| --- Log opened Sun Jan 08 00:00:04 2017 | ||
| shorne | nice... with musl 1.1.16 (Result: PASS) | 01:20 |
|---|---|---|
| shorne | stekern: I have been thinking about this next patchset | 05:00 |
| shorne | will the spinlocks code get used on UP kernel? | 05:00 |
| shorne | the futex, bitps/atomic and cmpxchg I can see its getting tested | 05:01 |
| shorne | but it looks like spinlocks are specifically for SMP | 05:01 |
| stekern | shorne: no, spinlocks are just achieved by disabling interrupts on UP | 05:50 |
| shorne | ok, ill exclude those patches from this series | 06:23 |
| olofk | shorne: Great to hear that you found a solution to the pthread stuff | 06:36 |
| wallento | shorne: 0x4 is only a boot barrier | 10:20 |
| wallento | normally that would be a pic, probably something we should get into the next arch rev | 10:21 |
| wallento | by the way, stekern, olofk and shorne maybe a goal for ORCONF, to finish up the next rev with the work shorne has done | 10:21 |
| stekern | shorne: I'm fairly confident about the spinlocks code though, I've been testing it fairly extensively | 12:12 |
| ZipCPU|Laptop | I don't get it ... can someone explain to me why all of the OpenRISC instructions being with an "l.", as in "l.j" for jump, or "l.jalr"? | 16:09 |
| shorne | ZipCPU|Laptop: l. are for the basic instructions | 16:40 |
| shorne | lf.ftoi is for floating point | 16:41 |
| shorne | l stands for something, I cant remember | 16:43 |
| shorne | stekern: ok, Ill include it but mention its for smp prep, and will not really get used on uniprocessor systems | 16:45 |
| shorne | wallento: I argee, I think having smp detail in the next arch revision would be good. I am not sure what else(other than tls), but keeping it inline with the kernel would be good | 16:47 |
| wallento | I remember address formation for 64bit seemed crucial | 16:59 |
| shorne | wallento: is it captured under Proposals? http://openrisc.io/architecture I see a lot of 64 bit requirements | 17:25 |
| shorne | I cant remember details on the address formation, also didn't think there were any 64bit implementations | 17:25 |
| shorne | stekern: one other question I was meaning to ask, for your vga setup, did you ever use vgacon driver? or always fbcon? | 19:16 |
| ZipCpu | shorne: Thanks. If you can ever remember what "l." stands for, I'm all ears. | 20:04 |
| -!- Netsplit *.net <-> *.split quits: heroux, shenki, olofk, fotis2, simoncook, rah | 20:12 | |
| -!- Netsplit over, joins: shenki | 20:12 | |
| -!- Netsplit over, joins: olofk | 20:12 | |
| -!- heroux_ is now known as heroux | 20:13 | |
| -!- Netsplit over, joins: fotis2 | 20:13 | |
| --- Log closed Sun Jan 08 20:23:30 2017 | ||
| --- Log opened Sun Jan 08 20:23:41 2017 | ||
| -!- Irssi: #openrisc: Total of 46 nicks [0 ops, 0 halfops, 0 voices, 46 normal] | 20:23 | |
| -!- Irssi: Join to #openrisc was synced in 25 secs | 20:23 | |
| -!- ZipCpu is now known as ZipCPU | 22:03 | |
| stekern | shorne: I don't think I've ever used vgacon | 23:23 |
| --- Log closed Mon Jan 09 00:00:05 2017 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!