| --- Log opened Thu May 30 00:00:21 2013 | ||
| stekern | _franck_: well, I'm living in finland, so cold is the default here ;) | 02:22 |
|---|---|---|
| stekern | hmm, looks like there's still some bugs in cappuccino to be hunted down | 03:29 |
| stekern | I'm playing with building a self-hosting gcc, but linux crashes all over the place on the board | 03:29 |
| stekern | right now I'm testing in or1ksim to see if it's even possible | 03:30 |
| stekern | it'll probably take a couple of months to build in or1ksim =P | 03:53 |
| ams | indeed.... | 04:16 |
| juliusb | olofk: just read your first blog post, pure gold mate, well done :) | 09:25 |
| juliusb | I limit scope creep by accepting mediocrity | 09:26 |
| juliusb | (but by telling myself I'll make it better later....) | 09:26 |
| stekern | that's actually how you should do it, because in the end, mediocre might be good enough | 09:42 |
| stekern | and then you haven't wasted time on fancy stuff that isn't needed | 09:43 |
| juliusb | stekern: in mor1kx_icache, what's the difference between cpu_adr_i and cpu_adr_match_i - is cpu_adr_i maybe a slightly earlier available version than ic_addr? | 22:51 |
| juliusb | i notice that's the one used to read the tag, and the match one, if you have no MMU, is the registered version of pc_fetch (out in the fetch_cappuccino module) | 22:51 |
| juliusb | so, that's used to do the compare? | 22:52 |
| juliusb | makes sense as you want the logic from the comparison not to snake back far, so using a registered version for that compare is good | 22:52 |
| juliusb | I'm just wondering if I can plumb it into the pronto fetch unit | 22:53 |
| --- Log closed Fri May 31 00:00:22 2013 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!