| --- Log opened Fri May 23 00:00:00 2014 | ||
| -!- Netsplit *.net <-> *.split quits: JakeUSC, jeremypbennett | 01:58 | |
| -!- Netsplit over, joins: jeremypbennett | 02:01 | |
| stekern | blueCmd: max 4 | 03:46 |
|---|---|---|
| stekern | I'm having some ideas about using the sockit, it has a lot bigger fpga | 03:50 |
| stekern | to give a rough estimate, mor1kx is ~2500 ALMs and the FPGA have ~40000 of them | 03:52 |
| stekern | so perhaps you could squeze in 8 on that | 03:54 |
| stekern | ...the said thing about the sockit is of course that it doesn't have ethernet | 03:55 |
| stekern | it's directly connected to the ARM | 03:55 |
| stekern | so I'll need to create a bridge ethernet device | 03:56 |
| stekern | shouldn't be that hard though | 03:57 |
| blueCmd | I played like 5 minutes with or32-system in qemu and I'm really excited | 08:57 |
| blueCmd | I didn't even know that worked, and it's blazingly fast! | 08:57 |
| stekern | why did you think I was saying stuff like "yes, but that will only work for qemu-user" when you posted the l.lwa/l.swa patch? ;) | 09:26 |
| stekern | we should provide an optimized version of this btw: http://lxr.free-electrons.com/source/lib/div64.c#L26 | 09:28 |
| stekern | I've got a vcd capture of that, and it's ~400 cycles | 09:28 |
| blueCmd | stekern: your statement is true even if -system doesn't work ;) | 09:33 |
| stekern | true | 09:37 |
| olofk | qemu-system, is that with guest arch kernel space as well? | 10:34 |
| olofk | _franck_: Just push the neek port if you're happy with it | 10:35 |
| olofk | stekern: Jesus christ. What the hell happened in #opencores? Just read a weird back log | 10:42 |
| olofk | Hard to tell if it was a bot or someone with multiple personalities | 10:42 |
| stekern | yeah, I saw that stekern guy in there too, he seemed mighty crazy! | 10:48 |
| olofk | :) | 10:54 |
| inigom | I'm fancing a problem using fusesoc | 12:20 |
| inigom | I have written this command | 12:21 |
| inigom | fusesoc sim --force de0_nano --elf-load ../tests/icarus_test/hello --vcd | 12:21 |
| inigom | but icarus is not working | 12:21 |
| inigom | Compiling /home/inigom/Desktop/pinksoc/build-fusesoc/build/de0_nano/src/elf-loader/elf-loader.c... | 12:22 |
| inigom | Compiling /home/inigom/Desktop/pinksoc/build-fusesoc/build/de0_nano/src/elf-loader/vpi_wrapper.c... | 12:22 |
| inigom | Making elf-loader.vpi from elf-loader.o vpi_wrapper.o... | 12:22 |
| inigom | Compiling /home/inigom/Desktop/pinksoc/build-fusesoc/build/de0_nano/src/jtag_vpi/jtag_vpi.c... | 12:22 |
| inigom | Making jtag_vpi.vpi from jtag_vpi.o... | 12:22 |
| inigom | /home/inigom/Desktop/pinksoc/build-fusesoc/build/de0_nano/src/wb_intercon/wb_mux.v:56: syntax error | 12:22 |
| inigom | I give up. | 12:22 |
| inigom | /home/inigom/Desktop/pinksoc/build-fusesoc/build/de0_nano/src/mor1kx/bench/verilog/mor1kx_monitor.v:42: Include file test-defines.v not found | 12:22 |
| stekern | simulation of the de0_nano system isn't supported yet | 12:24 |
| inigom | ok thanks | 12:24 |
| inigom | which one can I use?? | 12:25 |
| stekern | you can use mor1k-generic or or1200-generic for simulations | 12:25 |
| mafm | blueCmd: did you see that https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=749007 | 12:26 |
| inigom | Also reporting the same error with or1200-generic and mor1kx-sim | 12:31 |
| inigom | Compiling /home/inigom/Desktop/pinksoc/build-fusesoc/build/mor1kx-generic/src/elf-loader/elf-loader.c... | 12:31 |
| inigom | Compiling /home/inigom/Desktop/pinksoc/build-fusesoc/build/mor1kx-generic/src/elf-loader/vpi_wrapper.c... | 12:31 |
| inigom | Making elf-loader.vpi from elf-loader.o vpi_wrapper.o... | 12:31 |
| inigom | Compiling /home/inigom/Desktop/pinksoc/build-fusesoc/build/mor1kx-generic/src/jtag_vpi/jtag_vpi.c... | 12:31 |
| inigom | Making jtag_vpi.vpi from jtag_vpi.o... | 12:31 |
| inigom | /home/inigom/Desktop/pinksoc/build-fusesoc/build/mor1kx-generic/src/wb_intercon/wb_mux.v:56: syntax error | 12:31 |
| inigom | I give up. | 12:31 |
| inigom | ERROR: Failed to build simulation model | 12:31 |
| inigom | ERROR: Failed to compile Icarus Simulation model | 12:31 |
| blueCmd | mafm: I did yes, thanks! | 12:33 |
| stekern | inigom: please use pastie.org/pastebin.org/whatever | 12:33 |
| stekern | the error itself is because you have an old version of icarus that doesn't support vectored parameters | 12:36 |
| stekern | I thought we fixed that though... | 12:36 |
| stekern | maybe it was a similar error somewhere else | 12:37 |
| mafm | blueCmd: do you have any idea if 3, 4 or other is the right value? | 13:21 |
| blueCmd | mafm: I see no reason why it couldn't be 3 | 13:23 |
| blueCmd | and it probably should (less is better from what I can tell) | 13:23 |
| mafm | blueCmd: I don't know why all of them have 4 except __arm__ or __mips__ which have 3 | 13:30 |
| mafm | perhaps other people here know for sure? I can not even explain very well what it means | 13:30 |
| blueCmd | it's the structure alignment, I don't see why they would require 4, but we can certainly use 3 | 13:51 |
| blueCmd | mafm: did you see that libsdl1.2 built by itself in wannabuild? :) | 13:52 |
| mafm | blueCmd: no, didn't see that... but it's great news: over 400 rdeps :-) | 14:31 |
| ambro | hi, i discovered, that some of the links in the wiki don't work anymore. (e.g. the ftp server refuses connections). any ideas who is responsible / I should contact? | 15:13 |
| blueCmd | ambro: which links? | 15:17 |
| ambro | e.g.: ftp://ocuser:ocuser@openrisc.opencores.org/virtualbox-image/OpenRISC_Ubuntu_2011-12-15.vdi.bz2 | 15:17 |
| ambro | <-- quickly changing location (back in 10 minutes) | 15:19 |
| blueCmd | LoneTech: is that something you can poke someone for? ^^^^ | 15:20 |
| inigom | Stekern: Finally I was able to simulate with or1200-generic | 15:27 |
| inigom | the point is that if i compile the program to run with the option -g | 15:28 |
| inigom | it doesn't work | 15:28 |
| inigom | but if you omit this | 15:28 |
| inigom | it works | 15:28 |
| inigom | BTW you're right, i had to install icarus from git | 15:29 |
| inigom | If you try to execute the bin compiled with the option -g, I have obtained the next error | 15:30 |
| inigom | or1200_monitor : OR1200 crash detected (suspected CPU PC corruption) | 15:30 |
| stekern | interesting | 15:30 |
| ambro | re | 15:36 |
| blueCmd | mafm: I'm making good progress in transitioning the wanna-build environment to full VMs | 15:41 |
| blueCmd | it's going to be slower but should do almost perfect emulation of the CPU | 15:41 |
| blueCmd | so tests and pthreads should work just fine | 15:41 |
| blueCmd | uh-ooh, this is sloow | 17:34 |
| ambro | i'm still trying to get the toolchain, which fails sadly: | 20:39 |
| ambro | git submodule update --init | 20:40 |
| ambro | Cloning into 'linux'... | 20:40 |
| ambro | remote: aborting due to possible repository corruption on the remote side. | 20:40 |
| ambro | fatal: early EOF | 20:40 |
| ambro | fatal: index-pack failed | 20:40 |
| ambro | Clone of 'git://openrisc.net/jonas/linux' into submodule path 'linux' failed | 20:40 |
| olofk | Interesting with the PC corruption. Would like to get that test case | 20:45 |
| olofk | Can't reproduce it here | 20:45 |
| olofk | With a hello world | 20:45 |
| olofk | on or1200-generic | 20:45 |
| stekern | ambro: use the toolchain from github | 21:16 |
| --- Log closed Sat May 24 00:00:02 2014 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!