--- Log opened Wed Jan 27 00:00:04 2016 | ||
olofk | _franck_: I'm changing neek.core to use filesets instead. It's the only core I can see that uses the VHDL section, and I want get rid of that | 12:38 |
---|---|---|
olofk | ooohh... I never looked at that file. It's really pretty with all the comments :) | 12:39 |
_franck__ | olofk: yes, it is. It could become a *.core template :) | 12:44 |
olofk | Absolutely. And it will become even prettier when I have introduced the shiny new fileset features :) | 12:47 |
olofk | ah. fuck it. Forgot that I don't have Cyclone III support in my Quartus :( | 12:48 |
olofk | The generated tcl file looks ok though. | 12:50 |
olofk | _franck_: Do you have time to do a test build with the new .core file before I push it? | 12:51 |
olofk | Or anyone else with a Quartus installation that can build Cyclone III devices | 12:51 |
olofk | oh... do the latest free Altera Modelsim version support mixed-language? | 13:00 |
_franck__ | olofk: I was about to try but I'm stuck at: "Downloading vga_lcd from OpenCores" :( | 13:03 |
olofk | doh. cyclone III is not supported after quartus 13.1 | 13:03 |
olofk | _franck_: hmm.. opencores is accessible. Wonder if it's a FuseSoC bug | 13:04 |
_franck__ | well I'm behind a proxy and I had some trouble with fusesoc | 13:05 |
olofk | You can extract this to your fusesoc cache https://www.dropbox.com/s/rd3zv0pj2qgwx1f/vga_lcd.tar.gz?dl=0 | 13:06 |
olofk | I just hope it's not many other cores :) | 13:06 |
_franck__ | olofk: that's not easy, my virtualbox is in bad shape, I'll try to test that later | 13:09 |
_franck__ | plus, I need to install quartus 13.0 at home if I want to use my neek again (I got a new PC) | 13:10 |
olofk | _franck_: I'll just send you the updated .core file and you can add it whenever you got time to test it. ok? | 13:12 |
olofk | ...or actually. If the generated Quartus .tcl is similar to the old one, it should be ok to push | 13:37 |
olofk | Turns out that the neek core has actually been broken for a while :) | 13:40 |
olofk | I really really really should add CI to orpsoc-cores and fusesoc :/ | 13:40 |
olofk | So I'll push now. With the updated .core file FuseSoC will at least not crash | 13:44 |
* ams` looks at juliusb and olofk. | 14:27 | |
_franck__ | olofk: i didn't know we could do this kind of things: https://github.com/openrisc/orpsoc-cores/blob/master/systems/neek/neek.core#L38 | 14:28 |
_franck__ | append a [is_include_file] (or any other property) to a single file | 14:29 |
olofk | _franck_: It's pretty new. You like it? | 14:32 |
_franck__ | well, I prefer diffrent sections for different purpose | 14:36 |
olofk | _franck_: You can have as many sections as you want actually. You could do a fileset that only includes the include files instead, and set is_include_file=true for the fileset | 14:40 |
* olofk is trying to hide from ams`s seductive eyes | 14:41 | |
_franck__ | olofk: I thought setting is_include_file=true for the fileset for the entire section was the only option. FuseSoC user's guide was clear about that :) | 14:49 |
olofk | haha | 15:00 |
olofk | Google should have GSoD instead to get kids to do all the documentation :) | 15:01 |
_franck__ | THAT's a good idea ! | 15:03 |
_franck_ | olofk: there is a problem with sdc files, fusesoc generates set_global_assignment -name SDC_FILE data/neek.sdc | 20:15 |
_franck_ | but there is no data directory in bld-quartus | 20:16 |
_franck_ | ../src/neek/ is missing in front of data/neek.sdc | 20:17 |
--- Log closed Thu Jan 28 00:00:06 2016 |
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!