02:47:39 Does anyone here know which components of illumos are closed source? 03:06:13 there's a few kernel modules that aren't really used (some specific bits for some old intel cpus), pax, and some other old drivers for stuff people aren't likely to be using anymore 03:06:28 (why no one's really bothered to replace them, since they're not really needed) 03:06:52 thanks for the info 08:03:43 On the list of closed bits, for the history: 08:03:46 https://illumos.topicbox.com/groups/developer/T9694d83e5bcf84ed-M076701df50a68fff011b8962 08:03:57 Although a few of those have since been cleaned up 10:17:04 [illumos-gate] 16223 bhyve returns bogus cpuid 8000_001D leaf -- Andy Fiddaman 15:20:38 (set-frame-parameter (selected-frame) 'alpha '(75 . 100)) 15:23:15 I'm sorry wrong window 15:26:41 LISP window? 15:27:06 emacs? 15:35:07 Yeah, that's elisp alright. 15:46:24 i wanted to like emacs, but back when i was actually interested in trying to learn it, the tutorials were terrible 15:46:46 they either assumed you had never touched a computer before in your life 15:47:08 or assumed you've spent the past 10+ years programming lisp 16:16:56 jbk: that's not only true for emacs tutorials, though... 16:17:26 oh yeah, they're far from the only ones 16:17:45 at least back when i was first learning git, a lot of them were the same way 16:18:19 like, either they'd start off trying to explain what 'bytes' and 'files' were 16:18:47 or they just started off on a treatise on graph theory 16:19:34 that's one up on most of the documentation I find. The standard assumption is that you've been steeping in $thing for the past 40 years and know everything you could possibly need to know about the design and theory. 16:19:47 (i suspect things are probably better now) 16:19:48 * nomad glowers at shibboleth 16:19:58 at least for git 16:21:19 (the key insight was with other SCMs i always conceptualized things as versions of files, with git i found it much easier to understand as it managing a collection and ordering of diffs) 16:21:33 which you could argue the two are equivalent 16:22:01 but a lot of the commands you commonly use basically devolve into shuffling or combining diffs 16:23:30 true 16:25:25 so once I thought of it that way, i found it easier to use it 16:47:31 jbk: "they either assumed you had never touched a computer before in your life" 16:49:25 I think that's because lisp is very much unlike most of the languages that were common back around the time GNU emacs emerged (particularly BASIC which was a first language for many). advice I got about learning lisp was "forget what you know about programming and start from scratch" 16:52:37 jbk: the issue with git is that a lot of the underlying merkle tree implementation details are visible and sometimes they get in your face in confusing ways. 16:52:50 yes, but where lesson 1 is 'when you press the 'a' key, that's actually invoking a function that inserts that letter into the buffer and displays it' (ok), 'now for lesson 2, we're going to refactor this .c file and translate the comments into ancient greek by doing M-x some-long-ass-command' (uhhh ok that seems nice, but how would one discover such functionality? (silent) is there any sort of reference on 16:52:56 what functions/commands are included? (silent)' 16:59:16 that's where things kinda fell off... that's nice that your tutorial wants to show off all the really cool things you can do, but that does little in helping me actually be proficient in using it 17:00:18 aside from memorizing those specific things and not providing any means for actually being able to go beyond that 17:01:25 things again might be better now (this a long time ago, and never really looked into it again) 17:03:54 an yeah for git, prior to it, i always (maybe wrongly?) viewed diffs and such (in the context of an SCM) as more of an implementation detail/optimization whereas with git it's really feels like the fundamental bit (which I guess make sense given how linux dev is done) 18:22:11 Ironically git deals in diffs only begrudgingly as an aid to humans, given it's all snapshots of tree state underneath 18:24:55 yep. and sometimes the fact that it lets you look at the tree state has been useful to me - if you look at tree hashes you can confirm that a rebase didn't change the final state of the tree even though the path it took to get there differed. 18:35:48 is there any wip of importing openzfs to illumos? 18:38:49 so far, patches are being imported selectively rather than en masse. 19:24:44 [illumos-gate] 16217 Hold SCL_VDEV when counting leaves -- Olaf Faaland 20:47:04 danmcd: if nobody gets back to you on mastodon, poke me 20:47:23 sjorge: You are one of the few people who understand why I'm asking. 20:47:33 i'm pretty sure the answer is no 20:47:47 FreeBSD's driver doesn't do the thing the Linux Driver does, nor my fenix illumos#13230 fix does. 20:47:57 BUG 13230: i40e has duplicate traffic when used with bhyve/snoop running (New) 20:47:57 ↳ https://www.illumos.org/issues/13230 20:47:57 as i am currently using them via ppt 20:48:37 because if there is duplicate traffic my firewall would be broken 20:48:55 but already closed the laptop for tonight 20:52:52 assuming via ppt to a freebsd vm is fine 20:56:33 I *do* expect the answer to be no. I wanna hear it from someone not illumos-adjacent, just in case. 21:13:56 i guess you could try poking Kristof Provost, he might have, or know someone that has the hw