| --- Log opened Fri Dec 22 00:00:10 2017 | ||
| shorne | nooga: I dont see any reason the jvm wont run. | 08:56 |
|---|---|---|
| shorne | wallento: I am looking throught gcc in depth, you know I cant really find matjaz did much work | 10:15 |
| shorne | from 2000 to 2002 - damjan and simons did the initial bootstrap | 10:16 |
| shorne | in 2005 - matjaz did some optimization (cant figure out details) because I dont have the code from 2002 | 10:16 |
| shorne | 2010 -> jeremybennet did fp support, stack fixups, etc | 10:17 |
| shorne | 2012 -> on stekern did PIC, GOT, sync core, etc | 10:17 |
| shorne | 2015 -> rth did tail fall, more fixups for PIC, refactor of sync code, floating point fixups | 10:18 |
| shorne | there is really nothing matjaz did that I can see | 10:18 |
| shorne | Nothing significant | 10:18 |
| shorne | I would like to get the source code from 2002 to compare to matjaz's 2005 code dump (first code in opencores svn) | 10:20 |
| nooga | :D | 12:45 |
| nooga | I just wrote an elf loader so my vm can load test binaries and ran this https://github.com/openrisc/or1k-tests/blob/master/native/or1k/or1k-mul-basic.S | 12:46 |
| nooga | CPU panicked while setting timer mode but somehow 0x8000000d landed in r3 | 12:46 |
| nooga | mild success | 12:46 |
| nooga | looks like my MMU, lwz and lwa are seriously broken... | 13:07 |
| wallento | shorne: my understanding is that matjaz did some massive refactoring when bringing it to a new major version | 13:14 |
| wallento | I think blueCmd_or stekern may have the full history of the work | 13:15 |
| wbx | and matjaz is the developer responsible for not getting gcc support upstream? | 13:29 |
| nooga | looks like it passes some of the tests :D | 15:11 |
| shorne | wbx: yes | 16:55 |
| shorne | blueCmd_: do you have the history? | 16:57 |
| shorne | If its true I am thinking to start a clean room rewrite | 16:58 |
| shorne | but I am wondering if I can reuse things from Jeremy, stekern, blueCmd_ (because the fp support, PIC support, atomics is completely new work to me) | 16:59 |
| shorne | those code parts are independant | 17:00 |
| nooga | hah | 17:25 |
| nooga | looks like the kernel started and immediatly got to writing high memory counting down and hit a bunch of my dummy devices | 17:28 |
| --- Log closed Sat Dec 23 00:00:11 2017 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!