined.
15:07:16 -!- Sgeo has joined.
15:55:34 -!- djanatyn has quit (Ping timeout: 272 seconds).
15:56:13 -!- djanatyn has joined.
16:01:51 -!- craigo__ has quit (Quit: Leaving).
16:02:30 -!- craigo has joined.
16:03:24 -!- heavpoot has quit (Quit: Lost terminal).
17:29:39 -!- delta23 has joined.
18:05:58 -!- tromp has quit (Remote host closed the connection).
18:12:43 -!- tromp has joined.
18:24:03 -!- tromp has quit (Remote host closed the connection).
18:34:13 -!- craigo has quit (Ping timeout: 245 seconds).
18:38:03 <oren> what is the difference between aleatoric and stochastic
18:42:26 <int-e> . o O ( aleatoric sounds more musical. also I had to look it up. )
18:58:04 -!- tromp has joined.
19:02:48 -!- tromp has quit (Ping timeout: 265 seconds).
19:14:13 -!- atslash has quit (Ping timeout: 245 seconds).
19:14:37 -!- atslash has joined.
19:16:15 -!- atslash has quit (Excess Flood).
19:18:15 -!- tromp has joined.
19:18:23 <esowiki> [[Parse this sic]] M https://esolangs.org/w/index.php?diff=81246&oldid=81189 * Digital Hunter * (+0) /* Reverse cat */
19:19:16 <esowiki> [[Parse this sic]] M https://esolangs.org/w/index.php?diff=81247&oldid=81246 * Digital Hunter * (+0) /* 99 bottles of beer */ silly me. two-params are still confusing
19:20:28 <esowiki> [[Parse this sic]] M https://esolangs.org/w/index.php?diff=81248&oldid=81247 * Digital Hunter * (-49) /* FizzBuzz */ shorter one
19:21:05 <esowiki> [[Parse this sic]] M https://esolangs.org/w/index.php?diff=81249&oldid=81248 * Digital Hunter * (+0) /* FizzBuzz */ oops
19:21:23 <esowiki> [[Parse this sic]] M https://esolangs.org/w/index.php?diff=81250&oldid=81249 * Digital Hunter * (-187) /* FizzBuzz */ oops oops
19:32:33 -!- tromp has quit (Remote host closed the connection).
19:53:53 -!- tromp has joined.
20:05:42 -!- arcsor5 has quit (Quit: My MacBook has gone to sleep. ZZZzzz…).
21:24:39 -!- FireFly has quit (Ping timeout: 615 seconds).
21:25:34 -!- FireFly has joined.
22:10:32 -!- LKoen has quit (Quit: “It’s only logical. First you learn to talk, then you learn to think. Too bad it’s not the other way round.”).
22:13:17 -!- FireFly has quit (Quit: Goodbye).
22:16:39 -!- ubq323 has joined.
22:24:21 <esowiki> [[User talk:Ais523]] M https://esolangs.org/w/index.php?diff=81251&oldid=81236 * Ais523 * (-790) Reverted edits by [[Special:Contributions/Trump Bot|Trump Bot]] ([[User talk:Trump Bot|talk]]) to last revision by [[User:Ais523|Ais523]]
22:25:31 <esowiki> [[Special:Log/block]] block * Ais523 * blocked [[User:Trump Bot]] with an expiration time of 3 days (account creation disabled, email disabled): looks like this user is just here to troll / advertise other forums, not to post constructively; temporary block to give them a chance to cool down / improve, but I'm not hopeful
22:32:54 -!- FreeFull has joined.
22:38:54 -!- ais523 has joined.
22:39:16 <ais523> shachaf: MADV_FREE is, in effect, basically "make it so that reading from any address in this range before it is next written is undefined behaviour"
22:39:29 <ais523> (which is a useful thing to do to allow the kernel to optimise the use of the physical memory backing it)
22:39:49 <ais523> but, you can't have *actual* undefined behaviour at the user/kernel interface because that might lead to security bugs, so instead they document the possible behaviours
22:39:58 -!- zzo38 has quit (Ping timeout: 256 seconds).
22:40:24 <shachaf> There's a particular claim someone made that they read a zero at the beginning of the page, wrote to the page, and then read a nonzero from the same place.
22:40:32 <shachaf> Which is a situation I'm skeptical about.
22:41:20 <ais523> I think that a) the API definition should allow that as a possible sequence of events in theory, but b) there is no reason to implement the kernel in such a way that that sequence of events would happen in practice
22:41:39 <shachaf> Anyway, why not have it set up so that either a read or a write faults, and keeps the page as-is?
22:41:47 <ais523> with the possible exception of kernel builds on which non-zeroed memory can be allocated
22:42:22 <ais523> shachaf: modifying page tables is expensive, even taking into account the length of time a system call takes
22:42:23 <shachaf> As far as I can tell that should only affect people who read from "uninitialized" memory, making the system more predictable, at no particular cost to anyone else.
22:42:33 <shachaf> Aren't they modifying the page tables anyway?
22:42:35 <ais523> but I guess they need to be modified regardless so that you can detect writes
22:42:38 <shachaf> Right.
22:43:34 <ais523> (Linux has an API for allocating uninitialized memory, but it's normally ifdeffed out by default for security reasons, and is only really there for high-performance embedded systems which don't care about security between processes)
22:44:24 <shachaf> Yes. But I don't think that's the situation here.
22:45:10 <shachaf> A person is saying, maybe this is a dirty page allocated by the same process, and therefore it doesn't contain zeros, but you can still get it without a security problem.
22:45:17 <shachaf> But I don't think Linux does anything like that.
22:45:23 <ais523> nor do I
22:45:37 <shachaf> Other than MADV_FREE, I mean, which only works this way if you keep the page in place.
22:46:17 <ais523> MADV_FREE isn't an allocation API, it's a deallocation API
22:47:26 <shachaf> Well, you can think of it either way.
22:47:54 <shachaf> It's sort of an optimized version of mmapping a new anonymous page at the same address.
22:48:08 <ais523> yes
22:48:35 <ais523> my favourite madvise option is MADV_HUGEPAGE which seems to actually have a noticeable performance impact in some programs
22:49:05 <shachaf> Hmm.
22:49:07 <ais523> and yet basically nothing uses it
22:49:22 -!- arcsor5 has joined.
22:49:24 <ais523> it makes vmsplice(2) noticeably faster, but nobody uses vmsplice either
22:49:31 <shachaf> Last time I really tried to use huge pages for performance, you had to use hugetlbfs, I think.
22:49:42 <shachaf> And as I remember it didn't make much of a difference. But maybe I wasn't careful enough.
22:50:09 <shachaf> vmsplice is only for pipes, right?
22:50:12 <ais523> yes
22:50:38 <ais523> huge pages make pagewalks faster (most notably, recovering from a TLB flush, but some memory management operations make the kernel pagewalk)
22:50:59 <ais523> but they force memory management operations to have a larger granularity, so when you're using them, you have to be very careful to avoid anything that might cause a page copy
22:51:05 <ais523> otherwise you lose all your performance advantage
22:51:51 -!- FireFly has joined.
22:52:11 <ais523> something I've discovered recently is that the way to optimise I/O from the writer's point of view depends a lot on what is reading the data, and how it's being read
22:52:29 <ais523> in particular, whether the data is going to be read out of your L1/L2/L3 cache or not
22:53:12 <ais523> if it isn't then it makes most sense to stream the data directly to the physical memory that's going to back the kernel's view of it (using mmap for files and vmsplice for pipes, plus nontemporal writes)
22:55:40 -!- zzo38 has joined.
22:59:09 -!- arseniiv has quit (Ping timeout: 260 seconds).
23:00:46 <shachaf> Yes, but this was in a situation that should have benefited from huge pages (a database that managed its own cache).
23:03:52 <ais523> I actually think the largest benefits would be for, e.g., the memory backing large linked lists
23:04:15 <ais523> basically anything that makes a complete mockery of caches
23:04:45 <ais523> if something is cache-efficient then there won't be a large gain for hugepages because the TLB would be generally working pretty quickly anyway, so there isn't much to optimise
23:04:51 -!- Effilry has joined.
23:04:51 -!- FireFly has quit (Read error: Connection reset by peer).
23:05:06 <ais523> (other than recovery from TLB flushes, which I think the kernel does a pretty good job of reducing the frequency of nowadays)
23:26:42 -!- Effilry has quit (Ping timeout: 619 seconds).
23:38:30 -!- FireFly has joined.
←2021-03-08 2021-03-09 2021-03-10→ ↑2021 ↑all