| --- Log opened Tue Mar 31 00:00:04 2015 |
| --- Day changed Tue Mar 31 2015 |
| 00:00 | <&McMartin> | OK, that's what you want |
| 00:00 | < Turaiel> | Everything prior has a semicolon |
| 00:01 | < Turaiel> | (related: this project is killing me >.<) |
| 00:01 | <&McMartin> | http://stackoverflow.com/questions/23314409/getting-error-iso-c-forbids-declarat ion-of-with-no-type |
| 00:01 | <&McMartin> | I'm not sure if this is immediately relevant, but it may be |
| 00:02 | <&McMartin> | But this is definitely a "the parser is confused, thinks you're asking it for something else" |
| 00:02 | <&McMartin> | This kind of shit is part of the reason C++ is so full of spiders |
| 00:02 | < Turaiel> | I had a similar issue before when two header files are referencing each other |
| 00:02 | <&McMartin> | Oh |
| 00:02 | <&McMartin> | Um |
| 00:03 | <&McMartin> | Are you in fact including that semaphore header file before this? |
| 00:03 | <&McMartin> | synchconsole.h itself should be including it |
| 00:03 | < Turaiel> | It's included, yeah |
| 00:03 | < Turaiel> | Hm... synchconsole.h (this file) references synch.h, which references thread.h, which references synchconsole.h |
| 00:03 | < Turaiel> | Is that an issue? |
| 00:04 | <&McMartin> | Not as long as you've put #pragma once at the top, or done the #ifndef MYFILENAME_H_ #define MYFILENAME_H_ dance |
| 00:04 | <~Vornicus> | is there a #pragma o... |
| 00:04 | < Turaiel> | Yeah, I have the ifndef |
| 00:04 | < Turaiel> | On all of them |
| 00:04 | <&McMartin> | Then you shouldn't get true infinite regress |
| 00:04 | <&McMartin> | Which should be fine |
| 00:05 | <&McMartin> | However |
| 00:05 | <&McMartin> | Somebody may need a forward class declaration |
| 00:05 | <&McMartin> | (That's the recursive-reference problem) |
| 00:05 | | Meatyhandbag [sebastianfe@Nightstar-dk0.5eb.224.136.IP] has joined #code |
| 00:05 | <&McMartin> | That would be handled by saying "class Semaphore;" in the header file and making sure the .cc file includes the real version. |
| 00:06 | <&McMartin> | But that assumes recursive reference, which is not a great thing to have, especially given what your problem statement seems to be |
| 00:06 | < Turaiel> | Doesn't seem to be whining about that anymore when I add that |
| 00:07 | < Turaiel> | At this point, unfortunately, I'd rather get it working than get it right |
| 00:08 | <~Vornicus> | (you need forward class definitions when you refer to a class before its structure is declared. For instance: Vector {...Matrix asScalingMatrix();...} Matrix {...Vector[] eigenvectors();...} would require a forward declaration for one or the other, depending on which you put first.) |
| 00:09 | <&McMartin> | Turaiel: OK, in that case, the problem was indeed "you've got a recursive dependency somewhere" |
| 00:10 | <@gnolam> | A forward declaration is basically just telling the compiler "this is a class" or "this is a struct". That's all the information it needs for you to be able to declare pointers to the type (since all pointers are basically the same under the hood). |
| 00:10 | <&McMartin> | (At least since DOS4GW was released~) |
| 00:11 | <@gnolam> | If you need to actually instantiate the type or manipulate an instance of it in any way, then it needs the full definition of it. |
| 00:11 | <&McMartin> | Though if you pass a pointer into a manipulating function, that doesn't require anything special. |
| 00:12 | <&McMartin> | (Exploiting these properties are a standard way of getting public/private behavior in C) |
| 00:13 | | himi [fow035@Nightstar-dm0.2ni.203.150.IP] has joined #code |
| 00:13 | | mode/#code [+o himi] by ChanServ |
| 00:23 | < Turaiel> | Sigh. Of course the issues remain... I changed my setup in hopes of fixing an issue with Nachos, but alas it's still there. :( |
| 00:37 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 00:37 | | mode/#code [+o Checkmate] by ChanServ |
| 03:43 | | thalass [thalass@Nightstar-h1qmno.eastlink.ca] has joined #code |
| 03:43 | | mode/#code [+o thalass] by ChanServ |
| 04:25 | | Meatyhandbag [sebastianfe@Nightstar-dk0.5eb.224.136.IP] has quit [Client exited] |
| 04:43 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has quit [Ping timeout: 121 seconds] |
| 04:58 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 05:34 | | Turaiel is now known as Turaiel[Offline] |
| 05:49 | | thalass [thalass@Nightstar-h1qmno.eastlink.ca] has quit [Ping timeout: 121 seconds] |
| 06:26 | | celticminstrel [celticminst@Nightstar-c55sbu.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 06:34 | | Vash [Vash@Nightstar-uhn82m.ct.comcast.net] has quit [Connection closed] |
| 07:01 | | Kindamoody[zZz] is now known as Kindamoody |
| 07:01 | | himi [fow035@Nightstar-dm0.2ni.203.150.IP] has quit [Ping timeout: 121 seconds] |
| 07:05 | | macdjord|wurk is now known as macdjord|slep |
| 08:59 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 08:59 | | mode/#code [+o himi] by ChanServ |
| 09:08 | | Kindamoody is now known as Kindamoody|afk |
| 10:02 | | AverageJoe [evil1@Nightstar-2ofrtr.ph.cox.net] has joined #code |
| 10:35 | | AverageJoe [evil1@Nightstar-2ofrtr.ph.cox.net] has quit [[NS] Quit: Leaving] |
| 10:43 | | grindhold [quassel@Nightstar-0ona3l.cyan.servdiscount-customer.com] has quit [Connection closed] |
| 11:23 | | Derakon [chriswei@Nightstar-5fqf0m.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 11:29 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Connection closed] |
| 11:29 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 11:40 | <@Tarinaky> | D seems to be getting a fair amount of love atm. |
| 11:40 | <@Tarinaky> | At least, compared to Go. |
| 11:40 | <@TheWatcher> | That's a really unfortunate name these days |
| 11:42 | <@Tarinaky> | int The D() { lana = "hello world"! PHRASING lana! } |
| 11:42 | <@Tarinaky> | Sorry, I don't know why that entered my head. |
| 11:43 | <@Shiz> | D is just the worst |
| 11:46 | <@Tarinaky> | D or the D |
| 11:46 | <@Tarinaky> | ? |
| 11:46 | | grindhold [quassel@Nightstar-0ona3l.cyan.servdiscount-customer.com] has joined #code |
| 11:47 | <@Tarinaky> | Mostly I just like the fact that it has an import statement. |
| 11:50 | <@Tarinaky> | Also: Github's been under attack for 100+ hours now. |
| 11:50 | < [R]> | From? |
| 11:50 | <@Shiz> | D |
| 11:50 | <@Tarinaky> | This is getting a little silly, is this really going to be the new normal for hosting software China disagrees with? |
| 11:51 | <@Tarinaky> | [R]: The Great Firewall Of China is using MITM techniques to insert javascript requests for certain URIs into the header for webpages inside China being accessed from outside China. |
| 11:52 | <@Tarinaky> | e.g. expatriats and Proxy users using Baidu. |
| 11:52 | <@Tarinaky> | Basically turning innocuous traffic into their personal botnet. |
| 11:52 | <@Tarinaky> | At least, that was the original attack. It's changed up two or three times over the last 100 hours. |
| 11:53 | <@Tarinaky> | Given the Great Firewall of China is compromised this implicates the Chinese government since, well, who else has the means? |
| 11:54 | | Checkmate [Z@Nightstar-484uip.cust.comxnet.dk] has joined #code |
| 11:54 | | mode/#code [+o Checkmate] by ChanServ |
| 11:54 | <@Tarinaky> | And the target URIs are some greatfirewall circumvention tools hosted on github, as well as some Journalism resources. |
| 11:56 | <@Tarinaky> | Theories abound as to why and why now... but they're all blatant tinfoil twitching. |
| 12:03 | < [R]> | I'm impressed that they went to the effort of doing MITM for that |
| 12:03 | < [R]> | Rather than the obvious solution, given their resources. |
| 12:04 | < [R]> | +simpler |
| 12:04 | <@Tarinaky> | Well there's two advantages: 1) there's no pattern to the DDOS source address' |
| 12:04 | <@TheWatcher> | It's not MITM, it's MOTS actually |
| 12:04 | <@Tarinaky> | MOTS? |
| 12:04 | <@TheWatcher> | Man on the side |
| 12:04 | <@Tarinaky> | What's the distinction? |
| 12:05 | <@Tarinaky> | 2) Provides a degree of deniability. If it was a block of Chinese government IP addresses hurling packets at them 1) it'd be easier to just block the offending addresses and 2) it'd be less... discrete. |
| 12:06 | < [R]> | Just forge the requests using random IPs from in China, determined by randomly selected packets going outbound. |
| 12:07 | <@Tarinaky> | You'd still be limited to the packets passing out of china. Physically. |
| 12:08 | < [R]> | <Tarinaky> [R]: The Great Firewall Of China is using MITM techniques to insert javascript requests for certain URIs into the header for webpages inside China being accessed from outside China. |
| 12:08 | <@Tarinaky> | Nope. |
| 12:08 | <@Tarinaky> | The packets are being generated by a machine outside China |
| 12:08 | <@Tarinaky> | Which can be located anywhere in the world. |
| 12:08 | <@Tarinaky> | Client-side javascript. |
| 12:09 | < [R]> | You explained poorly, since I've no idea what's happening now. |
| 12:10 | <@TheWatcher> | Okay, someone outside china tries to access a site inside china |
| 12:10 | <@TheWatcher> | that site in china is using baidu's analytics fragment |
| 12:10 | <@TheWatcher> | so the client outside china tries to request that javascript file |
| 12:11 | <@TheWatcher> | the request goes to baidu, but something in the firewall is seeing that, passing the request one /but also generating a fake response/ |
| 12:11 | <@TheWatcher> | that fake response contains javascript that makes the client's machine attack github |
| 12:12 | <@TheWatcher> | baidu response with the real javascript, but by the time it does so the client has already handled the response it got from the firewall |
| 12:12 | <@TheWatcher> | s/request one/request on/ |
| 12:13 | <@Tarinaky> | +1 would explain again |
| 12:14 | < [R]> | Alright, seems odd to me still, I understand, but the fact there's enough incomming traffic to enable a DDoS is unexpected. |
| 12:17 | <@TheWatcher> | Yeah, somewhat surprising to me, too |
| 12:18 | <@TheWatcher> | http://www.netresec.com/?page=Blog&month=2015-03&post=China%27s-Man-on-the-Side- Attack-on-GitHub has a longer explanation |
| 12:18 | <@TheWatcher> | (same technique the NSA used as part of QUANTUM to try and break Tor, actually) |
| 13:09 | <@Tarinaky> | TheWatcher: What's the characteristic that makes this on the side rather than in the middle? |
| 13:09 | <@TheWatcher> | mitm generally involves interception and inspection or rewriting, while mots simply looks where the request is going and sneds a fake response |
| 13:10 | <@TheWatcher> | *sends |
| 13:10 | <@TheWatcher> | It doen't actually do anything to the request or real response, just injects a fake one |
| 13:19 | <@Tarinaky> | Isn't URI part of the HTTP request? |
| 13:19 | <@Tarinaky> | Rather than part of the header data though? |
| 13:19 | <@Tarinaky> | That kinda counts as inspection... |
| 13:20 | <@Tarinaky> | Well, not the URI but the path at any rate. |
| 13:21 | <@Tarinaky> | Normally when we talk about inspection we mean the body of the TCP packet... Rather than the body of the HTTP... |
| 13:30 | < [R]> | I think his point was more that it doesn't stop the real interaction, it just noisily intercepts it. |
| 13:35 | <@Tarinaky> | And now for something completely different: http://www.bbc.co.uk/news/uk-england-nottinghamshire-32117815 |
| 13:36 | <@Tarinaky> | Apparently Garlic, Onion and Cow's Intestines cures MRSA. |
| 13:49 | | thalass [thalass@Nightstar-h1qmno.eastlink.ca] has joined #code |
| 13:49 | | mode/#code [+o thalass] by ChanServ |
| 14:14 | <@gnolam> | Aaaaaaaaaaaaaan style done. |
| 14:14 | <@gnolam> | Here's hoping I never have to do that again. |
| 14:14 | <@gnolam> | (ahahahahahaha |
| 14:14 | <@gnolam> | ) |
| 14:38 | | macdjord|slep is now known as macdjord|wurk |
| 15:18 | | * Tarinaky grumbles at stupidity I saw on Facebook complaining about McDonalds containing such strange chemicals as cochineal(sp?) beetles. |
| 15:23 | <@gnolam> | At least they're not complaining that airplane air has nitrogen in it. |
| 15:23 | <@Tarinaky> | Hahahahahahahahahahahahahahahahahahahahahahahahahahahahahahahahahahahahaha |
| 15:24 | <@TheWatcher> | I hope airplane air doesn't, you might get high on it otherwise... ¬¬ |
| 15:24 | | * TheWatcher flrrd |
| 15:25 | <@gnolam> | (Actual complaint - now deleted - by Vani Hari a.k.a. "The Food Babe", who does exactly the kind of scaremongering you were talking about for a living) |
| 15:25 | <@Tarinaky> | My mission when I get home is to troll by complaining that organic, happy clappy, beef contains Titin. |
| 15:27 | <@gnolam> | ("The air you are breathing on an airplane is recycled from directly outside of your window. That means you are breathing everything that the airplanes gives off and is flying through. The air that is pumped in isnât pure oxygen either, itâs mixed with nitrogen, sometimes almost at 50%.") |
| 15:27 | <@TheWatcher> | gnolam: ... I'd say you're pulling my leg, but I know you're not. |
| 15:28 | < abudhabi> | I really, really like how XFCE is being fast. It hasn't felt this good since Win95. |
| 15:30 | <@gnolam> | Nope. Woo people. Stranger than fiction. |
| 15:30 | <@gnolam> | http://theness.com/neurologicablog/index.php/food-babe-misinformation-on-travel/ |
| 15:30 | <@Tarinaky> | 50% nitrogen in the air is loooow. |
| 15:30 | <@Tarinaky> | Air is close to 3/4 nitrogen at sea level |
| 15:31 | <@Tarinaky> | Oh wait. |
| 15:31 | <@Tarinaky> | I didn't see where this was a quote. |
| 15:40 | <@TheWatcher> | ... why did I look at that link |
| 15:40 | | * TheWatcher headshakes |
| 15:41 | <@TheWatcher> | But hey, no chemtrails, so she can't be /that/ bad, right? >.> |
| 15:41 | <@Tarinaky> | chemtrails? |
| 15:45 | <@TheWatcher> | You know your common or garden contrail? Well, it's apparently not condensed water vapour, but actually chemical or biological agents deliberately spread into the atmoshere by planes as partof a global conspiracy (which global conspiracy depends on which chemtrail woo you adhere to) |
| 15:48 | <@TheWatcher> | (In at least half of them, HAARP is involved in some neferious fashion) |
| 15:49 | | Meatyhandbag [sebastianfe@Nightstar-dk0.5eb.224.136.IP] has joined #code |
| 15:52 | | Meatyhandbag [sebastianfe@Nightstar-dk0.5eb.224.136.IP] has quit [Client exited] |
| 15:53 | <@Tarinaky> | : |
| 15:53 | <@Tarinaky> | :/ |
| 15:59 | | Meatyhandbag [sebastianfe@Nightstar-8l2.s7j.224.136.IP] has joined #code |
| 16:05 | < abudhabi> | Java Graphics2D. I'm trying to center the graphics around the clicked point. |
| 16:05 | < abudhabi> | The part where I get the coordinates of the click and set a property in concordance is fine. |
| 16:05 | < abudhabi> | But I'm having trouble with the "set center where the click is" part. |
| 16:27 | | Orthia [orthianz@Nightstar-mmv.dub.224.119.IP] has quit [Ping timeout: 121 seconds] |
| 16:32 | | Orthia [orthianz@Nightstar-n0o.kgp.224.119.IP] has joined #code |
| 16:32 | | mode/#code [+o Orthia] by ChanServ |
| 16:38 | <@Tarinaky> | http://morendil.github.io/folklore.html |
| 16:54 | <@gnolam> | https://twitter.com/ptone/status/582764080219320323 |
| 17:04 | | Meatyhandbag [sebastianfe@Nightstar-8l2.s7j.224.136.IP] has quit [Client exited] |
| 17:22 | | celticminstrel [celticminst@Nightstar-c55sbu.dsl.bell.ca] has joined #code |
| 17:22 | | mode/#code [+o celticminstrel] by ChanServ |
| 17:38 | | orth [orthianz@Nightstar-n0o.kgp.224.119.IP] has joined #code |
| --- Log closed Tue Mar 31 17:38:46 2015 |
| --- Log opened Tue Mar 31 18:01:38 2015 |
| 18:01 | | TheWatcher [chris@Nightstar-ksqup0.co.uk] has joined #code |
| 18:01 | | Irssi: #code: Total of 39 nicks [23 ops, 0 halfops, 0 voices, 16 normal] |
| 18:01 | | mode/#code [+o TheWatcher] by ChanServ |
| 18:02 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 18:02 | | mode/#code [+ao Reiver Reiver] by ChanServ |
| 18:02 | | Irssi: Join to #code was synced in 39 secs |
| 18:05 | | * thalass reads up |
| 18:05 | <@thalass> | the air you breath is taken from the engines. But before fuel is put in, so it's OK to breathe :P |
| 18:07 | <@gnolam> | But there's nitrogen in it! :o |
| 18:14 | <@thalass> | Yes! At the usual partial pressure for your altitude. :P |
| 18:14 | <@thalass> | But yes. Nutjobs. Such pain |
| 19:01 | | Netsplit *.net <-> *.split quits: @Orthia, abudhabi, @Alek, @himi, tripflag |
| 19:03 | | PinkFreud [WhyNot@Pinkfreud.is.really.fuckin.lame.nightstar.net] has quit [Ping timeout: 121 seconds] |
| 19:06 | | Netsplit over, joins: abudhabi |
| 19:08 | | Orthia [orthianz@Nightstar-n0o.kgp.224.119.IP] has joined #code |
| 19:08 | | mode/#code [+o Orthia] by ChanServ |
| 19:12 | | tripflag [im@Nightstar-1sfccr.ocv.me] has joined #code |
| 19:12 | | Alek [omegaboot@Nightstar-03ja8q.il.comcast.net] has joined #code |
| 19:12 | | mode/#code [+o Alek] by ChanServ |
| 19:15 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has joined #code |
| 19:15 | | mode/#code [+o himi] by ChanServ |
| 19:15 | | PinkFreud [WhyNot@Pinkfreud.is.really.fuckin.lame.nightstar.net] has joined #code |
| 19:15 | | mode/#code [+o PinkFreud] by ChanServ |
| 19:21 | < abudhabi> | This is strange. Mouse event points to coordinates relative to the component, but the component appears to draw relative to the window. |
| 19:42 | | sebastianferguson [sebastianfe@Nightstar-8l2.s7j.224.136.IP] has joined #code |
| --- Log opened Tue Mar 31 20:07:31 2015 |
| 20:07 | | TheWatcher [chris@Nightstar-ksqup0.co.uk] has joined #code |
| 20:07 | | Irssi: #code: Total of 40 nicks [23 ops, 0 halfops, 0 voices, 17 normal] |
| 20:07 | | mode/#code [+o TheWatcher] by ChanServ |
| 20:08 | | Irssi: Join to #code was synced in 38 secs |
| 20:08 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 20:08 | | mode/#code [+ao Reiver Reiver] by ChanServ |
| 20:56 | | Turaiel[Offline] is now known as Turaiel |
| 20:56 | < Turaiel> | I'm having the most absurd issue right now |
| 20:57 | < Turaiel> | I have a pointer to an object (C++) that's inexplicably changing out of nowhere. |
| 20:57 | < Turaiel> | In the same spot in the program, every time |
| 20:58 | < abudhabi> | Array index overflow? |
| 20:58 | < Turaiel> | There aren't any arrays nearby getting manipulated, I don't think |
| 20:58 | < Turaiel> | And it's still acting like the correct object type |
| 20:59 | < Turaiel> | It's a semaphore, and I'm referencing a list inside of it. |
| 20:59 | < abudhabi> | Valgrind? |
| 20:59 | < Turaiel> | Valgrind doesn't work on this project |
| 20:59 | <&McMartin> | Is the object changing or the pointer? |
| 20:59 | <&McMartin> | If the object is changing you may have a wild pointer |
| 21:00 | < Turaiel> | It seems to be changing into another object of the same type |
| 21:00 | < Turaiel> | Because the list inside it is functional, but empty and has some garbage elements in it |
| 21:00 | < Turaiel> | (such as the list counter I set) |
| 21:00 | < Turaiel> | SEM V 159257688 |
| 21:00 | < Turaiel> | !!!! (List 159257704) Something was removed from the list (there were 2 items) |
| 21:00 | < Turaiel> | SEM V 159257736 |
| 21:00 | < Turaiel> | !!!! (List 159257752) Something was removed from the list (there were -3 items) |
| 21:01 | < Turaiel> | The first one is the same all the way through up to this point |
| 21:02 | <&McMartin> | If there were -3 items it's not *that* functional. |
| 21:02 | < Turaiel> | It's functional in that its methods are working |
| 21:02 | <&McMartin> | If your class has no virtual methods, any random readable chunk of memory will appear to work |
| 21:02 | < Turaiel> | But the data is garbage |
| 21:02 | <&McMartin> | So a wild pointer is still a possibility. |
| 21:02 | < Turaiel> | I suppose that's true |
| 21:03 | <&McMartin> | (A class with no virtual methods is really just a C struct with some funny syntax around it) |
| 21:03 | < Turaiel> | What exactly is a wild pointer? |
| 21:03 | <&McMartin> | A pointer that isn't pointing where it should, in the most general case |
| 21:03 | < Turaiel> | What can cause that? |
| 21:04 | <&McMartin> | Use of one that hasn't been properly initialized yet, walking one off the end of an array or a structure, or doing unsafe pointer arithmetic that didn't get a correct answer |
| 21:04 | <&McMartin> | Hmmm, wait a sec |
| 21:04 | <&McMartin> | That "List 159257752" |
| 21:04 | <&McMartin> | Is that number the raw value of the pointer (that is, the virtual address itself)? |
| 21:04 | < Turaiel> | Yes. I'm using (int)this |
| 21:05 | <&McMartin> | kk |
| 21:05 | <&McMartin> | Then the object isn't being trashed, the pointer is indeed definitely being reassigned. |
| 21:06 | <&McMartin> | Actually while we're at it is it me or are those two different semaphores, too |
| 21:06 | < Turaiel> | Yes |
| 21:08 | < Turaiel> | Here's a snippet of my output if it helps, though I'm not sure how helpful it'll actually be. http://pastebin.com/vgDR10UW |
| 21:10 | < Turaiel> | Because of this issue, my semaphore isn't being released so the "main" thread gets stuck calling the same SYSCALL (join) forever |
| 21:10 | <&McMartin> | Yeah, it's not super useful there |
| 21:10 | <&McMartin> | Also am I right that you aren't logging SEM P events |
| 21:10 | < Turaiel> | Since it's trying to join "userthread" which is suspended |
| 21:11 | < Turaiel> | I am logging P |
| 21:11 | < Turaiel> | "### userthread sleeps - waiting for "write done"" |
| 21:11 | <&McMartin> | Aha, OK |
| 21:11 | <&McMartin> | Under these circumstances, especially if I can't get reasonable debugging hooks in, I would start logging the value of the list pointer at various points, to try to hammer down when it's mutating. |
| 21:12 | <&McMartin> | (If I *can* get debugging hooks in, I would set a "watchpoint" to break whenever that location is written) |
| 21:12 | < Turaiel> | How would I do that? |
| 21:12 | <&McMartin> | Do what? |
| 21:12 | <&McMartin> | I said a few things there :) |
| 21:12 | <&McMartin> | setting watchpoints depends on the precise debugger, so I'd hit up manuals or google |
| 21:13 | < Turaiel> | gdb is all I have. I'll have to look around. |
| 21:13 | <&McMartin> | Logging the value of the list pointer - whoever's *invoking* list::remove has changed its idea of what the list pointer is |
| 21:13 | <&McMartin> | https://sourceware.org/gdb/onlinedocs/gdb/Set-Watchpoints.html would be a place to start |
| 21:20 | | thalass [thalass@Nightstar-h1qmno.eastlink.ca] has quit [Operation timed out] |
| 21:21 | | thalass [thalass@Nightstar-h1qmno.eastlink.ca] has joined #code |
| 21:21 | | mode/#code [+o thalass] by ChanServ |
| 21:21 | < Turaiel> | I added a watchpoint, but it never stopped executing |
| 21:24 | < Turaiel> | Hm |
| 21:24 | < Turaiel> | It looks like the same thing is happening in other calls to this function |
| 21:27 | <&McMartin> | Mmm |
| 21:27 | <&McMartin> | Yeah, there's really no replacement for logging the shit out of things, when it comes to that |
| 21:27 | <&McMartin> | My guess is something's being initialized wrong (it thinks it's pointing to the same shared list but it isn't) or something is trashing the location where the "here's where the list is" is stored |
| 21:32 | < Turaiel> | Oh, this is weird. |
| 21:32 | < Turaiel> | I changed my callback to void SynchConsole::WriteDone() { printf("Received synchconsole %i\n", (int)this); writeDone->V(); }, but the print is not happening on the last V() |
| 21:32 | < Turaiel> | But nothing else calls V(), to my knowledge |
| 21:32 | <&McMartin> | Oh, are you using printf here? |
| 21:32 | < Turaiel> | Yeah |
| 21:32 | <&McMartin> | You should use %p instead of %i and a cast to int |
| 21:33 | <&McMartin> | That's the format code for "pointer" |
| 21:33 | | sebastianferguson [sebastianfe@Nightstar-8l2.s7j.224.136.IP] has quit [Client exited] |
| 21:33 | <&McMartin> | (Normally it puts it in hex, but it also means that if your pointers are not the same size as your ints - as they aren't on, say, the machine I'm typing on here - it'll do the right thing) |
| 21:34 | <&ToxicFrog> | Turaiel: also, use fprintf(stderr, ...); stderr is guaranteed to flush on write, stdout flushes "eventually". |
| 21:35 | <&McMartin> | Huh, I didn't know stderr had that guarantee |
| 21:35 | <&McMartin> | (fflush of course makes it happen right now) |
| 21:36 | < Turaiel> | ...whoops. |
| 21:36 | < Turaiel> | Just did a replace all. I thought it would only replace in the file.. |
| 21:37 | <&McMartin> | "Expected results: Squid is restarted" |
| 21:37 | <&McMartin> | "Actual results: All files are deleted on the machine" |
| 21:37 | <&McMartin> | https://bugzilla.redhat.com/show_bug.cgi?id=1202858 |
| 21:39 | <@Tamber> | eep |
| 21:40 | <&McMartin> | This was in unreleased software, though of course the Internet is not big on this notion |
| 21:40 | <@Wizard> | I fell in love with that expected/actual thing |
| 21:40 | <@Wizard> | It reminds me of that gif of the old man on titanic slowly having a drink while water bursts in |
| 21:41 | <@Tamber> | McM: It being unreleased is an insignificant detail compared to "AHAHAHAHA, LOOK SOMEONE SCREWED UP!" |
| 21:41 | <&McMartin> | Yes, and taking that attitude is dumb for a bug database for unreleased software |
| 21:41 | <&McMartin> | This is *literally* what testing is for |
| 21:42 | <@Tamber> | Yup. |
| 21:43 | < Turaiel> | Did you hear about the Steam bug on linux? |
| 21:43 | <&McMartin> | It's cited in that link |
| 21:43 | <&McMartin> | The speculation was "this may be the same sort of thing" |
| 21:43 | < Turaiel> | Ahh |
| 21:43 | <&McMartin> | Unfortunate symlink resulting in a process going rampant, etc. |
| 21:45 | <@thalass> | wut |
| 21:45 | < Turaiel> | I had a friend who was playing around with writing a neural net AI which got confused while running as root and wiped the computer.. |
| 21:45 | < Turaiel> | Including the only copy of its source |
| 21:46 | <&McMartin> | This is why you don't give your AI unlink access =P |
| 21:53 | | Kindamoody|afk is now known as Kindamoody |
| 21:57 | < Turaiel> | By the way, Nachos is the bane of my existence at this point |
| 21:57 | < Turaiel> | This assignment was due this morning but it's still nothing but problems |
| 21:58 | <&McMartin> | Back '98 the CS building at my university had a line-printer printout of the nethack death screen |
| 21:58 | <&McMartin> | marked "killed by NACHOS while helpless" |
| 22:00 | | [R] [rstamer@genoce.org] has quit [Ping timeout: 121 seconds] |
| 22:00 | <@gnolam> | Heh. |
| 22:01 | <@gnolam> | I can't see why anyone would still use NACHOS now that Pintos exists. |
| 22:02 | <&McMartin> | Curricula have strong inertia |
| 22:03 | <&McMartin> | Also hey I know that name |
| 22:04 | < Turaiel> | Sigh. I forgot that I wasn't using bash, because my university loves tcsh |
| 22:06 | <@gnolam> | McMartin: Reminds me of a slide Kevin, my "Programming, abstraction and modelling" (read: SICP) professor had on his first lecture. |
| 22:06 | <@gnolam> | "Why Scheme? |
| 22:06 | <@gnolam> | * LiU has a secret agenda to keep the language alive |
| 22:06 | <@gnolam> | * Kevin is too lazy to learn new course material |
| 22:06 | <@gnolam> | * Make up your own explanation" |
| 22:07 | < Turaiel> | I had a course last semester that used SICP |
| 22:07 | < Turaiel> | We were mostly using Racket though |
| 22:08 | <&McMartin> | I learned on SICP too, so I can name two very good reasons to use Scheme |
| 22:08 | <&McMartin> | Though those are also good reasons |
| 22:08 | <&McMartin> | * Humbles the braggarts who think they're hot shit because they learned C at age 12 (this was the mid-1990s; substitute "Javascript" at this point) |
| 22:09 | <&McMartin> | * You can teach enough of Scheme to teach the entire, terrifyingly wide-ranging course in like 3 lectures tops |
| 22:09 | <&McMartin> | The latter isn't actually true of the other major lisp dialects |
| 22:09 | <&McMartin> | (I'm treating Racket as a Scheme here even though it's got enough extra stuff it is rightfully its own language) |
| 22:10 | <@gnolam> | IIRC, the first one was almost literally the thinking behind it when they started that course. |
| 22:10 | < Turaiel> | I once got a job offer for learning Racket and completing a programming challenge.. turned it down though xD |
| 22:10 | < Turaiel> | It was fun though. |
| 22:10 | < Turaiel> | Okay, I guess I should go to dinner. Perhaps I'll poke at the code more before I concede defeat and strip down its functionality. |
| 22:11 | < Turaiel> | (I really don't want to do that though; this is 20% of my final grade T.T) |
| 22:11 | <&McMartin> | gnolam: As someone who took it at the height of the 90s tech bubble, surrounded by assholes |
| 22:11 | <&McMartin> | ... so. worth. it. |
| 22:11 | <@gnolam> | :) |
| 22:12 | <&McMartin> | I mean, by rights I should have been one of them, except for the part where I had a mathematician for a dad and therefore had a pretty good idea that I didn't already know everything |
| 22:12 | <&McMartin> | (That said, for various reasons both good and bad I tend to use Gambit, not Racket) |
| 22:19 | | himi [fow035@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 22:31 | | [R] [rstamer@Nightstar-d7h8ki.org] has joined #code |
| 22:33 | | sebastianferguson [sebastianfe@Nightstar-dk0.5eb.224.136.IP] has joined #code |
| --- Log opened Tue Mar 31 22:48:15 2015 |
| 22:48 | | TheWatcher [chris@Nightstar-ksqup0.co.uk] has joined #code |
| 22:48 | | Irssi: #code: Total of 38 nicks [21 ops, 0 halfops, 0 voices, 17 normal] |
| 22:48 | | mode/#code [+o TheWatcher] by ChanServ |
| 22:48 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 22:48 | | mode/#code [+ao Reiver Reiver] by ChanServ |
| 22:48 | | Irssi: Join to #code was synced in 38 secs |
| 22:50 | < abudhabi> | Is there any way to tell Netbeans to stop trying to run a custom paint() routine in a compnent? |
| 22:50 | < abudhabi> | *component |
| 22:50 | < abudhabi> | At least in its graphical form designer. |
| 23:00 | | Kindamoody is now known as Kindamoody[zZz] |
| 23:03 | | VirusJTG_ [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 23:03 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Connection reset by peer] |
| --- Log opened Tue Mar 31 23:11:34 2015 |
| 23:11 | | TheWatcher [chris@Nightstar-ksqup0.co.uk] has joined #code |
| 23:11 | | Irssi: #code: Total of 38 nicks [21 ops, 0 halfops, 0 voices, 17 normal] |
| 23:11 | | mode/#code [+o TheWatcher] by ChanServ |
| 23:12 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 23:12 | | mode/#code [+ao Reiver Reiver] by ChanServ |
| 23:12 | | Irssi: Join to #code was synced in 39 secs |
| 23:26 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has quit [Connection closed] |
| --- Log closed Tue Mar 31 23:27:04 2015 |