IRC logs for #openrisc Monday, 2016-11-28

--- Log opened Mon Nov 28 00:00:03 2016
wallentoshorne: the pgp web of trust was updated:
shornewallento: cool, it puts me at 4 levels from Linus, I heard from peter z that it needs to be 3 for linus to accept patches.08:42
wallentoSo, can I just mail them?08:42
wallentowho is this Linus that he decides what gets into Linux!08:42
shorneI met one guy yesterday that will sign my key to put me at 3, but it ended up he forgot his pasword08:42
shornehe forgot his pgp passphrase, so he was going to figure that out then sign08:43
shorneIll meet him again this friday.  If it doesnt work out I can try to send, and if he doesnt except we can fall back to you :)08:44
wallentoyes, that is good08:45
wallentodoes it help if I sign/ack/whatever the patches?08:45
shorneMeaning ill me the guy signing my key again, and then we can check with this linus guy08:45
shornewell, I guess al you really would need to do is create a signed tag (of my branch) probably best to host on openrisc/linux08:46
shorneI could push there, then you could tag and send the mail to the kernel08:47
shornelets give it a week though08:47
shornehmm, missed a few characters in those sentences, let me know if thats not clean08:48
shorneOn the other hand I am giving a presentation about openRISC, I wanted to cover some differnecs between openRISC and RISC-V08:49
shorneone is risc-v spec doesnt document an MMU08:50
shornealso, a lot is left up to implementation like reset vector08:50
shorneBut on the other hand, exception handling in risc is pretty flexible as there is a dynamic trap vector, and multiple trap status refisters08:51
shorneanyone have suggestion on any othe pros/cons vs openrisc?08:52
olofk_shorne: If you want to run 32-bit Linux, I think OpenRISC is a better idea. Not sure that works yet on RISC-V08:58
shornethats right as of now, but they have a set of patches for their rocket chip09:01
olofk_Thought they focused on the 64-bit stuff09:07
shorneoh, I didnt realize that09:15
shorne, right both their default configs are for 64 bit09:19
wallentoyeah, it is not a theoretical limitation, but a matter of focus09:34
wallentothe RISC-V stuff is a bit cleaner and better thought through09:34
wallentobut I like the OpenRISC community better09:34
-!- Netsplit *.net <-> *.split quits: stekern, rokka, rah, mwfc, rhythmx11:01
-!- Netsplit over, joins: rah, stekern, rokka11:02
kc5tjaolofk_: yt?12:30
kc5tjaolofk_: Just wondering, you'd mentioned that you had bus size adapters for Wishbone.  Is there one you are aware of that would be license-compatible with an MPLv2 project (namely mine)?12:31
olofk_kc5tja: I do have, but I'm not quite sure they fully work. Happy to relicense them if you find them useful though. You can find the interconnect tools here
kc5tjaUnfortunately, the resizer doesn't implement the state-machine I need.19:01
kc5tjaThanks though.19:01
--- Log closed Tue Nov 29 00:00:04 2016

Generated by 2.15.2 by Marius Gedminas - find it at!