| jonibo_ | derRichard: binutils-svn just mirrors the opencores svn repo... there have no changes there for about 15 months | 07:28 |
|---|---|---|
| axxlq1 | Hi. Is there any way to patch the linux kernel with rtai for openrisc? | 09:38 |
| axxlq1 | I want to port linuxCNC to or1k. Am I in my mind?:-D | 09:40 |
| jonibo | axxiq1: ah linuxCNC... that brings back memories | 10:31 |
| jonibo | i doubt if you'll be able to easily put the rtai patches on the openrisc kernel | 10:31 |
| jonibo | better approach may be the linux-rt kernel | 10:32 |
| jonibo | but I've never had time to look into that in any depth | 10:32 |
| jonibo | i would start by just trying to get linuxCNC to work on the kernel we have... and once that's running, running it on a realtime kernel is essentially an optimization | 10:33 |
| axxlq1 | well, thank you and on this one. | 10:37 |
| jonibo | is there any way to "reset" the or1200 via software... in order to cause it to "reboot"? | 12:15 |
| derRichard | jonibo: doesn't this depend on the board where you have your openrisc running? | 12:39 |
| jonibo | as far as I can tell, yes... was just hoping for a different answer | 12:44 |
| derRichard | maybe there is some nasty trick to trigger a machine reset :D | 12:48 |
| stekern | yeah, I don't think there is another way than jumping to the reset address, and that is implementation specific too | 13:16 |
| jonibo | stekern: i did try that, as per your suggestion... did not work, though | 14:07 |
| stekern | hmm, strange, what if you force npc? | 14:12 |
| jonibo | l.jr should have the same effect... haven't tried forcing npc | 14:13 |
| stekern | (that's not very elegant I know, but just as a test) | 14:13 |
| jonibo | will give that a try | 14:13 |
| stekern | I tried forcing npc over the debugger yesterday, and that did work | 14:14 |
| stekern | (you did turn off mmus etc before jumping, right?) | 14:15 |
| stekern | like, only let SM be set in SR | 14:16 |
| jonibo | yes, I turned off mmu.... didn't have gdb around so I couldn't tell where it was getting stuck | 14:17 |
| jonibo | it didn't jump to u-boot, in any case | 14:18 |
| jonibo | possible that it's not really in a "reset" state and that's why it doesn't work... nand controller or something needs to be reset? | 14:18 |
| stekern | could be | 14:21 |
| jonibo | I guess that's kind of my question... is there some general reset line that gets asserted for all the cores so that everything gets nicely reset and the system cleanly rebooted? | 14:21 |
| jonibo | or is that also implementation specific? | 14:22 |
| stekern | there is, but what it is connected to is implementation specific | 14:22 |
| jonibo | can it be asserted from software? | 14:23 |
| stekern | currently, no. but it wouldn't be hard to implement | 14:23 |
| jonibo | ok... i'll try to poke at the jump-to-reset with gdb later and see what it's doing... | 14:24 |
| jonibo | thanks | 14:24 |
| stekern | good luck! | 14:24 |
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!