| 00:10.58 | crdueck | <PROTECTED> |
| 00:11.03 | crdueck | ugh... |
| 00:55.40 | brlcad | petaflot: try this: http://vpaste.net/bCXSz |
| 03:37.28 | CIA-23 | BRL-CAD: 03Crdueck 07http://brlcad.org * r4125 10/wiki/User:Crdueck/log: |
| 03:50.03 | CIA-23 | BRL-CAD: 03phoenixyjll * r51446 10/brlcad/trunk/src/librt/primitives/dsp/dsp_brep.cpp: Add tolerance value to control the max error. |
| 04:03.43 | CIA-23 | BRL-CAD: 03Phoenix 07http://brlcad.org * r4126 10/wiki/User:Phoenix/GSoc2012/Reports: /* Week 8 */ |
| 04:06.18 | CIA-23 | BRL-CAD: 03Phoenix 07http://brlcad.org * r4127 10/wiki/User:Phoenix/GSoc2012/Reports: /* Week 8 */ |
| 04:07.02 | CIA-23 | BRL-CAD: 03Phoenix 07http://brlcad.org * r4128 10/wiki/User:Phoenix/GSoc2012/Reports: /* Week 8 */ |
| 04:12.37 | CIA-23 | BRL-CAD: 03Phoenix 07http://brlcad.org * r4129 10/wiki/User:Phoenix/GSoc2012/Reports: /* Test Results */ |
| 04:12.58 | andrei_ | brlcad : the script is at 600 now, not moving that fast. But I can try and finish all unit tests meanwhile |
| 04:14.31 | CIA-23 | BRL-CAD: 03Phoenix 07http://brlcad.org * r0 10/wiki/Special:Log/upload: uploaded "[[Image:Dsp reduced.png]]" |
| 04:15.14 | CIA-23 | BRL-CAD: 03Phoenix 07http://brlcad.org * r4131 10/wiki/User:Phoenix/GSoc2012/Reports: /* Test Results */ |
| 04:59.21 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 06:17.08 | *** join/#brlcad stas (~stas@188.24.33.157) | |
| 08:19.40 | *** join/#brlcad stas (~stas@82.208.133.12) | |
| 08:19.54 | CIA-23 | BRL-CAD: 03Mayanntelan 07http://brlcad.org * r0 10/wiki/Special:Log/upload: uploaded "[[Image:555470 494943477187884 490159780 n.jpg]]" |
| 10:42.32 | *** join/#brlcad Stattrav (u3131@gateway/web/irccloud.com/x-gptowzqdzktdqxst) | |
| 10:42.56 | *** join/#brlcad Stattrav (u3131@unaffiliated/stattrav) | |
| 10:42.56 | *** join/#brlcad Stattrav (u3131@gateway/web/irccloud.com/x-gptowzqdzktdqxst) | |
| 10:58.36 | *** join/#brlcad ksuzee (~ksu@46.149.81.166) | |
| 11:50.37 | CIA-23 | BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/delete: deleted "[[Image:555470 494943477187884 490159780 n.jpg]]": relevance? |
| 11:51.15 | CIA-23 | BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/block: blocked [[User:Mayanntelan]] with an expiry time of infinite (account creation disabled, e-mail blocked): Inserting nonsense/gibberish into pages |
| 11:56.56 | andrei_ | I finished a unit test for two functions. I merged both functions in one unit test because they are quite similar. However the compiler rants about itoa |
| 11:57.01 | andrei_ | I did include <stdlib.h> |
| 11:58.54 | andrei_ | this is the unit test I wrote : http://slexy.org/view/s21ZLIOwh9 |
| 12:09.02 | andrei_ | Found the issue. Need a solution tho |
| 12:10.11 | andrei_ | my gcc doesn't support itoa, I ll work it around. |
| 12:29.28 | andrei_ | using sprintf instead. |
| 14:32.38 | *** join/#brlcad cristina (~quassel@188.24.71.196) | |
| 16:07.44 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 17:03.50 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 17:08.09 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 17:40.25 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 17:56.45 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 17:57.06 | *** join/#brlcad anuragmurty (~anurag@14.139.128.11) | |
| 18:25.30 | *** join/#brlcad juan_man (~quassel@unaffiliated/juanman) | |
| 18:32.45 | brlcad | andrei_: you shouldn't use itoa |
| 18:36.02 | brlcad | andrei_: so those function name are terrible - it's not immediately obvious what they're testing |
| 18:36.28 | brlcad | and the testing has a variety of problems |
| 18:36.59 | brlcad | you shouldn't have blatent memory leaks even if it's a unit test |
| 18:37.52 | brlcad | you malloc and never free |
| 18:39.39 | brlcad | some minor style/formatting problems, lots of replication, doesn't test input value ranges for proper behavior |
| 19:06.06 | andrei_ | what do you mean by " doesn't test input value ranges for proper behavior? |
| 19:06.39 | andrei_ | as for rest of feedback, I ll start applying it now |
| 19:29.53 | andrei_ | the test got at 700 right now .. |
| 19:30.11 | brlcad | wow .. that's rather unexpected |
| 19:30.33 | brlcad | can you graph what you have so far as time vs packet size |
| 19:31.06 | brlcad | what I meant about testing input ranges is in reference to function parameters |
| 19:31.53 | brlcad | if I'm writing a unit test for a generic with this signature: int function(char *a, int b, int c) |
| 19:32.01 | brlcad | I should be testing what happens when a is null |
| 19:32.08 | brlcad | when a is non-null but empty "" |
| 19:32.13 | brlcad | when b is negative |
| 19:32.17 | brlcad | when c is negative |
| 19:32.33 | brlcad | when either is 0 .. really big values, etc |
| 19:32.56 | brlcad | you make sure it works when it should and behaves reasonably when the inputs are not valid |
| 19:37.02 | andrei_ | if I work on the unit tests |
| 19:37.08 | andrei_ | running the script isn't so counter productive |
| 19:38.09 | brlcad | yeah, and it's very useful regardless |
| 19:38.17 | andrei_ | however the difference between 678 and 679 is < 0.05 |
| 19:38.49 | brlcad | might want to follow one of the src/libbu/test_*.c examples on how to reduce duplication in your test |
| 19:39.00 | andrei_ | yeah, I just wanted to ask about duplication |
| 19:39.01 | brlcad | whether with a function or macro or both |
| 19:39.36 | brlcad | duplication isn't so much of a problem for unit tests, but does make it pain to change down the road |
| 19:40.37 | andrei_ | I m sorry that I ask you to repeat it, I have another spare machine to run the script. I should go for 3d Graphic with powers of two ? |
| 19:41.00 | brlcad | repeat what? |
| 19:41.08 | andrei_ | repeat what to run |
| 19:41.29 | andrei_ | I mean which test to run first, but I ll look into the logs. |
| 19:41.43 | brlcad | unit test or pkg test? |
| 19:41.55 | andrei_ | pkg test |
| 19:44.36 | andrei_ | I ll just look into logs. |
| 19:45.13 | brlcad | given how long it's taken: 1-2048 @ 2 byte increment, 2048-4194304 @ 4096 byte increments |
| 19:46.24 | brlcad | for file sizes 1 2 4 8 16 32 64 128 256 512 1024 2048 4k 8k 16k 32k 64k 128k 256k 512k 1M 2M 4M 8M |
| 19:48.17 | brlcad | basically just keep multiplying by 2: size_t file_size=1; for (int i = 0; i< 24; i++) { file_size *= 2; } |
| 19:49.16 | brlcad | those sizes are a little different from what I said yesterday, should be a little faster to run them as there are fewer |
| 19:49.25 | brlcad | 1024+1024 samples per file size |
| 19:50.05 | brlcad | 49152 invocations of a libpkg client/server connection |
| 19:51.08 | andrei_ | I don't know if I can make the process any faster |
| 19:51.20 | andrei_ | on the machine that I plan to run this script there is absolutely nothing running except OS |
| 19:51.42 | brlcad | it's okay if it takes a few days |
| 19:52.03 | *** part/#brlcad anuragmurty (~anurag@14.139.128.11) | |
| 19:52.23 | brlcad | that's one of the points for getting the 1-2048 graphed, to make sure it looks good before doing the longer one for more file sizes |
| 19:54.39 | andrei_ | I ll get to work then, thank you ! |
| 21:44.27 | *** join/#brlcad stas (~stas@188.24.33.157) | |
| 21:58.12 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 22:11.50 | *** join/#brlcad Al_DC_Best (~Al_Da_Bes@5e0e150d.bb.sky.com) | |
| 22:44.03 | *** join/#brlcad Al_Da_Best (~Al_Da_Bes@5e0e150d.bb.sky.com) | |