[00:13] I threw tsan at rakudo and I've been trying to figure out if I can do something about any of the data races it's reporting; there's surely some that aren't a problem because the rest of the system is built to take them into account [00:14] but I don't know enough about the nitty-gritty details of atomics and memory models to figure all of that out [00:32] *** kjp left [00:33] *** kjp joined [00:35] *** kjp left [00:35] *** kjp joined [10:50] *** sugarbeet left [11:00] *** sugarbeet joined [12:05] *** Nicholas left [13:06] *** Nicholas joined [13:51] hmmm ... looking with a fresh eye at ParaSeq code, found this: [13:51] if nqp::atomicload_i($!unhandled) { [13:51] nqp::atomicdec_i($!unhandled); [13:51] I assume that's a race condition, right ? [13:52] indeed, you would want to have a compare-and-set, or depending on what exactly the rest of the code does, use the result from an unconditional decrement? [13:53] sorry, compare-and-swap is the correct term [13:55] there's also test-and-set which is a similar kind of thing [13:55] right [14:02] *** apogee_ntv joined [14:04] not really in the right mindset for this atm.. will look at it tonight... [17:58] looks like in interesting underlying library: https://raku.land/zef:apogee/CRoaring [17:59] Yeah it's great, the API is really clean too. [18:00] does it build on Windows ? [18:00] It's used in most high perf tag search engines via various levels of indirection. [18:00] With great difficulty. [18:00] meh [18:01] It doesnt help that I don't know the Windows C ecosystem at all, so I don't even really know how to do a Build script for it. [18:02] My guess is if you have mingw it will probably be fine [18:06] https://raku.land/zef:apogee/Roaring::Tags I mostly wrapped it to build this [18:14] well, feels like this might be a useful addition to the MoarVM core [18:14] but **only** if it can be built on Windows [18:22] I will try to take a look at how to build on Windows later today or tomorrow. I should be able to spin up a VM. [18:24] ++apogee_ntv [18:25] What compiler should I be testing on? Best to use what's used in the moarvm release process? [18:38] lizmat: Looks like it builds fine, just have to make a defs file for the symbols you want to export: https://github.com/m-doughty/CRoaring-Raku/actions/runs/24098094861/job/70301908884 [18:39] and now I wonder how easy it could be made to support 1, 2, 4 bit native (shaped) arrays [18:41] i'm not sure this can do what you want it to [18:41] ah... ok [18:41] it only stores a large one bit array very efficiently [18:42] right... [18:42] you can't have the same number twice, or store an order of the numbers [18:42] hmmm...in that light: ah... ok, I guess I shoulda've read the docs :-) sorry for the noise here [18:43] i assume the algorithm doesn't generalize to multisets (which would allow one value to appear multiple times) or ordered sets (so you can have an order of items stored) [18:43] Otoh, I wonder if this could be of use in regexes ? [18:43] it could still be interesting for our unicode database [18:43] right [18:43] yeah, in regexes for character classes / sets [18:43] replace several deep hash lookups by a single bit union? [18:44] hm? [18:45] nvm... I would need to think about this more [18:45] which hashes were you thinking of? [18:49] the uniprop ones ? [18:50] ah, I'm not sure we have nested hashes there, isn't it mostly one huge flat table? [18:51] after you go from a name of the property to an index, i think that's where hashes end [18:52] ah... ok... that makes sense [18:53] ShimmerFairy is probably the most up-to-date in that area nowadays ? [18:56] not all properties are bits, of course, but in theory you can just assign multiple consecutive bits to a single property of a single codepoint, but then the operations like and/or/xor and such are perhaps not meaningful / useful any more [18:56] yeah... also makes sense :) [18:56] i'm not saying there's nothing to be gained though [19:00] with an ahead-of-time-known set of values you can also look into Perfect Hashing [19:01] not sure how well it applies to unicode properties though