| -!- Netsplit *.net <-> *.split quits: jonmasters, pgavin`, Daniel____, olofk | 13:32 | |
| -!- Netsplit over, joins: pgavin`, Daniel____, jonmasters, olofk | 13:32 | |
| pgavin` | hello, the FPMADDLO/FPMADDHI registers accessed with l.mfspr/mtspr, correct? | 14:51 |
|---|---|---|
| pgavin` | the manual doesn't give the indices for these registers though | 14:51 |
| pgavin` | nevermind, I guess I figured it out. It's clarified in the 1200 supplementary manual | 15:04 |
| juliusb | pgavin`: welcome to the world of the or1k spec | 16:17 |
| juliusb | lots to fix :( | 16:17 |
| pgavin` | lol | 16:17 |
| pgavin` | I'm working on getting FP into your cgen port | 16:17 |
| pgavin` | and there are some problems with the gcc 4.7 port I have to fix | 16:18 |
| pgavin` | some code isn't being put in libgcc for some reason | 16:18 |
| juliusb | is FP of big concern? | 16:18 |
| juliusb | i would worry more about getting the toolchain working | 16:19 |
| juliusb | ie. the gcc port | 16:19 |
| juliusb | the FP stuff isn't too important | 16:19 |
| juliusb | easy to do It hink | 16:19 |
| juliusb | think | 16:19 |
| pgavin` | I'm trying to get or1ksim's testsuite to build | 16:19 |
| pgavin` | it has some fp in it | 16:19 |
| pgavin` | but the binutils part of FP is pretty much done | 16:20 |
| pgavin` | at least, single precision | 16:20 |
| pgavin` | but double won't be hard | 16:20 |
| pgavin` | but, for some reason in giuseppe's gcc 4.7 tree, __udivsi3, __mulsi3, etc. are in the gcc/ subdirectory | 16:22 |
| pgavin` | and not libgcc like they should be | 16:22 |
| pgavin` | but those are for integer math | 16:23 |
| juliusb | ok cool | 16:28 |
| juliusb | sounds good, you appear to know what you're doing :) | 16:28 |
| pgavin` | well, sorta | 16:37 |
| pgavin` | I'm figuring it out :) | 16:37 |
| -!- Netsplit *.net <-> *.split quits: jonmasters, Daniel____, olofk | 19:55 | |
| -!- Netsplit over, joins: olofk, Daniel____, jonmasters | 19:56 | |
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!