| Fallenou | I think maybe someone should answer on the mailing list | 12:47 |
|---|---|---|
| Fallenou | maybe wolfspraul , about the guy wanting to contact a sale person | 12:47 |
| GitHub147 | [dmp] fallen pushed 1 new commit to master: http://git.io/vPulKg | 13:06 |
| GitHub147 | [dmp/master] Update README.md - Yann Sionneau | 13:06 |
| larsc | Fallenou: is it on purpose that there is no code in the repo yet? | 13:10 |
| Fallenou | yes I didn't have the time to put interesting stuff yet | 13:10 |
| Fallenou | but it will come very soon :) | 13:10 |
| larsc | ok | 13:10 |
| Fallenou | it will be pretty simple though, a simple python script to parse simulation output | 13:10 |
| Fallenou | and use pydot to generate a diagram | 13:11 |
| lekernel | aha, excellent project | 13:11 |
| Fallenou | that will help a lot in understanding all the mess in lm32 pipeline :) | 13:11 |
| Fallenou | oh, I just received an email from mwalle saying he is currently working on the lm32 pipeline diagram ^^" | 14:01 |
| Fallenou | he seems to be pretty advanced in his diagram | 14:01 |
| wpwrak | great ! i like the idea of a static diagram better than a diagram generator that examines traces | 14:31 |
| wpwrak | the letter looks more like a debugging tool. the former should be more like an architectural document. | 14:32 |
| wpwrak | ideally, you could use this when writing a whitepaper showing "before" and "after". | 14:32 |
| Fallenou | the more tool we have, the better :) | 14:42 |
| wpwrak | oh, sure. it can be very handy later. | 14:43 |
| wpwrak | it could also be useful to verify claims about the pipeline signaling, in case the path from code and drawing gets too unreliable (human error and such) | 14:45 |
| Fallenou | well yes the good point is that the generated diagrams should be 100% correct | 14:48 |
| Fallenou | they should not be wrong | 14:49 |
| Fallenou | it shows what actually happens | 14:49 |
| Fallenou | let's wait and see :) | 14:49 |
| mwalle | Fallenou: read my email regarding scriptsim? | 19:41 |
| mwalle | lets try tgif :) | 19:51 |
| wpwrak | do you like it ? | 20:19 |
| mwalle | nice and easy :) | 20:37 |
| mwalle | looks ugly, but its functional | 20:38 |
| Action: mwalle wonders what HyperSpace is ;) | 20:39 | |
| mwalle | In the hyperspace mode, certain objects are considered hot-links | 20:40 |
| mwalle | nice ;) | 20:41 |
| wpwrak | it also has a nice little enterprise icon :) | 20:41 |
| mwalle | wpwrak: happen to know how i resize the drawing area? | 20:42 |
| mwalle | mh drawing area seems to be exactly one page | 20:44 |
| wpwrak | hmm no. tgif seems to have matured quite a bit since i last used it (some 15+ years ago) | 20:45 |
| mwalle | i think it still looks the same ;) | 20:45 |
| Action: wpwrak likes long-lived software. vi, TeX, X11, ... | 20:46 | |
| mwalle | ah and its not multihead compatible | 20:47 |
| mwalle | hm i dont know how tgif does it, but the commands 'just work' :) | 21:07 |
| mwalle | eg align vertically.. somehow always the right reference object is picked | 21:07 |
| wpwrak | seems that you're good at predicting how tgif works, so your expectations are always met :) | 21:45 |
| mwalle | http://walle.cc/lm32-pipe.pdf | 21:53 |
| mwalle | rough sketch | 21:54 |
| mwalle | of the dataflow | 21:54 |
| mwalle | (still many modules missing) | 21:54 |
| Fallenou | mwalle: yeah but I didn't have time to have a look at your link unfortunately at the office | 22:38 |
| Fallenou | gn8! | 22:49 |
| wpwrak | hmm. without explanatory text, this confuses more than it explains. | 23:47 |
| wpwrak | looks roughly similar to lattice's diagram. but it isn't. i think the missing link is in that text surrounding the drawing ;) | 23:48 |
| --- Wed Oct 24 2012 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!