| 00:50.29 | jxqz | worst buy |
| 00:50.46 | jxqz | martwall |
| 01:10.17 | *** join/#utah tiwula (~lane@74-81-251-156.static.logl001.digis.net) |
| 01:35.25 | spangborn | You can have styles without class. |
| 01:35.27 | spangborn | #fail |
| 01:35.39 | spangborn | youtah { display: none;} |
| 01:39.17 | jxqz | ~napoleon youtah_ |
| 01:39.17 | infobot | ACTION eats all of youtah_'s steak |
| 02:29.20 | goozbach | geez, time for an upgrade |
| 02:48.05 | *** join/#utah tristanbob__l (~tristanbo@c-50-186-103-12.hsd1.ut.comcast.net) |
| 02:59.46 | *** join/#utah macnewbold (~macnewbol@c-24-2-87-227.hsd1.ut.comcast.net) |
| 03:19.34 | *** join/#utah olli (~olli@c-71-195-232-74.hsd1.ut.comcast.net) |
| 03:20.17 | *** join/#utah yukaia_ (~yukaia@174-23-113-160.slkc.qwest.net) |
| 03:25.06 | *** join/#utah smcquay (~smcquay@17.226.13.44) |
| 03:47.58 | *** join/#utah smcquay (~smcquay@17.226.13.44) |
| 03:49.14 | *** join/#utah Tristan_ (~Tristan@tristanrhodes-vpn.weber.edu) |
| 04:00.12 | *** join/#utah goozbach (~goozbach@fedora/goozbach) |
| 04:00.12 | *** mode/#utah [+v goozbach] by ChanServ |
| 04:07.17 | RyanE | What the? His mug shot has a BYU shirt in it? D'oh! |
| 04:11.53 | *** join/#utah Tristan__ (~Tristan@c-50-186-103-12.hsd1.ut.comcast.net) |
| 04:46.40 | *** join/#utah kendsnyder (~Adium@67-2-115-168.slkc.qwest.net) |
| 04:48.07 | *** join/#utah jalbretsen (~jalbretse@70-58-32-202.slkc.qwest.net) |
| 04:58.51 | bofhit | Don't say that youtah_ , it could cascade through the whole group! |
| 05:12.00 | *** join/#utah Nemus (~Nemus@unaffiliated/nemus) |
| 05:12.56 | *** join/#utah Nemus (~Nemus@unaffiliated/nemus) |
| 05:31.41 | *** join/#utah pdayton (~pjdayton@2607:fb90:40f:d24c:79ac:2ecf:6783:137c) |
| 05:43.24 | *** join/#utah pdayton (~pjdayton@63.251.sfcn.org) |
| 07:01.20 | jxqz | har har |
| 12:08.54 | *** join/#utah olli (~olli@c-71-195-232-74.hsd1.ut.comcast.net) |
| 12:36.08 | *** join/#utah bgerber (~gerberb@suse103.zenez.com) |
| 13:51.55 | *** join/#utah undertakingyou (~undertaki@neutroninteractive-xmission.xmission.com) |
| 13:55.27 | *** join/#utah smcquay (~smcquay@17.226.13.44) |
| 14:16.22 | *** join/#utah josephscott (~josephsco@192.0.65.252) |
| 14:30.47 | *** join/#utah tensai (~tensai@159.118.222.223) |
| 15:02.21 | *** join/#utah manithree (~blr@li4-76.members.linode.com) |
| 15:04.19 | *** join/#utah macnewbold (~macnewbol@inferno.apxalarm.com) |
| 15:10.54 | *** join/#utah bgerber (~gerberb@suse103.zenez.com) |
| 15:30.41 | *** join/#utah undertakingyou (~undertaki@neutroninteractive-xmission.xmission.com) |
| 15:39.18 | *** join/#utah undertakingyou (~undertaki@neutroninteractive-xmission.xmission.com) |
| 15:41.47 | *** join/#utah TimRiker (~TimRiker@bzflag/projectlead/TimRiker) |
| 15:41.48 | *** mode/#utah [+o TimRiker] by ChanServ |
| 15:42.07 | *** join/#utah kendsnyder (~Adium@67.214.245.39) |
| 15:42.30 | *** join/#utah jalbretsen (~jalbretse@sonic.itransact.com) |
| 15:42.44 | *** join/#utah pdayton (~pjdayton@67.21.63.146) |
| 16:04.38 | *** join/#utah resplin_ (~richard@esplins.org) |
| 16:06.17 | *** join/#utah romanovic_ (~romanovic@puritynine.com) |
| 16:06.54 | *** join/#utah macnewbold (~macnewbol@inferno.apxalarm.com) |
| 16:07.20 | *** join/#utah peno_ (~irc4me@c-98-202-160-116.hsd1.ut.comcast.net) |
| 16:07.46 | *** join/#utah jlp78 (jlp@2600:3c01::f03c:91ff:fedf:3238) |
| 16:08.48 | *** join/#utah siel (~siel@2605:6400:10:c07c:4231:4321:2341:2031) |
| 16:08.49 | *** join/#utah basepi_ (sid18756@gateway/web/irccloud.com/x-dovqysyvvlxkcyxa) |
| 16:08.49 | *** join/#utah dremex_work (~kerry@71.37.95.86) |
| 16:08.56 | *** join/#utah siel (~siel@unaffiliated/motley) |
| 16:10.36 | *** join/#utah fadein (~Erik@c-76-23-32-157.hsd1.ut.comcast.net) |
| 16:15.29 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-nulmkebdaokjhuww) |
| 16:23.46 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-kjderzlkansxutzl) |
| 16:32.34 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-fixdsgigiungrevt) |
| 16:41.04 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-gyjthffyrglaqpgz) |
| 16:49.35 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-kqzwqjgxeupdelsa) |
| 16:58.07 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-rhmjmfluzorzrvam) |
| 17:04.16 | *** join/#utah macnewbold (~macnewbol@inferno.apxalarm.com) |
| 17:06.32 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-jzdwumkbztgkbmwh) |
| 17:12.17 | *** join/#utah RyanE (rberick@shell.xmission.com) |
| 17:17.26 | *** join/#utah yukaia_work (~yukaia@198.183.219.9) |
| 17:20.18 | *** join/#utah Oberon4278 (~dcrooksto@38.112.155.163) |
| 17:43.09 | *** join/#utah robotarmy347 (~robotarmy@68.69.166.209) |
| 17:47.04 | Oberon4278 | So, I've started analyzing core files at work, and... I don't really know what I'm doing. |
| 17:47.22 | fadein | Oberon4278: analyzing them with GDB? |
| 17:48.11 | Oberon4278 | I have no idea. I think GDB is one of the tools used, but we have an analysis tool that does "a bunch of stuff" and then gives us a report. |
| 17:48.27 | fadein | that sounds nice |
| 17:48.40 | fadein | is your problem that you don't know exactly what you're looking for? |
| 17:49.26 | Oberon4278 | It shows the registers, a stack trace, a disassembly of the stack, and occasionally (if you give it enough raw materials) the C code that correlates to the disassembly. |
| 17:49.31 | Oberon4278 | It's that I don't know what I'm looking *at*. |
| 17:50.38 | Oberon4278 | My reach has exceeded my grasp. |
| 17:50.47 | fadein | it'll come |
| 17:51.02 | fadein | that's an area that I wish they spent some time on in school |
| 17:51.34 | fadein | personally, because I don't know much assembly or about how my CPU actually works, I ignore the registers and disassembly |
| 17:52.08 | fadein | I like to read the source code of the functions mentioned in the stack trace and try to work backwards up the chain to find where it began to go wrong |
| 17:52.10 | Oberon4278 | I know assembly in theory, but in practice... |
| 17:52.46 | Oberon4278 | for example, 00000000`002bc7d8addbyte ptr [rax],al |
| 17:52.55 | Oberon4278 | ...formatting got messed up |
| 17:53.29 | Oberon4278 | it's adding what's pointed to by rax to the low byte of register a |
| 17:53.43 | Oberon4278 | and rax is an invalid pointer, so it crashed |
| 17:53.53 | Oberon4278 | rax=0000000000000001 |
| 17:54.05 | Oberon4278 | but there's a lot of context that I don't understand |
| 17:54.07 | Oberon4278 | anyway, point being |
| 17:54.17 | Oberon4278 | what things am I going to need to learn to be able to do this? |
| 17:54.49 | fadein | if this were my problem, I'd start by understanding which variable rax corresponds to in the source code, |
| 17:55.03 | fadein | and walking up the stack in hopes that I can see where it takes on the value of 1 |
| 17:55.12 | fadein | hopefully that has happened in the current call chain |
| 17:55.31 | Oberon4278 | ahh, okay |
| 17:55.46 | fadein | if it happened previously and isn't in your stack trace, then you may be back to reading through more source code to figure out where that may have happened |
| 17:56.03 | Oberon4278 | my guess is it was here: pnt += GET_CBO_EXTC(extpnt)<<1; |
| 17:56.20 | Oberon4278 | I guess I need to learn C and assembly both. |
| 17:56.27 | Oberon4278 | Well, I've been meaning to do that anyway. |
| 17:57.37 | fadein | if you already have an understanding of assembly, I would think that you'll pick up C pretty well |
| 17:57.50 | fadein | it's the folks who come down from a higher-level language who really struggle with C |
| 17:59.28 | Oberon4278 | I actually know C about as well as I know assembly |
| 17:59.34 | Oberon4278 | which is to say I have passing familiarity with it |
| 17:59.56 | Oberon4278 | actually that's not true, I've taken a couple classes where we worked exclusively in C, and I've written one or two very very small programs outside of class in it |
| 18:04.43 | jxqz | ~napoleon |
| 18:04.43 | infobot | Wait, I forgot the crystals. |
| 18:50.10 | *** join/#utah macnewbold (~macnewbol@inferno.apxalarm.com) |
| 19:00.25 | *** join/#utah smcquay (~smcquay@17.226.13.44) |
| 19:40.17 | unum | randall munroe wiss be on colbert tonight |
| 19:47.18 | levi | Oberon4278: That line seems an unlikely culprit unless pnt was on the verge of overflowing. |
| 19:50.00 | *** join/#utah siel (~siel@unaffiliated/motley) |
| 19:52.34 | levi | Oberon4278: Anyway, helpful stuff to learn for debugging at that level are the grubby details of the platform function call conventions, the loadable library file format (probably ELF unless you're on OS X or something), and of course the basic CPU architecture details. |
| 19:54.15 | levi | A bit of an idea about how compilers and operating systems generally work wouldn't hurt either, though those hopefully won't often have too much impact on day-to-day debugging. |
| 19:56.10 | levi | Trying to piece together context information from post-mortem core dumps is often pretty difficult, especially when the compiler has optimizations enabled. |
| 19:56.42 | *** join/#utah SunSparc (~jonathan@162.219.43.50) |
| 20:00.10 | levi | And I guess Oberon4278 is back to debugging and not seeing what I'm saying. |
| 20:14.02 | Oberon4278 | oh hi levi |
| 20:15.56 | levi | Hi! |
| 20:16.36 | Oberon4278 | I was hoping you'd chime in :) |
| 20:16.47 | Oberon4278 | So our product runs on... just about every OS you can name. |
| 20:17.00 | Oberon4278 | I know you're about to say DEC Alpha and yes, back in the day it ran on that as well. |
| 20:17.14 | fugue88 | Oberon4278: "Every OS you can name..." Don't test levi. ;) |
| 20:17.28 | Oberon4278 | I did throw in the "just about" caveat ;) |
| 20:18.33 | Oberon4278 | anyway... when you say "platform call conventions" do you mean the OS or the processor? |
| 20:18.41 | Oberon4278 | oh, you added that later |
| 20:18.45 | Oberon4278 | awesome |
| 20:18.51 | Oberon4278 | I don't even know what those things *mean* |
| 20:19.39 | Oberon4278 | but we've got all the source code, and we keep debugging information internally for every build we ship, so at least there's that |
| 20:20.00 | Oberon4278 | anyway, is there a book or a Google keyword I can start with? |
| 20:42.33 | jlp_vnc | Oberon4278: does it run on FortuneOS? |
| 20:43.40 | jlp_vnc | http://www.old-computers.com/museum/computer.asp?c=767 |
| 20:45.04 | jlp_vnc | the OS was based on Research Unix System 7, as I recall |
| 20:49.40 | *** join/#utah jima (jima@laughton.us) |
| 20:54.18 | levi | Oberon4278: Calling conventions usually depend on both hardware and OS. |
| 20:55.14 | levi | It's basically a codification of the decisions about how to pass arguments to functions and how to get return values upon return. |
| 20:55.58 | levi | There's some impact/influence on language support as well. |
| 20:56.43 | levi | For example, C vararg functions put some limitations on what calling conventions you can use with them. |
| 20:57.23 | jlp_vnc | http://thedailywtf.com/Articles/The-Fizz-Buzz-from-Outer-Space.aspx |
| 20:58.17 | levi | Of course, your programs can use whatever sort of interprocedure calling conventions you want for private internal functions; the platform calling conventions only apply for programs that utilize platform-supplied functions or platform-native linking/loading mechanisms. |
| 20:58.59 | levi | But the more a compiler sticks to a standard calling convention for internal functions, the easier it is to debug the resulting code. |
| 21:03.18 | levi | Calling conventions boil down to stuff like which parameters go in registers, which go on the stack, how the call stack is handled, and who's responsible for saving/restoring context across call boundaries. |
| 21:04.55 | *** join/#utah tristanbob__l (~tristanbo@137.190.208.10) |
| 21:06.05 | levi | The wikipedia article for x86 conventions lists a bunch: http://en.wikipedia.org/wiki/X86_calling_conventions |
| 21:06.38 | *** join/#utah emcnabb (~emcnabb@cpe-107-015-189-013.nc.res.rr.com) |
| 21:06.39 | *** mode/#utah [+v emcnabb] by ChanServ |
| 21:09.25 | levi | So, calling conventions are just one aspect of the ABI, application binary interface. Other important bits are things like concrete data type representations and symbol naming conventions. |
| 21:10.15 | Oberon4278 | ok |
| 21:10.25 | Oberon4278 | so basically it's just a bunch of rules for how to read the call stack |
| 21:10.53 | *** join/#utah mkf (telex@gateway/shell/anapnea.net/x-rxcfttdmrzlvinvz) |
| 21:12.35 | levi | Most platform ABIs use some C-compatible data representation (C provides a lot of leeway in defining representations, which an ABI has to provide concrete decisions about) and a straightforward symbol translation, but languages with naming conventions less straighforward than C's (like C++) have to provide some translation to platform ABI names. |
| 21:14.14 | levi | From the perspective of someone trying to debug a program, you can look at the calling conventions that way. :) They're more usually used by compiler and associated tooling makers to emit the right object code for the platform. |
| 21:16.54 | levi | Oh, if you're going to be doing a lot of C, and trying to figure out if various bits of C *really* say what you think they say, I recommend getting this book: http://www.amazon.com/Reference-Manual-5th-Edition/dp/013089592X |
| 21:17.24 | levi | Especially if you are working with post-C89 C. |
| 21:30.48 | Oberon4278 | okay |
| 21:30.54 | Oberon4278 | thank you! |
| 21:35.02 | *** join/#utah smcquay (~smcquay@17.114.154.5) |
| 21:40.09 | *** join/#utah tristanbob__l (~tristanbo@137.190.208.10) |
| 22:01.52 | *** join/#utah sontek (~sontek@sontek.xen.prgmr.com) |
| 22:10.00 | *** join/#utah tristanbob__l (~tristanbo@137.190.208.10) |
| 22:12.14 | *** part/#utah Nemus (~Nemus@unaffiliated/nemus) |
| 22:19.09 | *** part/#utah Yorokobi (~Yorokobi@pdpc/supporter/professional/yorokobi) |
| 22:28.26 | *** join/#utah kendsnyder (~Adium@67.214.245.39) |
| 23:08.29 | *** join/#utah smcquay (~smcquay@17.226.13.44) |
| 23:09.27 | *** join/#utah macnewbold (~macnewbol@inferno.apxalarm.com) |
| 23:26.02 | jlp_vnc | so did anyone get bitten by the new ".prod" top level domain that went active today? |
| 23:33.13 | *** join/#utah pdayton (~pjdayton@67.21.63.146) |