Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2015-04-12

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

All times shown according to UTC.

Time Nick Message
00:39 vendethiel joined #moarvm
01:06 vendethiel joined #moarvm
01:31 vendethiel joined #moarvm
01:47 ilbot3 joined #moarvm
01:47 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
02:38 vendethiel joined #moarvm
03:02 vendethiel joined #moarvm
05:37 vendethiel joined #moarvm
05:48 arnsholt_ joined #moarvm
05:49 BinGOs_ joined #moarvm
06:00 BinGOs joined #moarvm
07:06 lue joined #moarvm
08:25 dalek MoarVM: 1b28af6 | nicholas++ | src/6model/serialization.c:
08:25 dalek MoarVM: Read the serialized reference type discriminator with read_discrim().
08:25 dalek MoarVM:
08:25 dalek MoarVM: This is the first stage of refactoring to permit storing the discriminator in
08:25 dalek MoarVM: one byte, not two.
08:25 dalek MoarVM:
08:25 dalek MoarVM: The reference type discriminator is the only thing written and written as a 16
08:25 dalek MoarVM: bit quantity. For now, keep read_int16() for the read of the discriminator that
08:25 dalek MoarVM: happens as part of *writing* the serialization blob.
08:25 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/1b28af697d
08:25 dalek MoarVM: 6433def | nicholas++ | src/6model/serialization.c:
08:25 dalek MoarVM: Refactor read_discrim() to take thread context and serialization reader as args.
08:25 dalek MoarVM: 4f55679 | nicholas++ | src/6model/serialization.c:
08:25 dalek MoarVM: Store the serialized reference type discriminator in 1 byte, not 2.
08:25 dalek MoarVM:
08:25 dalek MoarVM: This shrinks CORE.setting.moarvm by 5% (568K). It's now just over 11M.
08:25 dalek MoarVM: (We serialize a lot of references. 582038 to be exact, at the time of writing.)
08:25 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/4f556793c7
08:26 nwc10 in the end, I'm not sure how exact my exact was. But it was consistent with the size drop :-)
08:26 nwc10 jnthn++ # checking my stuff
08:27 vendethiel joined #moarvm
08:27 jnthn I *think* I have all of your Moar/NQP patches merged by now :)
08:27 nwc10 all the ones I've written and pushed
08:28 jnthn Also, this remote workflow works much better for me than email
08:28 nwc10 (it looks like it. I'll check soon, but the directory that it's easiest to use for that is busy doing more work)
08:28 nwc10 cool
08:28 nwc10 I just have to remember not to yank the rug from under you by force pushing a change while you're busy looking at the old one
08:30 jnthn *nod*
08:30 jnthn Leaving me with rebasing work to do is OK.
08:31 dalek MoarVM: 551e211 | jnthn++ | src/6model/serialization.c:
08:31 dalek MoarVM: Add missing IS_CONCRETE checks in serialization.
08:31 dalek MoarVM:
08:31 dalek MoarVM: We likely should not have gotten away with these before now; they can
08:31 dalek MoarVM: lead to illegal memory reads.
08:31 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/551e211384
08:33 dalek MoarVM: 42342d3 | jnthn++ | src/6model/reprconv.c:
08:33 dalek MoarVM: Make unbox int/num/str functions test IS_CONCRETE.
08:33 dalek MoarVM:
08:33 dalek MoarVM: We were failing to do this check in most places that called these
08:33 dalek MoarVM: functions, so it makes sense to move the check inside of them to fix
08:33 dalek MoarVM: the existing bugs and prevent a bunch of new and similar ones.
08:33 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/42342d3f60
08:36 dalek MoarVM: 1055bb7 | jnthn++ | src/core/args.c:
08:36 dalek MoarVM: Remove duplicate IS_CONCRETE check.
08:36 dalek MoarVM:
08:36 dalek MoarVM: Now that we moved this into the unbox functions themselves, we do not
08:36 dalek MoarVM: need to repeat it in args processing.
08:36 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/1055bb7dd5
08:37 nwc10 jnthn: yes, I think they're all there. Thanks
08:54 lizmat joined #moarvm
09:08 lizmat joined #moarvm
10:11 timotimo nwc10: i like your ~5% decrease
10:11 timotimo much fewer zeroes everywhere i bet
10:12 timotimo jnthn: +1/0/-1 on merging devirt_repr_ops into master?
10:13 jnthn 0; I'd like to review it first.
12:55 timotimo sure
12:55 timotimo on my machine it passed spec tests, but it wasn't a big speed increase, sadly
13:10 timotimo ... -Wpedantic does not help me figure out if this will build on msvc or not
13:11 timotimo a hundred million times warnings like src/gc/orchestrate.h:14:29: warning: comma at end of enumerator list [-Wpedantic]
13:11 timotimo for example
13:27 colomon joined #moarvm
13:50 zakharyas joined #moarvm
15:19 vendethiel timotimo: is moarvm c99 or c89?
15:19 jnthn C89
15:19 vendethiel ah, that explains the warning then
15:20 * vendethiel wrote the most basic possible VM in C (89) last week, and is very happy not having to maintain one...:)
15:21 vendethiel (even if I don't doubt moarvm's code to be literally a hundred times better than mine...)
15:35 dalek MoarVM: 23e9c98 | jnthn++ | tools/ucd2c.pl:
15:35 dalek MoarVM: Have Unicode DB generator emit NFC lookup table.
15:35 dalek MoarVM:
15:35 dalek MoarVM: These will be used in computing NFC/NFKC.
15:35 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/23e9c98afb
15:35 dalek MoarVM: b1e3ead | jnthn++ | src/strings/unicode_db.c:
15:35 dalek MoarVM: Update generated Unicode DB to have NFC tables.
15:35 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/b1e3ead1e5
15:38 dalek MoarVM: dadde5a | jnthn++ | src/strings/unicode (2 files):
15:38 dalek MoarVM: Implement codepoints -> primary composite lookup.
15:38 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/dadde5ae13
15:50 dalek MoarVM: febd736 | jnthn++ | src/strings/normalize.c:
15:50 dalek MoarVM: Implement most of Unicode canonical composition.
15:50 dalek MoarVM:
15:50 dalek MoarVM: Used by NFC and NFKC. Known to be missing the Hangul special cases,
15:50 dalek MoarVM: which is the likely reason for the remaining Unicode normalization
15:50 dalek MoarVM: test failures.
15:50 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/febd736d64
16:31 dalek MoarVM: 117cea2 | jnthn++ | src/strings/normalize.c:
16:31 dalek MoarVM: Factor out Hangul constants.
16:31 dalek MoarVM:
16:31 dalek MoarVM: So we can use them during canonical composition too.
16:31 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/117cea2470
16:31 dalek MoarVM: 3f29518 | jnthn++ | src/strings/normalize.c:
16:31 dalek MoarVM: Implement Hangul part of canonical composition.
16:31 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/3f295180fe
16:32 jnthn Neat, we pass all 74,324 test cases from NormalizationTests in the Unicode 7 spec.
16:47 jnthn dinner &
16:51 vendethiel jnthn++ # great work, as always :o)
16:52 camelia joined #moarvm
17:17 zakharyas joined #moarvm
18:53 brrt joined #moarvm
18:54 brrt \o
19:10 timotimo o/
19:10 timotimo brrt: did you see what sillyness i had done wrong in the devirt reprop branch?
19:10 brrt no
19:10 brrt i did not
19:10 brrt i'll go fetch it
19:11 timotimo it's embarassing, really :)
19:13 brrt i'm assuming you mean exchanging pop and push
19:13 brrt that's not embarassing, that just happens :-)
19:16 brrt that branch does still contain the patch i thought was unworthy :-P
19:17 brrt although it is correct
19:17 brrt and i'm wondering if we can't / shouldn't push that up somewhere in the chain
19:17 timotimo yes, i can still cut that out, no problem
19:19 dalek MoarVM/jit_devirtualize_reprops_3: 02ee703 | timotimo++ | src/jit/emit_x64.dasc:
19:19 dalek MoarVM/jit_devirtualize_reprops_3: Revert "special case initial TC and STable, obj, objbody cases in jit"
19:19 dalek MoarVM/jit_devirtualize_reprops_3:
19:19 dalek MoarVM/jit_devirtualize_reprops_3: This reverts commit a185b28d3d214afd4f648019a56dacdcabd1c50d.
19:19 dalek MoarVM/jit_devirtualize_reprops_3: review: https://github.com/MoarVM/MoarVM/commit/02ee703155
19:30 * brrt wants this: http://lvalsan.web.cern.ch/lvalsan/processor_benchmarking/presentation/#/available_arm_systems
19:31 brrt 40W for 8 cores @ 2,4Ghz is pretty awesome
19:31 brrt timotimo: do you notice any speedups yet?
19:31 brrt timotimo++ btw :-)
19:33 brrt it's AArch64 btw
19:35 timotimo the speedups were qiute sad
19:35 timotimo for a spec test run, that is
19:36 brrt hmm
20:43 cognominal joined #moarvm
21:05 * brrt afk
22:20 vendethiel joined #moarvm
23:05 vendethiel joined #moarvm

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