| --- Log opened Wed Dec 21 00:00:20 2011 |
| --- Day changed Wed Dec 21 2011 |
| 00:00 | <@Namegduf> | So really, inlining at O3 unless you are writing your C like it was Java |
| 00:00 | < McMartin> | Java stole that style from C++, so... |
| 00:00 | < himi> | I suspect that exactly how debuggability was affected would depend on what debug symbol format was used |
| 00:01 | <@Namegduf> | Recursion and stack traces, not really |
| 00:01 | < himi> | But you're right that keeping track of any deep tail-recursion will be ridiculous |
| 00:01 | | * ToxicFrog upreads, also hands McM a chocolate cigar. Congrats! |
| 00:02 | < ToxicFrog> | Regarding C optimization, the general assumption is that if you enable optimization at all debugging may get hairy and anything -O2 or above pretty much makes debugging impossible. |
| 00:02 | < ToxicFrog> | so tail call elimination isn't really an issue there. |
| 00:02 | <@Namegduf> | Stack traces *in general* are still helpful |
| 00:03 | <@Namegduf> | Even at quite high levels |
| 00:03 | <@Namegduf> | You will not have line numbers but you will have some direction to debugging. |
| 00:03 | < ToxicFrog> | (lua, which has TCE but also guarantees stack trace availability, does the fake-frame thing McM mentioned - there's a single frame for "some number of tail calls" and it just increments a counter in that frame every time you tail) |
| 00:03 | < celticminstrel> | Tail call elimination? |
| 00:04 | < McMartin> | celticminstrel: At the assembly level, replacing JSR _____; RTS with JMP _______ |
| 00:04 | < celticminstrel> | Optimizing tail recursion? |
| 00:04 | < McMartin> | Tail recursion is a special case |
| 00:04 | < celticminstrel> | I dunno what JSR and RTS are. |
| 00:05 | < ToxicFrog> | Jump to SubRoutine and ReTurn from Subroutine. |
| 00:05 | < McMartin> | Jump SubRoutine and ReTurn from Subroutine |
| 00:05 | < ToxicFrog> | Other architectures might use CALL and RET or similar. |
| 00:05 | < celticminstrel> | I think I knew those as call and ret. |
| 00:05 | < McMartin> | Tail call is like tail recursion but with no requirement that the call be to yourself. |
| 00:05 | < celticminstrel> | ...ninjas. |
| 00:05 | < celticminstrel> | Ah. |
| 00:06 | < McMartin> | The machine-code level optimization is similar if not identical (wipe out the calling stack frame before building the new one) but it's conceptually messier (because you can't say 'it turns recursion into iteration', it turns function calls into cross-function GOTOs that still work) |
| 00:17 | < himi> | I think it's safe to say that /any/ optimising compiler is a sneaky bastard - so you can safely call them ninjas |
| 00:19 | | * jerith has passed many a pleasant afternoon listening to Crazy Poles talk about the innards of JITs. |
| 00:19 | < jerith> | (Well, one Crazy Pole. But he's awesome enough for two.) |
| 00:22 | < celticminstrel> | Heh. I was actually calling ToxicFrog a ninja, but whatever. |
| 00:22 | < iospace> | jerith: https://lh5.googleusercontent.com/-0Confgel-Zg/Ttw_wQ5X2OI/AAAAAAAAItE/g8d5QynMR hg/h0/4536319810_973c657e6b_o.jpg |
| 00:25 | < himi> | celticminstrel: TF is indeed a code ninja |
| 00:43 | < gnolam> | jerith: a crazy monopole then? |
| 00:43 | | Derakon[AFK] is now known as Derakon |
| 00:43 | < jerith> | Yup. |
| 00:44 | < jerith> | But insomnia seems to be abating, so sleep time. |
| 00:47 | <~Vornicus> http://paste.ubuntu.com/776972/ my infix notes. |
| 01:03 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has joined #code |
| 01:16 | < sshine> | Vornicus, neat state machine. |
| 01:18 | < sshine> | Vornicus, have you compared your stack approach to how LR(1) parsers work? |
| 01:18 | <~Vornicus> sshine: I have precisely no training in actual parsers. |
| 01:18 | | Rhamphoryncus [rhamph@Nightstar-14eb6405.abhsia.telus.net] has joined #code |
| 01:21 | <~Vornicus> So no. |
| 01:33 | < McMartin> | Nobody actually uses LR(1) parsers |
| 01:33 | < McMartin> | They use LALR(1), a horribly-specified subset. |
| 01:44 | | Kindamoody[zZz] is now known as Kindamoody |
| 01:59 | | Attilla [Obsolete@Nightstar-9d19ccd5.as43234.net] has quit [Ping timeout: 121 seconds] |
| 03:01 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 03:04 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 03:12 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 03:17 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 03:42 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 03:46 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 05:05 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has quit [[NS] Quit: Z?] |
| 05:12 | | * Derakon spends a couple of hours sorting out why the Cocoa port of Angband looks like this: http://derakon.dyndns.org/~chriswei/temp2/splashscreen.png |
| 05:13 | < Derakon> | Turned out to be because when the platform-agnostic code was moved from using char arrays to wchar_t arrays, nobody informed the Cocoa code. |
| 05:13 | < ToxicFrog> | :catdrugs: |
| 05:13 | < Derakon> | So it was casting everything to char, printing only 1/4th of each string, and printing the characters that it did print four times further apart than it should have been. |
| 05:13 | < Derakon> | (Also it kept running into what it thought were string terminators but were just the blank parts of the wchar_ts) |
| 05:18 | | Kindamoody [Kindamoody@Nightstar-5507a6b5.tbcn.telia.com] has quit [Connection reset by peer] |
| 05:19 | | Kindamoody|out [Kindamoody@Nightstar-5507a6b5.tbcn.telia.com] has joined #code |
| 05:29 | | Kindamoody|out is now known as Kindamoody |
| 05:32 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has quit [[NS] Quit: I <3Lovecraft<3 Vorn!] |
| 05:51 | | Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 06:09 | | Taki^ [Taki@Nightstar-98d9afe5.consolidated.net] has joined #code |
| 06:13 | | iospace [Alexandria@Nightstar-635d16fc.org] has quit [Ping timeout: 121 seconds] |
| 06:22 | | iospace [Alexandria@Nightstar-635d16fc.org] has joined #code |
| 06:54 | | Derakon is now known as Derakon[AFK] |
| 06:56 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has joined #code |
| 07:14 | | You're now known as TheWatcher |
| 07:17 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has quit [[NS] Quit: I <3Lovecraft<3 Vorn!] |
| 07:30 | | eckse [eckse@Nightstar-6fa75dbe.dsl.sentex.ca] has quit [Connection reset by peer] |
| 10:29 | | Attilla [Obsolete@Nightstar-9d19ccd5.as43234.net] has joined #code |
| 11:05 | | Kindamoody is now known as Kindamoody|out |
| 11:41 | | celticminstrel is now known as celmin|sleep |
| 12:05 | | Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code |
| 13:32 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 13:34 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 14:10 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code |
| 14:40 | | * TheWatcher eyes this team status document his boss has produced |
| 14:40 | < iospace> | heh |
| 14:41 | < TheWatcher> | "Chris: .... bunch of stuff... Project: Megaphone, For: Toby, Status: Nearly complete..." |
| 14:41 | < TheWatcher> | AHAHAHAHAHA >.> |
| 14:41 | < iospace> | ... |
| 14:41 | | * iospace pats TheWatcher |
| 14:41 | < TheWatcher> | I think I need to have a word with her about the nature of software |
| 14:47 | < TheWatcher> | (Although I am probably going to be able to push out v0.7 (Langdon Park) soon after the Christmas/New Year holiday, but that's still a long way from current-requirements complete, let alone any ones that crop up before then >.< |
| 14:47 | < TheWatcher> | ) |
| 15:18 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has joined #code |
| 15:28 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 15:28 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 15:54 | < jerith> | We had a legitimate reason for naming this app "vodka". |
| 15:55 | < jerith> | The module inside is called "methanol" is mostly silliness, though. |
| 15:55 | < Tamber> | Hehehe |
| 15:56 | < jerith> | "Vumi Open Data Kit Application". |
| 15:56 | < Tamber> | At least you've not found the one called "ethylene_glycol" yet |
| 15:56 | < gnolam> | Heh |
| 15:56 | < jerith> | methanol contains stuff to deal with XML namespaces. |
| 15:56 | < gnolam> | And it makes you go blind? |
| 15:56 | < jerith> | Both in our XML and in ElementTree. |
| 15:57 | < gnolam> | ... ah. Tree => wood => wood alcohol => methanol? |
| 15:57 | < jerith> | gnolam: It's the messy bits of XML. Blindness is the least of our worries. |
| 15:57 | < jerith> | Nah, "makes you go blind" is the right reference. |
| 16:01 | < jerith> | Locally, meths has had a vile-tasting emetic added. |
| 16:01 | < jerith> | Also, purple colouring. |
| 16:02 | < jerith> | So people filter it through bread, which removes most of the emetic and colouring. |
| 16:04 | < ToxicFrog> | |
| 16:08 | < Tamber> | ! |
| 16:09 | < jerith> | http://www.urbandictionary.com/define.php?term=blue%20train |
| 16:13 | < gnolam> | Hah! |
| 16:13 | < gnolam> | I thought that was just a Swedish thing! |
| 16:16 | < ToxicFrog> | jerith: so...it looks horrible, tastes horrible, and is toxic. |
| 16:16 | < gnolam> | Although among our drunks, lore has it that one should use a specific type of bread. |
| 16:16 | < jerith> | ToxicFrog: But it's dirt cheap. |
| 16:16 | < ToxicFrog> | They filter it so that it no longer looks or tastes horrible...but is still toxic. |
| 16:17 | < jerith> | Because it gets them smashed on the few coins in their pockets. |
| 16:19 | < gnolam> | And our denatured spirits are pink rather than purple. |
| 16:19 | < gnolam> | And they no longer have any methanol in them, since that just didn't help. |
| 16:21 | < gnolam> | Nowadays it's denatured by a small percentage of acetone, butanone, ethyl acetate and bitrex. |
| 16:24 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 16:37 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 16:42 | | Kindamoody|out is now known as Kindamoody |
| 16:46 | < gnolam> | http://www.explosm.net/comics/2650/ |
| 17:11 | | Reivles [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 17:11 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 17:36 | | * Derakon[AFK] mutters at git. |
| 17:36 | | Derakon[AFK] is now known as Derakon |
| 17:37 | < Derakon> | I've mucked up my master branch with accidental commits (that were meant to go in other branches) and then rollbacks of those commits. |
| 17:37 | < Derakon> | I want to just replace my master copy with this: https://github.com/angband/v4 |
| 17:37 | < Derakon> | Ideally without touching my other branches. |
| 17:37 | < celmin|sleep> | git reset --hard? |
| 17:38 | | celmin|sleep is now known as celticminstrel |
| 17:38 | < Derakon> | Ahh. |
| 17:38 | < Derakon> | Thanks. |
| 17:41 | | * TheWatcher eughs as New Work Projekt hits 18 tables before he even has any UI working |
| 17:45 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has joined #code |
| 17:50 | | * Derakon amuseds at http://ridiculousfish.com/blog/posts/Angband-3.3.html |
| 17:53 | < celticminstrel> | How is it that I'm helping people with git now rather than the other way around... |
| 17:55 | | Reivles [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 17:55 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 17:58 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Client closed the connection] |
| 18:04 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 18:13 | | Derakon is now known as Derakon[AFK] |
| 18:17 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 18:25 | <~Vornicus> Derakon: I was actually disappointed that Chrome didn't put a thing at the top saying "this page is in Quenya. Would you like to translate it?" |
| 18:31 | < celticminstrel> | It's using web fonts? |
| 18:32 | | Kindamoody is now known as Kindamoody[zZz] |
| 18:41 | <~Vornicus> I assume so, I certainly don't have fonts for the high speech installed. |
| 18:43 | | * celticminstrel does, but recalls having more than one with different character mappings. |
| 18:44 | <~Vornicus> Yeah, it's downloading /blog/font/tngan.ttf |
| 18:58 | | Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 19:07 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 19:10 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has quit [[NS] Quit: Spaaaaaaaaaaaaaaaaace sleeeeeeeeeeeeeeeep!] |
| 19:12 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 19:48 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 19:52 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 20:11 | | Stalker [Z@Nightstar-5aa18eaf.balk.dk] has joined #code |
| 20:22 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 20:35 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 20:53 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 20:54 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 21:00 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has quit [Ping timeout: 121 seconds] |
| 21:00 | | Reiver [orthianz@3CF3A5.E1CD01.C6689C.33956A] has joined #code |
| 21:38 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 21:50 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 22:14 | | Vash [Vash@Nightstar-8697fea9.wlfrct.sbcglobal.net] has joined #code |
| 22:30 | | AnnoDomini [annodomini@Nightstar-86809d18.adsl.tpnet.pl] has joined #code |
| 23:03 | | AnnoDomini [annodomini@Nightstar-86809d18.adsl.tpnet.pl] has quit [[NS] Quit: I need some sleep. ] |
| --- Log closed Thu Dec 22 00:00:12 2011 |