| --- Log opened Mon Jan 13 00:00:51 2014 | ||
| stekern | did jonibo press some buttons on the openrisc.net list, seems like there was a bunch of dormant messages now coming through | 09:56 |
|---|---|---|
| olofk | stekern: Yeah, I noticed that too | 09:57 |
| olofk | One of these days, I'm going to explore branches in git. It's getting a bit messy to keep ten unfinished fixes/additions in my local master | 10:07 |
| stekern | olofk: it's never too late to learn an old dog new tricks, or what do they say ;) | 10:34 |
| olofk | stekern: Did you do anything with that wb fifo idea? I've just started doing a fixed clock dead simple uart decode that could be nice to hook up to that | 10:36 |
| _franck_web_ | olofk: about the verilator problem with or1200-generic.core. i think we could move bench/verilog/uart16550_model.v to tb_private_src_files in the verilog section. Then add a verilator compatible (using DPI) uart model with the same name in the verilator section | 10:36 |
| olofk | _franck_web_: I'm actually working on some stuff in that area right now :) | 10:37 |
| _franck_web_ | i did some simple tests with verilator and DPi this morning and it is quite easy | 10:37 |
| olofk | I also realized that I have some local fixes that takes care of the verilator problem, but I don't want to push it until I have an uart decoder in verilog | 10:38 |
| olofk | _franck_web_: That sounds nice. My approach is _not_ simple :) | 10:38 |
| _franck_web_ | ok, I'll prepare something and you'll tell me | 10:38 |
| olofk | awesome | 10:39 |
| olofk | The verilator parts need a lot of love | 10:40 |
| stekern | olofk: yes... and no... it's here, but I never completed it: https://github.com/skristiansson/orpsoc-cores/tree/d9e4999c6a654a06e83d4e8a812c914c8d2eb4f6/cores/wb_fifo | 10:41 |
| olofk | For example, I want it to be able to compile and use or1k-elf-loader with a c++ wrapper | 10:41 |
| stekern | it's just a 1-entry fifo right now | 10:41 |
| stekern | should be trivial to fix... | 10:42 |
| olofk | Right now it uses a local copy of (the old) or1k-elf-loader in the top level core | 10:42 |
| olofk | stekern: Sounds good. I'll hook it up to my simple UART when that is ready | 10:43 |
| olofk | baby IRQ | 10:43 |
| olofk | Staying home sick is not as productive as it used to be :) | 10:44 |
| _franck_web_ | olofk: yes, I've seen this code duplication. | 10:44 |
| _franck_web_ | :) | 10:45 |
| _franck_web_ | you may already know but look how simple is DPI: http://pastie.org/8628986 | 10:45 |
| stekern | olofk: you need to extend the fifo on your baby, seems like the irqs are coming in at high rate ;) | 10:46 |
| olofk | Looks like lekernel has already implemented exactly what I was looking for in the mmuart project | 13:12 |
| olofk | Just sucks that it's a tar file at opencores | 13:12 |
| olofk | Hmm.. but perhaps I can use the URL provider to fetch it through the websvn interface | 13:15 |
| olofk | yep, that should work | 13:17 |
| olofk | Ah.. it's on github as well, but I'm not very keen on pulling in the complete milkymist repo for a single file | 13:19 |
| olofk | Ahh.. I could use the URL provider to get that file instead | 13:19 |
| olofk | Damn, I'm clever | 13:19 |
| _franck_web_ | this URL provider is awesome | 13:21 |
| olofk | :) | 13:21 |
| olofk | The big-ass milkymist repo is exactly the kind of situations that I want to avoid with (the soon to be renamed) orpsocv3 | 13:22 |
| olofk | It would be fun to split up that in a couple of repos and see how hard it would be to pull back together as a bunch of orpsoc cores | 13:23 |
| olofk | I suspect there's a lot of good code there | 13:23 |
| olofk | stekern, ysionneau: You're a bit familiar with the milkymist stuff, right? Is it complicated to build an image from the repo? | 13:53 |
| stekern | olofk: the legacy or ng? | 16:01 |
| stekern | neither was hard though | 16:02 |
| olofk | the legacy one. No migen support in orpsoc yet | 16:04 |
| stekern | if you have ise ikstalled it was just a matter of make something iirc | 16:08 |
| olofk | ah crap. No xilinx support yet :( | 16:08 |
| mathias` | Hello, I tried to add an interrupt to the orpsoc_top by modifying the line `assign or1k_irq[3] = key1;' and key1 is assigned a push button on a DE1 board. Now I would like to enter an interrupt handler in my C code to execute some code whenever the button is pressed. I tried using the or1k_interrupt_handler_add and or1k_interrupt_enable functions. Using the interrupt id 3. Am I doing the right thing? | 18:14 |
| stekern | mathias`: sound right | 18:28 |
| jeremybennett | I see Sven is in EE Times again: http://www.eetimes.com/author.asp?section_id=36&doc_id=1320628 | 18:28 |
| blueCmd_ | busybox_1.21.0-1.1_or1k.deb | 21:23 |
| blueCmd_ | pretty sweet | 21:23 |
| stekern | nice ;) | 21:26 |
| --- Log closed Tue Jan 14 00:00:53 2014 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!