Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2015-07-26

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:09 colomon joined #moarvm
05:08 TEttinger joined #moarvm
08:47 rurban joined #moarvm
10:23 brrt joined #moarvm
10:24 brrt \o
10:24 nwc10 o/
10:25 brrt ehm
10:25 brrt hmm
10:26 brrt table generation looks like it'll take O(m*n^2) iterations
10:26 brrt n will at least be 100, 200
10:26 brrt so that... sucks, i think
10:27 brrt but it's probably worth it to do it at preprocessing time rather than at runtime
10:27 brrt maybe any of you know of a cheap way to represent a sparse 3-dimensional array
10:28 brrt or a cheap way to test filter impossible combinations
10:28 brrt (e.g. (add reg (const)) can be combined with (add reg reg) but not with (add reg (load mem))
10:29 jnthn brrt: Tree?
10:29 jnthn brrt: Bloom filter?
10:29 brrt i'm unfamiliar with a bloom filter
10:29 jnthn (for the "represent" and "filter" respectively)
10:30 jnthn Oh, they're cool!
10:30 brrt i thought about a tree, yes
10:30 jnthn https://en.wikipedia.org/wiki/Bloom_filter
10:30 brrt but i figure the key has to be at least three words long, and repeated
10:30 brrt or the value should be the marker for another tree
10:30 brrt let's read about it :-) (the bloom filter)
10:33 brrt ok, i agree, that's pretty cool
10:34 brrt not sure if necessarily  what i want, but cool :-)
10:35 brrt i think the problem to: i can combine rule a and rule b, if for all children c identified by i, a[i] can reduce to b[i] or b[i] can reduce to a[i]
10:36 brrt e.g. (const) can reduce to reg, as can (load mem), but (const) cannot be reduced to (load mem) or the other way arround
10:37 * brrt bbl
10:38 brrt thanks for that link, though :-)
10:52 lizmat joined #moarvm
11:00 bonsaikitten joined #moarvm
11:00 hoelzro_ joined #moarvm
11:11 timotimo joined #moarvm
11:12 retupmoca joined #moarvm
12:22 vendethiel joined #moarvm
12:46 nebuchadnezzar hello
12:46 lizmat nebuchadnezzar o/
12:47 nebuchadnezzar I'm preparing the 2015.07 Debian package which permit to enable mipsel arch using libffi (and possibly others too)
12:47 lizmat ++nebuchadnezzar
12:48 nebuchadnezzar Is there any reason not to use libffi for all arch or should I limit the libffi case to thoses dyncall does not support?
12:50 lizmat nebuchadnezzar: jnthn FROGGS moritz brrt might know
12:50 jnthn I'd do the latter for now.
12:50 jnthn As dyncall is what most folks are working against, and so that codepath is much better tested
12:59 nebuchadnezzar ok, I'll do that
12:59 nebuchadnezzar thanks
13:01 moritz lizmat: fwiw I don't know all that much about MoarVM; I just hang out here :-)
13:16 lizmat moritz: but you have some dyncall / libffi knowledge ?
13:22 moritz lizmat: negative
13:23 lizmat moritz++    # nonetheless  :-)
13:24 moritz :-)
13:26 * jnthn is glad moritz++ hangs out here :)
13:53 brrt joined #moarvm
13:53 FROGGS__ joined #moarvm
14:02 brrt left #moarvm
14:57 rurban joined #moarvm
15:13 ggoebel joined #moarvm
15:55 TEttinger joined #moarvm
17:08 zakharyas joined #moarvm
17:13 * timotimo is looking at the "suddenly decont" issue again a little bit
17:16 timotimo oh, look, there's a substr + eq in the compile_mastop method that can become an eqat instead
17:17 timotimo anyway. i put a debug print into the "if @deconts[$arg_num]" part of that method and it doesn't seem to trigger for the rpoblematic case i'm looking at
17:21 timotimo oh, aha, getting somewhere now
17:49 masak timotimo++
18:08 timotimo i'm not 100% sure how exactly code should look that references a local "into" a localref
18:08 brrt joined #moarvm
18:08 timotimo currently using a bind onto a localref will complain
18:09 timotimo but maybe that's exactly what we want?
18:38 brrt aw, yeah
18:38 * brrt has tiling working
18:39 brrt the only trick is efficient translation to a C header file
18:39 timotimo ooooh
18:39 timotimo tiling works?
18:40 timotimo oh, you probably mean just the preprocessor?
18:40 brrt well, the preprocessor, but the tiling algorithm, too :-)
18:40 brrt just implemented in perl rather than c
18:42 timotimo ah
18:42 timotimo neato :)
18:42 timotimo we're getting close to actual assembly code getting spit out?
18:44 timotimo it's food time locally
18:44 brrt hmmm... yeah, conceptually a lot closer, at least
18:44 timotimo buy maybe after that i'll look a bit more into the graph.c parsing thing
18:45 brrt don't worry about it too much
18:45 brrt :-)
19:00 brrt \o/
19:00 dalek MoarVM/even-moar-jit: 8be7d2a | brrt++ | / (3 files):
19:00 dalek MoarVM/even-moar-jit: Generate tiler tables in perl
19:00 dalek MoarVM/even-moar-jit:
19:00 dalek MoarVM/even-moar-jit: Add a testcase to check the suitability for tiling actual trees.
19:00 dalek MoarVM/even-moar-jit: The next step is the to represent this in some static data
19:00 dalek MoarVM/even-moar-jit: structure suitable for search in the VM.
19:00 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/8be7d2a5dc
19:00 brrt this was *such* a brain killer
19:02 brrt i'm wondering how it'll perform for a bigger grammar
19:09 brrt not perfectly, but i suspect that's due to grammar inconsistency
19:18 dalek MoarVM/even-moar-jit: 8761634 | brrt++ | tools/tiler-table-generator.pl:
19:18 dalek MoarVM/even-moar-jit: Ambiguous test case
19:18 dalek MoarVM/even-moar-jit:
19:18 dalek MoarVM/even-moar-jit: How do we deal with ambiguity? Well enough, it seems.
19:18 dalek MoarVM/even-moar-jit: review: https://github.com/MoarVM/MoarVM/commit/8761634ce9
19:20 brrt ooh... it is dramatically slow though
19:22 brrt hmmm
19:22 brrt well...
19:22 brrt we'll see what we do about that
19:23 zakharyas joined #moarvm
19:24 brrt worst case we can commit the stored preprocessed tiles
19:24 brrt and distribute them that way
19:27 * brrt afk for now
19:27 brrt or, wait afk in a few minutes
19:32 brrt left #moarvm
20:27 rurban joined #moarvm
21:34 jnthn brrt++ # JIT progress
21:34 jnthn timotimo: iirc binding to localref is needed; I think we find to lexicalref to handle "is rw" native parameters
21:38 jnthn uh, s/find/bind/
21:38 jnthn Apparently I made good use of The Bishop's Arms near my hotel :)
23:12 timotimo ok 4 - a localref'd var can have a local ref'd thing bound to it and accessed (str)
23:12 timotimo so the code had been right all along, but the test was b0rked?
23:13 timotimo i feel dumb :)
23:15 timotimo especially dumb for bothering jnthn about it so often
23:21 harrow joined #moarvm

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary